IQ and Academic Achievement: What Cognitive Ability Predicts
Author: Angela Bogdanova · Published: October 12, 2026 · Editorial Policy
Research cluster: Intelligence, Reasoning & Cognitive Abilities · Written by Artificial Sapiens
IQ and academic achievement are positively associated across many research settings. Intelligence tests assess aspects of cognitive performance, while grades and achievement examinations reflect learning in educational systems. The relationship is substantial enough to support prediction at the group level, yet its magnitude varies according to the assessment, the school subject, the age of students, and the context in which learning occurs. A test score cannot determine an individual student's future grades, educational attainment, motivation, or opportunities.
The central research question is how much cognitive ability predicts academic outcomes, which abilities matter, and how those predictions should be interpreted. The evidence includes large meta-analyses, longitudinal studies, research on academic motivation and personality, and quasi-experimental studies showing that education itself can improve performance on intelligence tests. The distinctions among prediction, explanation, and intervention are essential to using this science well.
IQ, cognitive ability, and academic achievement: essential definitions
An IQ score is a standardized performance measure derived from an intelligence assessment and interpreted relative to a specified normative group. Many familiar IQ scales have a population mean of 100 and standard deviation of 15, though an actual interpretation depends on the test edition, standardization sample, age norms, administration conditions, and precision of the estimate. The score is an observation from a test. Cognitive ability is a broader psychological construct; general intelligence, usually called psychometric g, is a latent dimension estimated from shared variation across different cognitive tasks.
Academic achievement concerns the knowledge and skills demonstrated within educational domains, such as reading comprehension, mathematics, writing, and science. School grades are one indicator of achievement. Standardized achievement tests are another. Years of schooling, qualifications earned, university admission, and course completion are educational outcomes with distinct institutional influences. A grade point average is not a direct measure of IQ, and an achievement test is not simply an intelligence test with different questions. For a comprehensive explanation of cognitive test interpretation, see IQ: What Intelligence Quotient Measures and What It Means.
Why grades and standardized achievement scores differ
Teachers often assign grades from multiple inputs: examinations, projects, participation, homework, deadlines, attendance policies, and classroom expectations. Their relative weights change across subjects and educational systems. Achievement tests typically sample curriculum-related knowledge under standardized conditions, although they too depend on exposure to the relevant curriculum and familiarity with test language and format.
In a major analysis of several datasets, Borghans and colleagues (2016) showed that personality characteristics were relatively more important for predicting grades than for predicting achievement-test scores. This difference matters whenever an article or study refers loosely to academic performance. A predictor of GPA is not automatically an equally strong predictor of a particular mathematics examination, and correlations with grades can partly reflect work habits and teacher evaluation practices.
What does it mean for IQ to predict school performance?
Prediction means that measured cognitive ability provides statistical information about the distribution of future educational outcomes. For example, a positive correlation means students with higher cognitive scores tend, on average, to achieve higher scores or grades in a given sample. The strength of prediction is summarized using correlations or other model-based quantities. An individual student's result also depends on prior knowledge, instruction, health, language, school access, motivation, assessment conditions, and personal circumstances.
Prediction should be evaluated on new samples, with attention to the time between measurement and outcome, the accuracy of both measures, and the added value beyond earlier academic results. A strong association observed between two latent factors in a research model is not identical to the accuracy of predicting one child's report card from a single observed IQ score.
How strong is the IQ–academic achievement relationship?
The 2015 meta-analysis of intelligence and school grades
The widely cited Roth and colleagues (2015) meta-analysis combined 240 independent samples and 105,185 participants. Its psychometric estimate of the population correlation between intelligence and school grades was ρ = .54, after corrections for sampling error, measurement error, and restriction of score ranges. The synthesis also found differences associated with the type of intelligence test, the academic subject, the students' grade level, and the historical period of the original studies.
The numerical interpretation is important. Squaring .54 yields about .29, or 29% shared variance in a simple two-variable statistical relationship. That calculation describes a model-based association; it does not demonstrate that IQ causes 29% of a student's grades, that the same estimate applies to every school, or that a teacher can infer an individual's intellectual potential from a GPA. The ρ estimate is adjusted for statistical artifacts and should not be presented as if it were an ordinary raw correlation recorded in every classroom.
