Chernozub O.L. Do indirect measures of attitudes improve our predictions of behavior? Evaluating and explaining the predictive validity of GATA. RUDN Journal of Sociology. 2024. Vol. 24. No. 1. P. 241-258. doi: 10.22363 ...

Текст статьи на сайте журнала URL: https://journals.rudn.ru/sociology/article/view/38512 (дата обращения 24.06.2024)

Abstract

The generalisation of a large number of accumulated results by this time showed that implicit measures of attitudes (some even suggest re-naming them "indirect", as it is less pretentious) demonstrate a disappointingly weak predictive potential in relation to real behaviour. Against this background, the predictive validity of the Graphical Association Test of Attitude (GATA), which also claims to be an indirect method, has been questioned. To test this, we analysed the results obtained by using GATA in 64 predictions where the predicted outcome can be verified by real action. Forecasts cover the domains of electoral, consumer and communicative behavior. In some cases, the prediction made on the basis of data from a representative sample has been checked against the actual behaviour of the group represented by the sample, e.g. the electorate, or the consumers of a certain category of goods, etc. In other cases, the accuracy of the forecast was checked for each individual respondent. This avoids the effect of "mutual compensation" of erroneous forecasts with opposite valence. The test method was a comparison of the prediction accuracy of pairs of “control” and “experimental” prediction models. The only difference between the latter and the former was that the latter used the data from the indirect measurements of GATA as an additional factor of action. All models are intentionally presented in their simplest and most transparent versions. It turned out that the results of our meta-analysis do not fully correspond to the general trend: the use of GATA data significantly and continuously improves the accuracy of predicting behaviour. In addition, the incremental effect on the accuracy of individual forecasts (for each respondent individually) turned out to be higher than that of sample-based group forecasts.