Time to be Creative? Self-Regulation of Time in Creative Professions. Flora Beeftink

Transcription

1 Time to be Creative? Self-Regulation of Time in Creative Professions Flora Beeftink

2 A catalogue record is available from the Eindhoven University of Technology Library ISBN: NUR 771 Keywords: Creativity / Time / Self-regulation / Time management Printed by Universiteitsdrukkerij Technische Universiteit Eindhoven Cover design: Anne Beeftink Cover photo: Anne Beeftink 2008, Flora Beeftink, Eindhoven All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording or by any information storage and retrieval system, without permission in writing from the author.

3 Time to be Creative? Self-Regulation of Time in Creative Professions PROEFSCHRIFT ter verkrijging van de graad van doctor aan de Technische Universiteit Eindhoven, op gezag van de Rector Magnificus, prof.dr.ir. C.J. van Duijn, voor een commissie aangewezen door het College voor Promoties in het openbaar te verdedigen op woensdag 15 oktober 2008 om uur door Flora Beeftink geboren te Pannerden

4 Dit proefschrift is goedgekeurd door de promotoren: prof.dr. C.G. Rutte en prof.dr.ir. J.W.M. Bertrand Copromotor: dr. W. van Eerde

5 Acknowledgements And now this PhD project is completed. It took some effort, but I enjoyed working on it very much. Doing research about time management I was very aware of the limited time I had to complete this dissertation and felt slightly threatened by the milestones and deadlines of my project. This urged me to work hard at times, which prevented extreme time pressure near the final deadline. Following the rules of time management as one could say, and indeed I experienced the feeling of being in control of time and of my project, which decreased the amount of experienced work stress considerably. And with great peace of mind I spend almost all my Saturdays (in addition to my already stretched work week) working on my dissertation for the past few months... This sounds like I have been working on this project all by my self, but that was definitely not the case. Although this dissertation carries my name, the manuscripts on which the chapters of this dissertation are based are written in co-authorship with Wendelien van Eerde, Christel Rutte, and Will Bertrand. Not only were they my co-authors, they were my teachers and coaches that transformed me slowly but surely from a engineer with a passion for work psychology into a work psychologist with a background in engineering. They gave me enough guidance to be able to get me started and enough freedom to make my own choices by the time I was up to it. In addition, I thank Wendelien for her indefinite inspiration and for being there at all times. My appreciation also goes out to Christel for recognizing my potential, then, but also now. Furthermore, the studies in this dissertation highly depended on the willingness of people to participate. Therefore, I would like to thank everyone that took the time and effort v

6 to be part of my investigations. I am particularly grateful to the 25 architects that participated in my interview study for being the 'voice of practice' and explaining to me in detail what it is like to be a creative professional under deadline pressure. Although this qualitative study absorbed an incredible amount of my scarce time, I cannot imagine me having done my other studies without knowing in the back of my head what the most urgent issues were that these architects were confronted with on a day-to-day basis. Also, I would like to show my gratitude to the participants of my diary study in particular, for their continued efforts and commitment that were necessary for this study. Furthermore, I thank the organizations that allowed me to approach all these people: the BNA (Organization of Dutch Architects), the Stan Ackermans Institute (School for Technological Design), KreaNET (Network of Creativity Enhancers), Festina Lente Creative Productions (and their Seaday project), and ARCAM (Amsterdam Centre for Architecture). A big help were also my colleagues from the HPM group, first and foremost my roommate Eric van der Geer-Rutten-Rijswijk (see also the 'dearest friends' section below). Second, my colleagues from the 'U-gang' (Ad Kleingeld, Anniek van Bemmelen, Josette Gevers, Tanja Bipp, Daphne Dekker, Marieke van der Tooren, and Marieke Habraken) for providing me with ample time for incubation during the afternoon tea sessions, which was a boost both for my positive affect and daily creativity. Anniek also deserves my gratitude for checking my English and being a very dedicated and patient teacher. Since feeling well and happy is one of the best predictors of job performance, I owe my friends, family, and pets. I would like to thank Sander in particular, for allowing me to be myself at all times, and because he was so very much not interested in my work that it was very easy to put things in perspective when I tended to be stressful. Also a big thank you to Gitta, for getting me through the worst episodes of my life, for being a great sport when it comes to sharing the most fun parts of life, and for being even slightly more confident in me then myself. Eric, thank you is not enough: I miss you every day! Lasts me to thank Gemma Zonneveld, my mum, for being so incredibly proud of me that it is almost impossible not to be proud of myself as well, and Wouter en Anne, my brother and sister, because them being family means more to me than all friendships together. Flora Beeftink, July 2008 vi

7 Contents Acknowledgements v Chapter 1 1 Introduction Chapter 2 9 The effect of interruptions and breaks on insight and impasses: Do you need a break right now? Chapter 3 23 Working on something else for a while: Uncovering patterns of task engagement and disengagement in creative professions Chapter 4 51 Managing one's creativity? A diary study among R&D engineers on the effect of time management on positive affect and daily creativity Chapter 5 77 Being successful in a creative profession: The role of having an innovative style, self-regulation of design work, and self-efficacy Chapter General discussion vii

8 References 117 Summary 133 Samenvatting 137 About the author 141 viii

9 Chapter 1 Introduction In general, a creative profession requires engaging in cognitive creative activities, taking time to let ideas sink in (i.e., incubation time), and making sure that project milestones and deadlines are met, while monitoring the timeliness, efficiency, and quality of work tasks. Although managing one's time in creative professions seems therefore to be highly relevant, up to now, little research has been conducted to investigate its added value. This dissertation presents five studies (one experiment and four field studies) in which we address the self-regulation of time (i.e., time management) in creative work. Particular emphasis is put on incubation time, task discretion, pacing patterns, deadlines, and self-regulation behaviors such as prioritizing, planning, and monitoring creative work activities. Knowledge based work, and creative work in particular, is becoming increasingly important in today's economies (DeFlillippi, Grabher, & Jones, 2007). Creative work requires specific research attention, because it differs from other work: It concerns the development of ideas that are novel and potentially useful (Amabile, 1996). Past research on creativity at work focused predominantly on the influence of personal characteristics and the work environment on creative performance (Shalley, Zhou, & Oldham, 2004). For example, it has been found that it is helpful to creativity to have an innovative cognitive style (i.e., an 1

10 Chapter 1 orientation towards creative problem solving) (e.g., Tierney, Farmer, & Graen, 1999), or a high level of self-efficacy (i.e., the extent to which individuals believe that they can successfully execute the behaviors required to produce an outcome (Bandura, 1977)). However, many questions still remain to be answered. In this dissertation we focus on a topic that has received little research attention so far: the role of time in creative work (Drazin, Glynn, & Kazanjian, 1999), and, specifically, the role of self-regulation behaviors that are focused at managing one's time at work, in more popular terms: time management. We investigate how creative professionals manage their time at work, whether they do it differently from people in other professions, and whether it is beneficial when they make an effort to manage it well. We built on Control Theory of Self-Regulation (Carver & Scheier, 1982). This theory suggests that task attainment requires cycles of feedback loops, in which 'feedback' refers to information about task progress. The purpose of going through these feedback loops is to manage the process of task execution and adjust the course of action if necessary. (For a more detailed description of this theory we refer to Chapters 4 and 5 of this dissertation.) Time management implies going through these self-regulation feedback loops, but, in addition, puts a particular emphasis on time use. Thus, in the term 'time management', it is not the time that is managed, but rather the process of task attainment with a focus on time use (Claessens, Van Eerde, Rutte, & Roe, 2007). In the following, we will first define creative work and elaborate on the role of time in creative work. Then, we will describe the research questions that are addressed in this dissertation. We will end with providing the outline of the dissertation in which we briefly illustrate the content of the different chapters and studies. Creative Work In this dissertation we use Amabile's (1996) well known definition of creativity: Creativity concerns the development of ideas that are novel and potentially useful. Our definition of creative work is therefore: Work in which it is required to develop ideas that are novel and potentially useful. In the work of a research and development engineer, for example, designing a new (novel) product that must meet functional specifications (usefulness) requires considerable creative efforts. Moreover, we extend our definition of creative work 2

11 Introduction to work in which novel and potentially useful ideas are required to solve ill-defined problems. Ill-defined problems are problems for which a clear path to solution is unknown (Sternberg & Davidson, 1999) and for which creative problem solving is necessary to find a solution (Cropley, 1999). An R&D engineer, for example, does not know beforehand what specific activities are needed to develop a certain new product. Similar to the definition of Amabile (1996), our definition does not distinguish between incremental and more radical creative contributions, because, in our view, creativity is needed for both. Our definition excludes work with only opportunities for voluntary creativity. In other words, we exclude work in which creativity is optional and not part of one's main responsibilities (Unsworth, 2001), and only include work in which creativity is required. Thus, this dissertation concerns the work of creative professionals. Specifically, this dissertation focuses at the role of time in their work. Time in Creative Work Time is latently present in our daily lives, and therefore, also at work: It passes while we engage in work activities. Thus, it also passes while engaging in creative work activities. Although in this perspective the role of time is passive, time can also have a more dominant role in creative work. First, time can be part of the creative process. Incubation time literature suggests that some time away from a creative task (i.e., incubation time) aids creative problem solving (Wallas, 1926). After a period of time during which an individual has been doing something else, suddenly and unexpectedly an insight (i.e., a helpful idea) can come to mind ('eureka!'). Thus, temporary disengagement from a creative task has purpose, and in this perspective, (incubation) time can be considered part of the creative process. Second, time can provide structure to the work through the use of project milestones and deadlines. Orlikowski and Yates (2002), for example, referred to project milestones and deadlines as being temporal structures that "are created and used by people to give rhythm and form to their everyday practices" (p. 685). In other words, individuals use them to make decisions with regard to the allocation of their time to their work activities. 3

12 Chapter 1 A third perspective views time as a scarce resource (Ancona, Okhuysen, & Perlow, 2001, p. 515). This perspective focuses on the economic side of time at work, in which time represents a valuable commodity. This perspective concerns aspects such as timeliness, efficiency, and quality. Timeliness refers to norms for when a certain task should be completed, and puts restrictions on the amount of available time. Striving for efficiency means trying to do as much as possible in as little time as possible. Being short of time may result in quality-time trade-offs. Thus, in concepts as timeliness, efficiency, and quality, time is scarce and valuable and should be used as effectively as possible. When considering the role of time in creative work, these three perspectives are all relevant: Creative professionals use their time at work to execute creative tasks, or for incubation, while monitoring the quality, efficiency, and timeliness of task completion, within time structures of project milestones and deadlines. Taking this into account, one can realize that it may require considerable efforts to manage time at work in creative professions. However, up to now, little empirical evidence has been provided to verify this. Self-Regulation of Time in Creative Work Research Questions This dissertation addresses a number of research questions concerning time management in creative work. The first concerns the role of the 'self' in the self-regulation of time in creative work. Specifically, whether it makes a difference if individuals can regulate when to work on which tasks themselves, compared to others deciding that for them. In other words, whether the discretion to choose when to work on which tasks contributes to creative performance. This also includes the discretion to choose when to stop and work on something else for a while (i.e., taking incubation time). Alternatively, tasks could be assigned in an optimal way so that incubation may occur, but without the individual's input as to when this is needed (see Chapter 2 of this dissertation). A first research question is therefore: Research question 1: Are individuals more creative if they have personal discretion over how to use their time? 4

13 Introduction The next two research questions concern the role of pacing in creative work. In this denotation pacing refers to the allocation of efforts for the execution of tasks over time (Gevers, Rutte, & Van Eerde, 2006), in other words, to the self-regulation of work efforts over time. Researchers have distinguished different pacing patterns that describe how work efforts can be distributed over time within tasks or projects (Gevers, Mohammed, Baytalskaya, & Beeftink, 2008). For example, individuals may (a) work on tasks with a moderate, constant level of engagement, without periods of very high levels of task engagement, or periods of disengagement; (b) include an intermediate period of disengagement, which may be used as incubation time; (c) do most of the work right after de start of a task or project; or (d) do most of the work at a looming deadline. (More details about these pacing patterns can be found in Chapter 3.) In this dissertation we address two research questions with regard to the pacing of creative work: Research question 2: Do creative professionals distribute their work efforts differently over time compared to others working in less creative professions? Research question 3: How do creative professionals pace their projects over time exactly? The fourth and fifth research questions concern the usefulness of time management in creative work. Although, in general, the self-regulation feedback loop (and thus also time management) seems to be highly useful in the process of task completion, its use in creative work is not self-evident. The difficulty lies in the specific behaviors of the self-regulation feedback loop: (a) deciding what to do, (b) planning for action, (c) task execution, and (d) monitoring task progress (Carver & Scheier, 1998). Since creative work consists of novel, ill-defined tasks (Cropley, 1999), it comprises a high level of task uncertainty. For some behaviors in this feedback loop, and for planning in particular, ill-defined tasks present a dilemma (Ormerod, 2005): Task uncertainty makes planning ahead difficult. On the other hand, the self-regulation feedback loop may be particularly useful in creative work because of this high level of task uncertainty: Adjusting and making new plans regularly during the course of a creative task or project may be required. Therefore, our fourth and fifth research 5

14 Chapter 1 questions regard the benefit of time management and the corresponding behaviors of the self-regulation feedback loop to creative work. Research question 4: Is time management beneficial to creativity? Research question 5: To what extent are self-regulation behaviors (setting priorities, planning work activities, and monitoring time and task progress) related to the success of creative professionals? Outline of the Dissertation The chapters of this dissertation present studies in which different research methods were applied: An experiment, two surveys, an interview study, and a diary study. Table 1.1 provides an overview of the chapters, the studies, their methods, and the topics addressed. Although the chapters are certainly related and may partly overlap, each chapter provides a unique contribution to the literature, and can be read independently of the others. Chapter 2 describes an experiment among 109 students of industrial engineering and management science. In this experiment it was investigated whether task discretion is helpful to creativity. Specifically, we tested whether the discretion to choose when to work on something else for a while (incubation time) is beneficial to creativity, over and above the effect of incubation time when others decide for you when to switch tasks. We investigated whether this would be beneficial to the number of insights conceived to solve problems in the tasks at hand and to prevent impasses (i.e., feelings of being 'stuck'). Chapter 3 contains two field studies that both focused on pacing of creative work. The first is a survey study among 88 individuals working in various professions. The purpose of this study was to investigate whether creative professionals distribute their work efforts differently over time than other professionals, and, if so, whether they have a preference for specific pacing patterns. The second study is an in-depth interview study among 25 architects. In this study we explored pacing patterns in more detail, taking into account some of the complexities of working on creative projects, such as the influence of project milestones and deadlines and of working on multiple projects simultaneously. Chapter 4 presents a diary study among 68 R&D engineers. This study investigated the effect of time management in creative work, over and above the impact of 6

15 Introduction Table 1.1. Overview of chapters, research methods, and topics in the dissertation Topics Chapter Research method Self-regulation of work time Creativity Other topics 2. Experiment Task discretion Incubation time Insight Impasse 3. Survey Pacing patterns Creative professions Interviews Pacing patterns Project milestones and deadlines Multi-tasking Deadline postponement Quality-time trade-offs Creative projects 4. Diary study Time management Daily creativity Fluctuations in daily creativity Innovative cognitive style Daily affect 5. Survey Self-regulation: - Prioritizing - Planning - Monitoring Innovative cognitive style Being a successful designer Self-efficacy Being a successful businessperson 7

16 Chapter 1 having an innovative cognitive style. Specifically, it investigated whether individuals who have a tendency to use time management are more creative during their work days. Furthermore, it was examined whether potential benefits of time management can be explained through its effect on positive affect. So, whether time management allows individuals to be more creative because they feel better. Finally, this study also examined whether time management was related to a more stable creative performance over several workdays. Chapter 5 describes a survey study among 339 architects. The aim of this study was to establish the relevance of three self-regulation behaviors (setting priorities, planning work activities, and monitoring time and task progress) to design work. Specifically, we assessed to what extent self-regulation and innovative cognitive style were related to being a successful designer and businessperson, and the mediating role of self-efficacy in these relationships. The last chapter, Chapter 6, provides an overview and integration of the results of each of the studies. We conclude with suggestions for future research and recommendations for practice. 8

17 Chapter 2 The Effect of Interruptions and Breaks on Insight and Impasses: Do You Need a Break Right Now? * Some time away from a problem, or incubation time, is found to be beneficial to creative problem solving. But are interruptions as equally helpful as breaks? An experiment was conducted to gain more insight into the differences between imposed and self-initiated breaks, and their effects on creativity, specifically on impasses and insights. There were three experimental conditions, (a) a continuous condition, in which participants were not allowed to switch back and forth between tasks, (b) an interruption condition, in which participants had to switch tasks at a predetermined moment, and (c) a break condition, in which participants could switch tasks at their own discretion. Results showed that taking breaks at moments chosen at one's own discretion led to solving more insight problems and reaching fewer impasses than at moments that were chosen by others. Furthermore, compared to working continuously, interruptions led to fewer impasses, but not to solving more insight problems. * This chapter is based on: Beeftink, Van Eerde, & Rutte (in press). The effect of interruptions and breaks on insight and impasses: Do you need a break right now? Creativity Research Journal. 9

18 Chapter 2 At work, most knowledge workers can switch between different activities or take breaks at their own discretion, and they are also regularly subjected to interruptions initiated by others (Czerwinski, Horvitz, & Wilhite, 2004; Nandhakumar & Jones, 2001). Jett and George (2003) theorized that self-initiated breaks from work can have positive effects on creativity, whereas interruptions initiated by others may not bring about these positive effects. This proposition has not yet been tested. The current chapter describes an experiment to gain more insight into the differences between imposed and self-initiated breaks, and their effects on creativity, specifically on insights (i.e., the distinctive and apparently sudden realization of a strategy that aids in solving a problem (Sternberg & Davidson, 1999, p. 58)), and on impasses (i.e., the feeling of being "stuck" and not being able to solve a problem (Fleck & Weisberg, 2004, p. 998)). Insight Problem Solving Having a sudden insight, or "eureka moment", is a well known phenomenon in everyday life and in the literature (Sternberg & Davidson, 1999). Individuals suddenly and unexpectedly get a good idea that brings them a great step further in solving a problem (Wallas, 1926). To be more specific: Individuals may suddenly become aware of a new strategy that may solve the problem at hand (Sternberg & Davidson, 1999). An often cited creative problem solving model that includes sudden insights is that of Wallas (1926). It consists of four stages: (a) The preparation stage: gain in-depth knowledge about a problem, (b) the incubation stage: take some time away from the problem, (c) the illumination stage: suddenly and unexpectedly an insight comes to mind, and (d) the verification stage: verify the applicability of the idea. The underlying hypothesis of this model is that incubation time aids the problem solving process. In a review article about experimental research on the effect of incubation time, Dodds, Ward, and Smith (in press) conclude that this incubation effect exists: roughly 75% of the experiments show positive effects of incubation time. In addition to Wallas' model, Smith and Blankenship (1991) and Smith (1995) suggested that incubation time is only valuable in specific situations: "Problems that are solved immediately require no incubation [time]" (Smith & Blankenship, 1991, p. 63). They argued that individuals try to tackle problems with a certain problem solving strategy, 10

19 Do you need a break right now? and that it is very difficult for them to approach a problem with a different strategy when the first one does not seem to be appropriate. They refer to this block to successful problem solving as a cognitive state of fixation. In four experiments, Smith and Blankenship showed that incubation time was only helpful in situations in which individuals needed to overcome this state of fixation. The insight problem solving model shown in Figure 2.1 combines Wallas' model (1926) with the ideas of Smith and Blankenship (1991) and Smith (1995). First, individuals start working on a problem, which results in some problem progress. This is similar to what Wallas called the preparation stage. If the problem is tackled with the appropriate problem solving strategy, it can be solved at once. If not, individuals reach a cognitive state of fixation. They can keep trying to solve the problem, but the chance that they will find a suitable strategy to tackle the problem is very small and, most likely, the problem will remain unsolved. Another option is to take some incubation time. A sudden insight might occur that reveals a new strategy which may solve the problem. Start working on problem Problem progress Fixation Continue to work on problem Incubation Problem solved Problem remains Insight Figure 2.1. Insight Problem Solving Model The Effect of Interruptions and Breaks on Insights Both interruptions and breaks give individuals the opportunity to take some time away from a problem, in other words, to have some incubation time. As previous research suggests (Dodds et al., in press), either form of incubation time should improve problem solving performance. Therefore, the first hypothesis of this chapter states that individuals who are 11

20 Chapter 2 interrupted benefit from the incubation effect, and solve more problems than individuals who continuously work on these problems. Similarly, a second hypothesis is that individuals who can take breaks at their own discretion solve more problems than individuals who continuously work on these problems. Hypothesis 1: Individuals who are interrupted solve more insight problems than those who work continuously on them. Hypothesis 2: Individuals who can take a break at their own discretion solve more insight problems than those who work continuously on them. Not empirically tested before is the question whether there is a difference between the effects of interruptions and breaks on creative performance, specifically, whether breaks should be preferred over interruptions. Interruptions are generally seen as negative and disruptive whereas breaks are not (Jett & George, 2003). In line with this belief, time management training advises to avoid interruptions as much as possible (Green & Skinner, 2005; Macan, 1994). Some research confirms the proposed negative effects on task performance. For example, in a field study in a commercial telecommunications office, Eyrolle and Cellier (2000) found that operators needed more time to complete tasks that were interrupted compared to tasks that could be carried out without interruptions. They also conducted an experiment that revealed that interruptions also have a negative effect on the number of errors individuals make. Recently, Bailey and Konstan (2006) have confirmed both of these findings in a similar experiment. They found that interrupted individuals needed more time to complete their tasks compared to uninterrupted individuals, and that they made twice as much errors. But not all research finds negative effects of interruptions on task performance. In a field study among 46 bank managers, König, Kleinmann, and Höhmann (2004) found no relation between daily performance and interruptions. Zijlstra, Roe, Leonora, and Krediet (1999) conducted two experiments in which they found that interruptions caused individuals to work faster while maintaining the same level of quality. In conclusion: Former research suggests that, opposed to the positive effect of incubation, interruptions might also cause negative effects on task performance. Self- 12

21 Do you need a break right now? initiated breaks, on the other hand, may not bring about such negative effects because individuals can choose a convenient point in time to take them. Therefore, a third hypothesis proposes that individuals who can take a break at their own discretion outperform individuals who are interrupted. Hypothesis 3: Individuals who can take a break at their own discretion solve more insight problems than those who are interrupted. The Feeling of Being Stuck Conceptually new to this field of research is the distinction between the cognitive state of fixation and the psychological state of impasse, and the use of impasses as a dependent variable in creativity research. An impasse is defined as the feeling of being "stuck" (Fleck & Weisberg, 2004), and can be differentiated from the cognitive state of fixation in two ways: First, the psychological state of impasse is affective and incorporates negative emotions such as confusion and frustration (Fleck & Weisberg, 2004), whereas the cognitive state of fixation is free of such emotions because it occurs unconsciously. Second, a lagged effect might exist between the moment of fixation and the perception of reaching an impasse: After fixation, individuals might not realize immediately that it will be difficult to solve the problem. Instead, it might take some time before they feel "stuck". In an experiment by Segal (2004), for example, participants worked up to twenty minutes on the presented insight problem before they stated that they had reached an impasse. To be more precise, we suggest that, after individuals get into a state of fixation, negative emotions gradually increase over time. When confusion and frustration reach a certain level, individuals decide that they do not want to work on the problem anymore and give up. At this point in time they reach the psychological state of impasse. In previous research on the incubation effect, both the cognitive state of fixation and the psychological state of impasse were predominantly treated as a necessary precondition for the incubation effect to occur (Dodds et al., in press). In this chapter the two are conceptually separated: Fixation is treated as a necessary precondition, but impasse as an affective outcome of the creative process, worthwhile to be studied in more detail. 13

22 Chapter 2 The Effect of Interruptions and Breaks on Impasses Both interruptions and breaks might be of influence on the process of reaching an impasse. After individuals reach the cognitive state of fixation, interruptions or breaks may bring the process of building up negative emotions to a temporary halt. This halt might prevent these emotions from reaching the critical level of impasse. Assuming that individuals reach a state of fixation if they are not able to solve an insight problem right away (Smith, 1995; Smith & Blankenship, 1991), Hypothesis 4 states that individuals who are interrupted report impasses less often than individuals who continuously work on problems. Similarly, a fifth hypothesis is that individuals who can take a break at their own discretion report impasses less often than individuals working continuously. Hypothesis 4: Individuals who are interrupted report fewer impasses than those who work continuously on an insight problem. Hypothesis 5: Individuals who can take a break at their own discretion report fewer impasses than those who work continuously on an insight problem. The last hypothesis focuses on the question whether there is a difference between the effects of interruptions and breaks on impasses reached, and which is to be preferred in order to prevent an impasse. Here the emphasis lays on the affective influences of interruptions and breaks, and whether these add to the negative feelings leading to the impasse. Breaks are not expected to be a source of negative emotions, but interruptions may be. For example, in the experiment of Bailey and Konstan (2006), interrupted participants felt more annoyed and anxious than uninterrupted participants. Similarly, Zijlstra et al. (1999) found that interruptions had a negative effect on their participants' emotions. On the other hand, interruptions may also prevent feelings of boredom when individuals are working on simple tasks (Fisher, 1998). To summarize: It is expected that, although interruptions may reduce feelings of frustration, they may also be a source of negative emotions. Therefore, the sixth hypothesis states that individuals who can take a break at their own discretion feel fewer negative emotions than individuals who are interrupted, and as a consequence they reach the state of impasse less often. 14

23 Do you need a break right now? Hypothesis 6: Individuals who can take a break at their own discretion report fewer impasses than those who are interrupted. Method Participants One hundred and nine Dutch undergraduate students (12 female) participated in the experiment. The average age of the participants was 20.8, the youngest was 19, and the oldest 27 years old. Their major was in Industrial Engineering and Management Science and they were highly educated in math and statistics. Task The creative task for each of the students was to solve three insight problems. Different types of insight problems exist: verbal, mathematical, and spatial problems (Dow & Mayer, 2004). We decided not to use mathematical or spatial insight problems, such as the well known nine dot problem (Runco & Sakamoto, 1999; Sternberg & Davidson, 1999), because of the risk that these would be solved mathematically instead of through insight (for comments on hybrid insight problems see: Weisberg, 1995). This risk was particularly large in our student sample. Therefore, verbal insight problems were used: Three relatively difficult cryptic crossword clues 2.1. An example is: "not seeing the window covering (5 characters)" is "blind". The three word puzzles were selected and pre-tested during a pilot study among 36 undergraduate students with the same educational background. Pre-testing showed answer patterns typical for insight problems: Solutions were generally found right away, in the first minute (55%), or when participants came back to a word puzzle after working on other word puzzles (21%). Design The design consisted of three conditions. Participants in all conditions had a total of three times six minutes to work on the word puzzles. Participants in the continuous condition 2.1 In cryptic crosswords, each individual clue is a word puzzle in and of itself. In this case the clues are the insight problems, not to be mistaken with clues or hints that are sometimes used as intervention to aid insight problem solving (e.g., Chronicle, Ormerod, & MacGregor, 2001). 15

24 Chapter 2 were not allowed to switch back and forth between the word puzzles, which meant that they worked six minutes on puzzle A, then six minutes on puzzle B, and then six minutes on puzzle C (see Figure 2.2). In the interrupted condition participants had to switch to the next puzzle after three minutes, and came back to it after they had worked on the other two puzzles for three minutes each. The sequential ordering of the puzzles was A, B, C, A, B, C (see also Figure 2.2). Note that both the participants in the continuous and the interrupted conditions worked continuously on a puzzle during the first three minutes. Differences between these conditions occurred only after the third minute. In the break condition participants could switch back and forth between the puzzles at their own discretion. An example of a possible sequence of puzzles in this last condition is also shown in Figure 2.2. Continuous Interrupted Break A A A B C A B C B A B C B C B C C time Figure 2.2. Experimental Design, Task Sequences Procedure Each participant was accompanied by an observer. Beforehand, the observers handed out and verbally repeated written instructions regarding the goal and procedure of the assignment. The participants were instructed to notify their observer when a puzzle was solved, or when they had reached an impasse. Participants were debriefed at the end of the experiment and the solutions to the puzzles were sent to them by the next day. Measures The observers recorded solutions and impasses. Since the continuous condition and the interrupted condition differed only after the third minute of a puzzle's allotted time, they distinguished between solutions and impasses that occurred during the first half (first three minutes) and the second half (last three minutes) of a puzzle's allotted time. As is customary when an experimental task consists of multiple insight problems (Dodds et al., in 16