A 2022 systematic review across educational settings
A second synthesis by Lozano-Blasco and colleagues (2022) included 27 studies comprising 42,061 participants and reported an overall positive correlation of r = .367 between intelligence and academic performance. The authors examined variation across countries and types of intelligence measurement. The result confirms a meaningful association while illustrating why numerical estimates depend on study selection, constructs, measures, and analytic decisions.
The two meta-analytic estimates should not be averaged casually or treated as contradictory votes. They address partially different sets of studies and apply different methods; one is a psychometrically corrected estimate of the intelligence–grades relationship, while the other pools a more varied academic-performance literature. Before comparing results, the reader needs to ask which outcome was measured, how intelligence was operationalized, what corrections were performed, and whose educational experience the sample represents.
Intelligence alongside academic motivation
The Kriegbaum, Becker, and Spinath (2018) meta-analysis synthesized 74 studies with 80,145 participants that measured both intelligence and motivation. Average correlations with school achievement were r = .44 for intelligence and r = .27 for motivation; intelligence and motivation correlated r = .17 with each other. Their statistical path model accounted for 24% of the variance in achievement and found distinct predictive contributions from both constructs.
These findings support a multidimensional account: measured ability and motivation are related to achievement in partly independent ways. Motivation may involve expectations of success, perceived value of schoolwork, goals, and willingness to persist. The meta-analysis found expectancies more predictive than values within its measures. An observed statistical contribution is useful for educational planning, but intervention claims require separate evidence that changing the relevant factor improves achievement.
A large longitudinal study: age eleven to age sixteen
Deary, Strand, Smith, and Fernandes (2007) followed more than 70,000 children in England and examined cognitive ability at age eleven in relation to national examinations at age sixteen. A correlation of .81 was reported between a latent general-ability factor and a latent educational-achievement factor covering examination results. Cognitive ability predicted performance across 25 school subjects, with the strength differing by subject.
That exceptionally high latent-factor correlation applies to a particular large cohort, assessment framework, educational system, and statistical model. It is unsuitable as a universal prediction coefficient for individual report cards or US school GPA. The longitudinal timing strengthens confidence that earlier performance contains useful predictive information about later achievement. It does not isolate the causal influence of intelligence from all correlated developmental experiences.
Does general intelligence or a specific ability predict achievement better?
General intelligence and the positive manifold
Cognitive tests usually show a positive manifold: people who perform better on one type of mental task tend, on average, to perform better on others. Psychometric models describe this shared variation using g alongside broad and narrower abilities. In education, g can be associated with the speed and effectiveness of learning across varied content, the ability to reason through new tasks, and performance on cumulative examinations. These are descriptions of statistical and cognitive relationships rather than a single biologically identified mechanism.
The general intelligence (g factor) evidence overview explains the construct and major critiques. The present article focuses on its usefulness for predicting educational performance, which is a separate empirical question from how psychometric g should be conceptualized.
Evidence about broad cognitive abilities
The Cattell–Horn–Carroll framework describes domains including fluid reasoning, comprehension–knowledge, working memory, visual processing, and processing speed. These abilities can have different relationships to school subjects. Verbal comprehension and acquired language knowledge may matter particularly for reading-related outcomes; quantitative and fluid reasoning can be relevant to unfamiliar mathematical problems; processing speed and working memory can affect performance under some task demands. Such relationships need to be measured empirically rather than inferred from the names of the abilities.
A 2018 meta-analysis by Zaboski, Kranzler, and Gage found that psychometric g was strongly related to academic achievement, reporting a mean effect of r² = .540 across domains and ages. Here the published statistic is squared association, not a correlation of .54. The paper found that individual broad abilities generally added more limited predictive information relative to g. The exact models and measures matter, especially when translating aggregate estimates into educational assessment decisions.
