Comparing reputation vs h-index rankings of doctoral programs

Abstract

Purpose This study builds on previous investigations on the scholarship of social work faculty using h-index scores. The purpose of this paper is to compare two methods of determining the excellence of social work doctoral programs. Design/methodology/approach This study compared rankings in 75 social work doctoral programs using h-index vs the US News and World Report (USNWR) list. The accuracy of predicting scholarly productivity from USNWR rankings was determined by joint membership in the same quantile block. Information on USNWR rankings, h-index, years of experience, academic rank, and faculty gender were collected. Regression analysis was used in creating a predictive model. Findings Only 39 percent of USNWR rankings accurately predicted which programs had their reputation and scholarly productivity in the same rating block. Conversely, 41 percent of programs had reputations in a higher block than their scholarly productivity would suggest. The regression model showed that while h-index was a strong predictor of USNWR rank (b=0.07, 95% CI: 0.05, 0.08), additional variance was explained by the unique contributions of faculty size (b=0.01, 95% CI: 0.01, 0.02), college age (b=0.002, 95% CI: <0.001, 0.003), and location in the southeast (b=−0.22, 95% CI: −0.39, −0.06). Originality/value For many programs, reputation and scholarly productivity coincide. Other programs have markedly different results between the two ranking systems. Although mean program h-indices are the best predictor of USNWR rankings, caution should be used in making statements about inclusion in the “top 10” or “top 20” programs.