Measuring evolution acceptance using the GAENE: influences of gender, race, degree-plan, and instruction

Abstract

Background: The evolution education research community has defined the construct of "evolution acceptance" in different ways and measured it using different instruments. One of these instruments-the GAENE-has not been analyzed across different student populations, demographic groups, degree plans, and instructional treatments. Such comparisons are crucial for examining whether the inferences drawn from instrument measures are valid, reliable, and generalizable. In this study, we attempt to replicate findings produced in the original validation study and explore aspects of the instrument not previously examined. Methods: We use Rasch analysis to study a large sample (n > 700) of undergraduates enrolled in standard introductory biology classes in the Northeastern USA. Participants completed the GAENE pre-and post-course for two semesters, and the MATE pre-and post-course for one semester. We assessed dimensionality, reliability, item fit, and rating scale functioning. We used regression analyses and generalized eta squared to evaluate the contribution of demographic and background variables to pre-course measures and pre-post course acceptance gains. Results: Our analyses of GAENE dimensionality and item properties were generally in line with prior work, including the finding that particular items displayed psychometric problems. Surprisingly, GAENE measures did not differ between biology majors and non-majors. Evolution instruction produced significant but small pre-post improvements in GAENE measures. GAENE measures were significantly associated with MATE measures (0.68-0.80). White and male participants had the highest evolution acceptance measures using both the MATE and the GAENE; race had a much stronger contribution to MATE measures as compared to GAENE measures. Race and gender acceptance differences were found to be as large as the differences produced in response to evolution instruction. Conclusions: Overall measures of acceptance change will be similar, but not identical, using the MATE and the GAENE. We make several recommendations for the modification or removal of some GAENE items, as well as future research directions for the measurement of evolution acceptance.

Do you see what I‐SEA? A Rasch analysis of the psychometric properties of the Inventory of Student Evolution Acceptance

“…The GAENE was found to be unidimensional; 11 of the 13 GAENE items had acceptable fit statistics, acceptable separation reliabilities, and generally good item functioning (Sbeglia & Nehm, ). Similar to the I‐SEA, GAENE measures increased significantly from the pre‐course to the post‐course ( b = 0.61, df = 771, p < 0.001).…”

How strongly does statistical reasoning influence knowledge and acceptance of evolution?

“…Smith et al () reported that validity evidence was established by expert panels and statistical analyses of student responses based on CTT and IRT. Further work examining validity inferences was conducted by Sbeglia and Nehm ().…”

“…However, to our knowledge, the field lacks large‐scale empirical research examining quantitative relationships among statistical reasoning, evolution understanding, and evolution acceptance (cf. Sbeglia & Nehm, , ). Prior work has noted that many learning difficulties in evolution education are related to abstract concepts such as randomness and probability (e.g., Mead & Scott, ; Tibell & Harms, ).…”

Acceptance and knowledge of evolutionary theory among third-year university students in Spain

“…The main advantage is that the MATE has been the most frequently-used test to date [19,21,34,35]. Moreover, when this instrument has been compared with alternatives, a moderate to high correlation between them has typically been observed [19,21,36,37].…”