25 Do you need a break right now? press), the data of the three puzzles were combined prior to data-analysis: For each participant, we added the number of puzzles that had been solved during the first half of their allotted time. Similarly, the number of solutions per participant during the second half was calculated. The same procedure was used to calculate the number of impasses during the first and second halves of the puzzles' allotted time. Results Solutions In total, 85 word puzzles were solved, which means that, on average, participants solved less than one out of the three puzzles. Sixty-three puzzles were solved during the first three minutes that had been spent on them, and 22 during the last three minutes. Specifically, half of the puzzles (n = 44) were solved right away, during the first minute, confirming that these were insight problems: Either a solution was found immediately, or it was unlikely that it would be found at all. Means and standard deviations of the number of solutions per participant for each of the three conditions during the first three minutes (first half) and the last three minutes (second half) are presented in Table 2.1. Significant differences between the conditions are marked with different superscripts. Table 2.1. Means and Standard Deviations of Solutions and Impasses Solutions Impasses 1st half 2nd half 1st half 2nd half M SD M SD M SD M SD Continuous (N = 36).62 a a a a.95 Interrupted (N = 36).67 a a a b.69 Break (N = 36).51 a b a c.45 Note: Different superscripts indicate significant differences between conditions, p <.05 (one-tailed). 17

26 Chapter 2 Although most puzzles were solved during the first half of their allotted time, no differences were expected between the three conditions during this first half. A One-Way ANOVA confirmed this: No significant differences were found between the groups, F (2, 108) =.47, p =.63. A second One-Way ANOVA revealed differences between the groups during the second half of the allotted time, F (2, 108) = 4.63, p =.01, ω =.30. Scheffé's tests were conducted to evaluate pairwise differences among the conditions during this second half. Participants in the continuous condition did not solve more puzzles than participants in the interrupted condition, p =.87, ns. This means that Hypothesis 1 was not confirmed; interruptions were not related to more problems solved. In the break condition, all but one participant took breaks from the puzzles they were working on: They started to work on a puzzle, worked on one or more other puzzles temporarily, and then got back to the first one. On average, they switched between puzzles nine times. During the second half of the allotted time, participants in the break condition solved more puzzles than participants in the continuous condition, p =.010 (one-tailed). Hypothesis 2 was confirmed, which means that positive effects of breaks on problem solving performance were found. Furthermore, participants in the break condition solved more problems than participants in the interrupted condition, p =.035 (one-tailed). Hypothesis 3 was also confirmed: Individuals who could take a self-initiated break outperformed the interrupted individuals. Impasses Ninety-two impasses were reported, which means that, on average, almost all participants reached an impasse. 2.2 Thirty-four impasses were reported during the first half of the allotted time and 58 during the second half. The fact that more participants felt stuck during the second half supports our assumption that it takes some time to reach the psychological state of impasse. Means and standard deviations of the number of reported impasses per participant for each of the three conditions during the first and the second half of the allotted time are presented in Table One outlier was excluded, because this participant reported nine impasses. 18

27 Do you need a break right now? Again, effects were only expected during the second half of the allotted time. A One-Way ANOVA showed no differences between the three conditions during the first half of the allotted time, F (2, 108) =.18, p =.84 ns, and did reveal differences between the groups during the second half of the allotted time, F (2, 108) = 8.66, p <.000, ω =.42. Scheffé's tests were conducted to evaluate pairwise differences among the conditions during this second half. Participants in the interrupted condition reported fewer impasses compared to the participants in the continuous condition, p =.004 (one-tailed). This means that Hypothesis 4 was confirmed; interruptions reduced the number of reported impasses. The participants in the break condition also reported fewer impasses than the participants in the continuous condition, p =.001 (one-tailed). This means that Hypothesis 5 was confirmed; breaks also reduced the number of impasses. Furthermore, the participants in the break condition reached fewer impasses than the participants in the interrupted condition, p =.035 (one-tailed). Thus, Hypothesis 6 was also confirmed: Taking a break at one's own discretion had additional positive effects over being interrupted. Discussion The purpose of this study was to gain more insight into the differences between imposed and self-initiated breaks, and their effects on creativity, specifically, on insight and impasses. The results reveal two major differences: Individuals who took breaks at their own discretion (1) solved more problems and (2) reached fewer impasses than interrupted individuals. It was also found that interrupted individuals reached fewer impasses than individuals who worked continuously on problems. Insight Problem Solving These results add to the literature on insight problem solving and incubation time. First, the findings confirm the existence of the incubation effect suggested by Wallas (1926), in line with earlier research on this topic (Dodds et al., in press; see also Segal, 2004). Second, they are a unique contribution, because they imply that the incubation effect is primarily evident when individuals can take incubation time at their own discretion, and not when others determine the moment at which the task is interrupted. 19

28 Chapter 2 One possible explanation might be that a self-chosen moment to switch tasks allows individuals to end their cognitive activities on one task and start working on the next, whereas an interruption may not allow for such closure, and temporarily forces individuals to divide their cognitive resources (Norman & Bobrow, 1975). This may have negative influences on task performance (Bailey & Konstan, 2006; Eyrolle & Cellier, 2000). If so, the beneficial effect of the incubation period on the creative problem solving process might be undone by the negative effect of being interrupted. A second possible explanation might be that the incubation effect is stronger when individuals intentionally choose to take some incubation time. In a survey among 213 college professors, Wells (1996) found that those who intentionally set manuscripts aside for some time produced more published articles than those who did not. More research is needed to replicate our results and investigate the precise mechanisms responsible. The results have important implications for further research on the incubation effect. In previous research, interruptions, rather than self-initiated breaks, were used in experimental designs: Generally, participants were not allowed to choose when they switched to another task and back. Instead, they had to switch tasks at predefined moments (Dodds et al., in press). Exceptions are some early studies on incubation (e.g., Olton, 1979; Patrick, 1938) and a recent study by Christensen and Schunn (2005), in which participants were free to move back and forth between tasks. Other exceptions are experiments in which participants were allowed to switch tasks after they had reached an impasse (e.g., Butler & Thomas, 1999; Segal, 2004; Seifert, Meyer, Davidson, Patalano, & Yaniv, 1995). Considering our results, it might be preferable to incorporate breaks instead of interruptions in future experimental designs when studying the incubation effect. The Feeling of Being Stuck A second unique contribution of this chapter is that it is one of the first experimental studies in which impasses, or feelings of being "stuck" (Fleck & Weisberg, 2004), are considered to be an affective outcome of the creative process, and have therefore been measured as a dependent variable. The results of this study suggest that both interruptions and breaks help to prevent individuals from reaching an impasse, and that individuals who take breaks at their own discretion reach even fewer impasses than interrupted individuals. 20

29 Do you need a break right now? Presumably, the underlying processes of these positive effects are associated with emotions rather than cognitions. If individuals use a faulty problem solving strategy, negative feelings, such as frustration and confusion, might build up gradually and result in an impasse when they reach a certain critical level. Both interruptions and breaks might prevent the negative emotions from reaching this critical level. Emotions may also explain why breaks are to be preferred over interruptions. The division of cognitive resources over the main task and the second, interrupting task (Norman & Bobrow, 1975) can cause an increased level of stress (Kirmeyer, 1988) and negative emotions (e.g., Bailey & Konstan, 2006; Zijlstra et al., 1999). When these are added to the negative emotions caused by the unsolvable problem, interrupted individuals might reach the critical level of impasse at an earlier moment than individuals who are able to take breaks at their own discretion. In future research, the psychological state of impasse might be an interesting concept for further investigation. Specifically, it may be interesting to focus on both the psychological state of impasse and the cognitive state of fixation and see how these relate to each other, to performance measures, and to other psychological constructs. Fixation might be an important construct with regard to problem solving and creative performance, and negative psychological effects are to be expected from impasses. To gain more insight into the difference between fixation and impasse, one could conduct an experiment similar to that of Fleck and Weisberg (2004), in which participants were asked to think aloud while solving an insight problem. Fleck and Weisberg primarily focused on the processes leading to insight, but a future experiment could focus on faulty problem solving strategies (fixation) and the development of emotions leading to the feeling of being stuck. Interruptions are generally considered to have an unfavorable effect on performance (Jett & George, 2003). Therefore, time management training advises individuals to prevent them as much as possible (Green & Skinner, 2005; Macan, 1994). Although the laboratory results of the current study may not be fully generalizable to creative tasks in a work setting, they imply that in certain situations interruptions may be beneficial. When individuals need to work on a problem on which they do not make much progress, an interruption may prevent high levels of frustration and a feeling of being stuck. Yet, we found that taking a break at one's own discretion is preferable to being interrupted. 21

30

31 Chapter 3 Working on Something Else for a While: Uncovering Patterns of Task Engagement and Disengagement in Creative Professions * In two field studies we investigated the dynamics of task engagement in creative work. We set out to develop an archetypical pattern of task engagement and disengagement over time, by linking the literatures about flow (high task engagement), incubation time (disengagement from tasks), and task pacing behavior (the tendency to distribute work efforts over time in a certain way). The first study, a survey study among 88 employees, showed that individuals with more creative work more often used patterns that included intermediate periods of disengagement and patterns that included most engagement right before a deadline. The second study, an explorative interview study among 25 architects, was used to explore patterns of task engagement and disengagement in more depth. Several suggestions for further research are presented. A substantive amount of empirical evidence reveals that creativity at work is related to organizational innovation, effectiveness, and survival (Amabile, 1996). Moreover, the * This chapter is based on: Beeftink, Van Eerde, & Rutte (2008). Working on something else for a while: Uncovering patterns of task engagement and disengagement in creative professions. Manuscript submitted for publication. 23

32 Chapter 3 current shift towards knowledge-based work has turned creativity into a source of strategic advantage (DeFlillippi et al., 2007). Therefore, an increasing number of research articles have examined the influence of personal and contextual factors on creativity (e.g. Miron, Erez, & Naveh, 2004; Ohly, Sonnentag, & Pluntke, 2006; Van Dyne, Jehn, & Cummings, 2002). Nonetheless, research that investigated creativity at work has been focused on creative outcomes predominantly and creative processes have been understudied (Drazin et al., 1999; Gilson & Shalley, 2004). For creative professionals, however, understanding these processes might be particularly important, because it is these processes (and not their outcomes) that they may manage and improve to optimize their creative performance. In general, little is known about processes that evolve over time in creative work (Drazin et al., 1999). A reason for this is that most of the past research on creativity at work investigated concurrent relationships between two or more variables, and did not investigate patterns of change within variables over time. In other words, most research followed a static research perspective and investigated "What is related?", and not a more dynamic perspective to investigate "What happens?" (Roe, 2008). This chapter will provide a unique contribution to the literature, because we use an innovative research approach by taking the dynamic research perspective and investigating patterns of change within one variable: task engagement. We focus on changing levels of task engagement over time during the course of creative tasks or projects. Task engagement is highly relevant for the execution of creative work, because it contributes to task completion of self-managed work tasks. It is an interesting subject for creative work in particular, because both being highly engaged in work tasks and being disengaged from them for a while provide advantages for creative processes: Research on flow experiences suggests that being highly engaged in work tasks (being in 'flow'; a state of total involvement in the work (Csikszentmihalyi & Csikszentmihalyi, 1988b)) contributes to creative performance (e.g., Csikszentmihalyi, 1996; Lefevre, 1988), and research on 'incubation time' shows that some time away from a problem (disengagement from a task) aids creative processes and helps to overcome impasses, i.e., feelings of being 'stuck' (e.g., Beeftink, Van Eerde, & Rutte, in press; Segal, 2004). Since a high level of task engagement and task disengagement cannot be combined at the same time, both phenomena are particularly interesting to study over time: It seems likely that creative professionals 24

33 Working on something else for a while benefit from work patterns in which periods of disengagement are alternated with periods of being highly engaged in work tasks. As past research did not explore these patterns, we considered it valuable to do so in two field studies. In the two studies we set out to investigate patterns of task engagement and disengagement over time in creative professions. Before we investigated these patterns in detail, we wanted to ascertain that creative professionals have patterns of task engagement that are distinct from those of less creative workers. Therefore, the research question of Study 1 was: Do creative professionals have different patterns of task engagement and disengagement over time than individuals in less creative professions? Subsequently, we investigated the patterns of task engagement of creative professionals in more detail in Study 2, using an in-depth explorative research method. The research question of Study 2 was: Can we provide more insights into the patterns of task engagement and disengagement over time in creative projects and develop an archetypical pattern of task engagement that takes into account some of the complexities of the creative work environment? To investigate patterns of task engagement we built on research on task pacing behaviors. In this denotation, pacing refers to the allocation of time and efforts for the execution of tasks (Gevers et al., 2006). The concept of task pacing behaviors was introduced by Gersick (1988; 1989). She applied the 'punctuated equilibrium' model (Eldredge & Gould, 1972) that describes patterns of relative stability and sudden radical change in the field of biology to social sciences. Similar to the idea of patterns of change in task engagement over time, Gersick (1988; 1989) described patterns of change in activity within teams over time. In her field study among teams working under deadline conditions (1988), she found that high levels of activity occurred at points in time when group members were highly aware of their deadline. Activity levels were relatively low at other points in time. Gersick argued that these fluctuating activity levels resembled the pattern of a punctuated equilibrium. Although her punctuated equilibrium model was intended to explain behavior at the group level, Seers and Woodruff (1997) showed empirically that the patterns reflected task pacing behaviors that also existed at the individual level, rather than behaviors that applied to groups only. Research on task pacing behaviors revealed that working on tasks or projects is generally characterized by alternate periods of activity (task engagement) and inactivity (disengagement), both in group work and in projects on which 25

34 Chapter 3 people work individually (e.g., König & Kleinmann, 2005; Lim & Murnighan, 1994; Seers & Woodruff, 1997; Waller, Zellmer-Bruhn, & Giambatista, 2002). However, up to now, these patterns have not been studied empirically among creative professionals. In the following, we will first explain why task engagement and disengagement are relevant concepts with respect to creativity. Then, we will use the literature about task pacing behaviors to provide some insights into the patterns of task engagement over time. In Study 1 we will argue that some patterns might suit creative work better than others, and formulate and test hypotheses accordingly. Study 2 will be used to uncover patterns of task engagement and disengagement over time in creative work in more detail. The Advantage of Being Disengaged from Creative Work Tasks The main advantage of being disengaged from creative work tasks is that it provides for incubation time. Incubation time literature is based on the model of creative problem solving of Wallas (1926). This model describes four stages: The first stage concerns preparation in which individuals gain in-depth knowledge about a problem. This stage ends if an impasse is reached, in other words, when one feels 'stuck' (Fleck & Weisberg, 2004). The second stage is the incubation period, the period of disengagement from the task. In the third stage, the illumination stage, suddenly and unexpectedly an insight or good idea comes to mind. In the last stage, the verification stage, the applicability of the idea is verified. The underlying hypothesis of Wallas's model was that incubation time aids creative processes. This hypothesis has been tested extensively through experiments, and has found considerable support (e.g., Beeftink et al., in press; Christensen & Schunn, 2005; Segal, 2004; Smith & Blankenship, 1991). Three possible explanations for the positive effect of incubation time have been provided (Segal, 2004). First, as was originally suggested by Wallas (1926), the effect can be explained by the unconscious integration of information that is stored in memory. A second, alternative explanation is that external cues encountered during the incubation period may trigger the emergence of new ideas. A third explanation questions the role of conscious and unconscious information processing during the incubation period, and suggests that the positive effect can be explained by the act of merely addressing a problem anew. Empirical evidence has been provided to support each 26

35 Working on something else for a while of these explanations (c.f., Dijksterhuis & Meurs, 2006; Segal, 2004; Sio & Rudowicz, 2007). Thus, research results on the underlying mechanisms of incubation time are inconsistent and each of the three explanations seem to be equally likely. Nevertheless, researchers generally agree that a period of disengagement from the work is helpful to creative performance. In contrast to the extent to which incubation time has been studied in experimental settings, incubation time at work has received very little research attention. Although anecdotes about incubation time in creative and scientific work are numerous (Smith & Dodds, 1999), we know of only one field study that addressed incubation time at work: A survey study of Wells (1996) among college professors. His research showed that more than 50% of the manuscripts the professors worked on had required a period of time in which they intentionally put them aside for a while. Furthermore, it showed that the creative productivity of the college professors was significantly related to the extent to which they used these periods of intentional disengagement. Although the support is limited, the results are a first indication that periods of disengagement are helpful, also in a work environment. Nevertheless, Wells's research did not provide a comparison of these periods of disengagement between creative professionals and individuals in less creative jobs. The Advantage of Being Highly Engaged in Creative Work Tasks Of course, only being disengaged from tasks does not lead to anything. Even the model of Wallas (1926) suggests that periods of task engagement are necessary: The period of disengagement should be preceded by a period of hard work during the preparation stage (Sternberg & Davidson, 1999), and should be followed by a period of task engagement during the verification stage (Smith & Dodds, 1999). The research of Kahn (1990) among architects and workers at a summer camp even suggests that creativity flourishes during high levels of task engagement, and not during periods of disengagement. Similarly, the research on flow suggests that being highly engaged in work tasks is beneficial to creativity. The state of flow can be described as a state of total involvement in the work (also known as task absorption) and is characterized by high levels of concentration, positive affect, and feelings of timelessness (Csikszentmihalyi & Csikszentmihalyi, 1988b; 27

36 Chapter 3 Quinn, 2005). This state of flow has been found to be highly related to creativity (Csikszentmihalyi & Lefevre, 1989; Lefevre, 1988). The concept of flow is related to the concept of work engagement. Work engagement can be defined as a positive, fulfilling state of mind that is characterized by vigor, dedication, and task absorption (Schaufeli, Bakker, & Salanova, 2006). Work engagement has been found to be related to positive work affect and good health (Demerouti, Bakker, de Jonge, Janssen, & Schaufeli, 2001; Rothbard, 2001). While most studies on work engagement investigated differences between individuals, Sonnentag (2003) showed that levels of work engagement varied over separate days within persons over time. This suggests that individuals may have been highly engaged in their work at certain moments in time, and less engaged at other moments in time. However, neither the research on flow, nor the research on work engagement provides more detailed insights into these fluctuations in task engagement over time. For example, a closer examination of these fluctuations may reveal that they are arranged in identifiable patterns of high and low task engagement from a creative task's or project's beginning to its deadline. To be able to provide some insights into the different patterns in which work efforts can be distributed over time over the course of a task or project we turn to the literature on task pacing behavior. Pacing Patterns: Patterns of Task Engagement over Time Since Gersick's (1988; 1989) work on task pacing behaviors, a number of researchers have distinguished specific pacing patterns over time. For the purpose of this chapter we will describe four distinct pacing patterns on how activity may be distributed over the course of tasks or projects with a deadline. The first is the 'constant' pacing pattern, introduced by Lim and Murnighan (1994, see also: Blount & Janicik, 2002; Claessens, 2004a; Gevers et al., 2006). This pattern represents a constant moderate level of task engagement, without periods of very high levels of work efforts or periods of disengagement. The second pattern, the 'incubation time' pacing pattern, was introduced by Claessens (2004a) and can be considered the counterpart of the constant pacing pattern. The incubation time pattern represents a work pattern in which a period of high task engagement is followed by a period of disengagement (that may represent a period of incubation time) and ends again with a 28

37 Working on something else for a while period of high task engagement towards the deadline. They are opposites because the constant pattern represents stability over time, whereas the incubation time pattern fluctuates from high task engagement to no task engagement and back. Two other pacing patterns that may be considered opposites are the 'early worker' pattern and the 'deadline worker' pattern (Gevers et al., 2008; Gevers et al., 2006). The early worker pattern represents a pacing pattern in which individuals highly engage in tasks right after the start of the task and finish them well before the deadline. In the deadline worker pattern periods of little task engagement are followed by increasing levels of effort and task engagement towards deadlines. Thus, these two patterns are opposites because an early worker tends to put a burst of effort into a task or project in the beginning, whereas a deadline worker does not do much in the beginning. Moreover, an early worker prevents high levels of task engagement towards deadlines, whereas a deadline worker spends high levels of effort close to deadlines. Study 1 Constant Pacing Patterns or Incubation Time Pacing Patterns? As we have argued above, creative professionals may benefit from both periods of high task engagement and periods of task disengagement. Therefore, certain pacing patterns may suit creative work well, whereas others may suit it less well. Specifically, the incubation time pattern might fit creative work well, because this pattern follows the stages of preparation (task engagement), incubation (disengagement), and verification (task engagement) of Wallas's (1926) model. This type of pacing pattern may be preferred over the constant pacing pattern, because the constant pattern does not provide an intermediate period of disengagement, and might thus less likely suit creative work. Therefore, we hypothesize that individuals working in creative professions more often use incubation time pacing patterns, and less often use constant pacing patterns. Hypothesis 1: Individuals working in more creative professions more often distribute their work efforts over time following incubation time pacing patterns than workers in less creative professions. 29

38 Chapter 3 Hypothesis 2: Individuals working in more creative professions less often distribute their work efforts over time following constant pacing patterns than workers in less creative professions. Early Workers or Deadline Workers? A combination of task engagement and disengagement can also be found in the deadline worker and the early worker pacing patterns: Deadline workers are highly engaged at the end of their tasks or projects, and not at the beginning; the early workers are highly engaged at the beginning of their tasks or projects, and not at the end. However, it may make a difference whether the burst of activity emerges at the start or at the end. Generally, the early worker pattern may be preferred over the deadline worker pattern, because deadline workers more often miss their deadlines (Gevers et al., 2006), and produce lower quality results (Gevers, Claessens, Van Eerde, & Rutte, in press). These effects may be the result of procrastination, the unintended delay of work (Van Eerde, 2003a), because the postponement of tasks may lead to increased time pressure, and consequently to difficulties meeting deadlines and to time-quality trade-offs (Van Eerde, 2000). However, "procrastination may lead to better task performance in creative tasks" (Van Eerde, 2000, p. 382), because creative work may benefit from some extra time in which new insights and new information may be gained. Furthermore, it is easier to incorporate changes along the way, which may be necessary giving the high level of task uncertainty of creative work (Cropley, 1999). Thus, despite the increase in time pressure deadline worker pacing patterns seem to be beneficial, particularly in the case of creative work. Therefore, we hypothesize that creative professionals may more often use the deadline worker pacing pattern, whereas individuals working in less creative professions may more often use the early worker pattern. Hypothesis 3: Individuals working in more creative professions more often distribute their work efforts over time following deadline worker pacing patterns than workers in less creative professions. 30

39 Working on something else for a while Hypothesis 4: Individuals working in more creative professions less often distribute their work efforts over time following early worker pacing patterns than workers in less creative professions. Method Procedure We approached participants of 4 workshops in the Netherlands. The topics of the workshops were time use and time management at work. Short questionnaires were completed before the start of the workshops. Two workshops were organized for an audience of mainly creative professionals (e.g., writers, architects), one workshop for employees in healthcare or education (e.g., internist, head nurse; head of student office, director of education), and one workshop for employees from a public housing department (e.g., urban planner, executive secretary). Participants In total, 94 participants filled out the questionnaires. Six of them were students and were therefore excluded from the sample. Forty-eight percent of the 88 remaining workers was male, and the average age was 41.8 years (SD = 9.4). Measures In a short questionnaire we measured age, gender, profession, and use of pacing patterns. The descriptions of the professions served to rate the extent to which creativity was likely to be required in the jobs of the participants. The questionnaire was in Dutch. Pacing patterns. To measure an individual's tendency to distribute work effort over time in a certain way, we used 4 items of the graphic scale derived from Claessens (2004a). Each of the items consisted of a graph that described a distinct pacing pattern. Of the original 5-item scale, one item (the inverted U-shaped pattern) was excluded because it proved to be a relatively rare pacing pattern in past research (Gevers et al., in press). The four graphs represented the pacing patterns that have been described in the introduction: (a) the 'incubation time' pattern, (b) the 'constant' pattern, (c) the 'deadline worker' pattern, and (d) the 'early worker' pattern (see also Table 3.1). On a 5-point Likert scale ranging from 31

40 Chapter 3 (almost) never (1) to (almost) always (5) participants were asked how often they used each of the four pacing patterns in their work. Although this scale included four items, each item represented a distinct pacing pattern that was represented by a graph. Thus, this 4-item scale is actually a 4-variant construct, consisting of four single-item graphic scales. As a consequence, no reliability measures can be calculated. However, reliability may be assumed acceptable, because a validation study of Gevers et al. (2008) revealed that the graphic items related highly to their newly developed multi-item textual counterparts, which, in turn, were sufficiently reliable (.72 < α <.90). Furthermore, this validation study showed that the scales had discriminant and convergent validity to other temporal constructs such as time urgency, polychronicity, and time perspective. Required creativity at work. The extent to which creativity was required at work was judged by two independent raters that categorized the professions of the participants. Each participant's profession was rated and assigned to one of three categories. Following Feist (1998), a distinction was made between: (a) routine work: work that required only little creativity (e.g., the work of executive secretaries or junior account managers), (b) moderately creative work: work that required some creativity for problem solving (e.g., the work of consultants or managers), and (c) highly creative work: work that required artistic creativity (e.g., the work of artists, designers, or architects) 3.1. Three participants were excluded because of missing data. The raters agreed on 83.5% of the professions, corresponding with a Cohen's kappa of.74, indicating substantial agreement (Landis & Koch, 1977). After the rating and calculation of interrater agreement, the professions the raters did not agree upon were discussed until full agreement was reached, and assigned to the categories accordingly. Results Means and standard deviations for the use of each pacing pattern at each level of required creativity at work are presented in Table 3.1. Different superscripts indicate significant differences between levels of required creativity at work for each of the pacing patterns 3.1 A list of the professions within each category will be provided by the first author on request. 32

41 Working on something else for a while Table 3.1. Means and Standard Deviations of the Use of Pacing Patterns at Different Levels of Creativity at Work Pacing patterns M SD M SD M SD M SD Routine work (N = 33) 2.58 a a a a.87 Moderately creative work (N = 36) 3.18 b a a,b a.71 Highly creative work (N = 16) 3.75 b a b b.58 Note: Different superscripts indicate significant differences between levels of creativity at work, p <.05 (one-tailed). 33

42 Chapter 3 5 average use of pacing patterns routine work moderately creative work highly creative work Figure 3.1. Average Use of Pacing Patterns at Different Levels of Creativity at Work separately. Figure 3.1 graphically depicts the extent to which each of the four pacing patterns (the four lines) were used on average at each level of required creativity at work. Hypothesis 1 stated that incubation time pacing patterns would be used more often by individuals working in more creative professions, and less often by individuals in less creative professions. A One-Way ANOVA revealed differences between the levels of required creativity at work for the use of these pacing patterns (F = 8.33, df = 2, p <.001, η =.17). The ascending solid line in Figure 3.1 shows that the difference was in the hypothesized direction: The more creativity was required at work, the more often this type of pacing pattern was used. Scheffé's tests showed that they were more often used in highly 34

