May, A. L., & Stone, C. A. (2014). An initial investigation into the role of stereotype threat in the test performance of college students with learning disabilities . Journal of Postsecondary Education and Disability , 27 (1), 89–106. https://www.ahead.org/professional-resources/publications/jped

[no doi reported]; also located on ERIC online database: https://eric.ed.gov/?id=EJ1029630

Stereotype threat and reduced threat conditions were the independent variable being studied; in addition, the association of extended-time accommodations was examined for their role in supporting students with learning disabilities.

Postsecondary students with (n=29) and without (n=62) learning disabilities, from two universities in the U.S. Midwest participated. Participants were males and females of European-American descent. Details were reported about the nature of the learning disabilities—such as reading-related, writing-related, language-related, math-related, attention-related, nonverbal, and other.

Postsecondary students completed a set of 45 sample and discontinued items from the Verbal section of the Graduate Record Examination (GRE). Participants were also asked to report their Verbal SAT or English ACT scores. Under the stereotype threat condition, students were told that the test was of verbal reasoning and reading abilities, and under the reduced threat condition, students were told that the test was of problem-solving style. Note: students' response accuracy for the GRE items was scored using an ETS approach, which awarded +1 point for each correct response and -1/5 point for each incorrect response; items left blank were not counted as incorrect. When compared with total correct and proportion correct out of items attempted, the score comparisons were similar.

Postsecondary students without disabilities scored similarly and took similar amounts of time per item in both the stereotype threat and reduced threat conditions. Postsecondary students with learning disabilities in the reduced threat condition had similar scores, and spent similar time per item, as students without disabilities. In the stereotype threat condition, students with learning disabilities spent more time per item yet scored no better or worse than in the reduced threat condition, on average, on the items they completed. This result was counter to researchers' expectations, which were that students with disabilities, when primed with stereotype expectation that they would not perform comparably when told that the task was ability-based, would meet the lower expectations. Other score patterns included that students with learning disabilities completed fewer items and skipped more items in both conditions than students without disabilities. This finding has implications on these students' possibly not benefitting from extended time accommodations—at least not with the ETS scoring approach, in that their test results were not significantly worse when not completing as many items. The researchers concluded that these research participants might not have held negative assumptions about their capabilities to the degree that it affected their scores under the different conditions. Limitations of the study were reported, and future research directions were suggested.