Intelligence Predicts Grades
Find and read a peer-reviewed research journal article using intelligence or achievement testing in research and share what you learned from this article with your classmates. Specifically (and in your own words):
1. Why was the study conducted? What hypotheses were being tested?
2. What test(s) were used?
3. What findings were reported, and what conclusions were drawn
Questions
-
Why was the study conducted? What hypotheses were being tested?,
-
What test(s) were used?,
-
What findings were reported, and what conclusions were drawn?,
Answers / What I Learned
1. Why was the study conducted? What hypotheses were being tested?
The study I looked at is “The Predictive Validity of Four Intelligence Tests for School Grades” (Frontiers, 2017) Frontiers.
-
Purpose / motivation: The authors wanted to examine how well different intelligence tests can predict later school performance (grades). Over many decades, psychologists have assumed that higher intelligence should lead to better academic outcomes, but the strength of that prediction can vary depending on which intelligence test is used and what subject or grade is being predicted.
-
Hypotheses tested: The authors hypothesized that each intelligence test would significantly predict average school grades over time, and that some tests might predict specific subject grades (e.g. math or language) better than others. They also expected that longitudinal prediction (predicting grades several years later) would hold for at least some of the tests.
2. What test(s) were used?
They used four intelligence tests commonly used in German-speaking countries:
-
Intelligence and Development Scales (IDS)
-
Reynolds Intellectual Assessment Scales (RIAS)
-
Snijders-Oomen Nonverbal Intelligence Test (SON-R 6-40)
-
Wechsler Intelligence Scale for Children (WISC-IV) Frontiers
These were administered to children around age ~9. Then the researchers collected their school grades (average grades, plus specific grades in mathematics and language) about 3 years later. Frontiers
So this was a longitudinal prediction design.
3. What findings were reported, and what conclusions were drawn?
Findings: Intelligence Predicts Grades
-
All four intelligence tests showed significant prediction of the average school grades measured 3 years later. Frontiers
-
For specific subjects:
• The IDS and RIAS predicted both mathematics and language grades. Frontiers
• The SON-R 6-40 test predicted math grades. Frontiers
• Interestingly, the WISC-IV did not show a significant association with later math or language grades when considered separately (though it did predict the overall averaged grades). Frontiers -
Their sample size for the 3-year follow up was modest (54 children for whom longitudinal data was available). Frontiers
-
The authors caution that because of this small follow-up sample, conclusions should be tentative. Frontiers
Conclusions:
-
Intelligence test scores have useful predictive validity for later academic performance (grades), especially when using tests like IDS, RIAS, and SON-R.
-
Some tests are better predictors for specific subjects than others.
-
The fact that WISC-IV failed to predict individual subject grades suggests that not all intelligence measures are equally good for all predictions.
-
In psychological practice (for guidance, placement, or interventions), intelligence tests can help anticipate academic difficulties or strengths, but one must interpret results carefully and in context.
-
Because of study limitations (small longitudinal sample, focus on German-speaking context), the results should be seen as preliminary evidence.
My reflections / what I learned generally
-
Intelligence tests are not perfect, but they do provide meaningful information about future academic success.
-
The choice of the test matters: some tests may be more predictive in certain domains (math, language) or contexts.
-
Longitudinal designs (testing intelligence early, then measuring achievement later) are powerful because they help us see causal potential rather than just correlations at one time point.
-
Even with significant predictive power, intelligence is only one piece of the puzzle — many other factors (motivation, teaching quality, environment, effort) also influence achievement.
-
When interpreting test results, especially in educational settings, one must consider sample sizes, cultural context, and whether the test was standardized in a comparable population.
