On Presentations INF2793 Research Design & Academic Writing Prof. Simone D.J. Barbosa simone@inf.puc-rio.br sala 410 RDC presentation @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 1
communication not decoration @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 2
slides notes handouts At a minimum, a presentation format should do no harm. (Tufte, 2003) @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 3
http://www.slideshare.net/garr/guy-kawasakis-foreword-for-presentation-zen http://www.slideshare.net/garr/guy-kawasakis-foreword-for-presentation-zen @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 4
ethos credibility of the speaker pathos emotional connection to the audience logos logical argument http://sixminutes.dlugan.com/ethos-pathos-logos/ preparation (Reynolds, 2008) @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 5
How much time do I have? What s the venue like? @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 6
Who is the audience? What is their background? What do they expect of me? What do I want them to do? @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 7
What is the fundamental purpose of my talk? What s the story here? @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 8
What s my absolute central point? content design delivery @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 9
structure natural chronological time progression sequential process or step-by-step spatial relations in a physical space climactic from least to most important contrast problem solution compare contrast cause effect advantage disadvantage (Duarte, 2010: 129) making your ideas stick @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 10
The SUCCESs framework Simplicity Unexpectedness Concreteness Credibility Emotions Stories core + compact idea attention + interest understanding belief care action (Heath & Heath, 2007) @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 11
story s sparkline what could be contrasting emotion and delivery what is (Duarte, 2010) questions content @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 12
content design delivery It s laziness on the presenter s part to put everything on one slide. (Duarte, 2008: 93) @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 13
rules of 3 abound 3 points 3 seconds 3 3 grid @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 14
Face your message. four design principles Contrast Repetition Alignment Proximity @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 15
contrast contrast contrast contrast contrast repetition Another text Some text Here, too Yet another @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 16
alignment and proximity Possibly Long Title Possibly Long Title Author s Full name Affiliation Author s Full name Affiliation Possibly Long Title Possibly Long Title Author s Full name Affiliation Author s Full name Affiliation picture superiority @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 17
simplify Cognitive Load Theory It is more difficult to process information if it is coming at you in the written and spoken form at the same time. (Sweller in Patty, 2007) @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 18
Cognitive Load Theory It is more difficult to process information if it is coming at you in the written and spoken form at the same time. Teachers should focus more on giving students the answers, instead of asking them to solve problems on their own. The human brain processes and retains more information if it is digested in either its verbal or written form, but not both at the same time. The findings show there are limits on the brain's capacity to process and retain information in short-term memory. John Sweller, from the university's faculty of education, developed the "cognitive load theory". "The use of the PowerPoint presentation has been a disaster," Professor Sweller said. "It should be ditched." "It is effective to speak to a diagram, because it presents information in a different form. But it is not effective to speak the same words that are written, because it is putting too much load on the mind and decreases your ability to understand what is being presented." "Looking at an already solved problem reduces the working memory load and allows you to learn. It means the next time you come across a problem like that, you have a better chance at solving it," Professor Sweller said. The working memory was only effective in juggling two or three tasks at the same time, retaining them for a few seconds. When too many mental tasks were taken on some things were forgotten. (Patty, 2007) cognitive load slides The human brain processes and retains more information if it is digested in either its verbal or written form, but not both at the same time. speech (inspired in article by Patty, 2007) @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 19
cognitive load slides The human brain processes and retains more information if it is digested in either its verbal or written form, but not both at the same time. (inspired in article by Patty, 2007) but @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 20
It is effective to speak to a diagram (Sweller in Patty, 2007) signal noise ratio @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 21
@ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 22
@ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 23
@ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 24
@ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 25
questions design @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 26
content design delivery rehearse! rehearse! rehearse! @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 27
prepare your gig bag the day before come in early test the equipment bring your own water (just in case) eventually, technology will fail relax! When time comes look at the audience and not at the screen. @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 28
take your hands of your pockets Whatever happens leave time for Q&A. Never, ever go beyond your time slot. @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 29
