Dissertation Defense Announcement Ph.D. in Education Program: Arminda Wey “Faculty Decision Making and Multiple Measures in Community College Mathematics Placement: A Mixed Methods Study”

Accurate mathematics placement is essential to student momentum in community colleges, yet institutions continue to struggle with placing students into the most appropriate entry-level course. Persistent barriers include limited access to timely and standardized high school records, uneven rigor and grading practices across secondary schools, and institutional reliance on single indicators that may be unreliable.

This mixed-methods dissertation investigates how faculty integrate multiple measures when making placement decisions for first-semester mathematics, focusing on decision environments, confidence, and agreement. The study is guided by Behavioral Decision Theory (BDT) and Effective Institutional Decision Making (EIDM), aiming to clarify when and how different information conditions shape placement decisions.

The study employed a sequential explanatory mixed-methods design. In the quantitative phase, five community college mathematics faculty members each assigned placements for 104 anonymized student profiles under three conditions: standardized scores only, high school records only, and both combined. Data sources included SAT and ALEKS scores, recalculated high school GPA, math GPA, and indicators of advanced or basic high school coursework. Chi-square tests examined variation in placement and confidence across conditions; interrater agreement was estimated using Fleiss’ kappa; and ordinal and binary logistic regression identified predictors of higher-level placement.

The qualitative phase used semi-structured interviews to explore faculty sense-making, perceptions of college readiness, and the value of information beyond grades and test scores. Quantitative findings showed that placement distributions differed significantly by information condition. Agreement was highest with standardized scores (κ = .73), lowest with high school data alone (κ = .18), and fair when both were combined (κ = .34). Confidence in placement decisions also varied significantly by condition, with the highest confidence reported when faculty had access to combined data. Logistic regressions indicated that ALEKS scores, math GPA, and advanced high school mathematics substantially increased the likelihood of higher placement, whereas completing only basic mathematics markedly increased the likelihood of developmental placement. Qualitative findings revealed that faculty prefer multiple measures but view standardized metrics as more stable and defensible cues. High school metrics, though valuable, were often discounted due to variability in rigor and grade inflation. Faculty highlighted the importance of noncognitive factors, such as study habits and self-advocacy, that are not captured in traditional metrics. They also emphasized the unreliability of self-reported data among lower-scoring students, underscoring the need for validated records and decision supports.

Overall, the study demonstrates that the structure, quality, and clarity of information significantly shape placement decisions, confidence, and interrater agreement. The findings support placement systems that combine standardized indicators with verified math-specific high school metrics and targeted decision supports. This research contributes a replicable methodological model for institutional inquiry and offers evidence-informed recommendations for improving mathematics placement accuracy, transparency, and equity in community colleges.