43 Working on something else for a while creative work than in routine work (p <.001), and more often in moderately creative work than in routine work (p <.05). No differences were found between highly creative work and moderately creative work (p >.10). The results suggest that the level of required creativity at work is related to the extent to which incubation time pacing patterns are used, and that these pacing patterns are more often used in highly and moderately creative work than in routine work. This means that support was found for Hypothesis 1. Hypothesis 2 stated that individuals working in more creative professions would use constant pacing patterns less often than individuals in less creative professions. A One- Way ANOVA showed no differences between levels of required creativity at work (F = 1.92, df = 2, p >.10, η =.05). Thus, the results do not support Hypothesis 2. Nevertheless, the descending line in Figure 3.1 may be an indication of an effect in the hypothesized direction, that we may not find due to a lack of power because one of our groups (the highly-creative-work group) contained merely 16 individuals. A power analysis (Faul, Erdfelder, Lang, & Buchner, 2007) revealed indeed a lack of power: Power of.38 was calculated, whereas power should be.80 or higher (Cohen, 1988, 1992). It was calculated that a sample size of twice the size (n ~ 160) might reveal differences between groups for this pacing pattern. Hypothesis 3 stated that deadline worker pacing patterns were more often used by individuals working in more creative professions than by individuals working in less creative professions. A One-Way ANOVA revealed a significant difference between creativity levels (F = 3.67, df = 2, p <.05, η =.08). The ascending dashed line in Figure 3.1 shows that the difference was in the hypothesized direction: The more creativity was required at work, the more often this type of pacing pattern was used. Scheffé's tests showed differences between highly creative work and routine work (p <.05), but not between highly creative work and moderately creative work (p >.10), nor between moderately creative work and routine work (p >.10). The results suggest that deadline worker pacing patterns are more often used in highly creative work than in routine work. Therefore, these results support Hypothesis 3. Finally, Hypothesis 4 stated that individuals working in more creative professions use early worker pacing patterns less often than individuals working in less creative professions. A One-Way ANOVA showed a significant difference between the levels of 35

44 Chapter 3 required creativity at work (F = 9.35, df = 2, p <.000, η =.19). The descending dashed line in Figure 3.1 shows that the difference was in the hypothesized direction: The more creative the work, the less often this type of pacing pattern was used. Scheffé's tests revealed that these pacing patterns were less often used in highly creative work than in routine work (p <.001), and also less often in highly creative work than in moderately creative work (p <.05). No differences were found between moderately creative work and routine work (p >.10). These results indicate that the level of required creativity at work is related to the extent to which early worker pacing patterns are used, and that these pacing patterns are less often used in highly creative work than in moderately creative work and routine work. Thus, support was found for Hypothesis 4. Discussion Study 1 confirmed that individuals working in more creative professions differ from individuals working in less creative professions in the ways in which they distribute their work efforts over time from start to deadline. In other words: They use different patterns of task engagement and disengagement over time. Specifically, the results suggest that individuals in more creative professions use incubation time and deadline worker pacing patterns more often than individuals in less creative professions, and early worker patterns less often. No differences were found for the use of constant pacing patterns, but this may be due to power limitations. Our findings with regard to the use of incubation time pacing patterns in creative professions are in line with incubation time literature, which suggests that it is helpful to creative problem solving to include intermediate periods of disengagement (Smith & Dodds, 1999; Wallas, 1926). Also, Study 1 provides a new contribution to the creativity literature, because it compared individuals in creative professions with individuals in less creative professions. Whereas Wells's research (1996) among college professors showed that these professors intentionally used intermediate periods of disengagement, our research results of Study 1 suggest that these patterns apply to creative work specifically, and to a lesser extent to other work. Moreover, this study provides a new contribution to the creativity literature because it shows that creative professionals are more often deadline workers and less often early workers. These results were to be expected on the basis of Van 36

45 Working on something else for a while Eerde's (2000) theorizing on the possible positive effects of procrastination on creative tasks. However, up to now, no empirical evidence had been provided to support this. Thus, this study shows that creative professionals have distinct pacing patterns. However, Study 1 did not provide in-depth insights into the actual patterns of fluctuations in task engagement over time in creative work tasks or projects. Rather, the graphical representations of the pacing patterns scale represented a simplified version of their pacing patterns at work. This is because work is generally represented as a single task with a single deadline in the literature on task pacing behaviors (e.g., Blount & Janicik, 2002; Claessens, 2004a; Gevers et al., 2008; Gevers et al., 2006; Lim & Murnighan, 1994). However, the actual patterns of fluctuations in task engagement over time may be more complex, because work may consist of more than one task with more than one deadline: Work may consist of a complex combination of several tasks with different deadlines (Louro, Pieters, & Zeelenberg, 2007; Schmidt & DeShon, 2007). Pacing patterns of different tasks may overlap and influence each other. Also, tasks and projects may have intermediate milestones that may influence pacing patterns. To be able to get a more realistic picture of the dynamics of task engagement over time, we conducted an in-depth explorative study (Study 2) to be able to examine these patterns in more detail, and to provide insights into the patterns of task engagement over time in creative work that go beyond the simplified representations provided by the literature on task pacing behaviors. Study 2 Method In this study we focused on task pacing behaviors in creative projects. Specifically, we focused on pacing patterns in design projects of architects. To explore in-depth the ways in which work efforts are distributed over time in these creative projects, a multi-method interview approach was used to provide the opportunity to investigate the topic from different angles. Elaborations on the graphic pacing patterns were combined with project timelines and anecdotes from recently finished project designs. 37

46 Chapter 3 Participants Twenty-five Dutch architects were interviewed. We chose to study the work of architects, because in their work creative problem solving is necessary to make designs that meet the technical specifications, and artistic skills are necessary to give designs their aesthetic value. Therefore, architects are generally perceived to be creative professionals (Feist, 1998; Kirton, 1994; Unsworth, 2001). Twenty-two of the participants were male, reflecting the male dominated population in architecture. In the Netherlands, over the past ten years, the average percentage of female workers in architecture and engineering has been 22.0% (Statistics Netherlands, 2008). Work experience varied from less than 5 years up to more than 30 years. Besides their architectural work, some architects held positions in management or middle management. Some others were self-employed running small businesses. The architects were recruited through 'snowball sampling' (Babbie, 2005, p. 190), meaning that the connections of the interviewed architects were used to approach other architects for the study. Procedure Interviews of between 60 and 90 minutes each took place at an interviewee's home or workplace. All interviews were conducted by the first author. To learn more about the patterns of task engagement and disengagement over time, three approaches to collect data were used: (a) pacing pattern graphs, (b) project timelines, and (c) project descriptions. Pacing pattern graphs. Similar to Study 1, we used the scale with the graphical representations of the different pacing patterns derived from Claessens (2004a). However, instead of using it as a survey scale, we used the graphs as a starting point for discussion, asking the architects to choose which of the pacing patterns fitted their design work best, and to elaborate on their choice. Project timelines. In the next part of the interview, we asked each architect to focus on one project for which they had recently finished the design. More specifically, we asked them to focus on one particular phase of the project, the design phase. This is the phase of the project in which the design is developed. The architects were asked to start with a brief description of the building(s) that had been designed. Then, we gave them a large (A3) piece of paper with a timeline on it. Timelines are "graphical representations 38

47 Working on something else for a while people can fill in with details; as visual artifacts, they are more easily processed than other forms of communication." (Barry, 1997, p. 279, in Yakura, 2002). The timeline simply consisted of two dots and a line: On the left side of the line, one dot marked the project start and on the right side another marked the design deadline. We asked each architect to fill out the timeline with major activities concerning the design they had recently finished, and, if applicable, also with milestones within the design phase. Project descriptions. Then, we used the timelines as a vehicle to elaborate on the different facets of the design process over time. This was a very open-ended discussion about the process of the development of the design, resulting in anecdotes of typical events that had happened during the design process. Data Analysis All interviews were audiotaped and notes were taken during the interviews. Also, the timelines were kept as graphical representations of the design projects. The essential excerpts of the tapes were transcribed and segmented, 1022 phrases in total. These phrases were categorized by two coders following a procedure of open coding, which is a systematic coding procedure known from its application in grounded theory methods (Babbie, 2005; Strauss & Corbin, 1990). The phrases were coded and assigned to one of three main categories: (a) pacing, concerning phrases that referred to patterns of task engagement and disengagement over time; (b) creativity, concerning phrases about creativity or the creative process, and (c) deadlines, concerning phrases about milestones and deadlines. The first coder's categorization of each phrase was compared with the second coder's categorization of that phrase. The average agreement of the two coders was 69.1%. The corresponding Cohen's kappa of.52 indicated moderate agreement (Landis & Koch, 1977). Therefore, all phrases on which the coders had been disagreeing were reexamined and discussed until both coders fully agreed on re-assignment to one of the three categories. The three categories included 258, 363, and 401 phrases, respectively. Finally, the data regarding the pacing patterns, the timelines, and the categorized phrases were used to develop an archetypical pattern of task engagement over time in creative projects. 39

48 Chapter 3 Results Before we present the results of Study 2, we provide some background knowledge about architectural projects in general. A full architectural project generally contains three major phases: (a) a design phase, (b) a drawing phase, during which detailed construction drawings are developed, and (c) a construction phase, during which the building(s) is (are) constructed. In this study we focused on the design phase, because this is the phase of the project in which both problem solving and artistic creativity are required, and we wanted to focus on creative work tasks specifically. In general, in the Netherlands, the design phase has a clear start and finish: It starts with the first meeting with a customer and ends on the day a design is approved by both the customer and the local authorities. Usually, a specific date is set on which the design should be sent in for approval. We refer to this point in time as the design deadline. In this study we investigated patterns of task engagement and disengagement over time from the first meeting until the design deadline. Incubation Time Pacing Patterns As starting point, we used the graphic representations of the pacing patterns from Claessens (2004a). Architects were asked how they had distributed their design efforts over time from the first meeting with the customer until the deadline. The majority (n =18) of the architects had been using incubation time pacing patterns. One of them, Martin 3.2, gave a striking description of his pacing pattern: "At the start I work a lot on it, so I know how much [work] it is but once I know, oh, well, that much work needs to be done, then I wait as long as possible, and then I do everything right before the deadline " (2, A ). Of the remaining seven architects, three said they had been deadline workers, and had only been highly engaged in the project close to the deadline. One said he had a constant pattern, which means that his project work was characterized by a constant level of moderate activity. Three architects did not want to choose one of the pacing patterns shown to them, because they thought their pacing patterns had been more complicated, and could not be represented by one of the figures. Specifically, they argued that their patterns were (a) not as abstract as the graphic representations, but fluctuated much more over time, (b) varied 3.2 All names are fictitious to provide anonimity to the participants. 3.3 Interview 2, location on tape: side A between couternumber 037 and

49 Working on something else for a while from project to project, or (c) could be combinations of pacing patterns (for example, a combination of a constant pattern with a heightened level of activity towards deadlines). Nevertheless, most of the architects immediately recognized themselves in one of the pacing patterns, which suggests that the graphic pacing pattern scale had face validity and was appealing to them. Most of the architects (n =15) referred to the creative process to explain their pacing pattern. William, for example, described: "It works rather well to free some continuous time [at the start], so you can thoroughly work your way into it, and thus become aware of the complexities and pitfalls. Well, then you let the material sink in for a while. Thus, the dip in activity is used for letting the material sink in, really." (9, A ). In addition, Martin said: " in this period of time, where I do nothing, that is really the period that I become aware of the problem, and where inspiration comes to me." (2, A ). Although we did not aim to investigate whether the period of disengagement was related to the creative process in this study, the comments of the interviewed architects point in that direction. Thus, most architects had very similar ways of pacing their design projects: They distributed their work efforts over time using incubation time pacing patterns, because, according to them, this supported their creativity. However, the fact that three architects argued that their pacing patterns did not fit the simple representations of the graphs presented to them may be an indication that more sophisticated representations of the pacing patterns may provide additional insights into the patterns of task engagement in creative work. Repeated Patterns A first suggestion for a possible extension of the incubation time pacing pattern was found in the timelines drawn by the architects. Each timeline included at least two milestones. This was due to the ways in which design phases of architectural projects are usually organized. After the first meeting that marks the start of the project, architects go and work on a concept design. This concept design is discussed with the customers and revised until they are satisfied. Then, a more detailed design is made, the preliminary design. This design is also discussed with the customers and revised until they approve. Then, a final design is 41

50 Chapter 3 made, which is of course also discussed with the customers and revised and then send to the local authorities for approval. The milestones in the timelines represented moments in time at which architects had meetings with customers to present their work. This means that they had to have completed a certain amount of design work before these moments in time. Therefore, these milestones were generally perceived as being intermediate deadlines. Phillip described the different deadlines: "There are two deadlines: The first is just the next meeting and the other is, in fact, the product [the final design]." (5, A ). Also, Michael explained that he viewed his milestones as intermediate deadlines. In addition, he told that each of the milestones influenced his work pace. Specifically, he asked "Are we allowed to draw another [graph]?" (16, A ). He drew a pattern of successive U-shapes, which resembled a pattern of recurring cycles of the incubation time pacing pattern with periods of task engagement around project milestones and the deadline and periods of disengagement in between. Thus, the incubation time pattern was repeated for each intermediate deadline. In light of the occurrence of intermediate deadlines in all of the projects described, repeated incubation time patterns may represent the patterns of task engagement and disengagement over time better than a single incubation time pattern. Working on Multiple Projects The original incubation time pattern, and also the repeated incubation time pattern, imply periods of disengagement. However, architects were not really doing nothing during these periods of time, instead, they were working on other projects. During the interviews, all but one architect mentioned that they worked on multiple projects, and that periods of disengagement from one project had been providing the opportunity to work on another project. This means that pacing patterns of multiple projects overlap. Blurred Deadlines In the original incubation time pattern the deadline was marked with a line. This line represented a specific point in time at which a certain amount of design work needed to be completed. However, a line may not have been the best representation of a deadline, 42

51 Working on something else for a while because lines assume a rather fixed character of deadlines, and the interviews revealed that they are not necessarily predefined and fixed and are also not always strictly met. Setting deadlines and milestones beforehand. Although most architects confirmed that they had set milestones and design deadlines when the projects started, six architects had made a conscious decision not to make such rigid appointments beforehand. Chris, for example, said: "I try to avoid deadlines." (16, A ). Instead, he promised his customers that he would work on the design and told them that they could be expecting it in about six weeks. He said that he would give them a call when he was ready. Surely, "in about six weeks" is quite a loose definition of a deadline. More rigid agreements were also made between architects and their customers. Seven architects mentioned that deadlines were imposed upon them by their customers. Richard, for example, who primarily dealt with professional customers, told us: "Usually, it [a deadline] is dictated by others." (24, A ). The other architects negotiated about their deadlines, or picked the dates for the deadlines themselves. For example, Martin explained that he had made a project planning at the project start. He told us: "That's something you just discuss for a moment [with the customer], after which you put it down in a calendar." (2, A ). Deadline postponement. Architects also differed in the rigidness with which they adhered to deadlines. Sarah, for example, stated that although she had finished her work on time at all project milestones, she had been aware that she could have moved her deadlines, since the customers she had made a design for had been "two people to whom you can say: 'I have not finished it yet, it will be ready next week.'" (21, A ). Marc sometimes found excuses to buy himself some more time: "For example, I can very easily ask a customer a certain question about something he had not thought about before, which I say I need to know, and so he will be busy for two weeks. Meanwhile I have some more time to think thoroughly about that project, or maybe do some work on another project that needs attention." (22, A ). About one-third of the architects (n = 9) mentioned the possibility of shifting deadlines. Four other architects had a very different view: They had taken their deadlines very seriously. André explained for example: "Deadlines are there for a reason. Often, very much depends on it. They not only have agreements with us, but with building constructors, structural engineers, and suchlike, and then you just have to 43

52 Chapter 3 finish your work within that amount of time." (3, A ). Two architects explained that some deadlines could be shifted, whereas others could not. Quality-time trade-offs. Some architects mentioned quality reduction decisions because of time limitations. In creative projects the solution is unknown beforehand, and this leaves room for architects to use their own interpretation of what needs to be presented at a deadline meeting. One fourth of the architects (n = 6) admitted that they sometimes had used this lack of clarity to do less than they initially had wanted to do. Kent, for example, told us: "When I am working under high time pressure and run out of time, then I lower my standards with regard to the quality of my work, not with regard to the quality of the final design, but I complete the concept design a bit less or a bit less detailed." (7, A ). Phillip explained: "You should reveal [to a customer] that you have thought about the problem and, at least, that you partly solved it. Maybe you made not as much progress as you had hoped for but if you can defend it well, than it's never a problem." (5, A ). Thus, these results show that architects do not necessarily approach their deadlines very rigidly. Discussion The results of Study 2 suggest that architects generally (a) pace their creative projects using incubation time pacing patterns, (b) repeat these patterns multiple times within one project, with peaks of activity around project milestones and the deadline, (c) work on multiple projects concurrently, which means that pacing patterns of different projects overlap, and (d) may not approach their deadlines very rigidly, which may also affect their pacing behavior. We will now use these results and relevant literature to develop an archetypical pattern of task engagement over time in creative projects. First, on the basis of the results of Study 2 we argue that an archetypical pattern of task engagement over time in creative work should include incubation time pacing patterns. The findings of Study 2 regarding the use of incubation time patterns confirm our findings of Study 1, in which we found that incubation time patterns were used more often in creative professions. Furthermore, these findings are in line with incubation time literature (e.g., Smith & Dodds, 1999; Wallas, 1926; Wells, 1996). Specifically, the incubation time pattern describes a pattern of task engagement and disengagement over time that fits the 44

53 Working on something else for a while stages in the creative problem solving model of Wallas (1926): preparation (task engagement), incubation (disengagement), and verification (task engagement). Therefore, we argue that the incubation time pattern provides a suitable base for an archetypical pattern of task engagement in creative work over time. A first extension of this pattern may be made on the basis of the timelines drawn by the architects. It was found that each timeline included at least two milestones, and that these were perceived to be intermediate deadlines. Tasks often are hierarchically ordered with a main task divided into multiple subtasks (Carver & Scheier, 1998, p. 72). Similarly, projects are often divided into project phases that each end with a project milestone (Nandhakumar & Jones, 2001; Yakura, 2002). As was explained by the architects, these milestones are generally perceived as intermediate deadlines and influence their distribution of work efforts over time. This is in line with the results of a case study on time management in a team that developed computer-based management information systems by Nandhakumar and Jones (2001). They found that "rather than a steady flow of time towards regular milestones, the team members' work was marked by significant changes in pace. Periods of relative inactivity [...] were matched by intensive efforts as deadlines for projects approached" (p. 207). Thus, milestones have similar impacts on patterns of task engagement and disengagement as deadlines. Therefore, we argue that an archetypical pattern of task engagement should follow a repeated incubation time pattern, with peaks of activity around milestones and the deadline, and periods of disengagement in between. While previous research typically described pacing patterns for single tasks or projects (e.g., Claessens, 2004a; Gersick, 1988, 1989; Gevers et al., 2006; Seers & Woodruff, 1997), at work individuals usually deal with multiple tasks and projects at the same time (Louro, Pieters, & Zeelenberg, 2007; Schmidt & DeShon, 2007). The architects in Study 2 also worked on multiple projects concurrently. Specifically, the periods of disengagement from one project provided periods of engagement in other projects. Therefore, an archetypical pattern of task engagement over time should represent overlapping patterns of task engagement and disengagement from multiple projects. Finally, we found that deadlines and milestones were not always fixed and strictly adhered. Some other field studies have also shown that deadlines are not always met and sometimes need to be shifted (e.g., Beech-Hawley, Wells, & Cole, 2004; Nandhakumar & 45

54 Chapter 3 Jones, 2001; Yakura, 2002). For example, in a field study among newspaper workers, Beech-Hawley et al. (2004) found that deadlines were sometimes missed. Furthermore, they found that meeting a deadline required the newspaper workers to work fast, sometimes at the cost of quality or accuracy. To include the different deadline approaches into an archetypical pattern of task engagement, deadlines and milestones should be depicted by blurred lines rather than by straight lines, which may infer that deadlines would be rather fixed moments in time. In sum, we argue that an archetypical pattern of task engagement in creative projects should (a) be based upon the incubation time pacing pattern, (b) follow a repeated pattern over time to take into account the effect of project milestones, (c) include overlapping pacing patterns of different projects, and (d) represent deadlines as blurred lines. The archetypical pattern that results is depicted in Figure 3.2. Figure 3.2. Archetypical Pattern of Task Engagement over Time in Creative Projects General Discussion In two field studies we investigated patterns of task engagement and disengagement over time in creative work. In study 1, we hypothesized and found that individuals in creative professions distribute their work efforts over time differently than individuals in less creative professions. Specifically, we found that creative professionals more often used 'incubation time' patterns (patterns in which individuals are first highly engaged in their 46

55 Working on something else for a while work, then disengaged for a period of time, and then highly engaged again when deadlines approach). Furthermore, it was found that they were more often 'deadline workers' (being little engaged at first, and increasingly engaged towards deadlines) and less often 'early workers' (being highly engaged right after the start of a task or project, and complete tasks or projects well before the deadline). In Study 2, we explored patterns of task engagement and disengagement in creative projects in more detail through interviews with architects. The interviews resulted in an archetypical pattern of task engagement over time. This pattern includes repeated incubation time patterns, with periods of high engagement around project milestones and the deadline, and periods of disengagement in between. It is a multi-project pattern, in which periods of engagement on one project match periods of disengagement from other projects. Furthermore, it takes into consideration that deadlines and milestones may be shifted, and also that they are sometimes rather vague agreements instead of clear-cut deadlines. These studies provide an important contribution to creativity literature because they provide insights into processes that evolve over time in creative work. This is particularly valuable for this field of research because the processes that lead to creative outcomes at work have not often been studied in the past (Drazin et al., 1999; Gilson & Shalley, 2004). We focused on task engagement, more specifically, on patterns of task engagement and disengagement over time in creative work. These patterns were particularly interesting to study, because previous research had shown that both being highly engaged in tasks and being disengaged from them contributes to creativity: Being in 'flow' helps to be more creative (Csikszentmihalyi, 1996; Lefevre, 1988), as does incubation time (Smith & Dodds, 1999). Therefore, it was expected that creative professionals would use patterns that included both periods of task engagement and periods of disengagement, and that is what was found in our two field studies. We built on literature on task pacing behaviors to be able to investigate patterns of task engagement over time. This literature has revealed that progress in tasks or projects is generally characterized by periods of activity and periods of inactivity (Gersick, 1988, 1989; König & Kleinmann, 2005; Lim & Murnighan, 1994; Seers & Woodruff, 1997; Waller et al., 2002), and suggested different pacing patterns to distribute work efforts over 47

56 Chapter 3 time (Claessens, 2004a; Gevers et al., 2008; Gevers et al., 2006; Lim & Murnighan, 1994). Our two field studies also provide a valuable contribution to the task pacing behavior literature. Whereas most of this literature has been focused on team processes (e.g., Gersick, 1988, 1989; Gevers et al., 2006) or individual differences (e.g. Claessens, 2004a), our research shows the relevance of task characteristics: Creative work is paced differently than other work. Furthermore, our research goes beyond the simplified representations of work as a single task with a single deadline: We developed a more sophisticated archetypical pacing pattern that has clear roots in both the creativity and the pacing literature. Strengths, Limitations, and Suggestions for Further Research A strength of this chapter is its multi-method approach. Different methods were used in Study 1 (survey) and Study 2 (in-depth interviews). Since the two studies both focused on patterns of task engagement and disengagement over time, the results are partly overlapping and thus strengthening each other's findings, whereas the differences in methods also have provided complementary results. Specifically, the use of a quantitative (Study 1) as well as a qualitative research approach (Study 2) has provided the advantage of statistically testing differences in pacing patterns between creative and less creative professions, and also the advantage of gaining in-depth insights into the complexities of pacing patterns at work. Also, both our samples comprised participants from a working population: Study 1 included participants from various professions, and Study 2 focused on architects. Student populations are often very homogeneous, and results from research using student samples might therefore be less well generalizable to work situations. Our two studies have the advantage of working samples, which makes it easier to translate their results to practice. However, the studies also have limitations. Maybe the most important limitation of the studies is that they both used single moment measurements, although we investigated patterns that evolve over time. Especially in Study 2, the architects had to recall the processes of the development of the designs, which were processes that had covered longer periods of time. Possibly, our data are biased because the processes may not have been remembered accurately. Thus, in future research it is important to use longitudinal research designs to measure patterns of task engagement over time. 48

57 Working on something else for a while Also, dividing the sample of Study 1 into three groups, each representing a level of required creativity at work, led to rather small group sizes. Although the results are noteworthy, the small sample size might have led to a limited amount of power in our analyses (Cohen, 1988, 1992). If we would have had a larger sample size, maybe clearer distinctions could have been made between the three levels of creativity at work (highly creative work, moderately creative work, and routine work). Maybe we then also would have found differences between the groups for the constant pacing pattern, which we failed to find now. Therefore, it might be worthwhile to replicate Study 1 with a larger sample. Further research might also focus on deadline worker patterns. Study 1 showed that, in addition to the incubation time pacing pattern, highly creative workers more often used deadline worker patterns. In Study 2, only three architects used these pacing patterns, but the narrative data suggested that deadlines and milestones sometimes posed problems for more than these three architects. More research is needed to investigate whether creative workers have more problems meeting deadlines, and if so, why? For example, it may be due to the creative character of the work. Creative tasks entail high levels of task uncertainty (Cropley, 1999), which makes time management of creative work relatively difficult (Britton & Glynn, 1989). More research is needed to explore the interactions between deadlines and creative work, and the factors influencing it. Also, it would be interesting to learn more about patterns of flow and incubation over time. Our studies only focused on patterns of task engagement and disengagement over time. A next step could be to study the cognitive processes underlying the periods of engagement en disengagement. Three alternative propositions may be tested in future research: (a) Flow and incubation time are counterparts; if the one occurs, the other one is unlikely to occur, (b) flow and incubation time are unrelated, because they tap into different cognitive processes that both contribute to creative performance independent of each other, (c) flow and incubation time are related to each other, but a lagged effect exists, in that incubation time may result in the sudden realization of a good idea, which may in turn provide new inspiration that may induce flow. In light of the results of this chapter regarding the patterns of task engagement and disengagement over time in creative work the third proposition may be likely, but more research is needed to test these propositions. 49

58 Chapter 3 Furthermore, we argue that it would be highly valuable to investigate relationships between the patterns of task engagement and creative outcomes. Future research might answer questions such as: Should the incubation time pattern be preferred over the constant pacing pattern in that the first results in better and more creative outcomes? Is the early worker pattern dysfunctional or only rarely used in creative work? Does an incubation time pattern lead to similar results as a deadline worker pattern? Practical Implications This chapter has important practical implications, especially for creative professionals. Our research shows that they generally use patterns of task engagement and disengagement over time. Being aware of these patterns may help them to manage their work more in a way that is beneficial for creativity. For example, they may try to work on multiple tasks or projects simultaneously, so that they have the discretion to work on something else for a while if they feel like it. This may prevent cognitive and emotional disengagement while still trying to complete the task, such as described by Kahn (1990). However, if one works on more than one task or project, attention may be paid to tuning the pacing patterns of different projects. Specifically, one may want to prevent two milestones of different projects close to each other. The two peaks in activity around the two milestones may then sum up to an uncomfortable high level of workload. Similarly, milestones and deadlines close to each other should be avoided. Rather, one would want to spread out the milestones and deadlines of all projects one works on evenly over time. This suggests that, in addition to planning the outline of their individual tasks and projects, creative professionals may benefit from a broad overall planning that includes milestones and deadlines of all the tasks and projects, or at least, that, if they set a new milestone meeting, they make sure that the efforts needed for the new project do not interfere with the pacing patterns of other projects. All-in-all, the two field studies in this chapter provide a valuable contribution to the literature, because they show that creative professionals distribute their work efforts differently over time than individuals with less creative jobs, and that they do so in repeated patterns of task engagement and disengagement over time. 50

59 Chapter 4 Managing One's Creativity? A Diary Study among R&D Engineers on the Effect of Time Management on Positive Affect and Daily Creativity * This study adds a new antecedent to the short list of personal characteristics that influence creative performance: the tendency to use time management. In a diary study among 68 R&D engineers we investigated whether time management, which we refer to as self-regulation with an emphasis on time use and workload, was associated with a higher and more stable level of daily creativity at work. Furthermore, we tested for a possible mediating effect of positive affect. Multilevel analyses showed that the tendency to manage time at the person level was positively related to positive affect and creativity at the daily level. In addition to these direct effects, an indirect was found between time management and daily creativity via daily positive affect, implying that managing time also helps to be more creative because well-organized individuals feel better. Furthermore, individuals who used time management to a larger extent had a more stable level of creative performance over time. * This chapter is based on: Beeftink, Van Eerde, & Rutte (2008). Managing one's creativity? A diary study among R&D engineers on the effect of time management on positive affect and daily creativity. Manuscript submitted for publication. 51

