Predictors of graduate employment outcomes: a study on universities of applied sciences in China using a machine learning approach

Identifying predictors of graduate employment outcomes is crucial for higher education institutions focusing on industry-ready graduates, such as the universities of applied sciences (UAS) in China. Using data from 1,817 graduates across three Chinese UAS, we employed the LightGBM machine learning model to identify key determinants of employment outcomes through the lens of the Graduate Capital Model. The results revealed seven key predictors, in descending order of feature importance: student leadership experience, professional certificates obtained, career counseling received, parental education level, academic practice, frequency of interaction with instructors, and perceived improvement from innovation and entrepreneurship education. MANOVA further showed that graduates in full-time roles significantly outperformed their non-employed and part-time counterparts on these factors, with those in major-unrelated full-time work often leading. The findings underscore that employability is not determined by a single factor but by the dynamic interplay of human, social, cultural and identity capital. This study provides empirical evidence for Chinese UAS to strategically improve their graduate employment outcomes. The findings may also benefit other higher education institutions prioritizing employability.

Authors

Institutions

Publication Details

Journal
Higher Education Research & Development
Published
2026-11-13
DOI
https://doi.org/10.1080/07294360.2026.2760275
Primary Topic
Higher Education and Employability
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Predictors of graduate employment outcomes: a study on universities of applied sciences in China using a machine learning approach

Hui Chen, Jian Li
Higher Education Research & Development
Higher Education and Employability
article

Predictors of graduate employment outcomes: a study on universities of applied sciences in China using a machine learning approach

Hui Chen, Jian Li
article en

Abstract

Identifying predictors of graduate employment outcomes is crucial for higher education institutions focusing on industry-ready graduates, such as the universities of applied sciences (UAS) in China. Using data from 1,817 graduates across three Chinese UAS, we employed the LightGBM machine learning model to identify key determinants of employment outcomes through the lens of the Graduate Capital Model. The results revealed seven key predictors, in descending order of feature importance: student leadership experience, professional certificates obtained, career counseling received, parental education level, academic practice, frequency of interaction with instructors, and perceived improvement from innovation and entrepreneurship education. MANOVA further showed that graduates in full-time roles significantly outperformed their non-employed and part-time counterparts on these factors, with those in major-unrelated full-time work often leading. The findings underscore that employability is not determined by a single factor but by the dynamic interplay of human, social, cultural and identity capital. This study provides empirical evidence for Chinese UAS to strategically improve their graduate employment outcomes. The findings may also benefit other higher education institutions prioritizing employability.

Higher Education Research & Development
Beijing Normal University (CN)
Higher Education and Employability
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon