The Psychological Record, 1999, 49, 3-14 EFFECTS OF EXTRINSIC REINFORCEMENT FOR READING DURING CHILDHOOD ON REPORTED READING HABITS OF COLLEGE STUDENTS STEPHEN R. FLORA Youngstown State University DAVID B. FLORA University of North Carolina The effects of participation in the "Book it!" reading program and parental pay for reading on reported reading habits of college students was investigated. College students were surveyed about the amount they read per week, their intrinsic interest in reading, if they participated in the Book It! program and if their parents paid them money to read during childhood. If they participated in Book It! and/or were paid to read, they were also asked what effects these factors had on their learning to read, their enjoyment of reading, and on the amount they read. Neither being reinforced with money or pizzas increased or decreased the amount college students read nor influenced their intrinsic motivation for reading. Answers to direct questions about Book It! and parental pay for reading suggest that when a child is extrinsically reinforced for reading the child will increase the amount read, enjoyment of reading may increase, and if they do not yet know how to read fluently, the programs may help the child to learn to read. These results provide no support for the myth that extrinsic rewards for reading undermine intrinsic interest in reading. Rather, extrinsic rewards for reading set the conditions where intrinsic motivation for reading may develop. Any concerns that reinforcement programs for reading will decrease later reading behaviors are unfounded. Behavior is a function of its consequences. Consequences that increase the rate of behavior are reinforcers. Increasing the rate of a behavior with reinforcers is the process of reinforcement. If a child is given a pizza for reading and consequently reads more often, then that child's reading behavior has been reinforced. If a child reads about fighting dragons and riding dolphins and consequently reads more, then reading has been reinforced by the stories that the reading reveals. Likewise, if a child reads a book on chess strategies, begins to beat his or her chess opponents, and consequently reads more chess strategy Correspondence concerning this article should be addressed to Stephen R. Flora, Department of Psychology, Youngstown State University, Youngstown, OH 44555. Electronic mail may be sent via Internet to srflora@cc.ysu.edu.
4 FLORA AND FLORA books, then that child's reading behavior has also been reinforced. In these examples reading has been reinforced by the consequences that reading produces. Psychology uses many terms to describe situations involving reinforcement. For example "intrinsically motivated behavior is said to be demonstrated when people engage in an activity primary for its own sake, whereas extrinsically motivated behavior is controlled by incentives that are not part of the activity" (Eisenberger & Cameron, 1996, p. 1154). Consistent with this jargon, in the first example the pizza was an "extrinsic reward." In the second example reading was "intrinsically motivating" because there was no contrived reinforcement for reading. The chess example is ambiguous as to whether reading was "intrinsically" or "extrinsically" motivating. Nothing was presented or given to the child for reading; this makes the motivation seem intrinsic. However, reading did not occur for 'its own sake,' but to help the reader defeat chess opponents, a purpose removed from the reading process itself From worker productivity in steel mills (Gleisser, 1998) to merit pay and performance-based funding for public universities (Carnevale, Johnson, & Edwards, 1998), the use of rewards to motivate performance is a fact of life in society (Daniels, 1994). Hundreds of studies with human subjects have revealed many beneficial effects of reinforcement programs (e.g., Martin & Pear, 1996; Miller, 1997). Despite the ubiquitousness of reinforcement and the documented benefits of reinforcement programs, a widespread myth (see Eisenberger & Cameron, 1996) exists in psychology and education that assumes if one engages in a task such as reading for extrinsic rewards like pizza or money, then any intrinsic motivation or interest in the activity will be undermined. That is, according to the myth that 'extrinsic motivation undermines intrinsic motivation,' the child who is given pizzas or money for reading will never want read to find out about dragons, dolphins, or chess strategy. The child will never 'read for reading's sake.' However, several theoretical reviews (Carton, 1996; Dickinson, 1989; Flora, 1990) and meta-analyses (Cameron & Pierce, 1994; Eisenberger & Cameron, 1996) of the empirical literature all have concluded the myth is just that, a myth. The belief that extrinsic reinforcement undermines intrinsic motivation is widely held but false in all but the most circumscribed, unrealistic situations. A slightly detrimental effect of extrinsic rewards sometimes occurs in artificial laboratory situations when tangible rewards like money (as opposed to nontangible rewards like praise). are given regardless of performance (Cameron & Pierce 1994). For example, if a child is given money for reading regardless of how little or how accurately, there is a slight chance that the child will read less in the future. Rewarding a child for reading irrespective of how how little, or inaccurately the child reads is likely a very rare occurrence and an obviously inappropriate educational practice. The actual practice of using rewards in applied settings has not been shown to have detrimental effects on task interest.
