This study investigates the impact of university-based full-time equivalent (FTE) research and development (R&D) personnel on the productivity of National Social Science Fund (NSSF) projects in China. Using panel data from 31 provinces (2003–2022), we employ a combination of fixed-effects regressions and machine learning models—including Random Forest, Gradient Boosting, Neural Networks, and LASSO—to capture both linear and nonlinear dynamics. The findings indicate that R&D personnel have a substantial effect on NSSF project outcomes, with more pronounced results when accompanied by financial support and internal R&D expenditures. Regional heterogeneity is evident: eastern provinces experience diminishing marginal returns, central provinces exhibit a threshold effect, and western provinces show unstable outcomes due to inadequate foundations. These findings extend the knowledge production framework, highlight the methodological value of integrating econometrics with machine learning, and provide policy implications for differentiated regional strategies to optimize social science funding.
Yu et al. (Fri,) studied this question.