Persuasive Advertising - J. Scott Armstrong [208]
We provide details on the WAPB analyses, along with full disclosure of the codings, at advertisingprinciples.com under the heading, “Analyses for Which Ad Pulled Best.” In addition, the website includes all the ads that we used along with the codings. In some cases, parts of the ads were not easily identifiable, although the text is always easy to read. The copy testing scores are available upon request to those who might want to conduct replications or extensions.
* * *
Appendix C
Non-experimental data on TV commercials
* * *
Many studies have used non-experimental data to compare the effectiveness of ads that use a principle with those that do not. Such studies are only useful when conditions (e.g., high-involvement or low-involvement product) play a small role. This applies although the sample sizes may seem large. Furthermore, the analyses are nearly always reported without regard for conditions. Even if the conditions were coded and included in the analyses, it would be impractical to consider many of them with only 1,000 commercials. More important, advertisers try to match their approach to conditions, so it is not really possible to assess the effects of the principles when important conditions apply.
The large-scale study by Stewart and Furse (1986) examined 1,059 TV commercials that had been submitted for testing to Research Systems Corporation. They involved mostly fast-moving consumer products, such as cereals, beverages, cleaners, and paper products. Almost 90 percent of the ads were of 30-second duration, and the others were of 60-second duration. Stewart and Furse used regression analysis to see how 160 features of the ads affected recall, comprehension, and persuasion. They provided a summary of their findings in their Appendix C, and I provide the full text at advertisingprinciples.com under “Developing ad campaigns.” (For related studies, see Stewart and Koslow 1989; Laskey, Fox, and Crask 1994; Stanton and Burke 1998; and Phillips and Stanton 2004.)
Dave Walker conducted extensive analyses of tested TV commercials and made these findings available for Persuasive Advertising. Dave is Senior Vice President, Global Research Director for Ipsos ASI, the advertising research division of the Ipsos Group (ipsos.com). His analyses look at one principle at a time and allow for an estimate of the importance of the principle.
Ipsos ASI tests ads in a context that simulates natural television viewing among respondents who are not initially aware that the advertising is the subject of the study. Multiple commercial spots (including the ads to be tested) are placed in a half-hour television show in a format typical of commercial programming. The program and the placement of ads are held constant across studies to provide a consistent context for test exposure. Respondents are screened and recruited by telephone to participate in a study to evaluate “new television material being considered for broadcast.” The day after watching the program, respondents are re-contacted by telephone for the follow-up interview to verify compliance with instructions, and to obtain information about recall and persuasion..
The recall score represents the number of respondents who recall an ad and describe it correctly. Ipsos ASI studies have shown that this measure is an indicator of the rate at which in-market exposure produces ad awareness, and is one indicator of the ad’s potential to impact consumer behavior. After the recall questions, respondents are re-exposed to selected ads, individually. For each ad, they answer a structured series of open- and closed-end questions including an in-depth profile of communication, reactions to the ad itself, purchase intent, and attitudes toward the brand. Ipsos ASI studies have demonstrated that their recall measure is correlated