Grit Scale: 12 items on sticking with a long goal
Talent gets measured early and often. Whether someone keeps going after the talent runs out is a different question, and Duckworth and colleagues wrote the Grit Scale in 2007 to get at it. They defined grit as perseverance and passion for long-term goals, and framed it as a marathon rather than a sprint.
Two versions exist. The original has 12 items, six per factor. A short 8-item version, the Grit-S, drops the four items with the weakest loadings. Both use a 5-point scale from 1 (not at all like me) to 5 (very much like me).
Two factors, and how well they hold together
| Factor | Items (original) | What the items ask |
|---|---|---|
| Consistency of Interests | 6 | Whether goals, interests and directions stay put over time |
| Perseverance of Effort | 6 | Whether effort continues through setbacks and stalled progress |
All six consistency items are reverse-scored. The two factors correlated at r = .45 in the original sample and at r = .37 in a later Dutch sample of 311 people. Related, but far from identical, which is why a report shows them separately. Confirmatory factor analysis of the two-factor model in the original paper fit only moderately: CFI = .83 and RMSEA = .11.
One caveat from a Korean validation: items 6 and 12 sometimes break off into a third factor the authors called diligence, so the two-factor answer is not universal.
Reliability
The original scale reported a total alpha of .85, with .84 for consistency of interests and .78 for perseverance of effort. The short version runs lower by construction: total alpha .73 to .83 across the four development samples, with a meta-analytic average of .79 across 46 samples. In one spelling bee sample the Grit-S came in at .80.
What it predicts
Across the original studies grit accounted for an average of 4% of the variance in success outcomes, after controlling for intelligence and conscientiousness. The outcomes were specific rather than vague: educational attainment in two adult samples (N = 1,545 and N = 690), grade point average among Ivy League undergraduates (N = 138), retention at West Point (N = 1,218 and N = 1,308), and final ranking in the National Spelling Bee (N = 175).
The overlap with conscientiousness matters more than the headline: r = .64. Grit is not a separate dimension of personality sitting off to the side. Most of what it measures is what a Big Five questionnaire already measures, and the incremental prediction, while real, is small.
Some studies find the perseverance items carry almost all of the predictive weight while consistency of interests contributes little. If your two factor scores point in different directions, that split is the useful part of the result.
What the scores look like
Adults in the short-version development sample averaged 3.4 out of 5. The range runs from 1.0 to 5.0. Older people tend to score higher than younger people, and some studies find women slightly higher than men.
There is no cutoff. Duckworth has never published a line separating high grit from low grit, and she recommends treating the score as a continuous variable or comparing it against a specific sample mean. Any band attached to a grit score came from somewhere other than the scale's authors.
Known problems
Three are worth knowing. The items are easy to fake, so a high score may reflect what someone wants to be true. Judgments are relative, meaning two people with the same behaviour can answer differently depending on who they compare themselves to. And the construct overlaps heavily with conscientiousness, which limits how much it adds.
The author's own terms of use exclude employee selection, college admissions, teacher evaluation, and comparisons between schools or countries. A number from this scale describes a habit rather than a person. This page is not a diagnosis and does not replace one.