READING INCENTIVES 5 Although the empirical evidence (e.g., Cameron & Pierce, 1994, 1996; Eisenberger & Cameron, 1996) has convinced many that rewards do not undermine intrinsic interest, but instead set the conditions where intrinsic interest may develop, others (e.g., Kahn, 1993) remain just as convinced that all rewards are inherently detrimental to the educational process. If rewarding children for acquisition of skills and competencies such as reading is beneficial to children, education, and society, then it is important to examine programs that currently exist so that their benefits can be maximized. Conversely, if rewards are detrimental, then it is just as critical to examine educational reward programs to advise on their termination or elimination of the reward portion of the program. "If a child receives repeated payments for reading... the more important question concerns the long-term effects of reward on intrinsic interest" (Eisenberger & Cameron, 1996, pp. 1163-1164). The current study examined possible long-term effects of reward (in the form of pizzas and/or money) for reading during childhood on later intrinsic interest in reading and amount of reading by college students. "Book It!" is a program run by Pizza Hut that rewards children in Grades K through 6 with a free individual pan pizza for meeting reading goals set by their teacher. When a child reaches the goal, the teacher provides the child with a certificate that the child can redeem at any Pizza Hut for the free pizza. According to Book It!, during the 1995-1996 school year over 22 million children in Australia, Canada, and the United States participated in the program. If rewards undermine intrinsic interest, then during the 1995-1996 school year, the reading habits of over 22 million children were aversely affected. If reinforcing reading sets the conditions where intrinsic interest in reading may develop (Flora, 1990), then Book It! may improve childhood and subsequent adult literacy. Likewise, parents who pay their children to read may be either harming or helping their children's possible interest in reading depending on the truth or falsity of the myth that rewards undermine intrinsic interest. Because the Book It! program began in 1985, many current college students participated in Book It! when they were children. Therefore it is possible to assess the effects, if any, that participating in Book It! may have had on the students' reading habits and intrinsic interest in reading at the college level. The current study surveyed college students on the amount they read per week, their intrinsic interest in reading, if they participated in the Book It! program, and if their parents paid them to read when they were children. If they participated in Book It! and/or were paid to read, they were also asked what effects these factors had on their learning to read, their enjoyment of reading, and on the amount they read. Method Participants One hundred seventy one undergraduate students at Youngstown State University participated in the study for extra course credit.
6 FLORA AND FLORA Procedure and Materials After reading and signing an informed consent form participants answered a brief survey. Participants were asked to mark male or female on the front of the survey and then to complete the survey in order. The first 10 questions were from the interest enjoyment and perceived competence dimensions of the Intrinsic Motivation Inventory (McAuley, Duncan, & Tammen, 1989) reworded to reflect interest in reading and shown in Table 1. The internal consistency of this reworded scale was Table 1 Items from the Interest-Enjoyment and Perceived Competence Dimensions of the Intrinsic Motivation Inventory Reworded to Reflect Interest in Reading 1. I enjoy reading very much. 2. I think I am pretty good at reading. 3. Reading is fun. 4. I would describe reading as very interesting. 5. I am satisfied with my performance in reading. 6. While reading, I think about how much I enjoy it. 7. After reading for awhile, I feel pretty competent. 8. I am pretty skilled at reading. 9. Reading does not hold my attention. 10. I can't read very well. high (Cronbach's Alpha =.90). The items were on a Likert scale from (1) strongly agree to (7) strongly disagree. The intrinsic motivation questions were followed with a question asking "not including school work, how much do you read (including books, newspapers, magazines, or other reading materials) on average per week?" This question was followed by a brief description of the Book It! program and the question, "when you were a child, did you participate in Pizza Hut's 'Book It!' Program?" If they answered "no" they were asked to skip the rest of the page and go to the next page. If they answered "yes" they were then asked: Circle ALL the grades that you participated in the Pizza Hut program. K, 1, 2, 3, 4, 5, 6. Do you think that, compared to when you were not in the program, the Pizza Hut program: a. Increased the amount read b. Had no effect on the amount you read. c. Decreased the amount you read. Do you think that, compared to when you were not in the program, the Pizza Hut program: a. Decreased your enjoyment of reading. b. Had no effect on your enjoyment of reading. c. Increased your enjoyment of reading.
