A QUANTITATIVE EXPLORATION OF MATH LEARNING CENTER VISIT BEHAVIOR

Abstract

Mathematics Learning Centers (MLCs) have been widely recognized as effective academic support environments in higher education, with substantial evidence linking student participation to improved academic performance, persistence, and engagement. However, much of the existing research relies on aggregate measures of attendance, such as total visits or hours spent. Thus, this limit understanding of the underlying behavioral patterns that drive these outcomes. This study addresses that gap through a quantitative analysis of student visitation behavior within a university-based MLC. Drawing on a novel theoretical framework that conceptualizes the MLC as a service-oriented “storefront,” this research integrates perspectives from foot traffic analysis, store layout and flow theory, and service operations management. Using de-identified attendance data comprising over 35,000 student visits across multiple semesters, the study examines patterns of usage across time, course, and physical space within the center. Analyses include descriptive statistics, data visualization, negative binomial regression modeling, clustering techniques, and predictive modeling to identify distinct patterns of student engagement. Findings reveal that MLC utilization is shaped by both course-level demand and temporal structures, but that traditional measures of participation are misleading when not normalized by enrollment. High-enrollment business math courses produced the largest number of total and unique visitors yet exhibited lower participation rates relative to enrollment when compared to smaller calculus courses. This demonstrates that raw attendance alone inflates perceived engagement and obscures meaningful differences in utilization across course contexts. Temporal analyses further reveal that demand is highly structured and predictable. Peak visitation occurred between 10:00 a.m. and 2:00 p.m., with pronounced surges in the weeks leading up to coordinated exams. These surges are not uniform across the center. Course spaces that served business mathematics courses and college algebra courses, such as the yellow and green rooms, experienced sharp pre-exam increases in demand. In contrast, course spaces that served calculus, such as the blue room, maintained relatively stable usage patterns. When measured using occupancy rather than check-in counts, these patterns provided a more accurate representation of real-time service demand, offering a more effective basis for staffing and operational planning. Using these occupancy-based models, the study identified optimized staffing thresholds across rooms and time blocks, demonstrating how staffing resources are aligned with predictable demand cycles. At the student level, distinct engagement patterns emerged. While early-semester visits are associated with higher likelihood of continued usage within the same semester, students who begin visiting later but engage at high frequency exhibited a strong probability of returning in subsequent semesters. This finding reframed traditional assumptions about early intervention by highlighting that sustained engagement, regardless of when it begins, plays a critical role in long-term participation. In practice, this suggests that initiating student engagement at any point in the semester and supporting continued usage once students enter the system becomes the primary objective of MLC operations. From an operational perspective, these findings underscore the importance of shifting from descriptive attendance metrics to behaviorally informed, service-oriented analytics. Effective MLC management requires not only attracting students, but strategically supporting patterns of repeated engagement, aligning staffing with predictable demand windows, and designing spaces that respond to course-specific usage patterns. By applying a service-oriented analytical lens, this study provides a more nuanced understanding of how students interact with academic support environments and offers actionable insights for optimizing staffing, space allocation, and service delivery. Ultimately, this work contributes to the research literature by linking detailed behavioral patterns of MLC usage to scalable, data-driven strategies for improving student success in mathematics.