60 Chapter 4 Considerable evidence shows that creativity is relevant and important for organizational innovation, competitiveness, and survival (Amabile, 1996). Therefore, an increasing amount of research focuses on the factors that influence creativity (Shalley et al., 2004), for example, on the role of personal characteristics (c.f., George & Zhou, 2001; Oldham & Cummings, 1996). While there are many possible personal characteristics that might be relevant, researchers have focused primarily on the influence of personality and cognitive style (Shalley et al., 2004). For example, it is well established that the Big Five dimension Openness to Experience is related to creativity (e.g., Feist, 1998, 1999). Also, an innovative cognitive style has been found to be beneficial (e.g., Kirton, 1994; Tierney et al., 1999). In this chapter, we extend the creativity literature by adding a new antecedent: the tendency to use time management. It has been theorized, but not empirically tested, that time management skills and the ability to manage and organize work processes contribute to creativity (Britton & Glynn, 1989; Isaksen & Treffinger, 2004). In addition, time management might help to reduce fluctuations in creative performance from one day to another, in other words, time management may lead to a more stable creative performance. This study set out to investigate whether the tendency to use time management in creative work provides a unique contribution to creative performance, over and above an individual's innovative cognitive style. Furthermore, we studied the possible mediating effect of positive affect between time management and creativity. Past research shows that, generally, positive affect is strongly related to creativity (Barsade & Gibson, 2007; Davis, 2007). We refer to selfregulation theory to argue that time management may lead to higher levels of creativity through higher levels of positive affect. A diary study was conducted to find support for our hypotheses. Time Management: Self-Regulation with a Focus on Time Use and Workload Self-regulation encompasses efforts of individuals to regulate their own emotions, actions, goals, and task performance (Carver, 2004; Vancouver & Day, 2005; Vohs & Baumeister, 2004). Time management refers to self-management or self-regulation processes that are specifically focused on time use and workload (Claessens et al., 2007). Control Theory of Self-Regulation (Carver & Scheier, 1982) suggests that for task pursuit it is required that 52

61 Managing one s creativity? individuals regulate their task execution behaviors through cycles of feedback loops. In this theory, feedback refers to information about task progress. A feedback loop starts with deciding that a certain task needs to be completed. The next step is to plan required activities. Then, these activities are executed. Monitoring task progress completes the feedback loop and makes it possible to adjust plans or behaviors (Carver & Scheier, 1998). Similar to self-regulation, time management functions through series of feedback loops. The difference between self-regulation and time management is its concern for time use: The purpose is not only to decide what activities to engage in, but also to allocate time to them, and not only to monitor the degree of progress that has been made, but also how much time has been used, and how much work is still to be done before a deadline. We define time management as setting and prioritizing goals, planning tasks, and monitoring progress (Peeters & Rutte, 2005) to actively manage and control multiple tasks within the available amount of time (Claessens et al., 2007). Most self-regulation literature has focused on reaching single goals, such as the well known goal setting literature (Locke & Latham, 1990, 2002), and it has only recently been recognized that individuals must often engage in multiple self-regulation processes simultaneously to juggle multiple tasks and goals (Vohs & Baumeister, 2004). The total workload of an individual in a work setting consists of a complex combination of several tasks and goals (Louro et al., 2007; Schmidt & DeShon, 2007). This means that, in addition to the planning activities for each individual task, efforts need to be made to manage this multi-task situation. Besides planning and monitoring separate work tasks, time management entails managing multiple tasks and reaching multiple goals within a limited amount of working hours (Claessens et al., 2007; Schriber & Gutek, 1987). Time management activities, such as making "to do" lists, help to get an overview of the tasks that need to be executed. This makes it possible to foresee workload problems due to, for example, conflicting deadlines, and to act upon them. Other time management activities, such as prioritizing tasks to their importance and urgency, are needed to decide which tasks to do first and which tasks to postpone, and to allocate work time to them accordingly. Together, these time management activities make it possible to distribute work activities over time thoughtfully and deliberately, or in other words, to regulate workload. 53

62 Chapter 4 In the case of workload regulation, the goal in the self-regulation feedback loop is to reach an optimal level of workload. Time management activities can be used to influence workload pressure and can be considered as the goal-directed behaviors to reach this optimal level. To what extent the optimal level of workload is really met can be monitored through the experienced level of workload. If the experienced workload pressure is too high, decisions with regard to time use can be made: Time may need to be allocated differently, and tasks may need to be rescheduled or postponed. The idea of workload regulation has been described in Activation Theory (Gardner, 1986; Gardner & Cummings, 1988). This theory suggests (a) that there is a certain optimal level of workload (neither too high or too low) at which individuals perform best, and (b) that individuals try to influence their workload towards the level at which they perform best. The latter concerns the selfregulation mechanism that helps to perform at an optimal level of activation. The underlying argumentation refers to the neural activities in the brain and spinal cord, and is based on the regulatory capability of the human brain to strive for an optimal level of activation for the cerebral cortex (Gardner & Cummings, 1988, p. 83). Although this specific self-regulation mechanism occurs unconsciously, Gardner and Cummings argued that these principles also hold for purposeful behavior. Thus, Activation Theory suggests that individuals actively and deliberately search for methods to reduce workload pressure if the workload they experience is too high. Since research has shown that time management leads to feelings of control over time and a decrease in work strain (e.g., Claessens, Van Eerde, Rutte, & Roe, 2004; Jex & Elacqua, 1999; Macan, 1994; Peeters & Rutte, 2005), time management might be a useful method to control and regulate workload as suggested by this theory. Time Management and Creativity Does the time management of creative work require actions other than the time management of any other type of work? By definition, creative tasks concern novel, illdefined problems (Cropley, 1999). This means that information is lacking with regard to what exactly needs to be the end product, and how it should be produced (Britton & Glynn, 1989). For certain time management activities, and for planning in particular, ill-defined problems present a dilemma (Ormerod, 2005): How to plan ahead if it is unclear what 54

63 Managing one s creativity? needs to be done? Task uncertainty makes detailed planning difficult. On the other hand, self-regulating activities, such as planning and monitoring, may be necessary and helpful for goal attainment, also for creative tasks and goals (Isaksen & Treffinger, 2004; Ormerod, 2005). We argue that creative work may even require a relatively high level of selfmonitoring and adjustment. As it is difficult to make detailed work plans beforehand, frequently adjusting plans along the way may be particularly important. Repeating the selfregulation feedback loop often might be necessary to guide the creative process and to obtain reassurance that the activities one engages in are indeed contributing to the completion of the creative task. Unfortunately, empirical evidence of the usefulness of the self-regulation feedback loop for creative performance is scarce. We know of only two studies that investigated part of it, and these were focused specifically on planning. Osburn and Mumford (2006) conducted an experiment among 174 undergraduate students in which planning was manipulated through a training program preceding the creative task. A control group received alternative training. The results showed that planning activities supported the refinement and implementation of new ideas. The second study concerns early work by Eindhoven and Vinacke (1952) and showed that experienced artists planned their creative work to a higher extent than novices. Indirect evidence of the role of self-regulation was found in a qualitative study among 151 undergraduates (Ruscio, Whitney, & Amabile, 1998). In this study, students engaged in creative tasks and were asked to think aloud while working on it. The end results were judged on creativity and their behavior was videotaped. It was found that, among other things, during their creative work students made goal statements and engaged in planning activities, which both were positively correlated to the level of judged creativity. Although the support is limited, the results are a first indication that planning and other time management activities may be beneficial for creative performance. This is in line with Britton and Glynn (1989), who pointed out the need for time management in creative professions, but also the difficulties because of unpredictability. In addition to the positive effects that time management activities might have on the execution and completion of separate creative goals and tasks, time management may contribute to creativity because it may help to control the multi-task work situation, and 55

64 Chapter 4 therefore the workload pressure. Workload pressure is as evident in creative professions as in any other profession (Elsbach & Hargadon, 2006). Following Activation Theory (Gardner, 1986; Gardner & Cummings, 1988), it can be argued that time management may be useful for individuals in creative professions to reach a certain optimal level of workload pressure at which they perform best. The idea that individuals are most creative at a certain optimal level of workload has found empirical support. Ohly and Pluntke (2006) collected survey data from 278 employees from a high-tech company. Their results showed an inverted U-shaped relationship between workload pressure and creativity and also between workload pressure and innovation. These results suggest that creativity is low when workload pressure is too low or too high, and that creativity reaches an optimum somewhere in between. Similar results were found by Janssen (2001) in a survey among 99 employees from an industrial organization in the food sector, and by Baer and Oldham (2006) in a survey study among 170 employees of a large organization, although the last study revealed an inverted U-shaped relationship only for employees who found support for their creative efforts from supervisors or coworkers. However, these studies did not investigate how this optimal level of workload could be reached. We suggest that time management might help to regulate the amount of workload pressure, and might therefore help individuals to perform at an optimal level during their workdays, with the result of higher creative outputs. In sum, we hypothesize that individuals who have a tendency to use time management are more creative during their working days, because (a) they are able to guide their creative processes to a higher extent and (b) they have more control over their workload pressure. Hypothesis 1: Time management tendency is positively related to daily creativity. Time Management and Fluctuations in Creativity over Time Furthermore, we argue that time management may lead to a more stable creative performance. Before we elaborate on the way in which time management may lead to a decrease in fluctuations in performance, we will first answer the question: What is the benefit of a more stable creative performance? The answer lies in the predictability of work processes and feelings of being in control of them. Creative tasks entail a relatively high 56

65 Managing one s creativity? level of task uncertainty (Cropley, 1999), and being in control of the process is therefore more difficult (Van Oorschot, Bertrand, & Rutte, 2005). Notwithstanding, it might make a difference to what extent creative performance fluctuates: If an individual is more or less equally creative on separate working days, it may be easier to make assumptions about the progress that will be made in the near future and, in turn, to guide the creative process. In contrast, if creative performance fluctuates heavily from one day to another, even the near future will be highly uncertain and the level of control over the process might be minimal. Having the feeling of not being in control may evoke work strain or job induced tensions (Claessens et al., 2004; Macan, 1994), which, in turn, may frustrate creativity. Thus, although creative tasks comprise uncertainty, it may still be beneficial to strive for a more stable level of creativity over time. We suggest that time management tendency may lead to less fluctuation. Our reasoning is that, if time management can be used as a self-regulation instrument to achieve a higher level of creativity at a certain moment in time, it may also be used to maintain this optimal level over a longer period of time. This may lead to a more stable level of creative performance and a decrease in fluctuations (or variation). Therefore, we hypothesize that the tendency to use time management reduces variation in daily creativity. Hypothesis 2: Time management tendency is negatively related to variation in daily creativity. Time Management, Positive Affect, and Creativity We are also interested in the processes through which time management might influence creative performance. In this chapter we argue that positive affect may at least partly mediate the effect of time management on creativity, because positive affect plays an important role in the self-regulation feedback loop: It carries the signal of the experienced distance between expected and experienced task progress (Carver, 2004, 2006; Carver & Scheier, 1998). Thus, if an individual perceives that task progress is less than expected, this is felt through negative affect, and if progress is as expected or above expectations, this is felt through positive affect. Research supports this mediating role of positive affect in selfregulation processes (e.g., Ilies & Judge, 2005). Transposed to the self-regulation of 57

66 Chapter 4 creative performance, this suggests that negative affect is closely related to poor creative performance, and positive affect to satisfactory creative performance. A recent metaanalysis (Davis, 2007) showed that affect, specifically positive affect, contributes to creative performance, which makes its mediating role between time management and creativity more likely. The argumentation above, that positive affect may play a mediating role in the self-regulation process to reach a single creative task, may also hold for the self-regulation of workload: If an individual experiences a low or optimal level of workload, this might be felt through positive affect, and if the experienced workload pressure is too high, this might be felt through feelings of negative affect. Past research confirmed that high levels of workload lead to negative affect, anxiety, and emotional exhaustion (Harris & Daniels, 2005; Teuchmann, Totterdell, & Parker, 1999; Totterdell, Wood, & Wall, 2006). Therefore, it may be possible that when creative professionals suffer from high workload pressures, not only the workload pressure itself, but also the related feelings influence their creative performance. Similarly, when they perceive to work under an optimal level of workload pressure, this may be experienced through positive affect, which may in turn contribute to a higher level of creativity. Therefore, we hypothesize that if time management leads to an increase in creativity, this may, at least partly, be due to the effect of time management on positive affect. Hypothesis 3: Daily positive affect mediates the relationship between time management tendency and daily creativity. Method We chose to use a diary study to be able to study within-person variations in creativity. Although diary studies have not been used extensively in work psychology, they have proven to be useful in the analysis of daily positive affect and creativity (Amabile, Barsade, Mueller, & Staw, 2005). Also, diary studies are particularly helpful to study the effect of personal characteristics on daily processes (Van Eerde, Holman, & Totterdell, 2005). Therefore, a diary method seemed to be suitable to study the influence of the tendency to use time management (person-level variable) on daily creativity and positive affect. 58

67 Managing one s creativity? Sample The sample comprised 68 junior research and development engineers. At the time of our study they were all full-time employed and took part in a postmaster educational program. Specifically, they were employed to perform individual, technical, in-company design projects as their final assignments of the postmaster educational program. On average, the projects lasted nine months, and the engineers had a meeting with a university supervisor once every month. They worked full-time in companies for the remainder of the time. The projects were in one of eight disciplines; (a) process and product design, (b) software technology, (c) information and communication technology, (d) design and technology for instrumentation (e.g., medical equipment), (e) design of management systems in architecture, (f) mathematical solutions for industry, and (g) user system interaction design. Thus, our sample consisted of junior R&D engineers working on specialized design projects in various disciplines. These engineers were "professional creatives" because they "are presented with a problem (the creative test) and have external demands placed on them to engage in creativity" (Unsworth, 2001, p. 291). The engineers were recruited through the postmaster educational program. At the time of measurement, 84 engineers were working on their individual design projects, and all were asked to participate. In the invitation it was stressed that the management of the educational program recommended participation. Furthermore, in return for their efforts, individual feedback reports would be sent to each of the participants after the study. A total of 68 engineers (46 male, 26 female) participated in the study; a relatively high response rate of 81%. Typical of this sample of participants was the variety in nationality; it included engineers from The Netherlands (18), Poland (3), Belgium (1), Finland (1), Ireland (1), Greece (10), Italy (1), Portugal (1), Serbia (2), Romania (2), Russia (1), Indonesia (6), Pakistan (3), China (9), India (3), Thailand (1), Malaysia (1), Bangladesh (1), Nigeria (1), Surinam (1), and Peru (1). Thus, about 26% of the participants came from the Netherlands, about 34% from other European countries, about 35% from Asia, about 2% from Africa, and about 3% from South America. All were working in companies in The Netherlands. Their mean age was 26.7 years (SD = 1.8). 59

68 Chapter 4 Measures Data were gathered with an online questionnaire and online daily surveys. The daily surveys started after the participants had completed the questionnaire. Both the questionnaire and the daily surveys were in English. All scales were scored on a 5-point Likert scale ranging from highly disagree (1) to highly agree (5). Questionnaire Data The questionnaire assessed three control variables (age, gender, and innovative style), and included a scale to measure the tendency for time management. Innovative style. We used the 4-item creativity subscale of the cognitive style measure of Miron, Erez, and Naveh (2004), which assesses innovative style as a personal characteristic. This subscale is a short version of the innovation style subscale of the Adaption Innovation Inventory developed by Kirton (1976; 1994). The items were: "I have a lot of creative ideas", "I prefer tasks that enable me to think creatively", "I am innovative", and "I like to do things in an original way". Cronbach's alpha was.82. Time management tendency. To measure the tendency for time management, we used a short version of the 33-item scale developed by Macan (1994). This 10-item version had shown sufficient reliability in previous research (Peeters & Rutte, 2005). Example items are: "I review my activities", "I complete my high priority tasks first", and "I feel in control of my time". Cronbach's alpha was.81. Daily Survey Data The daily surveys assessed creativity and positive affect. Participants completed the surveys over a period of 10 successive workdays. The surveys were sent out each afternoon, and participants were asked to fill them out at the end of their workday, before they left work. Participants notified to be out of office during, on average, 1.12 days of the 10 days of measurement. On average, they filled in the daily surveys on 6.13 days of the remaining 8.88 days they were able to participate, which corresponded with a response rate of 69% for the diary data. Daily creativity. For measuring creativity on a specific day, we adapted the 4-item creativity subscale of Miron, Erez, and Naveh (2004). The items were: Today during my 60

69 Managing one s creativity? workday, "I had a lot of creative ideas", "I was able to think creatively", "I was innovative", and "I did things in an original way". Cronbach's alphas were calculated for each of the days of measurement, and ranged between.78 and.89 (M =.84). To verify that we measured creativity on a daily level, we added one item measuring daily insight: "Did you have a 'Eureka!' moment regarding your work since the beginning of the workday? In other words: Did you have a novel and useful idea that came to mind suddenly and unexpectedly?", to be answered with yes (1) or no (2). An insight is an observable event and an outcome of creative problem-solving processes (Sternberg & Davidson, 1999). Our creativity scale highly correlated with the insight measure (r =.48, p =.001), providing support for the concurrent validity of this scale. Variation in daily creativity. Participants answered the daily creativity measure on multiple days. Therefore, variations in daily creativity could be calculated per person. Specifically, we calculated per person the deviation from the within-person mean level of creativity over the days that a participant had filled in the daily surveys. In other words, variation was calculated around the person mean. These calculations resulted in a personlevel variable indicating the amount of fluctuation in daily creativity. Daily positive affect. We used the diary measure for daily positive affect of Amabile, Barsade, Mueller, and Staw (2005). Of the original six items, two referred to teamwork. Since our participants worked on individual design projects, one of these items was adapted to the individual level, and the other was excluded. The remaining items were: "Today, I was happy", "I am satisfied about my work of today", "I enjoyed my work", "I am frustrated" (reversed scored), and "I feel positive about my project". Cronbach's alphas were calculated for each of the days of measurement, and ranged between.76 and.87 (M =.81). Data Analysis Procedure For each participant, data were available on two levels: At the person level (age, gender, innovative style, time management tendency, and variation in daily creativity), and at the day level (daily creativity and daily positive affect). For analyses between person-level variables (e.g., between time management tendency and variation in daily creativity), linear regression analyses were used. For analyses between person-level and day-level variables 61

70 Chapter 4 (e.g., between time management tendency and daily creativity), or between day-level variables (e.g., between daily positive affect and daily creativity), multilevel analyses were used, which is also known as hierarchical linear modeling (Bryk & Raudenbush, 2002). Since multilevel analysis takes into account the dependence of the day-level measurements within the person-level data, this method is the most appropriate for these types of data (Snijders & Bosker, 1999). As it is necessary to analyze day-specific effects (Hofmann & Gavin, 1998), independent day-level variables were centered around the person mean, and independent person-level variables around the grand mean. Results Means, standard deviations, and zero-order correlations among all variables can be found in Table 4.1. When appropriate, both day-level and person-level correlations are displayed. To correlate person-level variables with daily positive affect and daily creativity, the day-level variables were averaged across the 10 days of measurement. We performed a number of multilevel analyses to test our hypotheses concerning day-level variables. This resulted in a number of nested models, starting from null models that included only the intercepts as a predictor, followed by models of increasing complexity, including control variables, and both person-level and day-level predictors. The improvement of one model over the next was tested with a chi-square test, comparing the two likelihood ratios (the degrees of freedom are equal to the number of added variables). Control variables were all variables at the person level: age, gender, and innovative style. The other person-level variables were time management tendency and variation in daily creativity. Day-level variables were daily creativity and daily positive affect. Table 4.2 provides the results for daily creativity as the dependent variable, showing estimates, standard errors, and t values for all predictor variables in the models. Also, likelihood ratios and the differences between these ratios are displayed. Finally, daylevel and person-level intercept variances are shown. In the null model, the intercept variances of the day level and person level revealed that the variance in daily creativity was partly predicted by day-level variation (p <.001), and partly by person-level variation (p <.001). Model 1 included the control variables (age, gender, and innovative style). This model showed a significant improvement over the null model ( -2 log = 27.59, df = 3, p 62

71 Managing one s creativity? Table 4.1. Means, Standard Deviations, and Zero-Order Correlations Between Study Variables Variable M SD Age Gender Innovative style Time management tendency Daily creativity ***.42*** - 6. Variation in daily creativity *.29* Daily positive affect * a.34*** Note. Correlations are on the person level (N = 68), with day-level data averaged across 10 days, except for the correlation marked with a, which is on the day level (n = 417). Gender was coded as (1) male and (2) female. p <.05 (one-tailed), * p <.05, ** p <.01, *** p <

72 Chapter 4 Table 4.2. Multilevel Estimates for Models Predicting Daily Creativity Model and variable Estimate SE t -2 log -2 log (df) Day-level intercept variance (SE) Person-level intercept variance (SE) Null model *** (.02).19*** (.04) Intercept Model *** (3).32*** (.02).13*** (.03) Intercept Age Gender Innovative style *** Model *** (1).33*** (.03).10*** (.03) Intercept Age Gender Innovative style *** Time management tendency *** Note. n = 417. * p <.05, *** p <.001. Variables in italics were entered in an earlier model. 64

73 Managing one s creativity? Table 4.2. Multilevel Estimates for Models Predicting Daily Creativity (Continued) Model and variable Estimate SE t -2 log -2 log (df) Day-level intercept variance (SE) Person-level intercept variance (SE) Model *** (1).27*** (.02).10*** (.03) Intercept Age Gender Innovative style *** Time management tendency *** Daily positive affect *** Note. n = 417. *** p <.001. Variables in italics were entered in an earlier model. 65

74 Chapter 4 <.001), with only innovative style as a predictor of daily creativity: Individuals with a more innovative style were more creative on a daily level. Hypothesis 1 stated that time management tendency would be positively related to daily creativity. To test this hypothesis, we added time management tendency to Model 1, resulting in Model 2. Model 2 showed further improvement over Model 1 ( -2 log = 23.93, df = 1, p <.001). The estimate of time management tendency was highly significant, indicating that time management provides a unique contribution to daily creativity. Thus, the data supported Hypothesis 1. Because we controlled for innovative style, these results suggest that the tendency to manage one's time carefully helps to be more creative on a daily level, irrespective of a person's innovative style. Hypothesis 2 predicted that time management tendency would lead to a decrease in variation in daily creativity. Because time management tendency and variation in daily creativity are both variables on the person level, this hypothesis could be tested without multilevel analyses. Table 4.3 presents the results of the hierarchical regression analysis. In Model 1 we introduced the control variables into the regression equation. The results show that there was a gender effect: Women varied less in their daily creativity levels than men. Also, we found that innovative style predicted variation in daily creativity: The level of creativity fluctuated to a higher extent for individuals with a more innovative style. In Model 2 time management tendency was entered. Model 2 showed a significant improvement over Model 1 ( R 2 =.08, df = 1, p <.01). We found a negative effect of time management tendency on variation in daily creativity, suggesting that individuals who have the tendency to use time management have a more stable creative performance from one day to another. Thus, Hypothesis 2 was supported. Hypothesis 3 predicted a mediator effect of daily positive affect between time management and daily creativity. To test this mediator effect we followed the procedure suggested by Mathieu and Taylor (2007). This procedure is similar to the well known procedure of Baron and Kenny (1986), but specifically developed for testing mediation effects in multilevel datasets. According to this procedure, mediation only occurs if four conditions are met: (a) The independent variable (time management tendency) should be related to the mediator variable (daily positive affect); (b) the independent variable (time management tendency) should be related to the dependent variable (daily creativity); (c) the 66

75 Managing one s creativity? Table 4.3. Hierarchical Regression Results for Variation in Daily Creativity Model and variable Beta SE t R 2 R 2 Model 1.20** Age Gender * Innovative style * Model 2.28***.08** Age Gender * Innovative style ** Time management tendency * Note. N = 68. Gender was coded as (1) male and (2) female. Variables in italics were entered in an earlier model. * p <.05, ** p <.01, *** p <.001. mediator variable (daily positive affect) should be related to the dependent variable (daily creativity); (d) if both the independent variable and the mediator variables are entered in one model predicting the dependent variable, the relationship between the independent variable (time management tendency) and the dependent variable (daily creativity) should disappear for full mediation, and, if not, a Sobel test can reveal partial mediation or an indirect effect. Following these steps, we first tested whether time management tendency was positively related to daily positive affect. Table 4.4 shows the results for the multilevel analyses with daily positive affect as the dependent variable. In the null model, the personlevel intercept variance confirmed that the variance in daily positive affect was partly predicted by person-level variation (p <.01), although most variance was predicted by daylevel variation (p <.001). Model 1, including the control variables, showed a significant improvement over the null model ( -2 log = 13.06, df = 3, p <.001), but none of the control variables in themselves were significant predictors of daily positive affect. In Model 2 time management tendency was entered. Model 2 showed further improvement over Model 1 ( -2 log = 23.40, df = 1, p <.001). Time management tendency was a 67

76 Chapter 4 Table 4.4. Multilevel Estimates for Models Predicting Daily Positive Affect Model and variable Estimate SE t -2 log -2 log (df) Day-level intercept variance (SE) Person-level intercept variance (SE) Null model *** (.03).07** (.02) Intercept Model *** (3).37***(.03).06** (.02) Intercept Age Gender Innovative style Model *** (1).37*** (.03).05*(.02) Intercept Age Gender Innovative style Time management tendency ** Note. n = 417. ** p <.01, *** p <.001. Variables in italics were entered in an earlier model. 68

77 Managing one s creativity? significant, positive predictor of daily positive affect, suggesting that time management positively influences daily well-being: Individuals who tend to use time management feel better during their workdays. This means that the first condition for mediation was met. The second condition, that time management tendency was related to daily creativity, had already been tested in Hypothesis 1. As the results show, this condition was also met. Next, we tested the third condition: Whether or not daily positive affect was positively related to daily creativity. For this, daily positive affect was entered into the regression equation in the multilevel analyses predicting daily creativity (see Table 4.2) in Model 3. Model 3 showed further improvement over Model 2 ( -2 log = 77.14, df = 1, p <.001), and the estimate for daily positive affect was highly significant. These results suggest that individuals who feel better during a workday are more creative on that day. Thus, the third condition for mediation was met. Finally, we tested the fourth condition. When we take a second look at the results in Table 4.2, it shows that time management tendency remained to be a significant predictor of daily creativity, even after daily positive affect was entered into the model. Specifically, it shows that it did not only remain to be a significant predictor, it did not decrease in its predictive value. This means that we did not find full or partial mediation. However, a Sobel test showed an indirect effect of time management tendency on daily creativity via daily positive affect (Sobel = 2.75, p <.01). Specifically, our analyses show that no mediation exists, but, instead, that the direct effect and the indirect complement each other. The indirect effect provides support for Hypothesis 3: Time management tendency influences daily positive affect, and in turn, daily creativity. Discussion The results of this diary study suggest that time management has a positive effect on positive affect and daily creativity, and that it leads to a more stable level of creative performance. In addition, the results show an indirect effect of time management on daily creativity via daily positive affect, indicating that individuals who tend to use time management feel better during their workdays, which helps them to be more creative. In the analyses, we included innovative style as a control variable, which was found to be a significant predictor of daily creativity. Since we controlled for innovative 69

