In Europe, measures of Islamophobia cannot separate from prejudice against Turks

Evidence

The ATMS demonstrated high convergent validity with similar constructs (e.g., ethnic prejudice, r = .82), but did not demonstrate discriminant validity as the ATMS was highly correlated with prejudice against Turks (r = .91)6. — Ummul-Kiram Kathawalla, Fei Bi Chan, Quratulain Gulamhussein, Amy Riegelman, and Moin Syed, Conceptualization and Measurement of…, 542

Categorization

• XREF: Connects directly to the racialization argument noted in footnote 6 — Muslim and Muslim-like groups may be functionally interchangeable targets. Bolsters the recurring theme that Islamophobia is racialized rather than religious per se.
• QUESTION: Is a construct that correlates .91 with ethnic prejudice a distinct construct at all, or just a relabeled proxy? What r would be evidence of genuine discriminant validity?