Self-discipline Outdoes IQ Predicting Academic Performance in Adolescents

There might be affiliate links on this page, which means we get a small commission of anything you buy. As an Amazon Associate we earn from qualifying purchases. Please do your own research before making any online purchase.

Share this:

Some combination of personality, intelligence, and environmental factors determine success.

Intelligence and environment are usually considered to have a much larger role than personality. This study shows that personality may play a larger factor in academic performance than previously thought.

Some background:

The research behind IQ and performance in adolescents

Greater ability to delay gratification measured at age 4 predicted higher academic and social functioning more than a decade later. (H.N. Mischel & Mischel, 1983; W. Mischel, Shoda, & Peake, 1988; Shoda, Mischel, & Peake, 1990).

[See how motivation correlates to higher test scores.]


Wolfe and Johnson (1995) found self-discipline to be the only one among 32 measured personality variables (e.g., self-esteem, extraversion, energy level) that predicted college grade point average (GPA) more robustly than SAT scores did.

Hogan and Weiss (1974) found that high self-discipline distinguished Phi Beta Kappa undergraduates from non-Phi Beta Kappa students of equal intellectual ability.

In two large samples of undergraduates, Tangney, Baumeister, and Boone (2004) found that self-discipline correlated positively with self-reported grades, as well as a broad array of personal and interpersonal strengths.

The size of the effect of self-discipline on the short and long-term performance of young kids and college students may still be in doubt, but the idea that there is an effect should be taken more seriously.

What about with adolescents?

In this study, self-discipline better predicted the future GPA of two groups of eighth-grade students. Those students with the highest IQ had a GPA of only a few points higher than those with the lowest IQ (91 vs 85), but those with the highest self-discipline had a GPA of almost 15 points higher than those with the lowest (93.5 vs 80.5).

The authors present the idea that much of what gets graded in school and converted into GPA is based on effort, not intelligence. This study confirms that suspicion. In a later study, this hypothesis is further tested and further confirmed – GPA tests effort; standardized assessment (e.g. SAT) tests IQ.

Click here to learn more about self-control.

Study Details

Hypothesis: Self-discipline will predict improved academic performance in the future, over and above the effects of IQ.

Method: Two cohorts of eight-grade students (n = 304) from a magnet school in a NE city were assessed at two periods – during the first semester of their academic year, and seven months later.

Measures: Several measures of self-discipline and one measure of IQ was compared against several measures of academic performance.

A self-control, impulsiveness, and discounting rate questionnaire and IQ test were given to the students. For the second cohort, a less theoretical test of self-control and discounting was given, the delay choice task.

A self-control, observer-report questionnaire was given to both parents and homeroom teachers.

Academic performance was assessed using the following measures: the GPA of their first semester, their final GPA for the year seven months later, their score on a standardized achievement test (also several months later), their selection to a charter high-school, how many hours they spent doing homework and watching TV, how often they were absent, and the time of day they began doing their homework.

Results: The hypothesis was partially confirmed.

The correlation between academic performance and self-discipline was one to two times as large as the correlation between academic performance and IQ.

Those students with the highest IQ had a GPA only a few points higher than those with the lowest IQ (91 vs 85). On the other hand, those with the highest self-discipline had a GPA almost 15 points higher than those with the lowest (93.5 vs 80.5).

Concerns: The sample size was small, self-discipline may be correlated with a latent variable, rather than causing higher performance itself (e.g. household environment), and the population was socioeconomically and ethnically diverse, but not academically diverse (the school they were attending had an acceptance test).

Questions: Was there a correlation between IQ and self-discipline? What factors predict self-discipline?

Duckworth, A. L. & Seligman, M. E. P. (2005). Self-discipline outdoes IQ predicting academic performance in adolescents. Psychological Science, 16, 939-944

Share this:

Leave a Comment