78 Chapter 4 style, our results imply that the tendency to manage one's work time carefully helps to be more creative, in addition to the positive effects of having an innovative style. Zero-order correlations show no relation between innovative style and time management tendency, suggesting that these are two independent factors that both add to a person's daily creative performance. Also, we found that the level of creativity of individuals with a more innovative style was not only higher on average, but also fluctuated relatively much over their work days. This is in contrast with the effect of time management, which lead to a higher, more stable level of creativity. This diary study provides unique contributions to the literature on creativity. First, it broadens the scope of personal characteristics that have been researched. In their review of personal and contextual factors that influence creativity, Shalley et al. (2004) urged researchers in the field of creativity to expand the range of personal characteristics under research in order to gain a better understanding of the antecedents of creativity (p. 953). In addition to the well studied influence of personality and cognitive style, our study shows that the tendency to use time management is a new and distinct predictor of creativity. We provide empirical evidence where theoretical considerations had been made predominantly (Britton & Glynn, 1989; Isaksen & Treffinger, 2004). Second, our research approach, in which we collected data through multiple daily surveys, allowed us to study another understudied aspect of creative performance in organizations: Fluctuations in creativity. Whereas most researchers use the mean performance as a dependent variable, using variations in performance as a dependent variable often provides complementary insights into organizational behavior (c.f., Hofmann, Jacobs, & Baratta, 1993; Hofmann, Jacobs, & Gerras, 1992; Stewart & Nandkeolyar, 2007). Thus, it is interesting to study factors that predict a high mean level of creativity, but it might also be interesting to investigate to what extent creative performance fluctuates over time, and to study the factors that inhibit or encourage these fluctuations. In this diary study we have showed that individuals differ in the extent to which their creativity levels fluctuate over time, and that the creativity level of individuals who have the tendency to use time management is more stable, and the creativity level of individuals with an innovative style is less stable. 70

79 Managing one s creativity? Third, our study contributes to the discussion of the role of positive affect in creative performance (Davis, 2007; James, Brodersen, & Eisenberg, 2004), and confirms that positive affect is substantially related to daily creativity. Specifically, this study shows that, in addition to the positive direct effect that time management has on daily creativity, time management contributes to daily creativity indirectly via its effect on daily positive affect. We failed to find a mediating effect, though. This means that, instead of the mediator (positive affect) explaining the underlying mechanism of the direct effect, the direct and indirect effects of time management on daily creativity are complementary. Theoretically, this means that our argumentation that affect is part of the self-regulation feedback loop, and therefore a mediator, does not hold. Instead, the direct relationship may be explained by Activation Theory (Gardner, 1986; Gardner & Cummings, 1988). This theory states that the human brain functions best at an optimal level of activation. Since being creative is a cognitive act, an optimal level of activation may have a direct impact on creative performance. This may be the reason why time management has a direct effect on creativity: Time management may help individuals to influence their workload to attain an optimal level of activation for cognitive work. A spin-off of this cognitive mechanism may be positive affect, because this optimal level of activation may be the most pleasant state to work in. In turn, positive affect is related to being creative. Thus, the direct effect of time management on daily creativity may be explained by the cognitive mechanism itself, and the indirect effect by the spin-off of this mechanism: positive affect. Another possibility may be that the direct effect may be explained by other mediators than positive affect, for example, by self-efficacy. Self-efficacy concerns an individual's self-confidence that they can execute the activities that are required to complete a certain task (Bandura, 1977). Self-efficacy is particularly important when one encounters difficulties during task execution, because it induces persistence and coping behaviors. Time management may lead to a heightened level of self-efficacy, because time management behaviors provide feelings of being in control of time (Claessens et al., 2007). These feelings of control may be related to being more confident in case of difficulties, and thus to self-efficacy. In turn, self-efficacy has been shown to be related to creative 71

80 Chapter 4 performance (e.g., Tierney & Farmer, 2002, 2004). Thus, self-efficacy may also explain why time management has a direct effect on creativity. As in most field research, our diary study is not without its limitations. The first concern is that our research design did not allow for testing causal relationships. A strong point is that the tendency for time management was measured at an earlier moment in time than positive affect and daily creativity. This does not prove that time management leads to a higher level of positive affect and creativity, but makes it less likely that the causal relationship is in the opposite direction. Positive affect and creativity were measured at the same moments in time, though. This means that it is probable that positive affect predicts creativity, but that we cannot exclude the possibility that creativity predicts positive affect, or that the two occur simultaneously (Amabile et al., 2005). What we can conclude is that individuals who have the tendency to use time management are more creative and higher in positive affect, and that daily creativity and positive affect are closely related to each other. Second, the characteristics of our sample, such as our participants' ages, gender, nationalities, and professions, imply a limited generalizability of our results. The most important consideration is that our participants were young professionals at the very start of their careers. Nevertheless, a strong point of the sample was that it consisted of persons of various nationalities, with about two thirds of them coming from different East and West European countries, and about one third from Asian countries. Due to this broad sample, generalizing our results to other Western or Eastern countries may be less of a problem. Furthermore, although we have investigated only one specific group of professionals, our choice for R&D engineers was intentional. These engineers are a typical example of creative professionals (Unsworth, 2001): Being creative is one of the requirements of their job. A third point of concern is the use of self-rated measures. Preferably, additional data should have been collected from other sources, such as performance ratings from supervisors or coworkers. However, using a diary method provides considerable practical restrictions on collecting these. A major concern is the extent to which supervisors or coworkers have insight in the daily performance of the participants. Supervisors in particular, who may not work with the participants on a day-to-day basis, but rather meet them once or twice a week. The same may hold for coworkers, albeit to a lesser extent: 72

81 Managing one s creativity? Coworkers may share tasks with participants during only part of their working days. Our participants worked on individual design projects and, therefore, did not work very closely with others. For that reason, it was not possible to find other sources that had enough insight to judge their daily creativity. The practical restrictions to collect day-level data from supervisors or coworkers makes it very common in diary studies to use self-rated measures (c.f., Sonnentag & Zijlstra, 2006; Totterdell et al., 2006). The general notion is that the advantages of new insights through diary studies make up for the limitation of collecting data via self-rated measures. Suggestions for Further Research In future research, it would be interesting to disentangle the process of time management to be able to explain its effect on creative performance in more detail. Time management is a concept that covers many different aspects of the process of self-regulation of time use and workload (Claessens et al., 2007; Macan, Shahani, Dipboye, & Phillips, 1990; Schriber & Gutek, 1987). To gain a better understanding of the processes through which time management influences creativity, it might be valuable to distinguish between the different aspects of time management in future research. For example, it would be interesting to investigate whether its effect is due to (a) the management of multiple tasks within the limited amount of work time (the regulation of workload), (b) the process of going through self-regulation feedback loops for individual tasks separately, or (c) a combination. Also, it would be interesting to learn more about the cycle of goal setting, planning, task execution, monitoring and rescheduling in creative work. Because of high task uncertainty, creative tasks may need to be regulated differently from other tasks: Planning may be more difficult (Ormerod, 2005), and monitoring task progress and rescheduling may therefore be particularly important. In line with Shalley et al. (2004) we argue that, in general, future research should focus on a wider variety of antecedents of creative performance. The personal characteristics that have been examined in past research (predominantly the role of cognitive style and personality) seem to have been selected on their presumed effects on the cognitive processes leading to creativity. However, "both cognitive and motivational factors are involved in creative performance" (Hayes, 1989, p. 143), and intrinsic motivation has 73

82 Chapter 4 been found to be an important predictor of creativity in particular (e.g., Amabile, Hill, Hennessey, & Tighe, 1994; Dewett, 2007; Tierney et al., 1999). Therefore, future research might focus more on personal characteristics that have a positive effect on volition or motivation. Since time management taps into the volitional processes of self-regulation, this study may be considered a first move in that direction. In light of our results, and because time plays an important role in creativity (Mainemelis, 2002; Runco, 1999), it would also be interesting to elaborate more on personal characteristics that determine how individuals experience and organize their time at work. For example, it might be valuable to study the effect of individual differences in time urgency (i.e., the need to have control over deadlines and the feeling of being hurried) (Landy, Rastegary, Thayer, & Colvin, 1991). Another interesting concept is polychronicity, or the extent to which individuals prefer to be engaged in more than one task at the same time (Madjar & Oldham, 2006). Finally, pacing preferences, preferences with regard to the way in which workload is distributed over time (Gevers et al., 2006), might be an interesting new research topic in creativity research. Finally, it would be interesting to further investigate the fluctuations in creativity. To our knowledge, our study was the first to investigate fluctuations in creativity over time. It raises new research questions regarding creative processes during the course of creative projects. For example, our data suggest that individuals with a more innovative style have relatively high, but fluctuating creativity levels. Individuals that have a tendency to manage their time carefully also have a relatively high creativity level, but this level fluctuates little. Some individuals may have both an innovative style and a tendency to use time management. Their creativity levels may be very high and relatively stable. Time management may help them to go beyond other creative individuals, which may make them the more successful creative professions. More research is needed to confirm these premises. Theoretical and Practical Implications Our results have relevant theoretical implications. First, this study provides new insights for Activation Theory in the field of creativity research. With respect to the two principles of this theory ((a) creative performance peaks at a certain optimal level of workload; and (b) 74

83 Managing one s creativity? individuals actively try to reach this optimal level), empirical evidence has only been provided for the first principle (e.g., Janssen, 2001; Ohly et al., 2006). Up to now, no research efforts had been made to support the second principle. Specifically, the question how workload can be influenced in order to reach an optimal level of workload had remained unanswered. Our study provides one possible solution: time management. In fact, the results of this study not only suggest that time management can be used as an instrument to reach a higher level of creativity, but also indicate that it helps to stay in control to increase stability in performance over time. Therefore, in addition to providing some empirical ground for the second principle of Activation Theory, we suggest this principle may be extended to: Individuals actively try to reach and maintain an optimal level of workload. Second, our results imply that self-regulation theory applies to creative work. This was not self-evident because self-regulation, and planning in particular, requires a certain level of predictability in task execution, something creative tasks lack. Nonetheless, this field study suggests that, despite the unpredictable nature of creative work, self-regulation activities are helpful to creative performance. The results of this study also have clear implications for practice. Our results show that time management tendency, positive affect, and daily creativity are positively related to each other, suggesting that time management is associated with feeling better and being more creative, even if a person does not have an innovative cognitive style. Past research has shown that, although the tendency to use time management is related to one's personality, time management techniques can be improved through training (Green & Skinner, 2005; Orpen, 1994; Van Eerde, 2003b). This means that it might be worthwhile for companies that employ creative professionals to make their employees aware of the advantages of time management, and to offer training programs to improve their time management skills. Also, our research might have important implications for teaching. Students in, for example, industrial engineering or architecture might benefit from courses in time management, which could possibly be added to their curricula. Thus, also future creative professionals can benefit from the advantages of time management on positive affect and creativity. 75

84

85 Chapter 5 Being Successful in a Creative Profession: The Role of having an Innovative Cognitive Style, Self-Regulation of Design Work, and Self-Efficacy * In a field study among 339 architects we investigated to what extent self-regulation (setting priorities, planning work activities, and monitoring time and task progress) and having an innovative cognitive style contributed to the success of architects as designers and businesspersons, and examined the mediating role of self-efficacy in these relationships. The results suggest that self-regulation is directly related to being a successful businessperson, and indirectly, via self-efficacy, to being a successful designer. Furthermore, they suggest that having an innovative cognitive style is directly related to being a successful designer, and indirectly, via selfefficacy, to being a successful businessperson. The current prosperity of Western economies increasingly depends on knowledge-based work, and on creative work in particular (DeFlillippi et al., 2007; Thompson, Jones, & * This chapter is based on: Beeftink, Van Eerde, Rutte, & Bertrand (2008). Being successful in a creative profession: The role of having an innovative style, self-regulation of design work, and selfefficacy. Manuscript submitted for publication. 77

86 Chapter 5 Warhurst, 2007). Thus, presently, creative professionals, such as designers or architects, comprise a vital part of the work force. Them being successful brings wealth to our societies. Being successful in creative industries requires a multitude of skills and abilities, including the cognitive skills to be creative (Shalley et al., 2004). We argue that, in addition, skills with regard to the business aspects of the work may be important. Despite its relevance, these business aspects have hardly been taken into consideration in past creativity research. The present chapter aims to give a multifaceted view of design work by focusing both on the creative and business aspects in creative professions. Specifically, in a survey study among architects it was investigated to what extent having an innovative cognitive style and using self-regulation behaviors (setting priorities, planning work activities, and monitoring time and task progress) are related to the architects' success as designers and businesspersons, and examined the possible mediating role of self-efficacy in these relationships. In the following, we will first describe in more detail two complementary aspects of design work: being a designer and being a businessperson. Second, we will briefly discuss theory and past research on innovative style, self-regulation, and self-efficacy. Third, we will formulate and test hypotheses regarding the direct effects of innovative style and self-regulation on the success of individuals as designers and businesspersons, and their indirect effects via self-efficacy. Designer and Businessperson Designers have two purposes in their work: Being successful with regard to the designs they produce and being successful in business. Whereas the first puts more emphasis on a high level of creativity in design work, the other puts more emphasis on the management side of the work. Success as designer The very essence of the work of designers is the creative act of designing (Goldschmidt, 1999). Although there are several other relevant activities designers are engaged in during their work days (such as attending meetings, doing administrative tasks, or making phone 78

87 Being succesful in a creative profession calls), designing is their core business. Design processes may vary within and across disciplines with regard to, for example, the number of design phases, the duration of design projects, and the required innovativeness of designs. Nevertheless, all design work requires creative problem solving (Goldschmidt, 1999; Unsworth, 2001). Consequently, designers that have a talent for creative problem solving may be more successful designers. In addition, factors in the work environment, such as the influence of supervisors and coworkers, may inhibit or support the creativity of designers (Shalley et al., 2004). Success as businessperson Being successful in business in knowledge-based work incorporates managing the quality, efficiency, and timeliness of business processes (Wickramasinghe & Von Lubitz, 2007). With regard to design work, this means that not only should efforts be put into the design process itself, but also into the management and control of these processes. Of course, the business aspect of design work is particularly important when running a business, but even design work in employment has a business side, which means that this aspect applies to designers in general. If a designer is part of a design team, it is possible that these tasks are partly taken over by a team leader, who sets priorities, plans activities, and monitors task progress for the team (Amabile, Schatzel, Moneta, & Kramer, 2004). However, creative professionals perform worse if team leaders manage and control them too closely: Research has shown that close monitoring of their work activities is detrimental for their creativity (e.g., Amabile et al., 2004; George & Zhou, 2001; Oldham & Cummings, 1996; Zhou, 2003), and instead, team leaders should provide more general support, stimulation, and encouragement (Amabile et al., 2004; Shin & Zhou, 2003). Nevertheless, managing work processes is necessary, also in creative work (Isaksen & Treffinger, 2004). Therefore, it seems important that designers put effort into managing and controlling their work processes themselves. This should help them to perform well on the business aspects of their work. The Influence of Personal Characteristics Although the influence of the work context may also be important, in this chapter we focus on factors on the person-level: on the level of the designer. Person-level factors that 79

88 Chapter 5 influence the success of individuals as designers may not be the same factors that are of influence on their success as businesspersons. This study aims to investigate the effect of three personal characteristics on both aspects of the design work: (a) having an innovative cognitive style, (b) the use of self-regulation behaviors, and (c) the level of self-efficacy. Before we formulate and test hypotheses regarding the effects of these three personal characteristics on being a successful designer and businessperson, we will first briefly discuss theory and past research regarding each of them. Innovative cognitive style One of the personal characteristics that is well known for its positive effect on creativity is having an innovative cognitive style (Shalley et al., 2004). An innovative cognitive style refers to having an orientation to creative problem solving (Kirton, 1976, 1994). It is an individual's preferred way of gathering, processing, and evaluating information (Miron et al., 2004). In other words, it determines the ways in which individuals engage in cognitive processes. Kirton (1976) developed the Kirton Adaptor-Innovator Inventory (KAI theory), which differentiates between individuals with a more adaptive and individuals with a more innovative cognitive style. Whereas adaptors are characterized by doing things 'better' (incremental improvements), innovators try to do things 'differently' (radical improvements) (Kirton, 1994: 9). However, recent research has shown that being innovative and being adaptive are not opposites, but rather two independent aspects of an individual's cognitive style, and that only having an innovative cognitive style is predictive of creative performance (e.g., Miron et al., 2004). Thus, having an orientation towards creative problem solving contributes to creative performance. Self-regulation Recently, awareness has arisen that not only the cognitive processes of creative work itself, but also the meta-cognitions that provide the opportunity to manage and control these processes are highly relevant (Isaksen & Treffinger, 2004, pp ). These metacognitions concern self-regulation behaviors such as planning and monitoring time and task progress (e.g., Peeters, Van Tuijl, Reymen, & Rutte, 2007; Sonnentag, 1998). We refer to Control Theory of Self-Regulation (Carver & Scheier, 1982) to give more insight into self- 80

89 Being succesful in a creative profession regulation and the corresponding self-regulation behaviors. This theory suggests that task pursuit requires keeping track of task progress through repeated cycles of feedback loops. The feedback is gathered by monitoring task execution and consists of information about task progress. A full feedback loop contains the following sequence of behaviors (Carver & Scheier, 1998; see also: Vancouver & Day, 2005): (a) evaluating, prioritizing, and selecting tasks for completion, (b) planning for action, (c) executing the task, and (d) monitoring task progress (after which progress is evaluated and new decisions can be made with regard to task pursuit). The purpose of the feedback loop is to manage and control the execution process and to adjust the course of action if necessary. We argue that self-regulation is important to self-managed tasks, and is particularly important to creative work tasks. Creative tasks involve a high level of task uncertainty (Cropley, 1999). For example, in new product development projects the design activities to be executed are generally not fully known beforehand, and neither is the precise required amount of time to design the new product (Van Oorschot et al., 2005). Therefore, it is difficult to make detailed plans at the beginning of task execution (Britton & Glynn, 1989). Instead, task progress should be monitored, and feedback regarding the progress should be used to make and adjust further plans. Thus, the feedback loop of selfregulation seems to be especially important in creative work, because planning in advance is difficult, and adjusting and making new plans along the way is required. It is therefore to be expected that designers perform better in design work when they have the tendency to regulate their design activities well. However, up to now, little empirical evidence has been provided to support this proposition. Self-efficacy Literature on self-efficacy suggests that not only an individual's ability to accomplish a task predicts whether it will be completed successfully, but also the confidence that the required activities can indeed be executed (Bandura, 1977). This means that a lack of self-efficacy hinders task completion, and that a high level of self-efficacy facilitates it. Specifically, Bandura theorized that self-efficacy is particularly important when encountering problems or difficulties during task execution. In case of a low level of self-efficacy, individuals may not believe that they are able to overcome the problems at hand, and disengage from the 81

90 Chapter 5 task without completing it. A high level of self-efficacy, on the other hand, will induce coping behaviors and persistence in the face of obstacles. Meta-analyses have confirmed the importance of self-efficacy, and have shown that it is related to work outcomes such as job performance (Judge & Bono, 2001; Stajkovic & Luthans, 1998) and job satisfaction (Judge & Bono, 2001). Self-efficacy has also been found to be an important predictor of creative performance (e.g., Jaussi, Randel, & Dionne, 2007; Redmond, Mumford, & Teach, 1993; Tierney & Farmer, 2002). For example, a field study of Tierney and Farmer (2004) among R&D engineers showed that engineers with higher levels of creative self-efficacy were more creative, as judged by their supervisors. Furthermore, this study revealed that the level of self-efficacy of the engineers mediated the effect of their supervisors' expectations about the engineers' creative performance and their actual creative performance. Thus, this study also showed that supervisor behaviors are an antecedent of self-efficacy in creative work, similar to the effect of supervisor behaviors on general occupational self-efficacy (e.g., Schyns & Von Collani, 2002). Up to now, however, little research has been conducted to person-level antecedents of self-efficacy in creative work. This is in contrast with the ideas of Bandura (1977), who theorized that person-level factors are key predictors of self-efficacy. Specifically, he argued that one of the most important predictors of self-efficacy are an individual's experiences of past performance (p. 195). Repeated successful task completion results in mastery experiences and builds a steady level of self-efficacy that is not easily altered by an occasional failure to complete a task. Therefore, in addition to past performance itself, person-level antecedents of self-efficacy most likely are skills and abilities that have been beneficial to past performance, and person-level factors that are of influence on feelings of mastery and control. Innovative Cognitive Style and being a Successful Designer and Businessperson Research has confirmed that individuals with an innovative style are indeed more creative and innovative in their work (Miron et al., 2004; Tierney et al., 1999). Since design work requires creativity (Goldschmidt, 1999), having an orientation towards creative problem 82

91 Being succesful in a creative profession solving may be an advantage for design work. It may thus be expected that designers with a more innovative style are more successful designers. Hypothesis 1: Innovative cognitive style is positively related to being a successful designer. Individuals with an innovative style may also have had more positive past experiences with creative tasks. Specifically, they may more often have had the experience that they could master a task at hand, because their cognitive style was beneficial in these situations. Since the level of self-efficacy of individuals is gradually built through past experiences (Bandura, 1977), designers with a more innovative style may also have higher levels of self-efficacy. In turn, designers with a higher level of self-efficacy may be more successful designers. A higher level of self-efficacy enhances the persistence level and coping efforts of individuals when they encounter challenging situations (Bandura, 1977). To complete a creative task requires sustained efforts (Amabile, 1988; Tierney & Farmer, 2002), and persistence may therefore be beneficial in design work. As a result, a higher level of selfefficacy may be related to being a more successful designer. Combining the proposed effect of innovative style on self-efficacy, and the effect of self-efficacy on being a successful designer, we hypothesize that innovative style not only directly leads to being a more successful designer, but also indirectly via self-efficacy. Hypothesis 2: Innovative cognitive style is indirectly positively related to being a successful designer via self-efficacy (partial mediation). Persistence may be beneficial not only to task completion, but also to the process of striving for a high level of quality or when trying to deliver a design on time. Therefore, selfefficacy may contribute to quality and timeliness, and consequently, to the success of designers as businesspersons. If having an innovative style results in a higher level of self-efficacy, it may also indirectly contribute to success as businessperson, via a higher level of self-efficacy. Thus, 83

92 Chapter 5 we hypothesize an indirect effect of innovative style on success as businessperson, via selfefficacy. We have no grounds to expect that innovative style would directly lead to being a more successful businessperson. For example, in the survey study of Miron et al. (2004), innovative style was related to innovative performance, but not to other performance measures such as quality or efficiency. Thus, having an innovative style does not seem to affect business performance either positively or negatively on aspects other than creative performance. Therefore, it is hypothesized that self-efficacy fully mediates the effect of innovative style on success as businessperson. Hypothesis 3: Innovative cognitive style is indirectly positively related to being a successful businessperson via self-efficacy (full mediation). Self-Regulation and being a Successful Designer and Businessperson Isaksen and Treffinger (2004) acknowledged the importance of self-regulation in creative work. Their article described the evolution of a creative problem solving model that started with the creative problem solving model of Osborn (1952), which focused predominantly on the creative cognitive activities of creative problem solving. It ended with the creative problem solving model of Isaksen, Dorval, and Treffinger (2000), which included creative cognitive processes (e.g., generating ideas) as well as a creative-process-management component that consisted of activities such as "continuous planning, monitoring, managing, and modifying behavior during creative problem solving" (Isaksen & Treffinger, 2004: 92). Although little research has been conducted to investigate the guiding role of selfregulation behavior in creative or design work, some studies have confirmed its importance. A field study among individuals working in interdisciplinary design teams (Peeters et al., 2007), for example, explored relevant design behaviors, and found that both design creation behaviors (such as generating ideas and solutions) and design planning behaviors (such as planning time and keeping and adjusting schedules) were important. A study among software designers (Sonnentag, 1998) revealed that the designers engaged in planning and monitoring behaviors during designing, and that high performing software designers engaged in monitoring behaviors to a larger extent than moderate performers. In line with these two studies, we argue that self-regulation behaviors guide and support creative 84

93 Being succesful in a creative profession problem solving processes. Consequently, we hypothesize that individuals who engage in self-regulation to a larger extent are more successful designers. Hypothesis 4: Self-regulation is positively related to being a successful designer. Self-regulation may also lead to a higher level of self-efficacy. Engaging in self-regulation provides the opportunity to manage and control processes, which may, in turn, lead to the perception of being 'in control'. A study by Claessens et al. (2004) among R&D engineers, for example, showed that the planning behaviors of the engineers led to the perception of being in control of time and to higher levels of job performance and job satisfaction. Perceptions of being able to manage and control work processes may be related to perceptions of being able to overcome possible problems that might come up. In other words, they may be related to self-efficacy. Another field study of Claessens (2004b: ) indeed showed strong relationships between perceived control of time and selfefficacy, and between planning behaviors and self-efficacy. Since we expect self-efficacy to be related to being a successful designer, we hypothesize that self-regulation contributes not only directly to success as designer, but also indirectly, via a heightened level of selfefficacy. Hypothesis 5: Self-regulation is indirectly positively related to being a successful designer via self-efficacy (partial mediation). As has been mentioned before, self-regulation provides the opportunity to manage and control design processes. This also makes it possible to observe and influence aspects that are important for businesses in general, such as the quality, efficiency, and timeliness of work processes (Deming, 2000). Going through cycles of self-regulation behaviors of the self-regulation feedback loop encourages reflection on what has been accomplished already, and to develop plans for what still needs to be done. In other words, it encourages reflection on the design process. "Reflection provides an important possibility for improving the effectiveness and efficiency of design processes" (Reymen et al., 2006). In addition, moments of reflection may also remind designers of the limited amount of 85

94 Chapter 5 available time, and encourage them to use it well. This may prevent problems near deadlines that may result in quality-time trade-offs or missing deadlines. Thus, feedback loops of self-regulation may also help to monitor design processes on important business aspects such as quality, efficiency and timeliness. Therefore, we hypothesize that designers that engage in self-regulation more often may be better businesspersons. Hypothesis 6: Self-regulation is positively related to being a successful businessperson. Self-regulation may, in addition, also positively influence being a successful businessperson indirectly via a higher level of self-efficacy. We argued before that selfregulation may lead to a higher level of self-efficacy, because it may lead to an increased level of perceived control. Also, we argued that self-efficacy might contribute to being a successful businessperson, because persistence may help to sustain a high level of quality and timeliness of the design work. Combining these arguments, we hypothesize that selfregulation also has an indirect effect on being a successful businessperson, via its positive effects on self-efficacy. Hypothesis 7: Self-regulation is indirectly positively related to being a successful businessperson via self-efficacy (partial mediation). Method Sample Our sample consisted of 339 architects. We sampled architects because making architectural designs is a creative act (Goldschmidt, 1991), and therefore architects are generally perceived to be creative professionals (c.f., Feist, 1998; Kirton, 1994; Unsworth, 2001). Invitations to take part in this study were sent out by to the approximately 2100 members of the Organization of Dutch Architects (BNA). A week later a reminder was sent. The response rate was 16.1%, which is a moderate, but acceptable response rate for this type of study and respondents (Baruch, 1999). The majority of the participants were male (86.7%), reflecting that, in general, in the Netherlands there are many more male than 86

95 Being succesful in a creative profession female architects and engineers (c.f., the past ten years, the average percentage of female architects and engineers in the Netherlands was 22% (Statistics Netherlands, 2008)). The mean age of our participants was 44.7 years (SD = 10.2), and their average years of work experience was 16.9 years (SD = 9.4). The architects were either employees (30.7%), employers (44.8%), or self-employed and running a one-man business (24.2%). On average, they worked 43.0 hours per week (SD = 10.1), 33.6% (SD = 20.4) of which they indicated to spend on design work. Measures Data were collected with an online questionnaire. All items were in Dutch. Scales that originally were in English were translated using a procedure of back and forth translation with the help of a sworn translator. All questions were focused on the design work of the architects specifically. All scales were scored on a 5-point Likert scale ranging from highly disagree (1) to highly agree (5). Self-Regulation in Design Work The scale of self-regulation in design work consisted of three subscales, each representing a different self-regulatory behavior. We distinguished between (a) focusing on work priorities, (b) planning work time and activities, and (c) monitoring time and task progress. Focusing on work priorities. With 3 items from Tripoli's (1998) priority focus scale we measured the extent to which individuals focused on their work priorities with regard to their design work. An example item is: "I review my priorities before determining what design activities to work on next." Cronbach's alpha was.68. Planning work time and activities. The 4 items from the anchored planning scale of Tripoli (1998) measured the extent to which individuals engaged in planning activities to allocate their work time to design activities. An example item is: "I usually develop time tables for most designs on which I am working." Cronbach's alpha was.69. Monitoring time and task progress. The 6-item time monitoring scale of Claessens (2004b) was used to measure monitoring time and task progress. An example item is: "While executing my design tasks, I regularly check whether there is progress." Cronbach's alpha was

