Islamophobia measures split into two distinct dimensions, not one general construct.

Evidence

Two measurement models were tested. The one-factor model assumed that the IS items were driven by one general Islamophobia construct. The correlated two-factor model assumed that two distinct dimensions account for the relationship among the IS items. The results demonstrated salient differences between the hypothesized models (see Table 1). The one-factor model produced values outside of the recommended guidelines for all of the tests of model fit, indicating gross inadequate fit to the data. However, in support of our prediction, the two-factor model met all of Kline’s criteria for global model fit (SRMR .06, RMSEA .07, and CFI .98). Additionally, a scaled difference chi-square test (Satorra & Bentler, 2001) between the one-factor and two-factor model demonstrated significantly improved model fit from the one-factor to the two-factor model, SB 2 (1) 12.98, p .001. — Sherman A. Lee, Chelsea A. Reid, Stephen D. Short, Jeffrey A. Gibbons, Ruth Yeh, and Matthew L. Campbell, Fear of Muslims, 52

Categorization

• XREF: Connects to the reader’s interest in measuring contested social attitudes, echoing the measurement critique in The Reluctant Fundamentalist (Hamid) about reducing anti-Muslim sentiment to a single narrative.
• DEFINE: Defines a one-factor versus correlated two-factor measurement model and what fit statistics (SRMR, RMSEA, CFI) mean in practice.