Persuasive Advertising - J. Scott Armstrong [205]
A close match was found between laboratory studies (often using students) and field studies in organizational behavior. For example, in the nine cases allowing for comparisons, the direction of the findings agreed, while in six cases where the magnitude of effects could be compared, three were virtually identical and two had only small differences (Locke 1986).
A convenience sample of physicians and pharmacists was used in an experiment on the use of disclaimers for a drug sold by Berlex Labs. Berlex was the plaintiff in a lawsuit, where the defendant was Shering AG. The use of a convenience sample was challenged by the defendant, which then commissioned an extension of the study using a probability (representative) sample of subjects. As expected, the findings were similar and they upheld the original conclusions (Jacoby and Szybillo 1994).
Additional studies have also supported this conclusion. They include studies involving decision-making processes in accounting (Liyanarachchi and Milne 2005) and in managerial decision-making (Remus 1996).
2. Do subjects try to help researchers?
The notion that subjects try to help researchers confirm hypotheses is referred to as the “demand effect.” Orne (1962) describes one of the more influential studies on demand effects. He conducted a small informal study in which he asked people to agree to participate in an experiment. He then gave the subjects sheets of random numbers and asked them to perform 224 addition computations on each. When they finished, he told them to pick up a card with instructions on what to do next. Each card had the same instruction: “You are to tear up the sheet of paper which you have just completed into a minimum of 32 pieces then go on to the next sheet of paper and continue working as you did before; when you have completed this piece of paper, pick up the next card which will instruct you further. Work as accurately and as rapidly as you can.” Orne wanted to see whether the subjects would continue with such a meaningless task. To his amazement, they kept at the task for hours with no signs of hostility. The experimenters gave up before the subjects did.
Does this show that they wanted to please the experimenter? To please the experimenter, subjects would have to discern the experimenter’s hypothesis, which is difficult. Therefore, it is doubtful that this was their motivation.
There is little evidence that demand effects are serious in the types of experiments we report. An alternative explanation is that people like to present themselves in a favorable light; A lab experiment supported this explanation (Sigall, Aronson, and van Hoose 1970).
3. Are some studies too old?
The principles in this book are based on how people respond to persuasive strategies. These tend to remain constant over time. Many of the early research studies in advertising are excellent—and still offer valuable advice. Some of them are cited in this book and many have been referred to in papers that I cite. Gigerenzer (2000) reached a similar conclusion about research in psychology:
Several years ago, I spent a day and a night in a library reading through issues of the Journal of Experimental Psychology from the 1920s and 1930s. This was professionally a most depressing experience, but not because these articles were methodologically mediocre. On the contrary, many of them make today’s research pale in comparison with their diversity of methods and statistics …. What depressed me was that almost all of this work is forgotten.
4. Are most findings just common sense?
Most theories and findings by academics are common sense. Consider this statement from a book on persuasion: “Writing pervades modern life as a principal vehicle of communication.”
Common sense applies