Development and Initial Validation of Two Brief Measures of Left-Wing Authoritarianism: A Machine Learning Approach

Abstract

Although authoritarianism has predominantly been studied among political conservatives, authoritarian individuals exist on both “poles” of the political spectrum. A 39-item multidimensional measure of left-wing authoritarianism, the Left-wing Authoritarianism Index, was recently developed to extend the study of authoritarianism to members of the far-left. The present study coupled a fully automated machine learning approach (i.e., a genetic algorithm) with multidimensional item response theory in a large, demographically representative American sample (N = 834) to generate and evaluate two abbreviated versions of the Left-wing Authoritarianism Index. We subsequently used a second community sample (N = 477) to conduct extensive validational tests of the abbreviated measures, which comprise 25- and 13-items. The abbreviated forms demonstrated remarkable convergence with the full LWA Index in terms of their psychometric (e.g., internal consistency) and distributional (e.g., mean, standard deviation, skew, kurtosis) properties. This convergence extended to virtually identical cross-measure patterns of correlations with 14 external criteria, including need for chaos, political violence, anomia, low institutional trust. In light of these results, the LWA-25 and LWA-13 scales appeared to function effectively as measures of LWA.

Data availability statement

The data, analytic code, and supplementary materials that support the findings of this study are openly available at https://osf.io/thzs6/?view_only=a5be8872500447588be4097fae1d42b5.

Open Scholarship

This article has earned the Center for Open Science badges for Open Data through Open Practices Disclosure. The data are openly accessible at http://doi.org/10.17605/OSF.IO/THZS6.

Notes

1 We aired on the side of caution, such that participants who failed any one of our three attention checks were screened out. We adopted this strict approach to mitigate the possibility that our analysis of low base-rate outcomes, such as participation in political violence, were not distorted due to errant responding.

2 Indeterminacy of factor scores in the common factor model is a well-known issue. Problems pertaining to biased cross-factor correlations due to indeterminacy are resolved by our use of ESEM rather than CFA. Further, in IRT, the equivalent of indeterminacy is measured using reliability coefficients for θ, which we assessed using item- and test-information curves produced by MPlus (see online supplemental materials). Reliability of the estimates for all three factors was high for values of θ within +/- 2.5, further justifying our use of factor scores.