1. Writing the results, discussion & using graphics & visuals in research reports Communication Research Week 7

2. 2 Results Purpose of this section is to present a summary of the results and comment on the significance of the data Graphics are used in this section to present the full findings Text should highlight the most important results and help the reader interpret them

3. 3 Results For each piece of important information, a three (3) step process is useful: A statement which locates the figures or findings eg as shown in Fig 1, the trend … Statements which present the most important findings Comments by the author on these findings

4. 4 Results/comments May do the following Generalise from specific results to general trends in populations Explain the possible reasons or causes of such results Compare the results with results of other studies

5. 5 Discussion section Makes reference to the main purpose and/or hypothesis Reviews the most significant findings and whether or not they agree or disagree with the original hypothesis Explains/speculates about findings outlined or described in “Results” section Compares findings/outcomes with related studies already cited in literature review Outlines limitations of the study and the extent to which findings can really be generalised Makes recommendations for further research

6. 6 Difficulties with presenting & interpreting probability information from Berry (2004) “Fifty percent of the public doesn’t actually know what 50% means” (UK Secretary of State for Trade and Industry, 2002) A German study asked people: What does 40% mean” 1 in 4 4 out of every 10 Every 40th person Over a third of a sample of 1000 Germans tested did not select the correct answer (Gigerenzer, 2002, 23 cited in Berry, 2004, 5)

7. 7 Presenting probabilities as frequencies Gigerenzer (2002) argues that some of the misinterpretation by even experienced and educated readers can be overcome by using frequencies eg The probability that a woman of age 40 has breast cancer is about 1 percent. If she has breast cancer, the probability that she tests positive on a screening mammogram is 90 percent. If she does not have breast cancer, the probability that she nevertheless tests positive is 9 percent. What are the chances that a woman who tests positive actually has breast cancer? (Gigerenzer, 2002, 41 cited in Berry, 2004, 32)

8. 8 Presenting probabilities as frequencies Most people surveyed found the above problem confusing and believed the answer is 90 percent. Read the same problem expressed using natural frequencies: Think of 100 women. One has breast cancer, and she will probably test positive. Of the 99 who do not have breast cancer, 9 will also test positive.Thus a total of 10 women will test positive. How many of those who test positive actually have breast cancer? (Gigerenzer, 2002, 42) People found it easier to see that only one woman out of every ten tests positive will actually have cancer – that is, the probability is 10% not 90%

9. 9 Interpretation can be affected by Verbal expressions used eg ‘likely’, ‘rare’ Context – knowledge, experience, framing of info Different numerical formats eg Yamagishi (1997) presented people with two different statements about a certain type of cancer and asked them which they judged to be more risky Kills 1286 out of 10,000 people Kills 24.14 out of 100 people. The first statement was judged more risky even though the level of risk described in the second is twice as high

10. 10 Why is understanding how visuals communicate important? Our culture places much emphasis on the visual – seeing is believing Visual communication is faster and more easily processed Visuals and graphics add another layer of meaning and another way of communicating Visuals actively engage the brain in interpretation, making it more likely that readers will remember the information

11. 11 How do graphics & visuals enhance & supplement a report? Make points vivid and help readers “see” data Present information more compactly than words Convey/ simplify complex data Demonstrate contrasts/ comparisons Suggest movements/ trends over time Emphasise physical appearance

12. 12 How do graphics & visuals enhance & supplement a report? Analyse concepts/ processes/ abstract relationships Should not replace text Should be properly incorporated and referenced eg ‘ … as shown in figure 1 …’ Different graphics serve different purposes – choose the right visual for the story and the data Ensure each visual is accurate and ethical

13. 13 All visuals share certain conventions Identify an analytic perspective for the data with an interpretative title Clearly describe the type of data (survey or projection) Label the units (e.g. slices in a pie chart) Label the axes and use a legend List the source of the data or acknowledge the source of the visual (if copied) Integrate into text with table/ figure numbers