~- ~~ READING INCENTIVES 7 Do you think the Pizza Hut program helped you to learn how to read? a. Yes. b. It had no effect. c. It slowed my learning to read. If you have any other thoughts or feelings about how Pizza Hut's program affected your reading habits, learning to read, or enjoyment of reading, please write them below. The following page asked "did a parent or guardian ever reward you with money for reading, or for how well you did in English or language arts?" If they answered "no" they were asked to skip the rest of the page. If they answered "yes," they were asked to "circle ALL the grades that your parents rewarded you with money for reading, English, or language arts classes" (K- 12.) This question was followed by four questions identical to the Pizza Hut questions about amount, enjoyment, learning to read, and the open-ended question, except the questions reflected parental pay for reading (for example "Do you think that, compared to when you were not rewarded with money, rewarding you a. Increased the amount you read. b. Decreased the amount... "). Results Several statistical models of the data were tested using simultaneous regression analyses. All regression assumptions were reasonably met for each model, and each statistic was tested with an alpha level of.01. The models are discussed below and are summarized in Table 2. The first model used sex, participation in Book It! (simply "yes" or "no"), and parental pay (simply "yes" or "no") as predictor variables and Table 2 Summary of Regression Analyses Model Dependent Variable 3 Predictors b (df) F Ff 1. IMI 2', 3,4 (3, 147) 4.88'.09 2. Reading am!. 1', 2, 3, 4 (4, 146) 13.00'.26 3. IMI 2', 5 (2, 94) 6.60'.12 4. Reading am!. 1', 2, 5 (3, 93) 9.67'.24 5. IMI 2, 6 (2, 42) 1.53.07 6. Reading am!. 1', 2, 6 (3, 41) 9.02'.40 7. IMI 2, 5, 6 (3, 23) 0.84.10 8. Reading am!. 1', 2, 5, 6 (4, 22) 5.08'.48 ' p <.01. almi = Intrinsic Motivation Inventory; Reading am!.= amount read per week. b1 = Intrinsic Motivation Inventory; 2 = sex, 3 = participation in Book It! (yes/no); 4= parental pay (yes/no); 5 = number of years in Book It! ; 6 = number of years of parental pay.
