基于稀疏主成分回归的综合性人文社会科学类期刊质量的影响因素分析

本研究以最新公布的32种综合性人文社会科学类核心期刊为样本，根据2020年版《中文核心期刊要目总览》中的期刊排名，并借用功效系数法的思想测度了32种期刊的质量指数。本研究在此基础上基于多元线性回归和稀疏主成分回归方法实证研究了职称、机构、课题级别、单独署名论文比重、篇均参考文献数量、篇均论文页数和发表周期7个变量对综合性人文社会科学类学术期刊质量的影响。研究结果表明,就全样本而言，篇均论文页数在5％的显著性水平下对学术期刊质量有显著影响，其他变量对学术期刊质量都没有显著影响。从分样本回归结果来看，无论是多元线性回归还是稀疏主成分回归、无论是双一流大学样本还是非双一流大学样本，所有变量都对期刊质量没有显著影响。

An Analysis of the Factors Influencing the Quality of Comprehensive Humanities and Social Sciences Journals Based on Sparse Principal Component Regression

Taking the latest published 32 comprehensive core journals of humanities and social sciences as samples, the quality index of 32 journals was measured by using the idea of power coefficient method according to the journal ranking in the 2020 edition of Chinese Core Journals Overview. Based on multiple linear regression and sparse principal component regression,we studied the influence of 7 variables on the quality of comprehensive humanities and social science academic journals, including titles, institutions, topic levels, proportion of individual signed papers, number of references per paper and number of pages per paper and publication cycle. The results show that for the full sample, the number of pages per paper has a significant impact on the quality of academic journals at the significance level of 5%, while other variables have no significant impact on the quality of academic journals. From the results of sub-sample regression, no matter multiple linear regression or sparse principal component regression, no matter double first-class university sample or non-double first-class university sample, all variables have no significant impact on journal quality.