Fig. 2: Violin Plots Comparing Annotator and Patient Scores.
From: Validity of two subjective skin tone scales and its implications on healthcare model fairness
A, B – Violin Plots Comparing Annotator and Patient Scores. The Fitzpatrick (scale of I–VI) and Monk (scale of 1–10) refer to the two skin tone scales used for this study. Violin plot showing the distribution of differences between annotator consensus scores and patient self-reported scores, stratified by subject-reported scores. Positive values indicate that annotators assigned higher (darker) scores than the patients’ self-reported scores, while negative values indicate lower (lighter) annotator scores. The spread and median of differences highlight systematic discrepancies, with annotators tending to rate closer to the mid-range compared to the subject’s self-assessment.
