Scale items are modified only when theory justifies it, not for fit alone.

Evidence

An examination of the standardized covariance residuals (SCRs) revealed that the two-factor solution overestimated Items 3 and 8 (SCR 12.00) of the I-AB subscale, and underestimated the relationship between Items 10 and 15 (SCR 19.68) of the I-CG subscale. Because the two-factor model adequately fit the data and there was no theoretical rationale for improving the model’s fit, model modification was not implemented (Jackson, Gillaspy, & Purc-Stephenson, 2009; MacCallum, Roznowski, & Necowitz, Table 1 Confirmatory Factor Analyses Fit Statistics (Study 1) One factor 104 1155. — Sherman A. Lee, Chelsea A. Reid, Stephen D. Short, Jeffrey A. Gibbons, Ruth Yeh, and Matthew L. Campbell, Fear of Muslims, 53

Categorization

• SHARE: Niche psychometric detail about local misfit in the I-AB and I-CG subscales — the kind of thing only a methods-minded colleague would enjoy.