Examining measures of schizotypy for racial and gender bias using item response theory and differentiral item functioning
No Thumbnail Available
Authors
Meeting name
Sponsors
Date
Journal Title
Format
Thesis
Subject
Abstract
Schizotypy, a latent personality organization reflecting liability for schizophrenia, includes abnormalities and deficits in functioning and traits which cluster into positive, negative, and disorganized dimensions (Lenzenweger, 2018). Research within schizotypy provides useful information regarding etiological mechanisms and risk and protective factors for schizophreniaspectrum conditions without the presence of confounds such as medication effects and pronounced cognitive deficits associated with schizophrenia-spectrum disorders. The Magical Ideation Scale (MagicId), Perceptual Aberration Scale (PerAb), Revised Social Anhedonia Scale (RSAS), and the Cognitive Slipage Scale (CSS; Chapman et al., 1976; Chapman et al., 1980; Eckblad & Chapman, 1983; Eckblad et al., 1982; Miers & Raulin, 1987) are a collection of scales which have historically been used to measure schizotypy. A critical and unresolved issue is to what extent measures of schizotypy show racial and gender bias in their assessment with recent evidence suggesting these scales possess biased items. This current study examined racial and gender biases in items of the scales with the use of Item Response Theory (IRT) to test for levels of differential item functioning (DIF). Only two previous studies have used IRT analysis to test for racial and gender bias within the MagicId, PerAb, and/or SocAnh scales and this study was the first for the CSS. I had data for the following racial/ethnic categories: Black (African- American), Asian American, Hispanic, Multiracial, and White. Overall, I found some evidence for gender bias as well as racial bias (especially when comparing Black versus White participants) for some of these scales.
Table of Contents
DOI
PubMed ID
Degree
M.A.