presentation checklist (1/2) What is the key thing the audience should remember? Is there enough background material for the intended audience? Is any material unnecessary? Could some of the material be left for people to read about later'? Is the talk self-contained? Does the talk have a motivating preamble? Do you explain why the research is interesting or important? Have complex issues been explained in gentle stages? Are the results explained? What were the limitations of the research? Is there a clear conclusion? (Zobel, 2004: KL 2912-8) 60 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 30
presentation checklist (2/2) Are there enough examples? Are the numbers necessary? Are more diagrams needed? Are the slides simple? Do they have unnecessary ornamentation or distracting use of colour? Is there any unnecessary animation? Are the font sizes reasonable? If you are asked a question you can't answer, how will you respond? Have you rehearsed the talk? Have you prepared something to say about each slide? Have you rehearsed your manner? Will your enthusiasm show? Have you memorized the talk? Do you know how to use the equipment? (Zobel, 2004: KL 2912-8) 61 questions presentations @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 31
visually communicating your evidence William Playfair (1758-1823) 64 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 32
William Playfair 65 Florence Nightingale 66 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 33
Charles Joseph Minard 67 Charles Joseph Minard 68 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 34
Charles Joseph Minard 69 Charles Joseph Minard 70 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 35
Simple Data Booth et al., 2008: 213 71 Complex Data Booth et al., 2008: 214 72 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 36
Bar Charts and Line Graphs bar charts emphasize contrasts among discrete items (bar chart: emphasizes contrast between discrete values) line graph suggest continuous change over time (line graph: suggests continuous change over time) Booth et al., 2008: 215 73 Bar Chart - Clustered Column População Brasileira em 2010 (em milhões de habitantes) 80.0 Urbana Rural 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 74.7 5.7 38.8 14.3 23.3 4.1 12.5 1.6 11.7 4.2 Sudeste Nordeste Sul Centro-Oeste Norte IBGE 74 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 37
Bar Chart - Clustered Column População Brasileira em 2010 (em milhões de habitantes) 90.0 Urbana Rural Total 80.0 80.4 70.0 60.0 50.0 53.1 40.0 30.0 20.0 10.0 0.0 74.7 5.7 38.8 14.3 23.3 4.1 11.7 4.2 12.5 1.6 Sudeste Nordeste Sul Norte Centro-Oeste 27.4 15.9 14.1 IBGE 75 Bar Chart - Stacked Column População Brasileira em 2010 (em milhões de habitantes) 90.0 Urbana Rural 80.0 5.7 70.0 60.0 50.0 40.0 14.3 30.0 20.0 10.0 0.0 4.1 4.2 1.6 74.7 38.8 23.3 11.7 12.5 Sudeste Nordeste Sul Norte Centro-Oeste IBGE 76 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 38
But... POPULAÇÃO BRASILEIRA EM 2010 (EM MILHÕES DE HABITANTES) Norte, 11.7 Centro-Oeste, 12.5 Sul, 23.3 Sudeste, 74.7 Nordeste, 38.8 IBGE 77 Bar Chart 90 População Brasileira (em milhões de habitantes) 80 70 60 50 40 30 20 10 0 1960 1970 1980 1991 2000 2010 Sudeste Nordeste Sul Norte Centro-Oeste IBGE 78 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 39
Line Chart 90 População Brasileira (em milhões de habitantes) 80 70 60 50 40 30 Norte Nordeste Sudeste Sul Centro-Oeste 20 10 0 1960 1970 1980 1991 2000 2010 IBGE 79 Line Chart 90 População Brasileira (em milhões de habitantes) 80 Sudeste 80.4 70 60 50 Nordeste 53.1 40 30 20 10 Sul 27.4 Norte 15.9 Centro-Oeste 14.1 0 1960 1970 1980 1991 2000 2010 IBGE 80 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 40
Tables, Charts, and Graphs Labeling the Graphic Frame each graphic to help readers understand it Label every graphic in a way that describes its data Heads of households Changes in one- and two- parent heads of households, 1970 2000 Do not give background information or characterize what the data imply Weaker effects of counseling on depressed children before professionalization of staff, 1995 2004 Effect of counseling on depressed children, 1995 2004 Be sure labels distinguish graphics presenting similar data. Risk factors for high blood pressure Risk factors for high blood pressure Risk factors for high blood pressure among men in Cairo, Illinois Risk factors for high blood pressure among men in St. Louis, Missouri Booth et al., 2008: 216-7 81 Tables, Charts, and Graphs Annotating the Graphic Insert into the table or figure information that helps readers see how the data support your point. Booth et al., 2008: 217 82 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 41
Tables, Charts, and Graphs Introducing the Graphic Introduce the table or figure with a sentence that explains how to interpret it. Then highlight what it is in the table or figure that you want readers to focus on, particularly any number or relationship mentioned in that introductory sentence. Booth et al., 2008: 218 83 Tables, Charts, and Graphs Keep it Simple 1. Include only relevant data. 2. Keep the visual impact simple. 3. Use clear labels. Booth et al., 2008: 219-20 84 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 42
Tables, Charts, and Graphs Keep it Simple 1. Include only relevant data. If you include data only for the record, label it accordingly and put it in an appendix. 2. Keep the visual impact simple. 3. Use clear labels. Booth et al., 2008: 219-20 85 Tables, Charts, and Graphs Keep it Simple 1. Include only relevant data. 2. Keep the visual impact simple. Box a graphic only if you group two or more figures. Do not color or shade the background. for tables Never use both horizontal and vertical dark lines to divide columns and rows. Use light gray lines only if the table is complex or you want to direct your reader s eyes in one direction to compare data. For tables with many rows, lightly shade every fifth row. 3. Use clear labels. Booth et al., 2008: 219-20 86 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 43
