Extending validity testing of the Persuasion Principles Index

Abstract

PurposeThis study aims to independently test the predictive validity of the Persuasion Principles Index (PPI) for video advertisements for low-involvement products with a measure of in-market sales effectiveness. This study follows the inaugural test conducted by Armstrong et al. (2016) for print advertisements for high-involvement utilitarian products with a measure of advertising recall.Design/methodology/approachThe method was in line with that developed by Armstrong et al. (2016) for rating advertisements and assessing the reliability of ratings. Consensus PPI scores were calculated for a data set of 242 matched pairs of television advertisements. For each pair, the authors determined whether the advertisement that better adhered to the persuasion principles performed better in-market.FindingsConsensus PPI scores predicted the more sales effective television advertisement for 55% (confidence interval (CI) = 49%, 61%) of the 242 pairs. This result is no better than chance and much weaker than the result from the initial validation study, which found that the consensus PPI scores predicted the more recalled print advertisement for 74.5% (CI = 66%, 83%) of 96 pairs.Research limitations/implicationsThis study replicated the application of the PPI as per Armstrong’s guidelines and extended validity testing to a different set of advertising conditions. Findings indicate that better adherence to the persuasion principles produces only a weak, positive effect for predicting the performance of television advertisements for low-involvement products. A research agenda that flows from the results is discussed.Practical implicationsThe authors suggest that the PPI in its present form is best used to predict advertising performance under conditions as per the inaugural validation test (Armstrong et al., 2016).Originality/valueAdvertisers will require compelling evidence of the PPI’s predictive accuracy to adopt the tool for pre-testing advertising. This study is the first independent test of the predictive validity of the PPI and its generalisability across advertising conditions. Another contribution of this study is the assessment of Armstrong’s advice to remove unreliable ratings. The authors show that this procedure, surprisingly, does not improve the predictive accuracy of the PPI.