Composite Scientist Scores: How to Read a Number Out of 100

A score built from several indicators depends on the weights, transformations, missing-data rules and comparison group selected by its designers. It can organise complex evidence, but the number is not a direct natural measurement of scientific quality.

Changing the relative emphasis on career citations, recent activity or open science can reorder a leaderboard. No single weighting reflects every scientific discipline or every legitimate definition of contribution.

Researchers clustered near one another may exchange ranks after small data corrections or an updated snapshot. A tiny gap between scores should not be portrayed as a large difference in individual value or achievement.

For RSIS, examine the six-pillar design, field normalisation, evidence-completeness requirements and version label. A score is best understood as a documented comparison result for a specified dataset and time.