Embracing the Generalized Propensity Score Method: Measuring the Effect of Library Usage on First-Time-In-College Student Academic Success

Jingying Mao, Kirsten Kinsley



Objective – This research focuses on First-Time-in-College (FTIC) student library usage during the first academic year as number of visits (frequency) and length of stay (duration) and how that might affect first-term grade point average (GPA) and first-year retention using the generalized propensity score (GPS). We also want to demonstrate that GPS is a proper tool that researchers in libraries can use to make causal inferences about the effects of library usage on student academic success outcomes in observation studies.

Methods – The sample for this study includes 6,380 FTIC students who matriculated in the fall 2014 and fall 2015 semesters at a large southeastern university. Students’ library usage (frequency and duration), background characteristics, and academic records were collected. The Generalized Propensity Score method was used to estimate the effects of frequency and duration of FTIC library visits. This method minimizes self-selection bias and allows researchers to control for

demographic, pre-college, and collegiate variables. Four dose-response functions were estimated for each treatment (frequency and duration) and outcome variable (GPA and retention).

Results – The estimated dose-response function plots for first-term GPA and first-year retention rate have similar shapes, which initially decrease to the minimum values then gradually increase as the treatment level increases. Specifically, the estimated average first-term GPA is minimized when the FTIC student only visits the library three times or spends one hour in the library during his/her first semester. The threshold for first-year retention occurs when students visit the library 15 times or spend 21 hours in the library during their first semester. After those thresholds, an increase in students’ library usage is related to an increase in their academic success.

Conclusions – The generalized propensity score method gives the library researcher a scientifically rigorous methodological means to make causal inferences in an observational study (Imai & van Dyk, 2004). Using this methodological approach demonstrates that increasing library usage is likely to increase FTIC students’ first-term GPA and first-year retention rates past a certain threshold of frequency and duration.


Generalized Propensity Score; First-Time-in-College student; first-term GPA; first-year retention rate

Full Text:


DOI: http://dx.doi.org/10.18438/B8BH35

Evidence Based Library and Information Practice (EBLIP) | EBLIP on Twitter