8 FLORA AND FLORA scores on the Intrinsic Motivation Inventory as the dependent variable. The model was significant overall, F(3, 147) = 4.88, P <.01. However, only 9% of the total variance on the Intrinsic Motivation Inventory was accounted for by the model as model R2=.0906. Sex was the only significant predictor, B = -.28, P <.01, indicating women score higher on the Intrinsic Motivation Inventory. The second regression model used the amount each participant read per week as the dependent variable, and the Intrinsic Motivation Inventory score became one of the predictor variables. This model was significant overall, F(4, 146) = 13.0, P <.01, model R2=.26. The only significant predictor of the amount of reading was the score on the Intrinsic Motivation Inventory, B =.51, P <.01. Participants who reported reading more generally scored higher on the Intrinsic Motivation Inventory, indicating the inventory has a degree of predictive validity. Two models that considered only those people who participated in Book It! to determine whether number of years in Book It! predict Intrinsic Motivation Inventory score or amount read per week were significant overall, F(2, 94) = 6.6, P <.01, model R2 =.12, and F(3, 93) = 9.67, P <.01, model R2 =.24, respectively. But like the first model, the third model found only that females scored higher on the Intrinsic Motivation Inventory, B = -.31, P <.001. The fourth model found only that participants who reported reading more also scored higher on the Intrinsic Motivation Inventory, B =.50, P <.01. Two models that considered only those people who's parents paid them to read were evaluated to determine if number of years paid to read predict Intrinsic Motivation Inventory score or amount read per week. The fifth model, using score on the Intrinsic Motivation Inventory as the dependent variable, was not significant. The sixth model, with amount of reading as the dependent variable, was significant overall, F(3, 93) = 9.02, P <.01, model R2 =.3975. But like the fourth and second models, the sixth found only that participants who reported reading more also scored higher on the Intrinsic Motivation Inventory, B =.61, P <.01. Two models that considered only those people who participated in Book It! and whose parents paid them to read were evaluated to determine whether number of years in Book It! or being paid to read predict Intrinsic Motivation Inventory score or amount read per week. Like the fifth model, the seventh model used score on the Intrinsic Motivation Inventory as the dependent variable and was not significant. Like the second, fourth, and sixth model, the eighth model using amount of reading as the dependent variable was significant overall, F(4, 22) = 5.08, P <.01, model R2 =.48, but found only that participants who reported reading more also scored higher on the Intrinsic Motivation Inventory, B =.61, P <.01. Overall, the only consistent significant finding from the eight simultaneous regression analyses was that people who reported reading more also scored higher on the Intrinsic Motivation Inventory. Being
READING INCENTIVES 9 rewarded for reading with money or pizzas neither increased or decreased the amount college students read nor influenced their intrinsic motivation for reading. The answers to the direct questions about Book It! and being paid to read suggested strong beneficial effects of these procedures (Figures 1 Book It! 80 80 60 c ~ 40 0; a. 20 o _ Decreased = No Effect ~ Increased Amount Read 80 60 c ~ 40 0; a. 20 68 30 _ Decreased = No Effect _ Increased Enjoyment 80 60 C ~ 40 a. 20 53 53 _ Slowed = No Effect 11l!!l Yes Learn ing to Read Figure 1. Answers to direct questions about effects of participation in Book It! on amount read, enjoyment of reading, and learning to read. Percentage of people choosing each option is presented. The raw number of people selecting each option is printed in or above bars.
10 FLORA AND FLORA and 2). Specifically, of the people who reported being in the Book It! program none indicated that it decreased reading. Conversely, 27 responded that it had no effect, and 80 (74.8 % of those answering the item) indicated that the Book It! program increased reading amount. Eight people did report that participating in Book It! decreased their enjoyment of reading. However, 30 people (28%) reported that Book It! Parents Paid 60 25 c " ~ a.. " 40 20 2 4 2 - Decreased = No Effect Amount Read ~ Increased 100 80 C 60 1l Q; a.. 40 3 g 20 3 _ Decreased = No Effect Enjoyment ~ Increased 100 80 c 60 " ~ ~ 40 3 g 19 20 o _ Slowing = No Effect Learn ing to Read _ Yes Figure 2. Answers to direct questions about effects of parental pay for reading on amount read, enjoyment of reading, and learning to read. Percentage of people choosing each option is presented. The raw number of people selecting each option is printed in or above bars.
