Reading focus
Syllabus: What Can Individual Differences Predict?
Distinguish the latent correlation, subject-specific results, and examination-threshold analyses, then revisit the prediction criteria from Unit 1.
Why you’re reading it. Examine a substantial intelligence–achievement association without turning it into certainty about any individual.
A question to consider. What could an age-11 test association with age-16 examinations suggest about the programme’s reasoning measure, and what new validation would be required in its adult participants?
Why this reading
This large prospective study connects reasoning tests at the beginning of secondary school with later national examinations. It provides a concrete case for interpreting a strong predictive association without treating it as an individual destiny.
Before you read
The study follows English pupils from approximately age 11 to age 16. Cognitive test performance and educational achievement are distinct measures. A latent factor summarizes shared variation across indicators; a correlation between modeled factors is not the percentage of individual outcomes predicted correctly.
Read for
- Compare the overall association with differences across examination subjects and with the contribution of verbal ability beyond general ability.
- Distinguish correlations, modeled latent associations, and probabilities of reaching a particular examination threshold.
- Identify what the prospective design strengthens and which causal explanations remain untested.
Consider afterwards
- A school interprets the reported .81 latent correlation as an 81% chance of correctly predicting any pupil’s grades. What has it confused, and what evidence would support a forecast for a new pupil?
Editorial context
The results concern a particular cohort, testing system, and educational setting. The study lacked family-background data, and its discussion of possible genetic and environmental explanations does not test those mechanisms. The association is not an estimate of the effect of raising a test score. The later education meta-analysis asks a different causal question about whether additional schooling changes intelligence-test performance.