Adaptation of the System Usability Scale (SUS) questionnaire for Russian-speaking respondents

Abstract

The paper describes the methodology and presents the results of adapting the popular usability questionnaire System Usability Scale (SUS) for Russian-speaking users. The adaptation was carried out on a sample of 657 Internet users, Russian-speaking natives, based on the web interface of the subscription service. Since one of the most debatable issues in the context of SUS use is the factor structure of this questionnaire, a three-part exploratory factor analysis was conducted (sequentially, by the method of minimal residuals, by the method of maximum likelihood, and by principal component analysis). All the methods used unambiguously supported the two-factor structure of the questionnaire. Confirmatory factor analysis additionally compared the three most frequent factor models of the SUS in the literature: the original one-factor, two-factor semantic and two-factor tone models. According to the results, the latter displays factor loadings more adequately to the data. This model differentiates the highlighted factors by tonality: odd-numbered questions have positive tonality and even-numbered questions have negative tonality. Cronbach's reliability test also showed high consistency across the resulting questionnaire subscales (α for F1 = .816; α for F2 = .763). Thus, the adapted version of the SUS represents a valid and reliable measure of system usability for the Russian-speaking audience and can be used both by fundamental researchers of cognitive processes and the specifics of their manifestation when interacting with interfaces of various types, and by practitioners designing user-friendly interfaces.