96 Chapter 5 To gain more insight into the structure of the overall self-regulation scale and its subscales, we performed confirmatory factor analyses using LISREL 8.7 (Jöreskog & Sörbom, 1993). These showed that a three-factor model fitted the data best (see Table 5.1). An exploratory factor analysis in SPSS revealed that each subscale loaded on a separate factor. Eigenvalues of the focusing-on-work-priorities factor, the planning-work-time-andactivities factor, and the monitoring-time-and-task-progress factor were 1.28, 1.56, and 4.06, respectively. Nevertheless, zero-order correlations showed that the subscales were also related to each other (see Table 5.4). In the LISREL analyses to test our hypotheses, the three subscales were separate factors that together formed the self-regulation-in-designwork construct. Table 5.4 includes both the three separate subscales and a combined 13- item overall scale (α =.81). Table 5.1. Summery of Confirmative Factor Analyses for the Self-Regulation Subscales Competing Models χ 2 df χ 2 df RMSEA GFI CFI One-factor model Two-factor model *** Three-factor model *** Note. N = 333. *** p <.001, indicating a significant improvement over the previous model. RMSEA = root-mean-square error of approximation; GFI = goodness of fit index; CFI = comparative fit index Innovative Style We used the 4-item creativity subscale of the cognitive style measure of Miron, Erez, and Naveh (2004) to assess innovative style. This subscale is a short version of the innovation style subscale of the Adaption Innovation Inventory developed by Kirton (1976; 1994). An example item is: "I prefer tasks that enable me to think creatively". Cronbach's alpha was

97 Being succesful in a creative profession Design Related Self-Efficacy We used the 8-item occupational self-efficacy scale of Schyns and Von Collani (2002) to measure self-efficacy with regard to design work. An example item is: "When I am confronted with a problem in a design, I can usually find several solutions." Cronbach's alpha was.79. Success as Designer The scale for success as designer was specifically developed for this study. The scale consisted of three items (see Table 5.2). Cronbach's alpha was.65. To provide some concurrent validity for this scale with more objective measures of success as designer, we asked the architects to provide (a) the number of their designs published in daily press and (b) the number of their designs published in architecture journals. As expected, success as designer was related to the number of designs published in daily press (r =.21, p <.001) and in architecture journals (r =.19, p <.001). Success as Businessperson To measure success as businessperson we also developed a scale specifically for this study. The three items of this scale can be found in Table 5.2. Cronbach's alpha was.73. To be able to provide some concurrent validity for this new scale, we selected the group of architects that ran their own businesses (N = 225), and correlated the success-as-abusinessperson scale to a more objective measure of business success: the size of their firms. We reasoned that within this group, architects who were better businesspersons would employ more people, and this was confirmed (r =.30, p <.001). Confirmatory factor analyses in LISREL showed that the two scales for success were two separate factors (see Table 5.3), and could not be combined into one measure: The one-factor model did not fit the data, whereas the two-factor model fitted the data well and significantly better than the one-factor model ( χ 2 = , p <.001). Exploratory analyses in SPSS revealed that the items of the success-as-a-designer scale all loaded on one factor, and that all items of the success-as-a-businessperson scale loaded on the other factor (see Table 5.2). These factor analyses provide some discriminant validity for the success scales. 89

98 Chapter 5 Table 5.2. Factor Loadings and Eigenvalues of the Success Scales Factor Subscales with items 1 2 Success as designer If I take into account my design work only, I see myself as a successful architect My design ambitions make me a successful architect With respect to creativity, I am a successful architect Success as businessperson Business-wise, I am successful compared to other architects I am a successful architect thanks to my organizational abilities Looking at the financial side of it, I regard myself a successful architect Eigenvalues Note. N = 334. The highest factor loadings are shown in boldface. Table 5.3. Summery of Confirmative Factor Analyses for the Success Scales Competing Models χ 2 df χ 2 df RMSEA GFI CFI One-factor model Two-factor model *** Note. N = 334. *** p <.001, indicating a significant improvement over the previous model. RMSEA = root-mean-square error of approximation; GFI = goodness of fit index; CFI = comparative fit index 90

99 Being succesful in a creative profession Data Analysis Procedure The data were analyzed using structural equation modeling (SEM) by means of LISREL 8.7 (Jöreskog & Sörbom, 1993). This means that our hypotheses were tested by comparing the fit of different, nested models. In the models, each of the variables were included as latent factors that were operationalized by their corresponding items. To assess the overall model fit of each of the nested models, we used the chi-square statistic (χ 2 ), the root-meansquare error of approximation (RMSEA), the comparative fit index (CFI), and the goodness of fit index (GFI) (Bentler, 1990; Jöreskog & Sörbom, 1993). To test for (partial) mediation, we used the procedures described by Mathieu and Taylor (2006) for analyzing these types of relationships through structural equation modeling. These procedures are similar to those described by Baron and Kenny (1986), referring to the three conditions that should be met for (partial) mediation. First, the dependent and independent variables should be related. As suggested by Mathieu and Taylor (2006), this was tested by assessing the fit of a model including only direct effects. Second, we assessed the fit of a model including only indirect effects (the no-direct-effects model), making sure that all dependent and independent variables were related to the mediator. Third, we tested per mediating relationship whether full mediation, partial mediation, or no mediation fitted the data best by adding each direct effect separately to the no-direct-effects model and comparing it to the no-direct-effects model. Indirect effects were determined through Sobel tests. Since analyses with structural equation modeling include all variables at once, these methods are more vulnerable to the effect of missing data. After careful examination of our missing values, we concluded that the missing values in the main variables were limited to 9% of the data, whereas the missing values of the demographic variables concerned 27% of the data. Therefore, we decided to fill the missing values in the demographic variables with the overall means of the variables, and to apply listwise deletion for the missing values in the main variables. Results Means, standard deviations, and zero-order correlations among variables can be found in Table 5.4. This table shows that self-regulation was not related to innovative style (r =.02, 91

100 Chapter 5 Table 5.4. Means, Standard Deviations, and Zero-Order Correlations Between Study Variables Variable M SD a 4b 4c 1. Age Gender *** - 3. Work experience *** -.20*** - 4. Self-regulation (.81) 4a. Focusing on work priorities *** (.68) 4b. Planning work time and activities ***.34*** (.69) 4c Monitoring time and task progress ***.42***.38*** (.76) 5. Innovative style Self-efficacy *** ***.22***.24***.19***.11* 7. Success as designer * Success as businessperson ***.22***.19***.24*** Note. N = 325. p <.10,* p <.05, ** p <.01, *** p <.001. Cronbach's alpha reliabilities can be found within parentheses. Gender is coded as male (1), female (2). 92

101 Being succesful in a creative profession Table 5.4. Means, Standard Deviations, and Zero-Order Correlations Between Study Variables (Continued) Variable Innovative style (.65) 6. Self-efficacy.30*** (.79) 7. Success as designer.44***.30*** (.65) 8. Success as businessperson.02.31***.31*** (.73) Note. N = 325. *** p <.001. Cronbach's alpha reliabilities can be found within parentheses. p >.10) but to self-efficacy (r =.22, p <.001), success as businessperson (r =.28, p <.001), and success as designer (r =.13, p <.05). Innovative style was also related to selfefficacy (r =.30, p <.001), and to success as designer (r =.44, p <.001), but not to success as businessperson (r =.02, p >.10). Success as designer and success as businessperson were related to each other (r =.31, p <.001). Furthermore, the table shows an effect of age and work experience on self-efficacy (r =.22, p <.001; r =.28, p <.001). Since age and work experience were very highly correlated (r =.80, p <.001), we decided to include only the one with the strongest relation (work experience) as a control variable in our analyses. We did not include gender as a control variable, because hardly any gender effects appeared to exist. Following Mathieu and Taylor, we first tested the only-direct-effects model. This fitted the data sufficiently: The χ 2 /df-ratio was 2.07 (a χ 2 /df-ratio of < 3 is preferable in large sample sizes such as ours (Kline, 1998, pp )); the RMSEA was.06 (RMSEA of <.05 indicates a close fit, and between.05 and.08 a fair fit); the CFI was.93 and the GFI was.90 (CFI and GFI of >.90 indicate an adequate fit (Mathieu & Taylor, 2006)) (see Table 5.5). All paths between the independent and dependent latent variables were significant. The paths between self-regulation and success as businessperson (β =.43, t = 5.91, p <.001) and between innovative style and success as businessperson (β =.15, t = 93

102 Chapter 5 Table 5.5. Summery of Measures for Nested Path Models Competing Models χ 2 df χ 2 df RMSEA CFI GFI Only-direct-effects model No-direct-effects model Adding innovative style success as designer to no-direct-effects model Adding innovative style success as businessperson to no-direct-effects model Adding self-regulation success as designer to no-direct-effects model Adding self-regulation success as businessperson to no-direct-effects model *** *** *** Final model Note. N = 310. *** p <.001. χ 2 = chi-square difference compared to no-direct-effects model; RMSEA = root-mean-square error of approximation; GFI = goodness of fit index; CFI = comparative fit index. In all models work experience was included as a (not significant (p >.05)) control variable. 2.13, p <.05) explained 22% of the variance in success as businessperson (R 2 =.22). The paths between self-regulation and success as designer (β =.22, t = 2.98, p <.01) and between innovative style and success as designer (β =.71, t = 7.78, p <.001) explained 56% of the variance in success as designer (R 2 =.56). Thus, the first condition for (partial) mediation was met for all relationships between independent and dependent variables. The second baseline model, the no-direct-effects model, also fitted the data sufficiently (χ 2 /df-ratio = 2.25, MRSEA =.06, CFI =.92, GFI =.89) (see Table 5.5). In this 94

103 Being succesful in a creative profession model all paths between the independent variables and the mediator variable, and between the mediator variable and the dependent variables were significant. Self-regulation (β =.35, t = 4.67, p <.001) and innovative style (β =.45, t = 5.88, p <.001) explained 35% of the variance in self-efficacy (R 2 =.35). Self-efficacy explained 19% of the variance in success as businessperson (β =.44, t = 5.83, p <.001) (R 2 =.19), and 28% of the variance in success as designer (β =.57, t = 5.93, p <.001) (R 2 =.28). Also, the indirect effects of the independent variables and the dependent variables via the mediator were significant: from self-regulation to success as businessperson via self-efficacy: β =.07, Sobel = 3.66, p <.001; from innovative style to success as businessperson via self-efficacy: β =.09, Sobel = 4.11, p <.001; from self-regulation to success as designer via self-efficacy: β =.10, Sobel = 3.70, p <.001; and from innovative style to success as designer via self-efficacy: β =.13, Sobel = 4.17, p <.001. Thus, also the second condition for (partial) mediation was met. To test whether the third and final condition for mediation was met for the various combinations of variables, each combination of independent variable and dependent variable with their possible mediator was tested separately for mediating effects. We started with testing for partial mediation of self-efficacy between innovative style and success as designer (Hypothesis 2), which, at the same time, tested for the direct effect between innovative style and success as designer in this model (Hypothesis 1). Following Mathieu and Taylor (2006), the no-direct-effects model was used as a baseline model, to which a path between innovative style and success as designer was added. The new model fitted the data well (χ 2 /df-ratio = 2.04, MRSEA =.06, CFI =.93, GFI =.90), and was an improvement over the no-direct-effects model ( χ 2 = 45.15, p <.001) (see Table 5.5). All paths were significant: between innovative style and success as designer (β =.58, t = 6.11, p <.001), between innovative style and self-efficacy (β =.60, t = 6.06, p <.001), and between self-efficacy and success as designer (β =.24, t = 2.92, p <.01). These last two paths together formed the indirect effect of innovative style on success as designer (β =.14, Sobel = 2.54, p <.05). The results suggest both a direct and an indirect effect of innovative style on success as designer, supporting Hypothesis 1 and 2. Hypothesis 3 concerned full mediation of the effect of innovative style on success as businessperson via self-efficacy. For this, the path between innovative style and success as businessperson was added to the no-direct-effects model. The resulting model fitted the 95

104 Chapter 5 data sufficiently (χ 2 /df-ratio = 2.22, MRSEA =.06, CFI =.92, GFI =.89), and this model was a significant improvement over the no-direct-effects model ( χ 2 = 9.04, p <.001). However, instead of finding no direct effect between innovative style and success as businessperson, a negative effect was found (β = -.28, t = -3.02, p <.01). The paths between innovative style and self-efficacy (β =.48, t = 6.19, p <.001) and between selfefficacy and success as businessperson (β =.60, t = 6.06, p <.001) were both positive and significant, resulting in an indirect positive effect of innovative style on success as businessperson (β =.29, Sobel = 2.75, p <.01) that fully compensated its negative, direct effect. These results confirm the positive indirect effect of innovative style on business success via self-efficacy, and thus provide support for Hypothesis 3. However, also an unexpected, negative, direct effect was found that neither supports, nor opposes the full mediation hypothesis. Next, we tested the direct and indirect effects of self-regulation on success as designer (Hypotheses 4 and 5), by testing for partial mediation between self-regulation and success as designer via self-efficacy. Again, we used the no-direct-effects model as a baseline model, to which the path between self-regulation and success as designer was added. This model had an adequate fit (χ 2 /df-ratio = 2.26, MRSEA =.06, CFI =.92, GFI =.89), but was not a significant improvement over the no-direct-effects model ( χ 2 =.40, p >.10). Also, the path between self-regulation and success as designer was not significant (β = -.06, t = -.67, p >.10), whereas the paths between self-regulation and self-efficacy (β =.36, t = 4.79, p <.001) and between self-efficacy and success as designer (β =.60, t = 5.62, p <.001) were significant. A Sobel test confirmed an indirect effect (β =.22, Sobel = 3.60, p <.001). From these results we can conclude that the effect of self-regulation on success as designer was fully mediated by self-efficacy. Therefore, Hypothesis 4 is not supported, because we failed to find a direct effect. Hypothesis 5 is supported, because the results indicate an indirect effect, although it appears to be fully, instead of partially, mediating the relationship. Finally, Hypothesis 6 and 7 were tested regarding the direct and indirect effects of self-regulation on success as businessperson. A path between self-regulation and success as businessperson was added to the no-direct-effects model. This new model fitted the data adequately (χ 2 /df-ratio = 2.21, MRSEA =.06, CFI =.92, GFI =.90), and was a significant 96

105 Being succesful in a creative profession improvement over the no-direct-effects model ( χ 2 = 10.10, p <.001). All three paths (the path between self-regulation and success as businessperson (β =.26, t = 3.26, p <.001), between self-regulation and self-efficacy (β =.33, t = 4.51, p <.001), and between selfefficacy and success as businessperson (β =.33, t = 4.29, p <.001)) were significant. A Sobel test revealed partial mediation: β =.11, Sobel = 3.11, p <.01. These results support Hypotheses 6 and 7, suggesting that self-regulation contributes to success as businessperson, both directly and via its positive effect on self-efficacy. In sum, we found (a) positive direct and indirect effects of innovative style on success as designer, (b) a negative direct effect and a positive indirect effect of innovative style on success as businessperson, (c) only a positive indirect effect of self-regulation on success as designer and, (d) positive direct and indirect effects of self-regulation on success as businessperson. Combining these findings into one model (see Figure 5.1), we created a model that fitted the data well (χ 2 /df-ratio = 2.01, MRSEA =.05, CFI =.94, GFI =.90), and fitted significantly better than both the direct-effects model ( χ 2 = 13.79, p <.001) and the no-direct-effects model ( χ 2 = 55.38, p <.001). All paths that were significant in the separate tests above were also significant in the final model, except for the direct path between innovative style and success as businessperson (β = -.08, t = -.93 p >.01). This means that the negative direct effect of innovative style on success as businessperson disappeared, and only the positive indirect effect remained to be significant. In this final model, 24% of the variance in success as businessperson was explained (R 2 =.24) by the direct effect of self-regulation (β =.24, t = 2.96, p <.01), the indirect effect of selfregulation via self-efficacy (β =.12, Sobel = 2.99, p <.01), and the indirect effect of innovative style via self-efficacy (β =.15, Sobel = 3.22, p <.01). Also, 48% of the variance in success as designer was explained (R 2 =.48) by the direct effect of innovative style (β =.57, t = 5.95, p <.001), the indirect effect of innovative style via self-efficacy (β =.10, Sobel = 2.56, p <.01), and the indirect effect of self-regulation via self-efficacy (β =.08, Sobel = 2.45, p <.05). Thus, both self-regulation and innovative style are relevant predictors of success as designer and a businessperson, and self-efficacy plays an important role as mediator in these relationships. 97

106 Chapter 5 Focusing on work priorities Planning work time and activities Monitoring time and task progress Innovative style Self-regulation Self-efficacy Successful designer Succesful businessperson R 2 =.48 R 2 =.24 path is significant path is not significant Figure 5.1. Results for the LISREL Path Analyses of the Final Model. 98

107 Being succesful in a creative profession Additional Results The final model also shows the relative weight of each of the three specific self-regulation behaviors that contribute to the latent construct of self-regulation in design work. Each of them appears to be important (λ focusing on work priorities =.42, t = 9.61, p <.001; λ planning work time and activities =.37, t = 8.58, p <.001; λ monitoring time and task progress =.37, t = 10.10, p <.001). Discussion The results of this field study show that self-regulation (setting priorities, planning work activities, and monitoring time and task progress) and having an innovative cognitive style are important predictors of success as designer and as a businessperson, and that selfefficacy plays a key role in these relationships as a (partial) mediator. Specifically, it was found that self-regulation was directly related to being a successful businessperson, and indirectly, via self-efficacy, to being a successful designer. Also, innovative style was directly related to being a successful designer, and indirectly, via self-efficacy, to being a successful businessperson. In the past, research on creativity at work predominantly focused on the creative aspects of creative professions. This field study provides a distinct contribution to the literature about creativity at work, because it adopted a somewhat broader view on creative professions by including both the creative and business aspects of the work. Specifically, it showed that one antecedent (innovative style) predominantly influences the creative side of the work (being a successful designer), whereas another antecedent (self-regulation) mainly influences the business side of the work (being a successful businessperson). Self-efficacy explained existing spillover effects. Thus, our results suggest that being successful in creative professions requires different skills and abilities, and that some antecedents may influence the creative aspects of the work, and other antecedents the business aspects. Since past research mostly focused on antecedents predicting creative output (Shalley et al., 2004), this study contributes to the literature because it has added an antecedent of the business side of creative work, thereby extending the view on being successful in creative professions. A second contribution concerns the role of self-regulation behaviors in design work. Although awareness is growing that these kinds of behaviors may be beneficial to 99

108 Chapter 5 creativity (Isaksen & Treffinger, 2004), little research had been conducted to confirm their importance. Exceptions are the studies of Peeters et al. (2007) and Sonnentag (1998) that provided empirical evidence of the existence and role of planning behaviors and monitoring task progress during designing. Our study supports their findings and has added a new theoretical contribution, because we differentiate between the distinct behaviors within the self-regulation feedback loop of Control Theory of Self-Regulation (Carver & Scheier, 1982). This theory suggests that each feedback loop consists of a sequence of behaviors that include (a) decision making with regard to the tasks that are to be completed, (b) planning behaviors, (c) task execution, and (d) monitoring task progress. Our study is unique in that it included self-regulation behaviors concerning each of the phases of this cycle, with the exception of the execution phase. Specifically, our results show that each of these phases is more or less equally important, which confirms that all of the self-regulation behaviors of the cycle are relevant. These results have important theoretical implications, specifically with regard to the role of planning in creative work. Planning of creative work presents a dilemma. Unpredictability is an innate aspect of creative (or innovative) work (Koput, 1997, p. 529). This means that it is not known beforehand which activities are required to complete a task, which makes planning ahead difficult (Ormerod, 2005). Thus, until now, the relevance of planning in creative work was not self-evident. However, this study shows the value of planning behaviors as part of the self-regulation feedback loop, and (partly) solves the dilemma: On the one hand, the self-regulation feedback loop underlines the importance of planning behaviors in creative work, because it is part of the cycle that guides task attainment. On the other hand, this feedback loop provides the opportunity to adjust and further develop plans during the process of task completion, hence providing flexibility to the plans and offering a solution to cope with the task uncertainty that resides in creative tasks. A third contribution of this study concerns the role of self-efficacy in creative work. In a review, Shalley et al. (2004) argued that more research was needed on the role of self-efficacy in creative work, and specifically, that it was important to investigate the relationship between self-efficacy and other personal characteristics, such as having an innovative cognitive style (p. 946). We answered to this call. Specifically, our results 100

109 Being succesful in a creative profession suggest that self-efficacy serves a role as 'go-between' that passes on positive effects to otherwise unrelated constructs (from innovative style via self-efficacy to success as businessperson and from self-regulation via self-efficacy to success as designer). This has significant theoretical implications for the creativity literature, because it suggests that selfefficacy is more than only an antecedent: It may transpose the positive effects of other antecedents that affect a specific part of the work into a positive effect that affects creative work more in general. For example, our results suggest that innovative style predominantly influences the creative side of the work, but, through a heightened level of self-efficacy, it also affects other aspects such as being successful as a businessperson. A possible explanation of this role of self-efficacy as 'go-between' may be that it is built through positive mastery experiences (Bandura, 1977): There are many factors both on the person-level (skills and abilities) and in the work environment (enhancing and hindering factors) that may influence a mastery experience. Each specific antecedent may influence just a specific part of the work, but when it contributes to a mastery experience, it also contributes to the overall level of self-efficacy. Thus, via mastery experiences, self-efficacy may be built through the positive effects of many different antecedents. Since self-efficacy enhances coping behaviors and persistence (Bandura, 1977), the positive effects of all these different antecedents may be forwarded to task performance, and, ultimately, to being successful. Strengths and Limitations A first strength of this study is the use of structural equation modeling in our analyses. These sophisticated statistical analyses provide the opportunity to test hypotheses while including all variables in the model. This is particularly interesting when models contain more than one outcome variable, which is the case in our study (success as designer; success as businessperson), and when models contain paths via mediators (in our case via self-efficacy) (Mathieu & Taylor, 2006). A second strength is that it was conducted in a field setting, among a working population consisting entirely of creative professionals: architects. Since "architecture is considered [...] the mother of all design domains" (Goldschmidt, 1999: 526), architects may 101

110 Chapter 5 be considered suitable representatives of designers, and of creative professionals in general (c.f., Unsworth, 2001). Another strength of our sample is that it included individuals working in both large and small businesses. Often, for practical reasons, research on creativity at work is conducted in large organizations (c.f., Amabile et al., 2004; Hargadon & Bechky, 2006; for an exception see: Tschang, 2007). However, since a large part of creative professionals are working in smaller firms, it is preferable to include participants from organizations of different sizes. Since data were collected via the Dutch Organization of Architects (BNA) it was possible to send out our survey to about half of the Dutch population of architects. This means that we were able to reach architects working in large firms, but also in small firms, or one-man businesses. Notwithstanding the contributions of our study, we should also point out some limitations. A first limitation is the use of self-reports. For some scales this may be less of a problem, because these things can be best judged by the person itself (e.g., the use of selfregulation behaviors (e.g., Tripoli, 1998), having an innovative cognitive style (e.g., Miron et al., 2004), or the level of self-efficacy (e.g., Schyns & Von Collani, 2002)). Nevertheless, the measurement of the success of architects as designers and businesspersons would have been better if it had also been rated by supervisors or coworkers. However, collecting these additional ratings was difficult, since the self-employed architects in our sample did not have coworkers or supervisors, and about half of the architects were supervisors themselves. Thus, our sample had the advantage of including architects working in smaller firms, but this also resulted in limitations with regard to the possibilities for collecting peer ratings. Nevertheless, we were able to provide some concurrent validity for both of these success scales by relating them to more objective measures of success. In the group of architects who ran their own firms, the success of architects as businesspersons was related to the size of their firms. Also, being a successful designer was related to the number of designs that had been published in daily press and in architecture journals. A second limitation of this study was its moderate response rate. Non-response might have been relatively high because our sample included professionals, part of them with management tasks. This type of sample tends to result in lower response rates compared to, for example, samples from people working in education, or student samples 102

111 Being succesful in a creative profession (Baruch, 1999; Green, Boser, & Hutchinson, 1998). Furthermore, reaching acceptable response rates has become more difficult because, in general, the willingness to participate in survey research has declined through the years (Baruch, 1999; Dey, 1997). Taking into account our sample's characteristics, our response rate can be considered acceptable (Baruch, 1999). A final limitation is the cross-sectional design. This means that the causality of the relationships in our path model may be in the anticipated direction, but may also be in the opposite direction, or recurrent. For example, it may be possible that self-efficacy not only leads to success as designer, but that success as designer may also be an antecedent of selfefficacy, and that the relationship is dynamic. Even so, we can conclude that our variables (such as self-efficacy and success as designer) are related to each other. Suggestions for Further Research It may be interesting to further investigate the process of self-regulation in creative work. Specifically, we suggest it to be investigated over time and in more detail. Thus, it may be possible to investigate the interaction between the behaviors in the self-regulation feedback loop and the creative process. A think-aloud study may be a helpful method to investigate this. For example, a think-aloud study of Sonnentag (1998) provided interesting insights into the planning and monitoring activities of software designers during task execution. Where her study showed the importance of these behaviors during designing, a next thinkaloud study could focus specifically on the interaction of these self-regulation behaviors with cognitive creative processes. A think-aloud study of Fleck and Weisberg (2004) focused on cognitive creative processes, but not on self-regulation behaviors. We think that when a future think-aloud study could combine the studies of Sonnentag (1998) and Fleck and Weisberg (2004) (measuring both self-regulation behaviors and creative cognitive processes), this might provide theoretical as well as practical contributions. It may also be worthwhile to investigate the role of self-efficacy in creative work in more detail. For example, by means of longitudinal research designs it may be possible to find empirical evidence for the dynamic relationships between self-efficacy and possible antecedents and outcomes in the field of design work. Future research may also further investigate the business aspect of creative work. For example, it may be divided into 103

112 Chapter 5 different relevant subaspects, such as striving for a high level of efficiency, quality, or timeliness of the work. Practical Implications This study has significant practical implications. It emphasizes the importance of selfregulation in creative work. Not being aware of the relevance of self-regulation, designers may focus only on the creative side of their work when trying to improve their performance. This focus may involve, for example, to trying to improve sketching skills (Goldschmidt, 1991), or asking for the help of colleagues to brainstorm for more creative ideas (Valacich, Dennis, & Connolly, 1994). However, our study opens a whole new area of performance improvement possibilities that are focused on the business side of the work. Specifically, our study suggests that self-regulation is important and may help to perform better business-wise. Designers could, for example, try to improve their time management skills. Time management training generally focuses on the activities an individual is engaged in at a daily level, and tries to improve the effectiveness of time use at work (Claessens et al., 2007). Whereas all kinds of different practical rules are suggested (e.g., keep your desk tidy; avoid interruptions (Schriber & Gutek, 1987)), it also offers suggestions about how to prioritize, plan and monitor work activities; self-regulation behaviors that have been found to be useful in our study. Through their positive effects on self-efficacy, time management behaviors may even have spillover effects on creative performance. In sum, this field study has shown the importance of self-regulation (setting priorities, planning work activities, and monitoring time and task progress), having an innovative cognitive style, and self-efficacy in design work. Moreover, we revealed their relevance both for being a successful designer and a successful businessperson. 104

113 Chapter 6 General Discussion This dissertation presented five studies (one experiment and four field studies) in which we addressed the self-regulation of time (i.e., time management) in creative professions. In this concluding chapter the main findings of these studies are discussed. We provide answers to the research questions raised in the introduction, and discuss our contribution to the literature. Furthermore, we describe the practical implications of this dissertation and its strengths and weaknesses. We end with suggestions for future research and some concluding comments. Since little research has been conducted to the role of time in creative work (Drazin et al., 1999), and to the role of self-regulation of time (i.e., time management) in particular, this dissertation has investigated the benefit of managing one's time in creative professions. The foregoing chapters presented different, yet related studies regarding this topic, and addressed the five main research questions formulated in the introduction of this dissertation. The first research question addressed the role of the 'self' in self-regulation, and was: Are individuals more creative when they have personal discretion as to how to use their time? The results of an experiment among 109 students of industrial engineering and management science in Chapter 2 have confirmed this. Individuals that could decide 105

