•  
  •  
 

Keywords

propensity score, underrepresented students, ethnically diverse, low-income, first generation to college, diversity, retention, graduation, quasi-experimental analysis, logistic regression

Abstract

Learning communities are a high impact activity that can influence students’ likelihood for success. Colorado State University (CSU) created the Key Communities (Key) program, which is open to all students but targets students that have persistently lower graduation and retention rates. The majority of Key students are under-represented (ethnically diverse, low-income, and/or first generation to college) and/or students with lower levels of academic preparation. This paper describes the structure and purpose of Key and shares the results of an institutional level assessment of Key’s impact on graduation and retention. Since participation in Key is not randomly assigned, this analysis utilizes propensity score matching to estimate Key’s treatment effect. Results show that Key has a positive impact on graduation and retention for all students, but Key is incredibly effective for students who come to CSU with characteristics that have historically put them at risk for attrition.

Taé Nosaka is the Director of the Key Communities and University Learning Communities Coordinator at Colorado State University in Fort Collins, CO.

Heather Novak is a research analyst in the Office of Institutional Research at Colorado State University.