California Community College Administrators\u27 Use of Predictive Modeling to Improve Student Course Completions

Purpose: The purpose of this case study was to determine the impact of utilizing predictive modeling to improve successful course completion rates for at-risk students at California community colleges. A secondary purpose of the study was to identify factors of predictive modeling that have the most importance for improving successful course completion rates for at-risk students as perceived by California community college administrators. Methodology: This case study identified specific administrators at five community colleges within two California community college districts using predictive modeling to improve successful course completion rates. Participants were chosen based on specific criteria. The study was designed to collect data through interviews, documents and archival sources to answer the research questions. Findings: These findings were identified as impacts: (1) no discernable improvement in course completion rates; (2) student contact, (3) timely intervention strategies; (4) identify and monitor students; (5) sufficient support services; (6) successful completions and retentions to achieve educational goal; and (7) institutional metrics and reporting. The findings identified as important factors were: (1) planning and strategy; (2) communication and training; (3) resources; (4) outcomes; and (5) inclusion. Conclusions: It is too early to determine any impact on successful course completion rates by using predictive modeling software. A diverse population of stakeholders must jointly determine the outcomes desired from and identify the data needed to accurately analyze and model predictions. These data streams allow policy decisions to start with data. Predictive modeling software is a tool to identify students for timely and specific interventions. Increasing a student’s sense of belonging, engagement, and awareness is important to successful course completions. Administrators need assistance with and exposure to data analytics and predictive modeling to establish a data-driven decision-making culture. A culture of continuous review and improvement of the predictive models should be established. Recommendations: Provide administrators and other personnel with professional-development learning activities related to using data to inform policy and procedures that encourage student engagement, strategies for student success, and a cycle of continuous review and improvement