Confidence Intervals for Recursive Journal Impact Factors

TL;DR

This paper develops confidence intervals for recursive impact factors and journal ranks, accounting for citation prestige, and applies these methods to economics journals, revealing most differences in journal quality are statistically insignificant.

Contribution

It introduces methods to compute confidence intervals for impact factors and journal ranks that incorporate citation prestige, with application to economics journals.

Findings

The Quarterly Journal of Economics has the highest impact and rank.

Most differences in journal quality are statistically insignificant.

Bootstrap methods broaden confidence intervals for mid-ranking journals.

Abstract

We compute confidence intervals for recursive impact factors, that take into account that some citations are more prestigious than others, as well as for the associated ranks of journals, applying the methods to the population of economics journals. The Quarterly Journal of Economics is clearly the journal with greatest impact, the confidence interval for its rank only includes one. Based on the simple bootstrap, the remainder of the Top5 journals are in the top 6 together with the Journal of Finance, while the Xie et al. (2009), and Mogstad et al. (2022) methods generally broaden estimated confidence intervals, particularly for mid-ranking journals. All methods agree that most apparent differences in journal quality are, in fact, mostly insignificant.