In: Statistics and Probability
The positive impact on women holistic advancement from unequal pay system for women. Describe how you may use CFA and SEM to conduct this own research, and why may choose to do so.
Evaluating the psychometric properties of a newly developed instrument is critical to understanding how well an instrument measures what it intends to measure, and ensuring proposed use and interpretation of questionnaire scores are valid. The current study uses Structural Equation Modeling (SEM) techniques to examine the factorial structure and invariance properties of a newly developed construct called Superwoman Schema (SWS). The SWS instrument describes the characteristics of a superwoman (strong woman) which consists of 35 items representing five subscales: obligation to present an image of strength, obligation to suppress emotions, resistance to being vulnerable, intense motivation to succeed, and obligation to help others. Multigroup confirmatory factor analysis (CFA) and a multiple indicators multiple causes (MIMIC) model were the SEM approaches used to examine measurement invariance in the SWS instrument. Specifically in the multigroup CFA analyses, configural invariance, metric invariance, intercept invariance, residual variance invariance, and latent mean invariance are examined between a group of young (18-39 years old) women and middle-aged (40- 65 years old) women. In the MIMIC model, the hypothesized model of the SWS was used to investigate the group differences in the young and middle-aged women. Both SEM techniques provided a didactic discussion about the findings of the study, which confirmed that the SWS instrument could be broadly used (i.e., invariance held) to compare young and middle-aged African American women on superwoman characteristics. Further research is needed to better understand the possible contextual factors (i.e., racial or gender stereotyping, oppression, spiritual values, etc.) that may contribute to group differences on the SWS subscales and minor violation to invariance.