Abstract
INTRODUCTION: Executive function is important ability for individuals to deal with daily living activities, and the Adult Executive Functioning Inventory (ADEXI) with promising psychometric properties was developed to assess executive function in nonclinical adult populations. However, psychometric properties evidence for ADEXI was primarily based on classical test theory; therefore, the current study used Rasch analysis to examine ADEXI to provide additional psychometric properties evidence. METHODS: Participants aged 18-25 years (n = 1764; 55.9% women) were recruited using a social media platform (Dcard). They completed self-report Mandarin version of the ADEXI (of which includes three subscales of working memory, inhibition, and flexibility) and demographic characteristics (age, gender, and educational level) via online survey. Rasch analyses with differential item functioning (DIF) were used as analytic methods. RESULTS: All ADEXI items aligned with their intended subscales, with fit statistics of mean square ranging between 0.5 and 1.5. Moreover, the 5-point Likert scale of the ADEXI followed its order to indicate the difficulty (i.e., a higher score in the Likert scale indicates a poorer level of executive function). Moreover, no DIF items were displayed for the ADEXI across gender subgroup (men vs. women) or educational subgroup (college or above vs. high school or below). CONCLUSION: The psychometric properties of ADEXI were satisfactory, as indicated by the Rasch analysis findings. Unidimensionality of each ADEXI subscale was supported, categorical function was in order for the response, and no items displayed DIF. Therefore, the ADEXI can be used to evaluate executive function for general population and to enhance health promotion.