Date of Award

Spring 6-9-2023

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Industrial/Organizational Psychology (PhD)

Department

Industrial/Organizational Psychology

First Advisor/Committee Member

Lynette Bikos

Second Advisor/Committee Member

Helen Chung

Third Advisor/Committee Member

Julianne Tillmann

Keywords

decent work, psychology of working theory, psychometric evaluation, India

Abstract

Work has a high impact on an individual’s well-being, health, and the lives of their loved ones who depend on them (Ford et al., 2013). The Decent Work Scale, constructed by Duffy et. al. (2017), measures what it means to have access to decent work (safe conditions, time for leisure and rest, support for family and social values, adequate compensation, and access to adequate health care) in the United States. However, a gap in the social sciences is the lack of external validity. Psychological research has mainly focused its attention on western, educated, industrialized, rich, and democratic nations and then generalized to other nations (Henrich, et. al., 2010). The purpose of this study was to extend the research on the Psychology of Working Theory (PWT; Duffy et al., 2016) through psychometric evaluation of the Decent Work Scale (DWS; Duffy et al., 2017) within the context of English-speaking Indian workers. Indian workers (N = 311) were recruited via Amazon Mechanical Turk (MTurk) for this study. It was hypothesized that similar to Duffy et al., (2017) the Indian sample will fit a bifactor model where items correspond to a single common factor (decent work) while also controlling for the multidimensionality of the five components. After conducting a confirmatory factor analysis, results indicated that factor loadings were sometimes low and sometimes with the opposite magnitude. The source of the problem appeared to be four reverse-scored items. A post-hoc analysis was conducted where the reverse-scored items were removed. The confirmatory factor models all showed acceptable fit, all factor loadings began performed in the proper direction, and the results supported the anticipated five-factor model. Furthermore, the exploratory factor analysis indicated five interpretable factors similar to Duffy (2017). The reliability of the total and subscale scores of decent work was significant and positive. The scores for validity, however, did not go in the direction we hypothesized them go to. Reasons why the Indian sample may have provided different results compared to the United States sample are explored. This includes reverse-scored items, the need for translation, and using MTurk as a sampling method. The results of this study yield future research considerations.

Download

Share

COinS