PSYC 612, SPRING 2007Lecture 12: PersuasionApril 16, 2007Contents1 Part 1: A cursory review of the assigned readings 11.1 Gigerenzer (1993) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Part 2: Important aspects not covered in the readings 32.1 Sources and Standards of Evidence . . . . . . . . . . . . . . . . . . . . . . . 32.2 Argumentation and Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32.3 Skepticism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42.4 Weapons of persuasion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52.5 What do we know from the research on persuasion? . . . . . . . . . . . . . . 53 Part 3: Some ugly details 63.1 Read in your data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73.2 Check your data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83.3 Conduct your planned analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 133.4 Check your results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143.5 Correct any errors and rerun the analyses . . . . . . . . . . . . . . . . . . . . 173.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 Part 1: A cursory review of the assigned readingsLast week I introduced the idea of the simple minded and muddled headed continuum inthe context of decision making. Much of decision making in data analysis, however, oftenconcerns decisions related to the audience. In particular, your conceptualization of theaudience represents many aspects that ought to dictate what you do and how you describewhat you did. Imagine you conducted a study high school kids and you promised to tell themwhat you learned when the study concludes. So you collect data on them, analyze the data,and then prepare to present the results to this high school audience. What do you present?Certainly you would not present the data in the same format you might present it with a1scientific audience. You also might not present the same results to a scientific audience asyou might to a statistical audience. The audience matters. What matters most, however,is persuasion. In order to persuade a person, you must first be able to communicate yourideas. If the ideas are presented in a manner that is far to sophisticated for the audiencethen there is little hop e that the audience will be persuaded by your findings. What we willdiscuss today are the aspect of persuasion as they relate to data analysis and presentation.1.1 Gigerenzer (1993)Gigerenzer provides us with some insights into why decision making is not be relevant formany in psychology but why it ought to be much more relevant. He argues from the per-spective of a person who b elieves that scientific progress comes from conveying findings insuch a way as to capture the essence of the contribution. Hence, Gigerenzer holds that thereexists a hierarchy in statistical thinking that mimics Freud’s Superego, Ego, and Id. Thosethree aspects are similar to the statistical traditions of Neyman-Pearson, Fisher, and Bayes,respectively. The Superego dictates what we ought to do, the ego gets things done, and theid is our prime motive or drive.Gigerenzer walks us through a tour of how statistics has permeated every field in science.In particular, he reasoned that statistics followed this path:• probabilistic thinking became the de rigeur in science• that thinking manifested itself as null hypothesis significance testing (NHST)• NHST dominant in psychology• NHST dictates much of what psychologists do in researchBecause NHST became institutionalized in psychology, much of the decision makingwent away. Researchers merely selected which NHST procedure to run and then the restwas mechanical. That process was not the intent of Neyman, Pearson, or Fisher but that ishow things worked out.Where this chapter is relevant to persuasion is in understanding the rules of engagement.Our aim is to use data to test theoretical propositions (hypotheses) and then to communicateour uncertainty of our propositions through statistics. If the audience only understands orexpects NHST then it is our burden to communicate uncertainty in those terms and thenin the terms we deem more appropriate. So the rules of engagement are 1) to satisfy theaudience, 2) to satisfy ourselves, and 3) to push the audience into accepting our perspective.The first two are easily achieved but the latter is achieved with much more effort. Thateffort often involves a deeper understanding of logic, reasoning, and persuasion.22 Part 2: Important aspects not covered in the read-ingsThe second lecture in PSYC 611, I reviewed the following points that were relevant forpersuasion. Instead of just reminding you that I covered these points, I intend to go overthem again. Revisiting this material may give you an entirely different perspective on theprocess of design and statistics now that you have read and thought about the materialcovered to date.First I began with sources and standards of evidence. These are important for persuasionbecause you need to know what standard your audience expects. In some cases, the audiencewill simply defer to you as the expert or authority. In other cases, the audience will demandempirical evidence. The use of statistics is relevant to all of these sources.2.1 Sources and Standards of Evidence• Authority• Tradition• Lore• Consensus• Intuition• EmpiricalOnce you have identified an appropriate source and standard, you will then need tounderstand the nature of argumentation. To argue a p oint, you ought to be careful abouthow you proceed.2.2 Argumentation and Logic• Science: an inherently a human endeavor• Progress comes incrementally by first persuading the scientific community• Persuasion comes from eliminating skepticism• Eliminating skepticism comes from anticipating critiques• Anticipation of criticism comes from knowing...– the logic of inquiry– the potential alternative explanations3– the methods to account for those alternative explanations– how to retrace your steps and improve your methods for each incremental contri-bution– how to remain skeptical yourself• Critics fall prey to logical fallacies including...– faulty generalization - extending …