114 Chapter 6 themselves when to work on another task for a while (i.e., could take incubation time at their own discretion) conceived more insights to solve the creative problems at hand and reached impasses less often (i.e., felt less often 'stuck') compared to others for whom it was decided when to take their incubation time. Thus, the 'self' in self-regulation has been proven to be of relevance. The second and third research questions concerned pacing (i.e., the self-regulation of work efforts over time). The second question was: Do creative professionals use different pacing patterns compared to professionals in less creative professions? In a survey study among 88 individuals of different professions (Study 1 of Chapter 3) we found that individuals in creative professions more often used 'incubation time' pacing patterns, which included an intermediate period of disengagement from a task or project that may have been used as incubation time. They were also more often 'deadline workers' and less often 'early workers', which means that they generally did little of the work in the beginning of a task or project, and much of it towards the deadline. Thus, this study has confirmed that individuals in creative professions have distinct ways to distribute their work efforts over time. The third research question was an open question: How do creative professionals pace their projects over time exactly? Although Study 1 of Chapter 3 revealed that creative professionals use specific pacing patterns, the study's measurements of these patterns were simplified representations of their actual pacing behaviors. Therefore, questions with regard to the precise patterns still remained unanswered. Study 2 of Chapter 3 was an in-depth interview study among 25 architects through which we explored pacing patterns over the course of the design phases of their architectural projects. The results showed that almost all architects used the incubation time pacing pattern in their projects. Furthermore, this pattern was repeated several times within one project, specifically, it occurred before every project milestone and deadline. Furthermore, architects worked on multiple projects at the same time, which means that pacing patterns of these projects overlapped: Working on one project matched periods of disengagement from other projects. Finally, the study showed that some of the architects did not set strict project milestones and deadlines nor adhered to them very strictly. Whereas some architects accepted the deadlines that were imposed upon them by their clients, others negotiated about their deadlines, and yet others did not accept 106

115 General discussion strict deadlines at all. Also, our interviews revealed deadline postponement and quality-time trade-offs. Thus, the deadlines that mark the ends of the pacing patterns may not be that fixed and may be changed flexibly. Overall, this study has provided some valuable insights into the ways in which some of the complexities of creative work (project milestones, deadlines, working on multiple projects) and a preference for the incubation time pacing pattern are combined in practice. The fourth research question was: Is time management (i.e., the self-regulation of time) beneficial to creativity at work? Chapter 4 presented a diary study among 68 junior R&D engineers in which it was found that the tendency to use time management was related to being more creative on a daily level. Thus, this study provides empirical evidence of the benefit of time management in creative work. Furthermore, this study showed that this positive effect of time management was unrelated and complementary to the positive effect of having an innovative cognitive style. We also found that, next to this direct effect, time management had an indirect effect on daily creativity via daily positive affect, meaning that individuals were more creative on a certain work day when they felt better. Finally, the results showed that time management was related to a stable creative performance over time. A survey among 339 architects (Chapter 5) investigated the fifth research question: To what extent are self-regulation behaviors (setting priorities, planning work activities, and monitoring time and task progress) related to the success of creative professionals? Specifically, the study in Chapter 5 investigated the influence of innovative cognitive style, self-efficacy, and the three self-regulation behaviors on the success of the architects as designers and businesspersons. The results showed that the three selfregulation behaviors were about equally important, and that self-regulation was directly related to being a successful businessperson, and indirectly, via self-efficacy, to being a successful designer. Innovative cognitive style showed to be directly related to being a successful designer, and indirectly, via self-efficacy, to being a successful businessperson. Furthermore, the results showed that self-regulation and innovative cognitive style were unrelated. Similarly to the diary study of Chapter 4, this study suggests that innovative style and self-regulation of work time are complementary positive factors in creative work. However, this survey showed that innovative style predominantly influenced the design 107

116 Chapter 6 aspect of the work of the architects, and self-regulation the business aspect. Furthermore, it showed that self-efficacy explained spillover effects. Theoretical Implications This dissertation provides an important contribution to the creativity literature by addressing self-regulation of time in creative professions. We built on Control Theory of Self-Regulation (Carver & Scheier, 1982) that suggests that task attainment requires repeated cycles of feedback loops to plan and monitor task progress, allowing for adjusting the course of action during task execution. Up to now, little research efforts have been conducted to investigate the role of this self-regulation feedback loop in creative work, although its relevance has been acknowledged, for example, by Isaksen and Treffinger (2004). These authors described the evolution of a creative problem solving model over a period of 50 years that started with the creative problem solving model of Osborn (1952) (which was based on early creative problem solving models, such as the model of Wallas (1926)) and had a focus on cognitive creative processes. The evolution of the model ended with the creative problem solving model of Isaksen, Dorval, and Treffinger (2000), which included cognitive creative activities, but also creative-process-management activities, such as "continuous planning, monitoring, managing, and modifying behavior during creative problem solving" (Isaksen & Treffinger, 2004, p. 92). This dissertation has confirmed the added value of the self-regulation of time in creative professions. Furthermore, the experiment concerning personal discretion in creative work in Chapter 2 has provided a contribution to the literature on incubation time specifically. This literature is based on the early creative problem solving model of Wallas (1926). Wallas's model consists of four stages: (a) preparation; gaining in-depth insight into a creative problem, (b) incubation; taking some time away from the problem, (c) illumination; suddenly and unexpectedly an insight comes to mind, and (d) verification; verifying the applicability of the idea. The underlying proposition of this model is that incubation time aids creative processes, which has received considerable support in past experimental research, although the exact cognitive mechanism responsible for the incubation effect has not been fully understood yet (Smith & Dodds, 1999). The unique contribution of our experiment to this literature is that it shows the benefit of personal discretion with regard to 108

117 General discussion when to take incubation time. Thus, deciding personally when to take a 'break' is better than incubation time imposed by someone else (even though this last is to be preferred over not having any incubation time at all). Chapter 3, the chapter on pacing, has contributed to the literature because it provides insights into processes that evolve over time in creative work, which have not often been studied before (Drazin et al., 1999; Shalley et al., 2004). Pacing patterns are patterns of task engagement and disengagement over the course of tasks or projects. High levels of task engagement (i.e., being in flow, in other words, being totally absorbed in an activity) have been found to be beneficial to creative performance (Csikszentmihalyi, 1996; Lefevre, 1988). In addition, disengagement from a task (incubation time) is helpful, as was demonstrated in Chapter 2. High task engagement over prolonged periods of time may appear conductive to performance intuitively, however, the results of Study 1 in Chapter 3 imply that creative professionals use pacing patterns that include a period of disengagement between two periods of high task engagement more often than individuals working in less creative professions. Study 2 in Chapter 3 showed that in creative projects this 'incubation time' pattern was repeated prior to each new project milestone or deadline. Furthermore, this chapter contributes to the literature on pacing in general. Whereas most of this literature has focused on team processes (e.g., Gersick, 1988, 1989; Gevers et al., 2006) or individual differences (e.g., Claessens, 2004a; Gevers et al., 2008), the results of Chapter 3 imply that task characteristics may also be important: The fact that creative professionals use distinct pacing patterns may be an indication that creative tasks are paced differently than other tasks. Also, our interview study provides more in-depth insights into pacing patterns at work. Chapter 4, describing the diary study on the effect of time management on daily creativity, also provides a unique empirical contribution: It showed that the tendency to use time management was related to daily creativity. In their review of personal and contextual factors that influence creativity, Shalley et al. (2004) concluded that most research on person-level antecedents of creativity focused on personality and innovative style. They urged researchers to broaden the range of individual characteristics under study to be able to obtain more insight into the different person-level antecedents of creativity. Chapter 4 adds a new person-level antecedent: the tendency to use time management. 109

118 Chapter 6 Also, this study was innovative in that it investigated fluctuations in the level of creativity over time. Studying fluctuations in performance provides complementary insights to using the mean level of performance as a dependent variable (c.f., Hofmann et al., 1992; Stewart & Nandkeolyar, 2006). The results of our diary study suggest that the level of creativity fluctuates over time, and that it fluctuates more when individuals have an innovative cognitive style, and less when they have a tendency to use time management. Chapter 5, containing the survey study among architects, provides a complementary contribution because it differentiated between three different time management behaviors: prioritizing, planning, and monitoring time and task progress. Each of them appeared to be about equally important, which is in line with Control Theory of Self-Regulation: Going through an entire feedback loop requires prioritizing (deciding on which task to work first and which to postpone), planning for action, task execution, and monitoring task progress (Carver & Scheier, 1998). Thus, this study suggests that the separate behaviors in the feedback loop are relevant to design work. Moreover, the results of this study provide a theoretical contribution with regard to the role of planning in creative work specifically. Creative (or innovative) work is by definition fairly unpredictable (Koput, 1997, p. 529), which means that it is unknown beforehand which activities are required to complete a task, which makes planning ahead difficult (Ormerod, 2005). Therefore, planning of creative work presents a dilemma: On the one hand, planning is difficult because of a high level of task uncertainty, on the other hand, planning is an essential part of the self-regulation feedback loop. Thus, until now, the relevance of planning in creative work was not self-evident. The study in Chapter 5 (partly) solves the dilemma: The self-regulation feedback loop underlines the importance of planning behaviors in creative work, and provides the opportunity to adjust and further develop plans during the process of task completion, hence providing flexibility to the plans and offering a solution to cope with the task uncertainty that resides in creative tasks. Practical Implications This dissertation also has important practical implications. Together, the chapters stress the added value of self-regulation of time in creative professions: Self-regulation is associated with being more creative, feeling better, being more confident when problems arise, and 110

119 General discussion ultimately, with being a successful designer and businessperson. Individuals in creative professions may focus on the creative side of their work when they try to improve their performance, for example, by using brainstorming techniques (Valacich et al., 1994), or by trying to improve their sketching skills (Goldschmidt, 1991). This dissertation provides new opportunities for improvement that do not concern the cognitive creative process directly, but the process management activities around it. Specifically, the results of Chapter 4 and 5 suggest that improving time management skills is useful. Time management training generally aims at optimizing workday output by improving the effectiveness of time use at work (Claessens et al., 2007). Creative professionals may benefit from engaging in a time management course. Specifically, the results of Chapter 5 indicate that the time management behaviors within the self-regulation feedback loop are useful. The results of Chapter 3 imply that, in addition to engaging in the activities of the self-regulation feedback loop for each single creative task, it is important to take other creative tasks into account when making or adjusting plans. Specifically, the results of Study 2 in Chapter 3 showed that multiple pacing patterns of different tasks or projects cooccur. Therefore, it may be wise to tune pacing efforts of these tasks or projects, and to prevent that deadlines or milestones coincide. The results of the experiment in Chapter 2 suggest that it is also important to keep one's plans flexible to some extent, to make it possible to work on something else for a while at one's discretion. Strengths and Limitations The main strength of this dissertation is the combination of multiple research methods: an experiment, interviews, surveys, and a diary study. This diversity of methods has made it possible to approach the research topic from different angles, providing complementary insights. Also, each research method has its strengths and limitations, and applying multiple methods allowed the strengths to be complementary, and the limitations to be compensated. For example, the experiment in Chapter 2 had the advantage of testing causality, but the disadvantage of a student sample. This disadvantage was largely compensated by the four field studies in the dissertation. Furthermore, the dissertation includes qualitative as well as quantitative data. Whereas the qualitative, in-depth interviews in Chapter 3 have provided 111

120 Chapter 6 detailed insights, the quantitative methods in the other chapters have allowed for testing relationships statistically. Also, the scope of the studies varied: The diary study in Chapter 4 made it possible to investigate effects on a more micro level (day level), the interview study in Chapter 3 examined pacing at the project level, and the survey in Chapter 5 investigated more general effects with regard to the success of creative professionals. Finally, the interviews in Chapter 3 and the diary study in Chapter 4 allowed for investigating fluctuations over time in levels of task engagement and daily creativity respectively. Thus, the variety in research methods had the advantage of examining selfregulation of time in creative work from diverse perspectives. However, the studies in this dissertation have one important limitation in common: the reliance on self-reported data. This means that our results may be biased to some extent (Judd, Smith, & Kidder, 1991), and collecting additional data from coworkers or supervisors may have been preferable. Nevertheless, some constructs may be judged best by the person itself (Judd et al., 1991, p. 152). This may be the case for part of the constructs used in our studies, such as which pacing pattern one generally uses (c.f., Gevers et al., 2008), whether one has an innovative cognitive style (c.f., Miron et al., 2004) or a tendency to use time management (c.f., Peeters & Rutte, 2005), how one feels on a certain day (c.f., Amabile et al., 2005), and one's level of self-efficacy (c.f., Schyns & Von Collani, 2002). However, some other measures, such as daily creative performance and success as designer and businessperson should preferably have been rated by coworkers or supervisors. Unfortunately, practical considerations made collecting these additional measurements difficult. Measuring daily creativity with a second resource was problematic, because the R&D engineers were working on individual design projects, and it was unlikely that coworkers or supervisors would have had sufficient insights into their daily performance to make accurate ratings of their creativity. A similar practical problem occurred in the survey among architects in which their success was judged: Part of our sample was self-employed, and thus without supervisors or coworkers around to rate them, and another part were supervisors themselves. Thus, due to the restrictions of our field samples, our data may be biased by the use of self-report data. 112

121 General discussion Suggestions for Future Research Although the scientific contribution of this dissertation is noteworthy, more research on the role of time in creative work is desirable (Drazin et al., 1999), and on the role of selfregulation of time in creative professions specifically. In line with Roe (2008), we argue that it would be particularly interesting to investigate it by using research methods that allow for measurements over time. For example, it might be interesting to study pacing patterns in creative professions over time. Although our studies in Chapter 3 investigated patterns that occurred over time, the research methods (a survey and an interview study) measured them at one moment in time, thereby relying on the perception and memory of participants. A proper way to validate our results would be to measure actual task engagement and disengagement over a period of time, and to construe pacing patterns from that. These data can be collected with, for example, a diary study in which the work efforts spend on certain creative tasks or projects are recorded. Another possibility would be to use data readily available in companies, such as in hour registration systems of project administrations. Notwithstanding the fact that these systems may also be biased in some way, this would allow for exploring pacing patterns over longer periods of time without bothering participants with repeated effort investments for research purposes. These pacing patterns could also provide more insight into the 'tuning' of pacing patterns when individuals work on multiple projects. Furthermore, it would be interesting to investigate the cognitive processes underlying the pacing patterns in creative work in more detail. As argued in Chapter 3, peaks in the patterns may resemble periods of flow, and dips may resemble periods of incubation time. However, more research is needed to confirm that the pacing patterns indeed represent patterns in which periods of flow are alternated with periods of incubation time. An experience sampling study may be able to shed more light on these processes. Experience sampling is a research method that requires multiple measures per day for several days in which an individual is questioned about the specific experience of the moment (Csikszentmihalyi & Csikszentmihalyi, 1988a). Experience sampling has repeatedly been used to measure flow (c.f., Csikszentmihalyi & Lefevre, 1989; Lefevre, 1988), and we think it might also provide insights into the relationship between flow and creative problem solving processes. For example, into the ways in which flow interacts with 113

122 Chapter 6 the different stages in the model of Wallas (1926): preparation, incubation, illumination and verification. Experience sampling could also provide more insights into fluctuations in creativity over time. Specifically, it would be interesting to investigate the possible causes of fluctuations and their consequences. Experience sampling may help to identify specific triggers and obstacles to creativity. Furthermore, the duration of the effects of these triggers or hindering factors may be established. The duration of phenomena is not often studied, although it provides important information (Roe, 2008). For example, it might be relevant to know whether it takes a day or a week to recover from a certain hindering event. Also, it would be interesting to investigate whether certain events have positive consequences one moment in time, and negative consequences at another moment in time. For example, a distant deadline may be a motivator to get into action, and may therefore have a positive effect on creativity, whereas a close deadline may cause time pressure, which can be detrimental for creativity (e.g., Amabile, Hadley, & Kramer, 2002). Finally, we suggest it to be valuable to study the behaviors of the self-regulation feedback loop in creative tasks over time. Although our results of Chapters 4 and 5 showed that time management and self-regulation behaviors are not related to having an innovative cognitive style, Chapter 4 showed that time management was related to being more creative on a daily level. Future research may measure self-regulation behaviors and creative processes over time. A think-aloud study may be a helpful method to investigate this. We suggest a future think-aloud study might combine the studies of Sonnentag (1998) and Fleck and Weisberg (2004). Whereas the first investigated self-regulation behaviors (but not cognitive creative processes), the second explored the cognitive creative processes (but not self-regulation behaviors). A study that would investigate both at the same time could give more insights into the interaction of the creative cognitive processes and the guiding effect of the self-regulation behaviors. It might be valuable when such a study could point out specific moments in time when reflection (i.e., going through a self-regulation feedback loop) is particularly beneficial to creative performance. 114

123 General discussion Conclusions In sum, this dissertation has shown that there is more to being a creative professional than just being creative. In addition, time management skills are helpful. This dissertation has demonstrated the added value of the 'self' in self-regulation, the distinctness of pacing patterns in creative professions, the positive effects of time management on daily creativity, and the role of self-regulation in being a successful designer and businessperson. However, although our results have answered some important questions, they have also raised many new questions. Therefore, we invite other researchers to join us in future research on the self-regulation of time in creative professions. 115

124

125 References Amabile, T. M. (1988). A model of creativity and innovation in organizations. Research in Organizational Behavior, 10, Amabile, T. M. (1996). Creativity in context. Oxford, U.K.: Westview Press. Amabile, T. M., Barsade, S. G., Mueller, J. S., & Staw, B. M. (2005). Affect and creativity at work. Administrative Science Quarterly, 50(3), Amabile, T. M., Hadley, C. N., & Kramer, S. J. (2002). Creativity under the gun. Harvard Business Review, 80(8), Amabile, T. M., Hill, K. G., Hennessey, B. A., & Tighe, E. M. (1994). The Work Preference Inventory - Assessing Intrinsic and Extrinsic Motivational Orientations. Journal of Personality and Social Psychology, 66(5), Amabile, T. M., Schatzel, E. A., Moneta, G. B., & Kramer, S. J. (2004). Leader behaviors and the work environment for creativity: Perceived leader support. Leadership Quarterly, 15(1), Ancona, D., Okhuysen, G., & Perlow, L. (2001). Taking time to integrate temporal research. Academy of Management Review, 26(4), Babbie, E. R. (2005). The basics of social research (Third ed.). Toronto, Canada: Thomson Wadsorth. Baer, M., & Oldham, G. R. (2006). The curvilinear relation between experienced creative time pressure and creativity: Moderating effects of openness to experience and support for creativity. Journal of Applied Psychology, 91(4),

126 References Bailey, B. P., & Konstan, J. A. (2006). On the need for attention-aware systems: Measuring effects of interruption on task performance, error rate, and affective state. Computers in Human Behavior, 22(4), Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), Baron, R. M., & Kenny, D. A. (1986). The moderator mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), Barry, A. M. (1997). Visual intelligence. Albany: State University of New York Press. Barsade, S. G., & Gibson, D. E. (2007). Why does affect matter in organizations? Academy of Management Perspectives, 21(1), Baruch, Y. (1999). Response rate in academic studies: A comparative analysis. Human Relations, 52(4), Beech-Hawley, L., Wells, R., & Cole, D. C. (2004). A multi-method approach to assessing deadlines and workload variation among newspaper workers. Work, 23, Beeftink, F., Van Eerde, W., & Rutte, C. G. (in press). The effect of interruptions and breaks on insight and impasses: Do you need a break right now? Creativity Research Journal. Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107(2), Blount, S., & Janicik, G. (2002). Getting and staying in-pace: The "in-synch" preference and its implications for work groups. In Toward phenomenology of groups and group membership (Vol. 4, pp ). New York: Elsevier Science Ltd. Britton, B. K., & Glynn, S. M. (1989). Mental management and creativity: A cognitive model of time management for intellectual productivity. In J. A. Glover, R. R. Ronning & C. R. Reynolds (Eds.), Handbook of Creativity (pp ). New York: Plenum Press. Bryk, A. S., & Raudenbush, S. W. (2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). London: Sage Publications. 118

127 References Butler, D. L., & Thomas, K. M. (1999). Preliminary study of the effectiveness of some heuristics used to generate solutions to ill-structured problems. Psychological Reports, 84(3), Carver, C. S. (2004). Self-regulation of action and affect. In R. F. Baumeister & K. D. Vohs (Eds.), Handbook of self-regulation: Research, theory, and applications (pp ). London: The Guilford Press. Carver, C. S. (2006). Approach, avoidance, and the self-regulation of affect and action. Motivation and Emotion, 30(2), Carver, C. S., & Scheier, M. F. (1982). Control Theory: A useful conceptual framework for personality-social, clinical, and health psychology. Psychological Bulletin, 92(1), Carver, C. S., & Scheier, M. F. (1998). On the self-regulation of behavior: Cambridge University Press. Christensen, B. T., & Schunn, C. D. (2005). Spontaneous access and analogical incubation effects. Creativity Research Journal, 17(2-3), Claessens, B. J. C. (2004a). Individual pacing styles in relation to time management behaviors at work. In Perceived control of time: Time management and personal effectiveness at work (Dissertation) (pp ). Eindhoven, The Netherlands: Technische Universiteit Eindhoven. Claessens, B. J. C. (2004b). Perceived control of time: Time management and personal effectiveness at work: Dissertation Eindhoven University of Technology. Claessens, B. J. C., Van Eerde, W., Rutte, C. G., & Roe, R. A. (2004). Planning behavior and perceived control of time at work. Journal of Organizational Behavior, 25(8), Claessens, B. J. C., Van Eerde, W., Rutte, C. G., & Roe, R. A. (2007). A review of the time management literature. Personnel Review, 36(2), Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). New Jersey: Lawrence Erlbaum Associates, Publishers. Cohen, J. (1992). A Power Primer. Psychological Bulletin, 112(1), Cropley, A. J. (1999). Definitions of creativity. In M. A. Runco & S. R. Pritzker (Eds.), Encyclopedia of Creativity (Vol. 1, pp ). London: Academic Press. 119

128 References Csikszentmihalyi, M. (1996). Creativity: Flow and the psychology of discovery and invention. New York: Harper-Collins. Csikszentmihalyi, M., & Csikszentmihalyi, I. S. (1988a). Introduction to part IV: The measurement of flow in everyday life. In M. Csikszentmihalyi & I. S. Csikszentmihalyi (Eds.), Optimal experience: Psychological studies of flow in consciousness (pp ). New York: Cambridge University Press. Csikszentmihalyi, M., & Csikszentmihalyi, I. S. (1988b). Optimal experience: Psychological studies of flow in consciousness. New York: Cambridge University Press. Csikszentmihalyi, M., & Lefevre, J. (1989). Optimal experience at work and leisure. Journal of Personality & Social Psychology, 56(5), Czerwinski, M., Horvitz, E., & Wilhite, S. (2004, April). A diary study of task switching and interruptions. Paper presented at the CHI, Vienna, Austria. Davis, M. A. (2007). The impact of mood on creativity: A meta-analysis. Paper presented at the Academy of Management Conference, Philadelphia. DeFlillippi, R., Grabher, G., & Jones, C. (2007). Introduction to paradoxes of creativity: Managerial and organizational challenges in the cultural economy. Journal of Organizational Behavior, 28(5), Demerouti, E., Bakker, A. B., de Jonge, J., Janssen, P. P. M., & Schaufeli, W. B. (2001). Burnout and engagement at work as a function of demands and control. Scandinavian Journal of Work Environment & Health, 27(4), Deming, W. E. (2000). The new economics for industry, government, education (2nd ed.). Cambridge, MA: MIT Press. Dewett, T. (2007). Linking intrinsic motivation, risk taking, and employee creativity in an R&D environment. R&D Management, 37(3), Dey, E. L. (1997). Working with low survey response rates: The efficacy of weighting adjustments. Research in Higher Education, 38(2), Dijksterhuis, A., & Meurs, T. (2006). Where creativity resides: The generative power of unconscious thought. Consciousness and Cognition, 15(1),

129 References Dodds, R. A., Ward, T. B., & Smith, K. G. (in press). A review of experimental research on incubation in problem solving and creativity. In M. A. Runco (Ed.), Creativity research handbook. Cresskill, NJ: Hampton Press. Dow, G. T., & Mayer, R. E. (2004). Teaching students to solve insight problems: Evidence for domain specificity in creativity training. Creativity Research Journal, 16(4), Drazin, R., Glynn, M., & Kazanjian, R. (1999). Multilevel theorizing about creativity in organizations: A sense-making perspective. Academy of Management Review, 24(2), Eindhoven, J. E., & Vinacke, W. E. (1952). Creative processes in painting. Journal of General Psychology, 47, Eldredge, N., & Gould, S. J. (1972). Punctuated equilibria: An alternative to phyletic gradualism. In S. T. J. (Ed.), Models in paleobiology (pp ). San Francisco: Freeman, Cooper and Co. Elsbach, K. D., & Hargadon, A. B. (2006). Enhancing creativity through "mindless" work: A framework of workday design. Organization Science, 17(4), Eyrolle, H., & Cellier, J. M. (2000). The effects of interruptions in work activity: field and laboratory results. Applied Ergonomics, 31(5), Faul, F., Erdfelder, E., Lang, A. G., & Buchner, A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39(2), Feist, G. J. (1998). A meta-analysis of personality in scientific and artistic creativity. Personality and Social Psychology Review, 2(4), Feist, G. J. (1999). The influence of personality on artistic and scientific creativity. In R. J. Sternberg (Ed.), Handbook of creativity (pp ). Cambridge: Cambridge University Press. Fisher, C. D. (1998). Effects of external and internal interruptions on boredom at work: two studies. Journal of Organizational Behavior, 19(5), Fleck, J. I., & Weisberg, R. W. (2004). The use of verbal protocols as data: An analysis of insight in the candle problem. Memory & Cognition, 32(6),

130 References Gardner, D. G. (1986). Activation Theory and task design: An empirical test of several new predictions. Journal of Applied Psychology, 71(3), Gardner, D. G., & Cummings, L. L. (1988). Activation Theory and job design - Review and reconceptualization. Research in Organizational Behavior, 10, George, J. M., & Zhou, J. (2001). When openness to experience and conscientiousness are related to creative behavior: An interactional approach. Journal of Applied Psychology, 86(3), Gersick, C. J. G. (1988). Time and transition in work teams: Toward a new model of group development. Academy of Management Journal, 31(1), Gersick, C. J. G. (1989). Marking time: Predictable transitions in task groups. Academy of Management Journal, 32(2), Gevers, J. M. P., Claessens, B. J. C., Van Eerde, W., & Rutte, C. G. (in press). Pacing styles, personality and performance. In R. A. Roe, M. J. Waller & S. Clegg (Eds.), Time in organizations: Approaches and methods. London: Routledge. Gevers, J. M. P., Mohammed, S., Baytalskaya, N., & Beeftink, F. (2008). Measuring pacing preferences: The development and validation of a new measurement instrument. Paper presented at the Academy of Management Conference, Anaheim, California. Gevers, J. M. P., Rutte, C. G., & Van Eerde, W. (2006). Meeting deadlines in work groups: Implicit and explicit mechanisms. Applied Psychology: An International Review, 55(1), Gilson, L. L., & Shalley, C. E. (2004). A little creativity goes a long way: An examination of teams' engagement in creative processes. Journal of Management, 30(4), Goldschmidt, G. (1991). The dialectics of sketching. Creativity Research Journal, 4(2), Goldschmidt, G. (1999). Design. In M. A. Runco & S. R. Pritzker (Eds.), Encyclopedia of creativity (Vol. 1, pp ). New York: Academic Press. Green, K. E., Boser, J. A., & Hutchinson, S. R. (1998). Response-rate differences and response-enhancement effects by population type. Psychological Reports, 83(1),

