“Can You Ever Be Too Smart for Your Own Good? Linear and Nonlinear Effects of Cognitive Ability”, Matt Brown, Jonathan Wai, Christopher Chabris2020-01-30 (, ; backlinks; similar)⁠:

Despite a long-standing expert consensus about the importance of cognitive ability for life outcomes, contrary views continue to proliferate in scholarly and popular literature. This divergence of beliefs among researchers, practitioners, and the general public presents an obstacle for evidence-based policy and decision-making in a variety of settings. One commonly held idea is that greater cognitive ability does not matter or is actually harmful beyond a certain point (sometimes stated as either 100 or 120 IQ points).

We empirically test these notions using data from 4 longitudinal, representative cohort studies comprising a total of 48,558 participants in the US and U.K. from 1957 to the present.

We find that cognitive ability measured in youth has a positive association with most occupational, educational, health, and social outcomes later in life. Most effects were characterized by a moderate-to-strong linear trend or a practically null effect (mean R2 = 0.002 to 0.256). Although we detected several nonlinear effects, they were small in magnitude (mean incremental R2 = 0.001). We found no support for any detrimental effects of cognitive ability and no evidence for a threshold beyond which greater scores cease to be beneficial.

Thus, greater cognitive ability is generally advantageous—and virtually never detrimental.

Figure 5: Locally weighted regression plots for annual income (top) and educational attainment (bottom) regressed on cognitive ability. Annual income is displayed in dollars (or pounds) without a log transformation. A practically important nonlinear effect was found for annual income within the National Longitudinal Survey of Youth1997 (NLSY97; Bureau of Labor Statistics, 2019b) cohort (survey waves 2008, 2009, 2010, and 2011) but not in any of the remaining three cohorts. Educational attainment is reported in number of years (Wisconsin Longitudinal Survey [WLS; Herd et al 2014], National Longitudinal Survey of Youth1979 [<a href=NLSY79; Bureau of Labor Statistics, 2019a], and NLSY97) or using the National Vocational Qualification (NVQ; 1970 British Cohort Study [BCS70; Elliott & Shepard2006]). A practically important nonlinear effect was found for educational attainment in the BCS70 but not in any of the remaining cohorts.” />
Figure 5: Locally weighted regression plots for annual income (top) and educational attainment (bottom) regressed on cognitive ability. Annual income is displayed in dollars (or pounds) without a log transformation. A practically important nonlinear effect was found for annual income within the National Longitudinal Survey of Youth1997 (NLSY97; Bureau of Labor Statistics, 2019b) cohort (survey waves 2008, 2009, 2010, and 2011) but not in any of the remaining three cohorts. Educational attainment is reported in number of years (Wisconsin Longitudinal Survey [WLS; Herd et al 2014], National Longitudinal Survey of Youth1979 [NLSY79; Bureau of Labor Statistics, 2019a], and NLSY97) or using the National Vocational Qualification (NVQ; 1970 British Cohort Study [BCS70; Elliott & Shepard2006]). A practically important nonlinear effect was found for educational attainment in the BCS70 but not in any of the remaining cohorts.
Figure 6: Summary of linear and nonlinear cognitive ability effects by outcome. Red bars represent the percentage of variance explained (R2) by the linear effect of cognitive ability. In ~90% of all models, linear effects indicated that greater cognitive ability is predicted to yield better occupational, educational, health, or social outcomes (194⁄214 models). Offset black bars represent the incremental percentage of variance accounted for by the quadratic effect of cognitive ability.

View PDF:

Can You Ever Be Too Smart for Your Own Good? Linear and Nonlinear Effects of Cognitive Ability