As part of ACT's Holistic Framework, ACT researchers developed a Behavioral Skills Framework that can be used to assess behavior in education and work settings. By utilizing specific statements of effective behaviors, known as performance level descriptors (PLDs) (e.g., communicating well in group tasks), the Behavioral Skills Framework is…

This study compares two methods of defining groups for the detection of differential item functioning (DIF): (a) pairwise comparisons and (b) composite group comparisons. We aim to emphasize and empirically support the notion that the choice of pairwise versus composite group definitions in DIF is a reflection of how one defines fairness in DIF…

Building upon the research findings in an earlier ACT Research Report (Westrick, 2016), this study used data from 119,131 students at 26 four-year institutions to make comparisons between STEM majors earning semester GPAs of 3.0 or higher and their STEM peers earning semester GPAs less than 3.0. The results indicate that the higher-performing…

ACT reporting categories and ACT Readiness Ranges are new features added to the ACT score reports starting in fall 2016. For each reporting category, the number correct score, the maximum points possible, the percent correct, and the ACT Readiness Range, along with an indicator of whether the reporting category score falls within the Readiness…

Using date from 120,612 students at 26 four-year institutions, this report provides a description of STEM majors entering four-year postsecondary institutions and those who persist through four years of study. The results indicate that persisting STEM majors enter college with very high levels of pre-college academic achievement as measured by the…

The use of categorical dependent variables with the classical linear regression model (CLRM) violates many of the model's assumptions and may result in biased estimates (Long, 1997; O'Connell, Goldstein, Rogers, & Peng, 2008). Many dependent variables of interest to educational researchers (e.g., professorial rank, educational attainment) are…

Presented at the national conference for AERA (American Educational Research Association) in April 2009. The large variability of SAT taker population across states makes state-by-state comparisons of the SAT scores challenging. Using a mixture modeling approach, therefore, the current study presents a method of identifying subpopulations in terms…

Distinctions were made between three classes of statistical linkage: equivalence, concordance, and prediction. These distinctions were based on rational content considerations and empirical statistical relationships. A large database involving SAT I and ACT scores was used to determine which type of linkage was best suited for different scores and…

Objective: To identify those factors that influence freshmen application to Commission on Accreditation of Allied Health Education Programs (CAAHEP)-accredited undergraduate ATEPs having a secondary admissions process. (All undergraduate ATEPs are now accredited by the Commission on Accreditation of Athletic Training Education (CAATE)). Design and…