131 References Green, P., & Skinner, D. (2005). Does time management training work? An evaluation. International Journal of Training and Development, 9(2), Hargadon, A. B., & Bechky, B. A. (2006). When collections of creatives become creative collectives: A field study of problem solving at work. Organization Science, 17(4), Harris, C., & Daniels, K. (2005). Daily affect and daily beliefs. Journal of Occupational Health Psychology, 10(4), Hayes, J. R. (1989). Cognitive processes in creativity. In J. A. Glover, R. R. Ronning & C. R. Reynolds (Eds.), Handbook of creativity (pp ). New York: Plenum Press. Hofmann, D. A., & Gavin, M. B. (1998). Centering decisions in hierarchical linear models: Implications for research in organizations. Journal of Management, 24(5), Hofmann, D. A., Jacobs, R., & Baratta, J. E. (1993). Dynamic Criteria and the Measurement of Change. Journal of Applied Psychology, 78(2), Hofmann, D. A., Jacobs, R., & Gerras, S. J. (1992). Mapping Individual-Performance over Time. Journal of Applied Psychology, 77(2), Ilies, R., & Judge, T. A. (2005). Goal regulation across time: The effects of feedback and affect. Journal of Applied Psychology, 90(3), Isaksen, S. G., Dorval, K. B., & Treffinger, D. J. (2000). Creative approaches to problem solving (2nd ed.). Dubuque, LA: Kendell/ Hunt. Isaksen, S. G., & Treffinger, D. J. (2004). Celebrating 50 years or reflective practice: Versions of creative problem solving. Journal of Creative Behavior, 38(2), James, K., Brodersen, M., & Eisenberg, J. (2004). Workplace affect and workplace creativity: A review and preliminary model. Human Performance, 17(2), Janssen, O. (2001). Fairness perceptions as a moderator in the curvilinear relationships between job demands, and job performance and job satisfaction. Academy of Management Journal, 44(5),

132 References Jaussi, K. S., Randel, A. E., & Dionne, S. D. (2007). I am, I think I can, and I do: The role of personal identity, self-efficacy, and cross-application of experiences in creativity at work. Creativity Research Journal, 19(2-3), Jett, Q. R., & George, J. M. (2003). Work interrupted: A closer look at the role of interruptions in organizational life. Academy of Management Review, 28(3), Jex, S. M., & Elacqua, T. C. (1999). Time management as a moderator of relations between stressors and employee strain. Work and Stress, 13(2), Jöreskog, K. G., & Sörbom, D. (1993). LISREL 8: Structural equation modeling with the SIMPLIS command language. Chicago: Scientific Software International. Judd, C. M., Smith, E. R., & Kidder, L. H. (1991). Research methods in social science (6th ed.). Fort Worth: Holt, Rinehart, and Winston. Judge, T. A., & Bono, J. E. (2001). Relationship of core self-evaluations traits - self-esteem, generalized self-efficacy, locus of control, and emotional stability - with job satisfaction and job performance: A meta-analysis. Journal of Applied Psychology, 86(1), Kahn, W. A. (1990). Psychological conditions of personal engagement and disengagement at work. Academy of Management Journal, 33(4), Kirmeyer, S. L. (1988). Coping with competing demands - Interruption and the Type-A pattern. Journal of Applied Psychology, 73(4), Kirton, M. (1976). Adaptors and innovators. Journal of Applied Psychology, 61, Kirton, M. (Ed.). (1994). Adaptors and innovators: Styles of creativity and problem solving (Revised ed.). London: Routledge. Kline, R. B. (1998). Principles and practice of structural equation modeling. New York: The Guilford Press. König, C. J., & Kleinmann, M. (2005). Deadline rush: A time management phenomenon and its mathematical description. Journal of Psychology, 139(1), König, C. J., Kleinmann, M., & Höhmann, W. (2004). Is a quiet hour an effective time management technique? Paper presented at the Academy of Management, New Orleans. 124

133 References Koput, K. W. (1997). A chaotic model of innovative search: Some answers, many questions. Organization Science, 8(5), Landis, J. R., & Koch, G. G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33, Landy, F., Rastegary, H., Thayer, J., & Colvin, C. (1991). Time urgency - The construct and its measurement. Journal of Applied Psychology, 76(5), Lefevre, J. (1988). Flow and the quality of experience during work and leisure. In M. Csikszentmihalyi & I. S. Csikszentmihalyi (Eds.), Optimal experience: Psychological studies of flow in consciousness (pp ). New York: Cambridge University Press. Lim, S. G.-S., & Murnighan, J. K. (1994). Phases, deadlines, and the bargaining process. Organizational Behavior and Human Decision Processes, 58, Locke, E. A., & Latham, G. P. (1990). A theory of goal setting and task performance. New Jersey: Prentice-Hall. Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of goal setting and task motivation - A 35-year odyssey. American Psychologist, 57(9), Louro, M. J., Pieters, R., & Zeelenberg, M. (2007). Dynamics of multiple-goal pursuit. Journal of Personality and Social Psychology, 93(2), Macan, T. H. (1994). Time management: Test of a process model. Journal of Applied Psychology, 79(3), Macan, T. H., Shahani, C., Dipboye, R. L., & Phillips, A. P. (1990). College-students time management: Correlations with academic performance and stress. Journal of Educational Psychology, 82(4), Madjar, N., & Oldham, G. R. (2006). Task rotation and polychronicity: Effects on individuals' creativity. Human Performance, 19(2), Mainemelis, C. (2002). Time and timelessness: Creativity in (and out of) the temporal dimension. Creativity Research Journal, 14(2), Mathieu, J. E., & Taylor, S. R. (2006). Clarifying conditions and decision points for mediational type inferences in organizational behavior. Journal of Organizational Behavior, 27(8),

134 References Mathieu, J. E., & Taylor, S. R. (2007). A framework for testing meso-mediational relationships in organizational behavior. Journal of Organizational Behavior, 28(2), Miron, E., Erez, M., & Naveh, E. (2004). Do personal characteristics and cultural values that promote innovation, quality, and efficiency compete or complement each other? Journal of Organizational Behavior, 25(2), Nandhakumar, J., & Jones, M. (2001). Accounting for time: Managing time in projectbased team working. Accounting, Organizations and Society, 26, Norman, D. A., & Bobrow, D. G. (1975). Data-limited and resource-limited processes. Cognitive Psychology, 7(1), Ohly, S., Sonnentag, S., & Pluntke, F. (2006). Routinization and its relationship with creative and proactive outcomes. Journal of Organizational Behavior, 27, Oldham, G. R., & Cummings, A. (1996). Employee creativity: Personal and contextual factors at work. Academy of Management Journal, 39(3), Olton, R. M. (1979). Experimental studies of incubation: Searching for the elusive. Journal of Creative Behavior, 13, Orlikowski, W. J., & Yates, J. (2002). It's about time: Temporal structuring in organizations. Organization Science, 13(6), Ormerod, T. C. (2005). Planning and ill-defined problems. In R. Morris & G. Ward (Eds.), The Cognitive Psychology of Planning (pp ). London: Psychology Press. Orpen, C. (1994). The effect of time-management training on employee attitudes and behavior - A field experiment. Journal of Psychology, 128(4), Osborn, A. F. (1952). Wake up your mind: 101 ways to develop creativeness. New York: Scribners. Osburn, H. K., & Mumford, M. D. (2006). Creativity and planning: Training interventions to develop creative problem-solving skills. Creativity Research Journal, 18(2), Patrick, C. (1938). Scientific thought. The Journal of Psychology, 5,

135 References Peeters, M. A. G., & Rutte, C. G. (2005). Time management behavior as a moderator for the job demand-control interaction. Journal of Occupational Health Psychology, 10(1), Peeters, M. A. G., Van Tuijl, H. F. J. M., Reymen, I. M. M. J., & Rutte, C. G. (2007). The development of a design behaviour questionnaire for multidisciplinary teams. Design Studies, 28, Quinn, R. W. (2005). Flow in knowledge work-high performance experience in the design of national security technology. Administrative Science Quarterly, 50(4), Redmond, M. R., Mumford, M. D., & Teach, R. (1993). Putting creativity to work: Effects of leader-behavior on subordinate creativity. Organizational Behavior and Human Decision Processes, 55(1), Reymen, I. M. M. J., Hammer, D. K., Kroes, P. A., Van Aken, J. E., Dorst, K., Bax, M. F. T., et al. (2006). A domain-independent descriptive model and its application to structured reflection on design processes. Research in Engineering Design, 16(4), Roe, R. A. (2008). Time in applied psychology: The study of "what happens" rather than "what is". European Psychologist, 13(1), Rothbard, N. P. (2001). Enriching or depleting? The dynamics of engagement in work and family roles. Administrative Science Quarterly, 46(4), Runco, M. A. (1999). Time. In M. A. Runco & S. R. Pritzker (Eds.), Encyclopedia of creativity (Vol. 2, pp ). New York: Academic Press. Runco, M. A., & Sakamoto, S. O. (1999). Experimental studies of creativity. In R. J. Sternberg (Ed.), Handbook of creativity (pp ). Cambridge: Cambridge University Press. Ruscio, A. M., Whitney, D., & Amabile, T. M. (1998). Looking inside the fishbowl of creativity: Verbal and behavioral predictors of creative performance. Creativity Research Journal, 11(3), Schaufeli, W. B., Bakker, A. B., & Salanova, M. (2006). The measurement of work engagement with a short questionnaire - A cross-national study. Educational and Psychological Measurement, 66(4),

136 References Schmidt, A. M., & DeShon, R. P. (2007). What to do? The effects of discrepancies, incentives, and time on dynamic goal prioritization. Journal of Applied Psychology, 92(4), Schriber, J. B., & Gutek, B. A. (1987). Some time dimensions of work: Measurement of an underlying aspect of organizational culture. Journal of Applied Psychology, 72(4), Schyns, B., & Von Collani, G. (2002). A new occupational self-efficacy scale and it's relation to personality constructs and organizational variables. European Journal of Work and Organizational Psychology, 11, Seers, A., & Woodruff, S. (1997). Temporal pacing in task forces: Group development or deadline pressure. Journal of Management, 23(2), Segal, E. (2004). Incubation in insight problem solving. Creativity Research Journal, 16(1), Seifert, C. M., Meyer, D. E., Davidson, N., Patalano, A. L., & Yaniv, I. (1995). Demystification of cognitive insight: Opportunistic assimilation and the preparedmind perspective. In R. J. Sternberg (Ed.), The nature of insight. Cambridge, MA: The MIT Press. Shalley, C. E., Zhou, J., & Oldham, G. R. (2004). The effects of personal and contextual characteristics on creativity: Where should we go from here? Journal of Management, 30(6), Shin, S. J., & Zhou, J. (2003). Transformational leadership, conservation, and creativity: Evidence from Korea. Academy of Management Journal, 46(6), Sio, U. N., & Rudowicz, E. (2007). The role of an incubation period in creative problem solving. Creativity Research Journal, 19(2-3), Smith, S. M. (1995). Fixation, incubation, and insight in memory and creative thinking. In S. M. Smith, T. B. Ward & R. A. Finke (Eds.), The creative cognition approach. Cambridge, MA: MIT Press. Smith, S. M., & Blankenship, S. E. (1991). Incubation and the persistence of fixation in problem-solving. American Journal of Psychology, 104(1), Smith, S. M., & Dodds, R. A. (1999). Incubation. In M. A. Runco & S. R. Pritzker (Eds.), Encyclopedia of creativity (Vol. 2, pp ). London: Academic Press. 128

137 References Snijders, T. A. B., & Bosker, R. J. (1999). Multilevel analysis: An introduction to basic and advanced multilevel modeling. London: Sage Publications. Sonnentag, S. (1998). Expertise in professional software design: A process study. Journal of Applied Psychology, 83(5), Sonnentag, S. (2003). Recovery, work engagement, and proactive behavior: A new look at the interface between nonwork and work. Journal of Applied Psychology, 88(3), Sonnentag, S., & Zijlstra, F. R. H. (2006). Job characteristics and off-job activities as predictors of need for recovery, well-being, and fatigue. Journal of Applied Psychology, 91(2), Stajkovic, A. D., & Luthans, F. (1998). Self-efficacy and work-related performance: A meta-analysis. Psychological Bulletin, 124(2), Sternberg, R. J., & Davidson, J. E. (1999). Insight. In M. A. Runco & S. R. Pritzker (Eds.), Encyclopedia of creativity (Vol. 2, pp ). London: Academic Press. Stewart, G. L., & Nandkeolyar, A. K. (2006). Adaptation and intraindividual variation in sales outcomes: Exploring the interactive effects of personality and environmental opportunity. Personnel Psychology, 59(2), Stewart, G. L., & Nandkeolyar, A. K. (2007). Exploring how constraints created by other people influence intraindividual variation in objective performance measures. Journal of Applied Psychology, 92(4), Strauss, A., & Corbin, J. (1990). Basics of qualitative research: Grounded theory procedures and techniques. London: Sage Publications Ltd. Teuchmann, K., Totterdell, P., & Parker, S. K. (1999). Rushed, unhappy, and drained: An experience sampling study of relations between time pressure, perceived control, mood, and emotional exhaustion in a group of accountants. Journal of Occupational Health Psychology, 4(1), Thompson, P., Jones, M., & Warhurst, C. (2007). From conception to consumption: Creativity and the missing managerial link. Journal of Organizational Behavior, 28(5),

138 References Tierney, P., & Farmer, S. M. (2002). Creative self-efficacy: Its potential antecedents and relationship to creative performance. Academy of Management Journal, 45(6), Tierney, P., & Farmer, S. M. (2004). The Pygmalion process and employee creativity. Journal of Management, 30(3), Tierney, P., Farmer, S. M., & Graen, G. B. (1999). An examination of leadership and employee creativity: The relevance of traits and relationships. Personnel Psychology, 52(3), Totterdell, P., Wood, S., & Wall, T. (2006). An intra-individual test of the demands-control model: A weekly diary study of psychological strain in portfolio workers. Journal of Occupational and Organizational Psychology, 79, Tripoli, A. M. (1998). Planning and allocating: Strategies for managing priorities in complex jobs. European Journal of Work and Organizational Psychology, 7, Tschang, F. T. (2007). Balancing the tensions between rationalization and creativity in the video games industry. Organization Science, 18(6), Unsworth, K. L. (2001). Unpacking creativity. Academy of Management Review, 26(2), Valacich, J. S., Dennis, A. R., & Connolly, T. (1994). Idea generation in computer-based groups: A new ending to an old story. Organizational Behavior and Human Decision Processes, 57(3), Van Dyne, L., Jehn, K. A., & Cummings, A. (2002). Differential effects of strain on two forms of work performance: individual employee sales and creativity. Journal of Organizational Behavior, 23(1), Van Eerde, W. (2000). Procrastination: Self-regulation in initiating aversive goals. Applied Psychology - An International Review / Psychologie Appliquee - Revue Internationale, 49(3), Van Eerde, W. (2003a). A meta-analytical derived nomological network of procrastination. Personality and Individual Differences, 35, Van Eerde, W. (2003b). Procrastination at work and time management training. Journal of Psychology, 137(5),

139 References Van Eerde, W., Holman, D. J., & Totterdell, P. (2005). Editorial: Special section: Diary studies in work Psychology. Journal of Occupational and Organizational Psychology, 78, Van Oorschot, K. E., Bertrand, J. W. M., & Rutte, C. G. (2005). Field studies into the dynamics of product development tasks. International Journal of Operations & Production Management, 25(8), Vancouver, J. B., & Day, D. V. (2005). Industrial and organization research on selfregulation: From constructs to applications. Applied Psychology-an International Review-Psychologie Appliquee-Revue Internationale, 54(2), Vohs, K. D., & Baumeister, R. F. (2004). Understanding self-regulation: An introduction. In R. F. Baumeister & K. D. Vohs (Eds.), Handbook of self-regulation: Research, theory, and applications (pp. 1-9). New York: Guilford Press. Wallas, G. (1926). The art of thought. New York: Harcourt. Waller, M. J., Zellmer-Bruhn, M. E., & Giambatista, R. C. (2002). Watching the clock: Group pacing behavior under dynamic deadlines. Academy of Management Journal, 45(5), Weisberg, R. W. (1995). Prolegomena to theories of insight in problem solving: A taxonomy of problems. In R. J. Sternberg & J. E. Davidson (Eds.), The nature of insight (pp ). London: The MIT Press. Wells, D. H. (1996). Forced incubation. Creativity Research Journal, 9(4), Wickramasinghe, N., & Von Lubitz, D. (2007). Knowledge based enterprise: Theories and fundamentals. London: Idea Group Publishing. Yakura, E. (2002). Charting time: Timelines as temporal boundary objects. Academy of Management Journal, 45(5), Zhou, J. (2003). When the presence of creative coworkers is related to creativity: Role of supervisor close monitoring, developmental feedback, and creative personality. Journal of Applied Psychology, 88(3), Zijlstra, F. R. H., Roe, R. A., Leonora, A. B., & Krediet, I. (1999). Temporal factors in mental work: Effects of interrupted activities. Journal of Occupational and Organizational Psychology, 72,

140

141 Summary Today's economy depends more and more on innovation, and therefore on the work of creative professionals. Science has picked up on this development and researchers increasingly investigate the antecedents of creativity at work. However, some interesting and highly relevant research topics have been understudied, for example, the role of time in creative work. This dissertation addressed this topic, and, specifically, the role of selfregulation behaviors focused at managing one s time at work, in more popular terms: time management. The dissertation presented five studies (one experiment and four field studies) in which particular emphasis was put on incubation time, task discretion, pacing patterns, deadlines, and self-regulation behaviors such as prioritizing, planning, and monitoring creative work activities. First, the role of the 'self' in self-regulation was investigated. It was tested whether individuals are more creative when they have personal discretion as to how to use their time. The results of an experiment among 109 students of industrial engineering and management science have confirmed this. Individuals that could decide themselves when to work on another task for a while (i.e., could take incubation time at their own discretion) conceived more insights to solve the creative problems at hand and reached impasses less often (i.e., felt less often 'stuck') compared to others for whom it was decided when to take their incubation time. Thus, the 'self' in self-regulation has been proven to be of relevance. Second, the role of pacing (i.e., the allocation of efforts for the execution of tasks over time) in creative work was explored. In a survey study among 88 individuals of 133

142 Summary different professions we found that individuals in creative professions more often used 'incubation time' pacing patterns, which included an intermediate period of disengagement from a task or project that may have been used as incubation time. They were also more often 'deadline workers' and less often 'early workers', which means that they generally did little of the work in the beginning of a task or project, and much of it towards the deadline. Thus, individuals in creative professions have distinct ways to distribute their work efforts over time compared to individuals in other professions. Furthermore, an in-depth interview study among 25 architects was conducted through which we explored pacing patterns over the course of the design phases of their projects. The results showed that almost all architects used the incubation time pacing pattern in their projects. Furthermore, this pattern was repeated several times within one project, specifically, it occurred before every project milestone and deadline. In addition, architects worked on multiple projects at the same time, which means that pacing patterns of these projects overlapped: Working on one project matched periods of disengagement from other projects. Finally, the study showed that some of the architects did not set strict project milestones and deadlines nor adhered to them very strictly. Whereas some architects accepted the deadlines that were imposed upon them by their clients, others negotiated about their deadlines, and yet others did not accept strict deadlines at all. Also, our interviews revealed deadline postponement and quality-time trade-offs. Thus, deadlines that mark the ends of the pacing patterns may be rather flexible. Overall, this study has provided some valuable insights into the ways in which a preference for the incubation time pacing pattern fits in with some of the complexities of creative work (project milestones, deadlines, working on multiple projects). Third, it was examined whether time management is beneficial to creativity at work. A diary study was conducted among 68 junior R&D engineers in which we investigated whether the tendency of individuals to use time management in their work was related to being more creative on a day-to-day basis. As hypothesized, this relationship existed. Thus, this study provided empirical evidence of the benefit of time management in creative work. Furthermore, this study showed that this positive effect of time management was unrelated and complementary to the positive effect of having an innovative cognitive style. We also found that, next to this direct effect, time management had an indirect effect 134

143 Summary on daily creativity via daily positive affect, meaning that individuals were more creative on a certain work day when they felt better. Finally, the results showed that time management was related to a stable creative performance over time. Fourth, a survey among 339 architects investigated to what extent self-regulation behaviors (setting priorities, planning work activities, and monitoring time and task progress) were related to the success of creative professionals. Specifically, this study investigated the influence of innovative cognitive style, self-efficacy, and the three selfregulation behaviors on the success of the architects as designers and businesspersons. Similarly to the diary study, this study suggests that innovative style and self-regulation of work time are complementary positive factors in creative work. However, this survey showed that innovative style predominantly influenced the design aspect of the work of the architects, and self-regulation the business aspect. In addition, it showed that self-efficacy explained spillover effects. Thus, this dissertation revealed that there is more to being a creative professional than only being innovative: In addition, regulating your time at work is important. 135

144

145 Samenvatting De huidige economie is in een steeds grotere mate afhankelijk van innovatie en daarom van mensen met creatief werk. De wetenschap is deze ontwikkeling niet ontgaan: er wordt steeds meer onderzoek gedaan naar de antecedenten van creativiteit op het werk. Toch zijn er bepaalde relevante en interessante onderwerpen die nog maar weinig onderzocht zijn, zoals de rol van tijd in creatief werk. Dit proefschrift betreft dit onderwerp en dan met name de rol van zelfregulatiegedragingen (self-regulation behaviors) die erop gericht zijn de eigen tijd op het werk te beheersen, in meer populaire termen; de rol van time management in creatief werk. Het proefschrift bevat vijf studies (een experiment en vier veldstudies) waarin het onder andere gaat over incubatietijd, keuzevrijheid bij het werken aan taken (task discretion), patronen in het verdelen van werklast over de tijd (pacing patterns), deadlines en zelfregulatiegedragingen zoals prioriteiten stellen, plannen en het in de gaten houden van de voortgang van het werk (monitoring) tijdens het uitvoeren van creatieve taken. Eerst werd de zelf in zelfregulatie onderzocht: We hebben getoetst of mensen creatiever zijn als ze zelf invloed hebben op hoe ze hun tijd indelen. De resultaten van een experiment onder 109 Technische Bedrijfskunde studenten bevestigen dit. Studenten die zelf konden beslissen wanneer ze even aan iets anders werkten (d.w.z., zelf invloed hadden op het moment dat ze incubatietijd namen) kregen meer ingevingen om de creatieve taken die hen waren voorgelegd op te lossen en raakten minder vaak in een impasse (d.w.z., ze hadden minder vaak het gevoel dat ze vast zaten) vergeleken met anderen voor wie beslist 137

146 Samenvatting werd wanneer ze incubatietijd moesten nemen. Dus, dit bewijst dat de zelf in zelfregulatie relevant is. Een tweede onderwerp van studie was de rol van pacing in creatief werk. Uit een vragenlijstonderzoek onder 88 mensen met allerlei verschillende beroepen bleek dat mensen met creatieve beroepen vaker incubatietijd patronen volgen om hun werk over de tijd te verdelen; patronen die tussentijds een periode bevatten waarin niet aan de taak gewerkt wordt die als incubatietijd gebruikt kan worden. Ze zijn ook vaker deadline werkers en minder vaak vroege werkers, wat betekent dat ze over het algemeen relatief weinig werk verzetten aan het begin van een taak of project en veel vlak voor de deadline. Dus, uit dit onderzoek blijkt dat mensen met creatieve beroepen een andere manier van werken hebben als het gaat om het verdelen van hun werklast over de tijd vergeleken bij mensen met andere beroepen. Verder voerden we een studie uit waarbij we diepte-interviews hielden met 25 architecten. Hiermee onderzochten we pacing patronen gedurende de ontwerpfase van projecten. De resultaten laten zien dat bijna alle architecten in hun projecten het incubatietijd patroon gebruikten. Ook laten ze zien dat dit patroon meerdere malen voor kwam binnen één project, het herhaalde zich namelijk voorafgaande aan elke projectmijlpaal of deadline. Daarnaast werkten architecten aan meerdere projecten tegelijkertijd, wat betekent dat pacing patronen van verschillende projecten overlappen: Terwijl er aan het ene project gewerkt wordt liggen andere projecten even stil. Ten slotte liet deze studie ook zien dat sommige architecten niet erg strikte projectmijlpalen of deadlines afspraken, noch heel erg strikt waren in het naleven ervan. Hoewel bepaalde architecten de deadlines accepteerden die hen door hun klanten opgelegd werden, waren er anderen die onderhandelden over deze deadlines en weer anderen die voor hun werk helemaal geen strikte deadlines accepteerden. Verder bleek uit de interviews dat architecten soms deadlines uitstelden of dat het halen van een deadline ten koste ging van de kwaliteit van hun werk. Hieruit kunnen we concluderen dat deadlines die het einde van een pacing patroon markeren misschien enige mate van flexibiliteit bevatten. Over het geheel genomen heeft de interviewstudie waardevolle inzichten gegeven in de manier waarop in een creatief beroep een voorkeur voor het incubatietijd pacing patroon gecombineerd wordt met de 138

147 Samenvatting complexiteit van de praktijk (zoals projectmijlpalen, deadlines en aan meerdere projecten tegelijkertijd werken). Daarna werd onderzocht of time management in positieve zin bijdraagt aan creativiteit. We voerden een dagboekstudie uit onder 68 jonge R&D ingenieurs waarbij we uitzochten of mensen die de neiging hebben time management te gebruiken in hun werk ook creatiever zijn. In lijn met onze hypothese bleek deze relatie te bestaan. Deze studie levert dus het empirische bewijs dat time management gunstig is voor creatief werk. Bovendien liet de studie zien dat de positieve invloed van time management onafhankelijk is van, en complementair is aan, de positieve invloed van het hebben van een innovatieve denkstijl (innovative cognitive style). Naast dit directe effect vonden we ook dat time management een positief indirect effect op creativiteit heeft omdat time management een positieve invloed heeft op het gevoel (positive affect). Dit wil zeggen dat mensen die time management toepassen creatiever zijn tijdens hun werkdag mede omdat ze zich beter voelen. Tenslotte blijkt ook uit de resultaten dat time management gerelateerd is aan een stabielere creatieve prestatie over de dagen heen. Ten slotte werd een vragenlijstonderzoek gedaan onder 339 architecten. Hierin onderzochten we in welke mate zelfregulatiegedragingen ((a) het stellen van prioriteiten, (b) het plannen van werktaken en (c) het in de gaten houden van de tijd en voortgang tijdens het uitvoeren van taken (monitoring)) samenhingen met het succes van mensen met creatief werk. Om precies te zijn, deze studie onderzocht de invloed van een innovatieve denkstijl, van het vertrouwen om problemen op te kunnen lossen (self-efficacy) en van de drie eerder genoemde zelfregulatiegedragingen op het succes van architecten als ontwerpers en zakenmensen. Net als bij de dagboekstudie blijkt uit dit vragenlijstonderzoek dat het hebben van een innovatieve denkstijl en zelfregulatie factoren zijn die beiden een positieve invloed hebben op creatief werk en elkaar aanvullen. Toch laat deze studie zien dat een innovatieve denkstijl vooral van invloed is op het ontwerpaspect van het werk en zelfregulatie op het zakelijke aspect. Verder blijkt dat zowel een innovatieve stijl als zelfregulatie een positief effect hebben op het vertrouwen mogelijke problemen op te kunnen lossen, waardoor ze beiden ook indirect een effect hebben op andere aspecten van het werk (innovatieve denkstijl op zakelijk succes en zelfregulatie op ontwerpsucces). 139

148 Samenvatting Al met al blijkt uit dit proefschrift dat er meer nodig is dan alleen innovativiteit in creatief werk: Daarnaast is de regulatie van de eigen tijd op het werk belangrijk. 140

149 About the Author Flora Beeftink was born in Pannerden (The Netherlands) on December the 4 th in She completed her pre-university education (VWO) in Zevenaar at the Liemers College in In 2004 she received her Master of Science degree in Industrial Engineering and Management Science (Technische Bedrijfskunde) from Eindhoven University of Technology. Her master thesis concerned an intervention study to the effects of a time management tool among 64 employees of an electro-engineering company. After graduation she stayed at Eindhoven University of Technology and started to work on a PhD project regarding time management in creative professions. This dissertation is the result of that work. Currently, Flora works at Tilburg University as a postdoctoral researcher at the Department of Social Psychology doing research in the field of work and organizational psychology, specifically, she continues to do research on creativity and time at work. 141