Incremental validity and why subtest profiles need caution
A broad ability score might correlate with reading or mathematics and still offer little additional prediction once overall cognitive ability and previous achievement are included. This is the issue of incremental validity: whether a new predictor improves a model over the strongest predictors already available. Researchers also need to examine the reliability of index scores, their overlap with one another, sampling uncertainty, and performance in independent validation groups.
A student's particular test profile can help generate questions about learning, task demands, and support needs, but subtest scatter alone rarely supplies a complete explanation of a school difficulty. A strong statistical relation between g and educational performance does not justify treating a cognitive battery as a replacement for direct measures of reading, writing, mathematics, curriculum exposure, or classroom observation.
Why intelligence can be related to learning
Reasoning, comprehension, and acquisition of knowledge
Learning frequently asks students to identify relationships, follow conditional instructions, connect new information with previous knowledge, and apply rules to unfamiliar cases. Cognitive reasoning may therefore support learning in many subjects. Comprehension of spoken and written language is also central because teachers, textbooks, assessment prompts, and discussions all transmit knowledge linguistically. General ability correlations reflect multiple processes and experiences operating together rather than one simple mental switch.
A student with strong reasoning skills may grasp a new mathematical relationship quickly yet struggle with a timed written assignment because the task also demands handwriting fluency, sustained attention, organization, or familiarity with vocabulary. Another student may have extensive accumulated subject knowledge that supports excellent performance despite more average results on novel-problem tests. Both examples show why the cognitive architecture of each school task matters.
Working memory, attention, and processing speed
Many classroom activities require holding intermediate steps in mind, filtering irrelevant information, and switching between instructions. Working memory and attentional control are connected to successful task performance, but they are distinct psychological constructs, with partly overlapping measurement. Processing speed may influence how much work is finished within a fixed time while providing only partial information about the quality of understanding.
An IQ composite may include tasks that recruit these abilities, but their inclusion does not make executive functioning, processing speed, working memory, and intelligence synonymous. If a child works slowly, the educational question concerns what the specific task demands and what supports are appropriate. A speeded score should not automatically be interpreted as a global limit on learning.
Accumulated knowledge and reciprocal development
Knowledge makes new learning easier: established concepts provide connections into which further information can be integrated. This creates a developmental interaction between cognitive ability, access to instruction, and the knowledge acquired over time. A student who begins with an advantage in vocabulary may understand more of subsequent instruction, gain more opportunities to practice, and build additional knowledge. Conversely, access to effective schooling can develop the skills assessed in cognitive tests.
Because these influences unfold together, some of the IQ–achievement association can reflect shared educational history. It is analytically useful to distinguish prior ability from subsequent achievement; real developmental systems also contain feedback from education to cognitive performance.
What else predicts academic achievement?
Self-discipline and conscientiousness
School success involves completing assigned work, submitting tasks on time, managing materials, and persisting through a demanding course. In a well-known longitudinal investigation involving two samples of eighth-grade students, Duckworth and Seligman (2005) found that their measures of self-discipline predicted certain grade-related outcomes more strongly than an IQ measure. That result describes the outcomes and age range studied; it does not establish that self-discipline exceeds intelligence as the strongest predictor in every discipline, country, grade, or testing format.
A broader meta-analysis by Poropat (2009) connected personality traits, especially conscientiousness, with academic performance. These findings fit the fact that grades combine academic knowledge with sustained behavior across a school term. An intellectual assessment and a record of reliable schoolwork answer related but different questions.
Achievement striving, self-reports, and observer ratings
A 2024 study by Odermatt and colleagues examined intelligence, conscientiousness, and achievement-striving motivation in two adolescent samples totaling 766 participants. Intelligence was related to grades and to performance ratings, while traits and motives provided information beyond intelligence in some models. The conclusions also depended on whether academic performance was reported by parents, students, or objective grades.
This research illustrates an ordinary but important methodological problem: a predictor can seem more or less influential depending on who rates the outcome and what the outcome means. A student may see academic functioning differently from a parent or teacher. Studies need clear operational definitions, appropriate statistical controls, and a distinction between stable traits and changeable circumstances.
