Does the average JIF percentile make a difference?
- 652 Downloads
Average journal impact factor (JIF) percentile is a novel bibliometric indicator introduced by Thomson Reuters. It’s of great significance to study the characteristics of its data distribution and relationship with other bibliometric indicators, in order to assess its usefulness as a new bibliometric indicator. The research began by analyzing the meaning of average JIF percentile, and compared its statistical difference with impact factor. Based upon factor analysis, the paper used multivariate regression and quantile regression to study the relationship between average JIF percentile and other bibliometric indicators. Results showed that average JIF percentile had changed the statistical characteristic of impact factor, e.g. improved the relative value of impact factor, having smaller variation coefficient and distribution closer to normal distribution. Because it’s non-parametric transformation, it cannot be used to measure the relative gap between journals; Average JIF percentile had the highest regression coefficient with journal impact, followed by timeliness and lastly the citable items; The lower the average JIF percentile, the higher the elastic coefficient of journal impact; When average JIF percentile was extremely high or extremely low, citable items were not correlated with the average JIF percentile at all; When average JIF percentile was low, elastic coefficient of timeliness was even higher; Average JIF percentile was not a proper indicator for multivariate journal evaluation; Average JIF percentile had both the advantages and disadvantages of impact factor, and thus had the same limitation in applying as the impact factor.
KeywordsAverage JIF percentile Impact factor Quantile regression
This research was supported by China Scholarship Council (No. 201506270024) and the National Natural Science Fund of China (NSFC) (No. 71303179). The authors would thank anonymous reviewers for their helpful comments.
- Bradford, S. C. (1985). Sources of information on specific subjects. Journal of information Science, 10(4), 173–180.Google Scholar
- Hintze, J. L., & Nelson, R. D. (1998). Violin plots: A box plot-density trace synergism. The American Statistician, 52(2), 181–184.Google Scholar
- Thomson Reuters. JCR Reports. (2016). http://ipscience-help.thomsonreuters.com/incitesLiveJCR/glossaryAZgroup/g8/9586-TRS.html. Accessed 19 Mar 2016.