In latent-variable SEM models, intelligence — not working memory — is the significant direct predictor of academic outcomes in both childhood and early adulthood¶

Claim · All claims
Evidence · 1 study (2 entries) · associational r2 · q2
Currency · newest study 2017 · checked against its sources 2026-10-24 (automated, 2 of 2 entries) · not reviewed by a person

Subclaims¶

q2 i? In SEM models for Grade 2 outcomes, intelligence was the only significant predictor of all academic outcomes, although WMC and intelligence were highly intercorrelated latent variables. → Schneider Wolfgang 2017 q2 i? In SEM models for later development, intelligence remained a significant predictor of later reading competencies, whereas WMC did not predict any of the outcome measures. → Schneider Wolfgang 2017 (2)

Evidence¶

Schneider Wolfgang 2017¶

Schneider Wolfgang, Niklas Frank. (2017). Intelligence and Verbal Short-Term Memory/Working Memory: Their Interrelationships from Childhood to Young Adulthood and Their Impact on Academic Achievement. Journal of Intelligence. https://doi.org/10.3390/jintelligence5020026

q2 · i? · associational · r2

Full-forward latent variable SEM (FIML estimation) predicting Grade 2 achievement from WMC, intelligence, and controls; the model showed a very close and good fit. The article states "intelligence was the only significant predictor of all academic outcomes in Grade 2".

"WMC and intelligence were highly intercorrelated as latent variables, and intelligence was the only significant predictor of all academic outcomes in Grade 2."

Schneider Wolfgang 2017 (2)¶

Schneider Wolfgang, Niklas Frank. (2017). Intelligence and Verbal Short-Term Memory/Working Memory: Their Interrelationships from Childhood to Young Adulthood and Their Impact on Academic Achievement. Journal of Intelligence. https://doi.org/10.3390/jintelligence5020026

q2 · i? · associational · r2

SEM models for the t4/t5 wave of the LOGIC sample, including academic precursors assessed at t2/t3, again showing good fit. The article reports "WMC did not predict any of the outcome measures" while intelligence predicted later reading.

"Intelligence remained a significant predictor for later reading competencies, whereas WMC did not predict any of the outcome measures."

Discussion¶

Related Claims¶

• Non-verbal intelligence measures (Raven, WISC Block Design) are non-significant math predictors when well-specified spatial and memory measures are in the model — reports the opposite
• Visuospatial memory did not predict word reading achievement in any grade, providing a discriminative contrast with mathematics — related
• IQ, short-term memory and study habits each correlate positively and significantly with academic achievement in both 3rd and 4th grade — related
• Full-scale IQ predicts reading comprehension more strongly than word reading, and working memory and processing speed remain significant predictors after IQ is controlled — reports the opposite
• Synchronous observed-variable correlations between WMC and intelligence are modest (r = 0.17 to 0.40), lower than latent-variable correlations
• Neither IQ nor WMC significantly predicted early mathematics when both were in the same model, explaining only about 5% of variance — reports the opposite
• Spontaneous flexibility, but not reactive flexibility, significantly predicts children's reading comprehension and writing — related
• Previous domain knowledge accounts for most of the variance in later achievement, considerably reducing the predictive impact of IQ and WM — related
• When previous academic achievement is controlled, intelligence predicts only later reading comprehension and age-10 mathematics, and WMC predicts nothing additionally — related
• Intelligence and working memory predicted seventh-grade mathematics before prior achievement was controlled, but their direct effects were masked by sixth-grade competence — related
• The central executive component of working memory is the most important domain-general predictor of later-grade mathematics achievement — reports the opposite