Research on the evaluation of academic journals based on structural equation modeling

摘要

This paper classifies common journal evaluation indicators into three categories, namely three first-level indicators. They are respectively the indicators on journal impact, on timeliness, and on journal characteristics. The data used here is drawn from the medical journals in CSTPCD, a citation database built by the Institute of Scientific and Technical Information of China. The three categories of indicators are correlated with one another, so a structural equation may be established. Then we calculate the value of three first-level indicators and give subjective weights to the indicators. The comprehensive evaluation upon the medical journals yields satisfactory results. By simulating the complex relationship among journal indicators, the structural equation can be used for the estimation of some implicit indicators and the screening of indicators. This approach provides a new perspective for scientific and technological evaluation in general sense. It should be noted that the availability of basic data and the rationality of modeling bear much upon the evaluation results.