14. 14 What’s wrong with this graphic? Has there been a growth in the number of bananas between 1960 and 1980? Or have the bananas grown increased in size? Source: Sadler & Tucker, 1981, 116

15. 15 Line graphs Indicate movements over time, compare frequency, identify correlations 6 Inappropriate labels and scales can make them difficult to interpret Q:What is the difference between these two graphs? Source: Sadler & Tucker, 1981, 116

16. 16 Line graphs n Put time on the horizontal axis n Avoid more than 3 or 4 lines n Use only 2 lines if they cross a lot n Use different colours and a legend n Label the axes n Avoid perspective Source: Gould www.rpi.edu/~goulde/co_su02/viscom.ppt

17. 17 Poor example Source: Gould www.rpi.edu/~goulde/co_su02/viscom.ppt

18. 18 Column or vertical bar graph 4 Compare items, show distributions or highlight correlations 4 Different bar charts for different purposes  Grouped (compare aspects of each item across time  Segmented, subdivided or stacked (helps compare totals but cannot compare segments  Deviation (identify opposites)  Paired (show correlation between two items) 6 Difficult for the eye to interpret size and proportions Source: Eunson, 1995, 79

19. 19 Grouped bar charts allow comparison Source: Gould www.rpi.edu/~goulde/co_su02/viscom.ppt

20. 20 Segmented, subdivided, stacked bar charts show different relationships Source: Gould www.rpi.edu/~goulde/co_su02/viscom.ppt

21. 21 Deviation bar charts show exceptions Source: Gould www.rpi.edu/~goulde/co_su02/viscom.ppt

22. 22 Bar chart design Use a logical order Chronological By region Put bars close enough for comparison Label both axes and make increments consistent Make all bars the same width Use colours for coding (not just “to look good”) Avoid “chart junk” – especially 3D views

23. 23 Chart junk Source: Gould www.rpi.edu/~goulde/co_su02/viscom.ppt

24. 24 Pie charts 4 Show relative proportions and the importance of each part to the whole 4 Label segments and proportions outside the pie 4 Limit segments to 5-7 6Can be difficult to judge area and size differences therefore should not be used to exactly compare segments Source: Eunson, 1995, 78

25. 25 Good example Source: www.rpi.edu/~goulde/co_su02/viscom.ppt]

26. 26 Poor example Source: Gould www.rpi.edu/~goulde/co_su02/viscom.ppt

27. 27 Tables Offer clear comprehensive detail Allow comparison between large amounts of data Make readers focus on the raw data not your interpretation of the data How you set out the table can affect interpretation 6 Difficult to read quickly 6 Hard to recognise relationships

28. 28 Even tables can tell different stories – compare … Source: Gould www.rpi.edu/~goulde/co_su02/viscom.ppt

29. 29 Summary of charts n Use charts to simplify data n Pick an appropriate style – bar charts are most common for business audiences n Provide an interpretative title – you want your readers to understand the data in a way that supports your arguments, not theirs n Avoid “chart junk”

30. 30 Visual Ethics in Accident Reporting (from Dragga & Voss, 2003)

31. 31

32. 32

33. 33

34. 34 Bibliography & further reading Berry, D.(2004) Risk, Communication and Health Psychology UK, Open University Press Dragga, S. & Voss, D. (2003) Hiding Humanity: Verbal and Visual Ethics in Accident Reports Technical Communication Vol 50, No 1, Feb 2003 pp.61- 82 Dragga, S. & Voss, D. (2001) Cruel Pies: The Inhumanity of Technical Illustrations Technical Communication Vol 48, No 3, Aug 2001 pp.265-274 Eunson, B. (1995) Writing and Presenting Reports Melbourne, John Wiley Gould, E (2002) www.rpi.edu/~goulde/co_su02/viscom.ppt [accessed May 20, 2003]www.rpi.edu/~goulde/co_su02/viscom.ppt Mohan, T., McGregor, H., Saunders, S., & Archee, R. (2004) Communicating as Professionals Melbourne, Thomson Sadler, R.K & Tucker, K (1981) Common Ground: A Course in Communication Melbourne, Macmillan