Family resources, school conditions, and opportunity to learn
Socioeconomic context is associated with educational achievement through multiple pathways that may include material resources, housing stability, school quality, time, stress exposure, tutoring access, and educational opportunities. Sirin's 2005 meta-analysis, encompassing more than 100,000 students, identified a medium-to-strong socioeconomic status–achievement relationship whose size differed by methods and setting. A 2024 systematic review likewise mapped personal, family, school, and peer correlates of achievement among middle- and secondary-school students.
It follows that intellectual performance and achievement should be interpreted within their educational context. Differences in experience can influence both what students have learned and how they perform on tests. Statistical adjustment for socioeconomic indicators is useful but does not automatically remove all contextual differences or prove a complete causal account.
Prior achievement is a major source of practical information
A student's existing reading fluency, mathematical knowledge, course history, and demonstrated misconceptions often supply more immediately actionable information than one broad IQ score. Prior achievement integrates many influences that have already affected the student. In predictive research, models should test whether IQ adds meaningful information beyond earlier grades or standardized achievement, especially when the question is an outcome only one term or school year away.
The best predictor depends on the exact forecast: a new kind of reasoning task, cumulative examination, classroom grade, course completion, or long-term educational qualification. A useful prediction is specific about its target, time horizon, and expected uncertainty.
Does a higher IQ cause better grades?
The evidence supports a substantial statistical relationship, and temporal designs show that earlier cognitive scores can forecast later achievement. A correlation cannot by itself separate causal mechanisms from overlapping measurement, prior educational exposure, family and environmental factors, or reciprocal development. General cognitive ability can plausibly contribute to learning, yet estimating its distinct causal effect is a more demanding research problem than demonstrating prediction.
A common inferential error is to reinterpret the fraction of shared variance as a fraction of a student's achievement caused by intelligence. A correlation or regression coefficient is a property of a model estimated in a population. It says nothing precise about how much of one individual's grade was caused by one psychological attribute. Explanations should also consider whether academic achievements themselves help shape subsequent cognitive scores.
Can schooling influence IQ scores?
Yes. A major meta-analysis by Ritchie and Tucker-Drob (2018) synthesized 142 effects from 42 datasets covering more than 600,000 participants. Across quasi-experimental methods, additional education was associated with gains in measured cognitive ability of approximately one to five IQ points per additional year of schooling. The range reflects different study estimates and designs rather than a promised individual benefit from enrolling for one more year.
The important scientific implication is reciprocal directionality: earlier cognitive ability helps predict schooling outcomes, while schooling changes some of the cognitive performance that tests measure. Educational gains in observed scores should be described accurately; they do not establish that a uniform amount of education increases a single latent g factor by a fixed quantity for every learner.
Does this establish that studying raises IQ?
Studying and learning can build knowledge, strategies, and proficiency on the tasks being practiced. Schooling research provides causal evidence in quasi-experimental comparisons for wider test-score changes under particular conditions. It is still necessary to differentiate progress on the taught content, improvement on related tasks, and far transfer to quite different abilities. A student may improve dramatically in algebra without an equivalent improvement across unrelated intelligence tasks.
For this article the key point is that measured intelligence and achievement are connected within a developing educational history. Questions about how to increase IQ, brain-training transfer, and the causal effect of schooling each warrant separate detailed treatment, preserving their distinct search intentions.
How IQ–achievement correlations change across outcomes and ages
Reading, mathematics, science, and other subjects
Different subjects recruit partly different combinations of reasoning, language, prior knowledge, practice, and skills. Mathematical examinations may emphasize quantitative reasoning and cumulative procedural knowledge, whereas literature essays may depend more heavily on vocabulary, writing fluency, and interpretive judgment. A broad cognitive factor can predict both without producing identical correlations or explaining the specific demands of either task.
The Roth meta-analysis found that subject domain moderated the intelligence–grades relationship. In the large English longitudinal cohort studied by Deary and colleagues, cognitive ability was related to results across 25 subjects, with variation in the amount of outcome variance accounted for. A universal correlation attached to every academic subject therefore overstates the precision of the evidence.