READING INCENTIVES 11 increased their enjoyment of reading and 68 (64%) reported Book It! had no effect on enjoyment. Fifty three people (49.5%) reported that the Book It! program helped them learn how to read. Fifty three people reported that Book It! had no effect on their learning to read. Only one person reported that Book It! slowed them in learning to read. The answers to the direct questions about the effects of being rewarded with money for reading were similar to the answers to the questions about Book It! Two people answered that being paid to read decreased reading amount. Conversely, 25 people (49% of those answering the question) answered that being paid to read increased the amount they read. Twenty four people (47%) reported that being paid to read had no effect on the amount they read. Thirty nine people (76.5%) indicated that being paid to read had no effect on their enjoyment of reading. Three people reported that being paid to read decreased their reading enjoyment. But nine people (17.6%) responded that being paid to read increased their enjoyment of reading. No one indicated that being paid to read slowed their learning to read. Conversely, 19 people (37%) responded that being paid to read helped them to learn how to read. Thirty nine people (63%) indicated that being paid to read had no effect on learning to read.. Twenty five people responded to the open-ended questions. Five responses suggested the programs had no effects (e.g., "Money had no influence - I love reading."). Three responses indicated cheating in the Book It! program. Beneficial effects of being paid to read or of participating in Book It! were given by the remaining 17 people (68%) who responded to the open-ended questions (e.g., "I believe that being rewarded for reading showed me how much an education pays off. It has provided me with a motivation to study and pay attention in school that I still have today."). Discussion The current study found no reliable effect of either participating in the Book It! reading program or of being rewarded with money for reading as children on either intrinsic motivation for reading, or on the self-reported amount of reading per week of college students. Direct questions about the effects of Book It! and/or of being paid to read found the procedures to be beneficial or at worst benign. Indeed, the results suggest that when a child participates in Book It! or is rewarded for reading with money the child will increase the amount read, enjoyment of reading may increase, and if they do not read fluently, then the programs may help the child to learn to read (Figures 1 and 2). There is some argument that 'intrinsic motivation' for an activity occurs when the behavior involved becomes fluent (Daniels, 1994, pp. 183-187; Lindsley, 1996, pp. 199-210). The reader who is fluent (Le., reads at a high rate with accuracy) is likely to find reading intrinsically
12 FLORA AND FLORA motivating. The student who struggles to read three and four letter words will not enjoy reading. Because students who were in Book It! or were paid to read generally reported that the procedure(s) helped them learn to read, increased the amount they read, and increased their enjoyment of reading, it is likely that these programs helped many of them become fluent readers. That is, extrinsic motivators for reading may have set the conditions for reading to become intrinsically motivating. At least two other studies support the conclusion that contrived reinforcers, or extrinsic motivators, increase rather than decrease later intrinsic motivation for reading. The Earning by Learning Program pays academically at-risk children two dollars for each book they read. Children are quizzed and may be required to read parts of the book out loud before they receive their payment. An evaluation of the program (McNinch, Steely, & Davidson, 1995) found that participation increased total positive attitude toward reading, attitude toward recreational reading, and attitude toward academic reading using both pretestposttest measures and comparisons against nonparticipating controls. A preliminary evaluation of the Book It! program, commissioned by Pizza Hut during the initial year, 1985, of the program, was conducted by the Institute of Human Science and Services of The University of Rhode Island (1986). Surveys, nonsystematic observations from 2,741 teachers representing 16,130 students were used in the evaluation. According to the teachers, participation in Book It! generally improved attitude toward learning (61%), reading level (69%), enjoyment of reading (80%), and the longer children were in the program the more their reading level rose and enjoyment of reading increased. According to the report, ''the basis behind the program was to offer immediate positive reinforcement to reward individual accomplishments.... It was this rewarding of effort and not ability that probably made Book It! so attractive to both teachers and students" (p. 17, emphasis in original). Furthermore, "the