Tables, Charts, and Graphs Keep it Simple 1. Include only relevant data. 2. Keep the visual impact simple. for charts and graphs Use background grid lines only if the graphic is complex or readers need to see precise numbers. Make them light gray. Color or shade lines or bars only to show a contrast. Use color only if the text will be printed in color and not photocopied later. (Black- and- white photocopies make many colors look alike.) Never use iconic bars (for example, images of cars to represent automobile production) or add a third dimension merely for effect. Both look amateurish and can distort how readers judge values. Plot data on three dimensions only when your readers are familiar with such graphs and you cannot display the data in any other way. 3. Use clear labels. Booth et al., 2008: 219-20 87 Tables, Charts, and Graphs Keep it Simple 1. Include only relevant data. 2. Keep the visual impact simple. 3. Use clear labels. Label all rows and columns in tables and both axes in charts and graphs. Use tick marks and labels to indicate intervals on the vertical axis of a graph. If possible, label lines, bar segments, and the like on the image rather than in a legend set to the side. Use a legend only if labels would make the image too complex to read. When specific numbers matter, add them to bars or segments in charts or to dots on lines in graphs. Booth et al., 2008: 219-20 88 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 44
Tables Tables with lots of data can seem dense, so organize them to help readers. Order the rows and columns by a principle that lets readers quickly find what you want them to see. Do not automatically choose alphabetic order. Round numbers to a relevant value. If differences of less than 1,000 don t matter, then 2,123,499 is irrelevantly precise. Sum totals at the bottom of a column or at the end of a row, not at the top or left. Booth et al., 2008: 220-1 89 Bar Charts Booth et al., 2008: 221-2 90 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 45
Stacked Bars When you want readers to compare whole values for different bars rather than their divided segments Arrange segments in a logical order. If possible, put the largest segment at the bottom in the darkest shade. Label segments with specific numbers and to assist comparisons, connect corresponding segments with gray lines to compare whole values for different bars Booth et al., 2008: 223-4 91 Line Graphs Choose the variable that makes the line go in the direction, up or down, that supports your point. If the good news is a reduction (down) in high school dropouts, you can more effectively represent the same data as a rising line indicating increase in retention (up). If you want to emphasize bad news, find a way to represent your data as a falling line. Plot more than six lines on one graph only if you cannot make your point in any other way. If you have fewer than ten or so data points, indicate them with dots. If only a few are relevant, insert numbers to show their exact value. Do not depend on different shades of gray to distinguish lines. Booth et al., 2008: 225-6 92 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 46
Communicate Ethically Do not manipulate a scale to magnify or reduce a contrast. Do not use a figure whose image distorts values. Do not make a table or figure unnecessarily complex or misleadingly simple. If the table or figure supports a point, state it. Booth et al., 2008: 227 93 Communicate Ethically Booth et al., 2008: 227 94 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 47
Communicate Ethically Booth et al., 2008: 228 95 avoiding visual misrepresentation Do not manipulate a scale to magnify or reduce a contrast. Do not use a figure whose image distorts values. Do not make a table or figure unnecessarily complex or misleadingly simple. If the table or figure supports a point, state it. Booth et al., 2008: 229 96 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 48
1960 1970 1980 1991 2000 2010 Line Chart 90 População Brasileira (em milhões de habitantes) 80 Sudeste 80.4 70 60 50 Nordeste 53.1 40 30 20 10 Sul 27.4 Norte 15.9 Centro-Oeste 14.1 0 1960 1970 1980 1991 2000 2010 IBGE 97 Line Chart - distorted to emphasize growth 90 80 População Brasileira (em milhões de habitantes) Sudeste 80.4 70 60 50 Nordeste 53.1 40 30 20 10 Sul 27.4 Norte 15.9 Centro-Oeste 14.1 0 IBGE 98 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 49
Line Chart - distorted to de-emphasize growth 100 80 60 40 20 0 População Brasileira (em milhões de habitantes) Sudeste 80.4 Nordeste 53.1 Sul 27.4 Norte 15.9 Centro-Oeste 14.1 1960 1970 1980 1991 2000 2010 IBGE 99 Booth et al., 2008: 230-1 100 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 50
Booth et al., 2008: 230-1 101 Booth et al., 2008: 230-1 102 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 51
Booth et al., 2008: 230-1 103 Booth et al., 2008: 230-1 104 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 52
Booth et al., 2008: 230-1 105 Booth et al., 2008: 230-1 106 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 53
Booth et al., 2008: 230-1 107 Booth et al., 2008: 230-1 108 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 54
Booth et al., 2008: 230-1 109 Booth et al., 2008: 230-1 110 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 55
questions communicating visually references Duarte, N. (2008) slide:ology. O Reilly. Duarte, N. (2010) resonate. John Wiley & Sons. Heath, D. & Heath, C. (2007) Made to Stick. Random House. Patty, A. (2007) Research points the finger at PowerPoint. In The Sidney Morning Herald, April 4, 2007. Available online at http://www.smh.com.au/articles/2007/04/03/1175366240499.html Tufte, E. (2008) Powerpoint is Evil. Wired Magazine, Sep 2003. Available online at http://www.wired.com/wired/archive/11.09/ppt2.html Zobel, J. (2004) Writing for Computer Science. Kindle Edition. slides and examples http://www.slideshare.net/garr/sample-slides-by-garr-reynolds http://www.slideshare.net/philtoland/presentation-zen-1655196 112 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 56
simone@inf.puc-rio.br http://www.ideias.inf.puc-rio.br/aulas 113 @ 2013 Simone DJ Barbosa, Departamento de Informática, PUC-Rio 57