Download this Paper Open PDF in Browser

How an Artificially Intelligent Virtual Assistant Helps Students Navigate the Road to College

27 Pages Posted: 25 Mar 2017 Last revised: 14 Oct 2017

Lindsay C. Page

University of Pittsburgh School of Education

Hunter Gehlbach

University of California, Santa Barbara

Date Written: October 11, 2017

Abstract

Deep reinforcement learning using convolutional neural networks is the technology behind autonomous vehicles. Could this same technology facilitate the road to college? During the summer between high school and college, college-related tasks that students must navigate can hinder successful matriculation. We employ conversational artificial intelligence (AI) to efficiently support thousands of would-be college freshmen by providing personalized, text-message based outreach and guidance for each task where they needed support. We implemented and tested this system through a field experiment with Georgia State University (GSU). GSU-committed students assigned to treatment exhibited greater success with pre-enrollment requirements and were 3.3 percentage points more likely to enroll on-time. Enrollment impacts are comparable to those in prior interventions but with substantially reduced burden on university staff. Given the capacity for AI to learn over time, this intervention has promise for scaling personalized college transition guidance.

Keywords: Artifical intelligence; college access; summer melt; randomized controlled trial

JEL Classification: I20, I21

Suggested Citation

Page, Lindsay C. and Gehlbach, Hunter, How an Artificially Intelligent Virtual Assistant Helps Students Navigate the Road to College (October 11, 2017). Available at SSRN: https://ssrn.com/abstract=2940297

Lindsay Page (Contact Author)

University of Pittsburgh School of Education ( email )

Pittsburgh, PA 15260
United States
412-648-7166 (Phone)

Hunter Gehlbach

University of California, Santa Barbara ( email )

Santa Barbara, CA 93106
United States

Paper statistics

Downloads
382
Rank
64,737
Abstract Views
2,355