Skip to main content
. 2025 Oct 15;16:1592658. doi: 10.3389/fpsyg.2025.1592658

Figure 2.

Chart consisting of three panels. Panel A shows a bell curve with psychological resources on the x-axis and percentage of population on the y-axis, depicting states from “mental disorder” to “flourishing.” Panel B illustrates a scatter plot comparing mental health assessments, highlighting patients in red and controls in black across population variance. Panel C presents a similar scatter plot for trait assessment, indicating high traits in purple and low traits in black.

Group differences as a function of population variance. (A) The dimensional view of mental disorders; adapted from Huppert et al. (2005). (B) The relationship between patients vs. controls group differences and population variance, assuming that patients are defined as 10% of the population with the poorest mental health. (C) A more general case illustrating the group differences resulting from the median split of the data (based on some cognitive measure) as a function of population variance. In both (B) and (C) we see that as true population variance increases, raw group difference increase too, while adding measurement error to the true scores results in misclassification of some individuals—which will end up attenuating observed group differences in the measures of interest.