A Closer Look at the Items within Three Measures of Evolution Acceptance: Analysis of the MATE, I-SEA, and GAENE as a Single Corpus of Items

Background Current direct Likert measures for evolution acceptance include the MATE, GAENE, and I-SEA. Pros and cons of each of these instruments have been debated, and yet there is a dearth of research teasing out their similarities and differences when they are used together in a single context beyond the fact that their measures tend to be highly correlated. We administered these to 452 college students in non-major biology classes at two research-intensive universities from the Midwestern and Western United States to investigate the measurement properties of the items within these instruments when combined as a single corpus. Results Factor analysis using exploratory and confirmatory methods, and Rasch analyses, suggested that a two-dimensional factor structure best describes the corpus of items. Whether the item was positively or negatively worded was the key delimiter in its factor assignment. Examination of the highest loading items on the respective factors indicates that the first factor measures acceptance of the truth of evolution and the second factor measures rejection of incredible ideas about evolution. The correlation of these two factors is 0.73, indicating that they share 53% of their variance with each other. When treated unidimensionally, eleven items exhibited potential misfit with the Rasch model. This number dropped to nine items when the two factors were considered. These items, and implications for future use of the MATE, GAENE, and I-SEA together, are discussed in detail. Conclusions This study is the first analysis of the MATE, GAENE, and I-SEA as a single corpus of items, and yet corroborates previous work showing that these instruments yield measures with highly similar quantitative interpretations. This study also corroborates the effect of negative item wording on how college students interpret the item. While this finding can be applied to college-level students taking undergraduate non-majors biology coursework, work with more advanced biology students has demonstrated that this apparent item wording effect tends to disappear as students advance and become more accepting of evolution. We conclude that despite apparent epistemological differences between the MATE, GAENE, and I-SEA, these can be treated as a single set of items measuring a single factor or two factors without significant loss of quantitative interpretability