Dear community,
In a recent study we published, we conducted a nationwide, longitudinal analysis (2013–2021) of 3,575 formally accredited research groups within the Colombian science system, using open governmental data from the Ministry of Science, Technology, and Innovation (MinCiencias).
The work contributes to literature in four principal ways:
Contextual expansion beyond high-income systems: The analysis addresses the persistent geographical bias in SciTS by examining a middle-income national research system using non-proprietary, public datasets.
Team-level measurement of disciplinary diversity: Disciplinary diversity is operationalized using the DIV indicator (Leydesdorff et al.), decomposed into variety, balance, and disparity. The study focuses on formal research groups rather than co-authorship networks, allowing assessment of stable organizational science teams.
Non-linear diversity–prestige relationship: The relationship between disciplinary diversity and scientific prestige (national rank A1–C) is non-monotonic and field-dependent. Higher diversity does not uniformly correspond to higher prestige. Instead, low diversity consistently characterizes groups with declining trajectories. Groups advancing in rank exhibit statistically similar diversity structures to volatile groups. And, effect sizes are statistically significant but small, reinforcing findings from meta-analytic work suggesting diversity alone explains limited variance in performance.
Trajectory-based analysis of team performance: Longitudinal classification (advancement, stagnation, decline, volatile) reveals that disciplinary diversity structures are more strongly associated with avoiding decline than with guaranteeing advancement.
We add insights on the discussion surrounding: "more interdisciplinary diversity is better” assumptions and suggest that optimal, field-specific diversity configurations may exist.
Thanks for your time.