Search results

The argumentativeness measure has been used in more than a 100 studies since 1982. The measure was developed and validated within a US university/college student sample. Despite its intended use, the measure is regularly used outside of the US and outside of the university/college setting without tests of validity. There is also intense debate as to the dimensionality of the measure, with one camp defending the bi-dimensionality of the measure and another proposing uni-dimensionality. Therefore, the purpose of this study was to assess the utility of the measure outside of its intended population.

A total of seven samples were collected (n = 1860) from the UK, Germany, France, Turkey, the Philippines, Nicaragua and the US. In this study, Infante and Rancer’s (1982) original 20-item argumentativeness measure was used to assess argumentativeness. Confirmatory factor analyses was used to test content validity.

Fit statistics were consistently poor for the unidimensional factor structure. As there is debate as to whether the measure is uni or bi-dimensional, a bi-dimensional fit was also analysed. The measure performed slightly better in each sample using a bi-dimensional factor structure. However, fit statistics were still poor for each sample.

Specifically, the seven samples are convenience samples. While such a sampling technique does limit the generalizability of a study’s findings, convenience samples are common when using the argumentativeness measure. These results present avenues for exploring the dimensionality of the argumentativeness measure and for revisiting cross-cultural examinations of argumentativeness.

Factor structure is a critical issue in validity. Whether authors specify their prediction or not, factor structure is always hypothesized as part of a study when measurements are used, and therefore, should be examined in every study as part of the scientific process. Making claims about human behaviour based upon measures with mis-specified factor structures or other validity issues can lead to the perpetuation of misinformation within the literature.

This is one of the few studies to empirically explore the psychometric properties of one of the most used measures in argument/conflict research. In doing so, this study enhances the understanding of decades of argumentativeness research.