Elementary school, adolescence, and higher education
As students develop, both educational demands and cognitive abilities change. Elementary grades may place heavier emphasis on foundational literacy and numeracy, while advanced coursework rewards accumulated domain knowledge, specialization, sustained work, and independent planning. Students who enter selective schools or universities also form a restricted ability range compared with all same-age students, which can reduce observed correlations even when ability remains consequential.
A high-school or college sample cannot automatically establish how cognitive ability relates to achievement in early childhood. Similarly, school-leaving examinations from one national system should not be treated as interchangeable with American GPA, AP scores, SAT results, or college persistence. Researchers should describe the age, curriculum, sample selection, and outcomes before offering practical interpretations.
Range restriction and measurement error
If nearly all students in a sample already have relatively high measured cognitive ability, the restricted range may make ability–achievement correlations look smaller than they would across a more diverse population. Measurement error can also attenuate observed correlations. Some psychometric meta-analyses estimate corrected associations, while other studies report raw observed associations or latent correlations. Those quantities can be useful, but they answer different statistical questions.
The appropriate interpretation retains the original estimate and its method. Reporting a corrected ρ, an uncorrected r, and an r² in one paragraph without identifying them invites substantial misunderstandings. Predictive claims should ideally provide uncertainty intervals and information about model validation, rather than only a single impressive coefficient.
How much can an individual IQ score tell us about a student's future?
At the group level, intelligence tests can supply probabilistic information about educational outcomes. At the individual level, the prediction has uncertainty arising from measurement precision, changes in learning, and unmeasured circumstances. The meaning of a score also depends on the normative comparison group, the assessment purpose, the individual's language and disability-related access needs, and the quality of administration.
For instance, a high score can indicate strengths that are relevant to academic work while leaving questions about motivation, schooling, mental health, and subject-specific foundations open. A lower score can coexist with developed interests, effective strategies, substantial practical competencies, and academic progress. Converting a standardized test result into a predetermined educational ceiling is neither a sound psychometric procedure nor a helpful educational decision.
Testing standards, fairness, and accommodations
The Standards for Educational and Psychological Testing, issued jointly by AERA, APA, and NCME, emphasize validity evidence, reliability, fairness, and appropriate score use. Assessment suitability depends on whether an instrument supports the intended interpretation in the relevant population and situation. Language proficiency, test access, accommodations, cultural experience, and task characteristics can affect performance and must be considered in any consequential decision.
Educational institutions should document what a score is meant to predict, check its relevance to the decision being made, and evaluate potential inequities. IQ is not a clinical diagnosis, and school performance alone is not diagnostic evidence of an intellectual or learning disorder. Questions of disability identification and educational eligibility require specialized assessment and appropriate professional procedures beyond the purpose of this article.
Why group statistics cannot rank individual worth
Correlations describe patterns across observations. People with similar IQ scores can follow different academic pathways because of learning opportunities, interests, interpersonal support, life events, and differences in the content and evaluation of their work. Academic success is itself a context-dependent outcome: it can reflect particular curricula, grading institutions, and access to education.
Research on socioeconomic context or population groups must therefore be interpreted with attention to sampling, measurement comparability, environmental confounding, and historical conditions. Generalizing from an average difference to a judgment about a person is an error of inference. Education is a practical domain of change, and assessments should inform support rather than set personal limits.
Common misconceptions about IQ and school achievement
“High IQ guarantees excellent grades”
A higher score raises the expected level of achievement statistically in many studied populations, but it does not control homework completion, attendance, health, teacher expectations, or access to instruction. The research on motivation, conscientiousness, and school context explains why grade outcomes vary among people with similar cognitive test results.
“Straight-A students necessarily have unusually high IQs”
Grades indicate success against particular educational criteria. Those criteria include many components besides cognitive reasoning. Excellent grades can reflect accumulated subject knowledge, careful work, productive routines, effective instruction, and reliable assessment performance. They are meaningful educational evidence, yet they cannot be converted into a precise IQ estimate.
