Architecture of a virtual simulation teaching platform designed for college students' innovation and entrepreneurship training and its empirical study
Year: 2026
Type: article
Abstract: Innovation and entrepreneurship education requires experiential and adaptive learning environments, but evidence is often limited by weak theoretical integration, incomplete intervention reporting, and non-randomized designs. This study evaluated a theory-informed virtual simulation platform for entrepreneurial learning while explicitly accounting for allocation, instructional exposure, measurement quality, and clustering. A two-arm quasi-experimental study included 480 students from 12 intact classes. Baseline assessment preceded non-random classroom allocation to the virtual simulation platform (six classes; n = 240) or conventional training (six classes; n = 240); students did not self-select. Two instructors taught both conditions using common objectives, schedules, and assessment tasks over eight weeks. Multi-item measures underwent reliability and confirmatory factor analysis, performance tasks were scored by blinded independent raters, and primary analyses used covariate-adjusted ANCOVA with classroom-, instructor-, exposure-, missing-data-, and alternative-composite sensitivity analyses. Baseline characteristics were comparable, although post-allocation attendance, active learning time, assignment completion, and participation were higher in the platform condition. Platform assignment was associated with a 6.96-point higher composite learning score (95% CI: 6.34–7.57, p < 0.001); the estimate remained 5.38 points (95% CI: 4.69–6.07) after exploratory adjustment for observed instructional exposure. Classroom-clustered and instructor-adjusted models were consistent. Engagement was the only significant specific indirect pathway, but it explained only 5.41% of the total association. Scale reliability, construct validity, and inter-rater reliability were acceptable to strong. The findings provide supportive, but not definitive causal, evidence that a structured virtual simulation environment is associated with better entrepreneurial learning, engagement, and acceptability. The magnitude of benefit should be interpreted alongside exposure differences, possible novelty and Hawthorne effects, assessment alignment, classroom clustering, and the single-institution context.
Source: Scientific Reports
Authors: Tingyu Chen, Jie Liu, Guocai Yin, Xiaoyi Xuan
Institution: North China Institute of Aerospace Engineering
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: gold
Estimated APC paid: $2,690
OpenAlex ID: W7227323018