program did not detract from or hinder students' education" (p. 18). Because Pizza Hut contracted the commissioned study the results from it should be taken cautiously. However, the results of the current study, which was conducted independently and without the support or consent of Pizza Hut, are consistent with the commissioned study. The 1986 initial commissioned study, the Earning by Learning study (McNinch et ai., 1995), and the current study all converge on the same conclusion that extrinsic reinforcers or incentives for reading set the conditions where intrinsic motivation may develop. These results suggest that the question "can reinforcement be used to increase intrinsic task interest?" (Eisenberger & Cameron, 1996, p. 1164) may be answered in the affirmative. Reconsidering Intrinsic and Extrinsic The terms "intrinsic interest," "intrinsic reward," or "intrinsic motivation" all refer to a hypothetical construct. Dropping the first word
READING INCENTIVES 13 and calling intrinsic motivation a construct makes it no less hypothetical. Intrinsic motivation only exists in the verbal behavior (spoken and written) of the researchers and theorists. What exists for the subject are behaviors (e.g., reading behaviors and both overt and covert verbal behavior related to reading such as the "expression of interest or attitude toward reading") and environment, both physical (the availability of books and possible reinforcers for reading) and social. The social environment may provide both discriminative stimuli (e.g., "why don't you read me a story?") and reinforcers for reading. Intrinsic interest in reading is actually "intrinsic to the behavior-environment interaction, not the organism" (Flora, 1990, p. 337). When extrinsic reward and intrinsic motivation are reconsidered as environment and behavior, the question becomes one of how environmental manipulations affect behavior. When the question is a more specific one of how a reward manipulation later affects the rate of the previously rewarded behavior the empirical answer is clear. Cameron and Pierce's (1994) review of approximately 100 experiments covering 20 years of research remains definitive; "In classroom situations... rewards can be used to maintain or enhance students' intrinsic interest in schoolwork... When tangible rewards are offered contingent on level or performance..., students remain motivated in the subject area" (Cameron & Pierce, 1996, p. 40). That is, after contrived reinforcement (reward) is removed, natural or noncontrived reinforcers (intrinsic motivators) maintain the behavior. Results of the current study support this conclusion. Any concern that reinforcing reading will decrease later reading is unfounded. References CAMERON, J., & PIERCE, D. (1994). Reinforcement, reward, and intrinsic motivation: A meta-analysis. Review of Educational Research, 64, 363-423. CAMERON, J., & PIERCE, D. (1996). The debate about rewards and intrinsic motivation: Protests and accusations do not alter the results. Review of Educational Research, 66, 39-51. CARNEVALE, A. P, JOHNSON, N. C., & EDWARDS, A. R. (1998, April 10). Performance-based appropriations: Fad or wave of the future? The Chronicle of Higher Education, pp. B6, B7. CARTON, J. S. (1996). The differential effects of tangible rewards and praise on intrinsic motivation: A comparison of cognitive evaluation theory and operant theory. The Behavior Analyst, 19,237-255. DANIELS, A. C. (1994). Bringing out the best in people. New York: McGraw-HilI. DICKINSON, A. M. (1989). The detrimental effects of extrinsic reinforcement on "intrinsic motivation:' The Behavior Analyst, 12, 1-15. EISENBERGER, R., & CAMERON, J. (1996). Detrimental effects of reward: Reality or myth? American Psychologist, 51, 1153-1166.
14 FLORA AND FLORA FLORA, S. R. (1990). Undermining intrinsic interest from the standpoint of a behaviorist. The Psychological Record, 40, 323-346. GLEISSER, M. (1998, April 10). LTV steel co. workers to diwy up $32.6 million. The Plain Dealer, B6. INSTITUTE OF HUMAN SCIENCE AND SERVICES OF THE UNIVERSITY OF RHODE ISLAND (1986). Pizza Hut Inc.'s The BOOK IT national reading incentive program. Final evaluation report. KOHN, A. (1993). Punished by rewards. Boston: Houghton Mifflin. LINDSLEY, O. R. (1996). The four-free operant freedoms. The Behavior Analysis, 19, 199-210. MARTIN, G., & PEAR, J. (1996). Behavior modification (5th ed.). Upper Saddle River, NJ: Prentice Hall. McAULEY, E., DUNCAN, T., & TAMMEN, V. V. (1989). Psychometric properties of the Intrinsic Motivation Inventory in a competitive sport setting : A confirmatory factor analysis. Research Quarterly for Exercise and Sport, 60,48-58. McNINCH, G. w., STEELY, M., & DAVIDSON, T. J. (1995). Evaluating the Earning by Learning program: Changing attitudes in reading. Paper presented at the annual meeting of the Georgia Council of the International Reading Association, Atlanta, GA. MILLER, L. K. (1997). Principles of everyday behavior analysis (3rd. Ed.). Pacific Grove, CA: Brooks/Cole.