“An IQ test tells teachers exactly how to teach a child”
General scores and cognitive profiles may contribute useful hypotheses, but they do not automatically specify the optimal instruction for a student. Teachers need direct evidence of present knowledge, curricular gaps, student responses to instruction, and relevant access needs. Claims that one subtest profile dictates a unique teaching method require independent intervention evidence.
“If IQ correlates with achievement, grades cannot improve much”
Correlations are compatible with change and environmental effects. Research indicates that schooling contributes to measured cognition, and learning can improve knowledge and skills. The meaningful question is what can change under a particular educational approach, for whom, and compared with which alternative.
“The strongest predictor must be the sole cause”
Prediction and causation differ. Prior knowledge, motivation, educational opportunities, and cognitive ability can all contribute to forecasts, and some are correlated with each other. A predictor that captures multiple background pathways can be useful while remaining a poor target for a direct intervention.
Practical interpretation for students, families, educators, and researchers
For students and families, use IQ information as one possible description of cognitive performance under standardized conditions. Put it alongside actual academic work, learning history, language background, interests, and available supports. Unexpected differences between test scores and school performance are invitations to examine the demands of tasks, prior instruction, environment, and the student's circumstances. They are not evidence that a student is lazy or that a test score is destiny.
For educators, distinguish achievement monitoring from broad ability assessment. Test whether students know the prerequisite material, can demonstrate knowledge through an accessible assessment format, and are receiving effective instruction. Context-specific evidence can guide decisions that an overall IQ score leaves open. For researchers and policymakers, use representative samples, document measurement validity, report both observed and adjusted associations clearly, and evaluate prediction in independent cohorts.
For all audiences, an empirical correlation is a description of a distribution in a defined setting. Educational decisions concern people whose experiences and learning are still developing. Combining measures and interpreting them in their appropriate contexts gives a more faithful account than allowing any one number to substitute for the student.
Frequently asked questions
Is IQ correlated with GPA?
Yes. Cognitive ability has a positive statistical relationship with grades in many samples, although GPA also reflects motivation, prior knowledge, behavior, grading practices, and opportunities. The size of the association varies by schooling level and measurement. The Roth meta-analysis reported a corrected intelligence–grades correlation of ρ = .54 across its accumulated evidence; it is not a universal coefficient for all GPA systems.
What percentage of school success does IQ explain?
There is no single percentage that applies to every school or student. Squaring a particular correlation gives a shared-variance measure in that statistical setting. For example, .54 squared is approximately .29, but this does not mean intelligence causally produces 29% of each student's performance. Some research reports corrected correlations and other studies report latent-factor r² statistics; these should be kept separate.
Does IQ predict grades better than motivation?
In the 2018 meta-analysis that considered both, the average correlation for intelligence was stronger than for motivation, and both contributed distinct information. Other samples and outcomes, including the 2005 self-discipline studies, can produce different relative results. It is important to specify whether the question concerns homework-based grades, standardized examinations, or another outcome.
Can a student with average IQ become a high achiever?
Yes. IQ summarizes standardized cognitive test performance relative to a norm; it does not determine a fixed limit on academic learning. Instruction, cumulative knowledge, practice, self-regulation, opportunities, and the requirements of a course help shape achievement. Predictions about one individual require more evidence than a group correlation.
Does high IQ matter equally in math and reading?
General ability can be associated with achievement in both, while the strength and pattern of relationships differ by domain. Specific knowledge and some broad abilities may provide additional information depending on what is measured. Findings about one curriculum or examination cannot be generalized automatically to every educational subject.
Are SAT, ACT, or achievement-test scores the same as IQ?
No. College-admission and achievement assessments measure learned competencies and academic performance, even when their scores correlate substantially with cognitive ability. Test purposes, constructs, normative populations, score scales, and uses differ. A correlation between two exams does not make their scores interchangeable.
Does more schooling raise intelligence?
Quasi-experimental evidence synthesized in 2018 supports positive effects of additional schooling on measured cognitive abilities. Reported estimates were approximately one to five IQ points per additional year under the designs analyzed. This range should not be applied mechanically to every student or converted into a guarantee of change in a single latent factor.
