Two scholarly publishing cultures? Open access drives a divergence in European academic publishing practices

Abstract

The current system of scholarly publishing is often criticized for being slow, expensive, and not transparent. The rise of open access publishing as part of open science tenets, promoting transparency and collaboration, together with calls for research assesment reforms are the results of these criticisms. The emergence of new open access publishers presents a unique opportunity to empirically test how universities and countries respond to shifts in the academic publishing landscape. These new actors challenge traditional publishing models, offering faster review times and broader accessibility, which could influence strategic publishing decisions. Our findings reveal a clear division in European publishing practices, with countries clustering into two groups distinguished by the ratio of publications in new open access journals with accelerated review times versus legacy journals. This divide underscores a broader shift in academic culture, highlighting new open access publishing venues as a strategic factor influencing national and institutional publishing practices, with significant implications for research accessibility and collaboration across Europe.

Figures (6)

• Figure 1: Distributions of publications ratio at university and country level. (A) $\rho$ (see eq. 1) distribution for 2022 at the university level, Full line shows two-gaussian mixture fit, individual gaussians are plotted with dashed lines. (B) $\rho$ distributions for 2022 at the country level.
• Figure 2: Publications ratio at country level.$\rho$ for 2022 for countries with universities ranked in CWTS Leiden Open Ranking. The thick horizontal line corresponds to the minimum of the two-gaussian mixture fit. The dotted lines are the means $\mu_1$ and $\mu_2$ of the fitted two gaussians and the shaded bands are the corresponding widths $\sigma_1$ and $\sigma_2$ of the distributions.
• Figure 3: Evolution of publications ratio. Mean $\rho$ vs years for $\rho_{high}$ (open circles) and $\rho_{low}$ (open squares) countries. Mann-Wittney U test shows statistically significant differences in mean $\rho$ between $\rho_{high}$ and $\rho_{low}$ countries with $p<0.01$.
• Figure 4: Statistical significance of $\rho_{high}$ and $\rho_{low}$ separation. Mean p-values for Mann-Whitney U test for differences in mean $\rho$ between $\rho_{high}$ and $\rho_{low}$ countries. The p-values are calculated for 10000 random permutations of countries between the two groups in which 0 (open circles), 1 (open squares) or 2 (open triangles) countries are randomly switched between the two groups.
• Figure 5: Publication ratio and ranking. Relationship between $\rho$ and the rank of universities (panel A) and countries (panel B) in CWTS Leiden Open Ranking. The individual data points in the figure are the $\rho$s of universities computed as averages in bins 50 ranks wide.