Studying the context effect of family norms on gender role attitudes: An experimental design

The measurement of gender role attitudes has been found to be problematic in previous studies, especially in comparative perspective. The present study adopts a novel approach and investigates the position of the gender role attitudes scale in the questionnaire as a potential source of bias. In particular, the present study aims at assessing the context effect of the family norms question on the measurement of gender role attitudes by adopting the theoretical perspective of the construal model of attitudes, according to which the adjacent questions constitute the context for interpreting and answering a stimulus. The study employs data from the CROss- National Online Survey panel, which was fielded in 2017 and contained an experiment where the order of the questions under investigation varied. The reliability, invariance and validity of the measurement of gender role attitudes across experimental settings and countries (Estonia, Great Britain and Slovenia) are explored adopting several analytical techniques within the Structural Equation Modelling (SEM) framework. Whereas the measurement of gender role attitudes resulted stable between experimental settings, some differences emerged in terms of criterion and, to a lesser extent, construct validity. Differences among the countries emerged, indicating that the cultural context may interact with the question context. Finally, we stress that the measurement is overall poor, urging survey infrastructures to investigate more in depth the formulation of the items measuring gender role attitudes.