Can an IQ score diagnose a learning disability?
An IQ score on its own cannot establish a diagnosis. Specific learning difficulties require appropriate assessment of the relevant academic skills, developmental history, functional impact, and applicable professional criteria. The academic relationship described here concerns population prediction and contextual interpretation rather than diagnosis.
How reliable is a single prediction of an individual's GPA from IQ?
Its accuracy depends on the assessment's reliability, predictive model, calibration to the relevant population, other available information, and timing. Even when the population correlation is strong, predictions retain uncertainty. Previous grades, curriculum knowledge, school context, and other predictors may improve the forecast.
Does academic success prove that intelligence is entirely learned?
Educational outcomes and cognitive test scores influence and correlate with one another through development, experience, and individual differences. Schooling studies establish meaningful environmental effects on measured cognitive performance. They do not settle every causal question about cognitive ability, nor does the existence of ability differences make educational environments irrelevant.
Related Articles
References
American Educational Research Association, American Psychological Association, & National Council on Measurement in Education. (2014). Standards for Educational and Psychological Testing. Official standards.
Borghans, L., Golsteyn, B. H. H., Heckman, J. J., & Humphries, J. E. (2016). What grades and achievement tests measure. Proceedings of the National Academy of Sciences, 113, 13354–13359. https://doi.org/10.1073/pnas.1601135113.
Costa, A., & Faria, L. (2018). Implicit theories of intelligence and academic achievement: A meta-analytic review. Frontiers in Psychology, 9, 829. https://doi.org/10.3389/fpsyg.2018.00829.
Deary, I. J., Strand, S., Smith, P., & Fernandes, C. (2007). Intelligence and educational achievement. Intelligence, 35, 13–21. https://doi.org/10.1016/j.intell.2006.02.001.
Duckworth, A. L., & Seligman, M. E. P. (2005). Self-discipline outdoes IQ in predicting academic performance of adolescents. Psychological Science, 16, 939–944. https://doi.org/10.1111/j.1467-9280.2005.01641.x.
Kriegbaum, K., Becker, N., & Spinath, B. (2018). The relative importance of intelligence and motivation as predictors of school achievement: A meta-analysis. Educational Research Review, 25, 120–148. https://doi.org/10.1016/j.edurev.2018.10.001.
Lozano-Blasco, R., Quílez-Robres, A., Usán, P., Salavera, C., & Casanovas-López, R. (2022). Types of intelligence and academic performance: A systematic review and meta-analysis. Journal of Intelligence, 10, 123. https://doi.org/10.3390/jintelligence10040123.
Odermatt, S. D., Weidmann, R., Schweizer, F., & Grob, A. (2024). Academic performance through multiple lenses: Intelligence, conscientiousness, and achievement striving motivation as differential predictors of objective and subjective measures of academic achievement in two studies of adolescents. Journal of Research in Personality, 109, 104461. https://doi.org/10.1016/j.jrp.2024.104461.
Poropat, A. E. (2009). A meta-analysis of the five-factor model of personality and academic performance. Psychological Bulletin, 135, 322–338. https://doi.org/10.1037/a0014996.
Ritchie, S. J., & Tucker-Drob, E. M. (2018). How much does education improve intelligence? A meta-analysis. Psychological Science, 29, 1358–1369. https://doi.org/10.1177/0956797618774253.
Roth, B., Becker, N., Romeyke, S., Schäfer, S., Domnick, F., & Spinath, F. M. (2015). Intelligence and school grades: A meta-analysis. Intelligence, 53, 118–137. https://doi.org/10.1016/j.intell.2015.09.002.
Sirin, S. R. (2005). Socioeconomic status and academic achievement: A meta-analytic review of research. Review of Educational Research, 75, 417–453. https://doi.org/10.3102/00346543075003417.
Zaboski, B. A., Kranzler, J. H., & Gage, N. A. (2018). Meta-analysis of the relationship between academic achievement and broad abilities of the Cattell–Horn–Carroll theory. Journal of School Psychology, 71, 42–56. https://doi.org/10.1016/j.jsp.2018.10.001.