Developing an Empirically Valid Function Acquisition Speed Test for Assessing Attitudes to and Predicting Real-world Behaviour

The current research was designed to validate a newly developed Function Acquisition Speed Test (FAST; O’Reilly et al., 2012) developed to measure attitudes to abortion among the general population, and to retrospectively predict a real-world behaviour in relation to voting for or against the availability of abortion services, in a national referendum. Participants, consisting of those with openly expressed pro-life or pro-choice attitudes, were exposed to an Implicit Association Test (IAT), a FAST procedure, an explicit attitudes to abortion questionnaire, and a demographic survey. This procedure allowed for a comparison of FAST and IAT outcomes, inter-correlations between the explicit and implicit measures, as well as the predictive validity of the implicit measures of referendum voting behaviour and self-reported group affiliation. The first aim of this research was to assess the FAST’s utility in providing converging test results to predict both group affiliation and voting behaviour in the 2018 referendum on the 8th amendment concerning the legalisation of abortion. In seeking to validate the FAST, the gold standard IAT, as well as explicit test measures (i.e., the Attitudes to Abortion Questionnaire and Demographic survey), were administrated to assess convergent and predictive validity. The second aim of this research was to meaningfully compare the two implicit test measures, by employing common data scoring methods across the tests (i.e. using Rate-Fluency Differential or RFD scores, D scores and slope differential scores). Results showed that the FAST predicted group affiliation and voting behaviour satisfactorily. In particular, FAST RFD scores were found to be the optimal scoring metric when predicting voting behaviour, whereas FAST slope scores were found to be the optimal scoring metric when predicting group affiliation. Cut-off points were also provided for both scoring metrics, contributing to the production of a meaningful scoring system for the FAST. In terms of the implicit measures, the IAT D scores was found to have the best overall predictive validity. However, neither test outperformed the almost perfectly predictive explicit test score, regardless of the scoring metric used. Nevertheless, the FAST offers a functional model of implicit attitudes that is wanting for the IAT, and additional research is needed in sensitive contexts, in order to increase the FAST’s predictive validity up to and beyond that of explicit measures. Several methodological and conceptual issues are considered in an effort to accelerate fruitful research towards this end.

Level dc:type.qualificationlevel

masters

Grantor dc:publisher.institution

National University of Ireland Maynooth

Year dc:date.issued

2019

Author dc:creator

• Lalor, Isabella

Language dc:language

en