The Impact of Student Success Predictions and Dedicated Course Supports on Community College Student Academic Success.

How to cite

Castleman, Ben; Bird, Kelli. 2025. "The Impact of Student Success Predictions and Dedicated Course Supports on Community College Student Academic Success.." AEA RCT Registry. AEARCTR-0015886. https://doi.org/10.1257/rct.15886-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Ben Castleman* , Kelli Bird

* Primary investigator

Registry ID
AEARCTR-0015886
DOI
https://doi.org/10.1257/rct.15886-1.0
First registered
Last updated
Status
On going
Trial dates
2023-08-01 – 2025-12-31
Countries
Private
JEL codes
I23, I24
Registry record
http://www.socialscienceregistry.org/trials/15886

Abstract

We conduct a multi-semester randomized controlled trial (RCT) with a community college in Virginia of an intervention designed to improve student academic success in foundational Math, English, and Biology courses with historically-high course failure rates. The intervention consists of three distinct components: (1) Early-term predictions for instructors of which students may struggle to succeed academically; (2) embedded academic tutors; and (3) dedicated advisors to support students with broader challenges. Randomization is conducted at the instructor section-level. The primary outcome of interests are whether students pass the course, either with a "C" or higher or a "B" or higher, and whether students withdraw from the course prior to the end of the semester. We will investigate mechanisms underlying program impacts using college administrative data on individual-level take-up of college support services and learning managment system data on student engagement with course resources.

Intervention

The intervention consists of three distinct components: (1) Early-term predictions for instructors of which students may struggle to succeed academically; (2) embedded academic tutors; and (3) dedicated advisors to support students with broader challenges. Early-term predictions were generated by Student Success staff at the college, with guidance from the research team, during the 3rd and 4th week of the semester. Student Success staff and treated instructors reviewed these predictions to identify a priority list of students to intervene with over the balance of the semester. Embedded academic tutors and the dedicated advisor, coordinating with the instructor, proactively reached out to priority students to offer academic and more holistic support over the remainder of the semester.

Experimental design

Several months prior to each semester, we worked with PVCC to identify gateway courses (i.e. courses typically taken by students near the start of their time at PVCC, and satisfies pre-requisites for higher-level courses) to include in the pilot study, and then recruit instructors to participate in the pilot who would be teaching these gateway courses in the upcoming term. At the start of each semester, we randomized gateway course sections to either receive the intervention or to a control condition in which instructors pursued standard instructional practice and standard outreach to students; standard practice could have included instructors supporting students individually as well as referring them to PVCC advising and tutoring resources. Randomization occurred within gateway course, at the section level, e.g., if there were four English Composition I instructors signed up to participate in the project, we first randomly selected two instructors to receive the intervention and two instructors to be in the control condition. Then, for each instructor assigned to the intervention, we randomly chose one of their gateway course sections to receive the intervention; any other sections those instructors taught were not included in the pilot.

Design details

Randomization method
Computer random number generator
Randomization unit
Course section
Clusters
52 course sections
Observations
1133 student x course section observations
Arms
27 course sections in treated condition; 25 course sections in control condition
Primary outcomes
The primary outcome of interests are whether students pass the course, either with a "C" or higher or a "B" or higher, and whether students withdraw from the course prior to the end of the semester.
Secondary outcomes
We will investigate the impact of the intervention on end-of-semester course grades in the participating courses. We will investigate mechanisms underlying program impacts using college administrative data on individual-level take-up of college support services and learning management system data on student engagement with course resources. These include student use of tutoring and other support services (e.g. advising) at the institution; student log-ons, assignment submissions, activity participation, and instructor engagement to and via the Learning Management System.

Cited by

No citation data available for this registration yet.

Keywords