Attitude toward numbers: A better predictor of financial literacy and intelligence than need for cognition

Abstract

In a recent meta-analysis, financial literacy – understanding basic financial terms and concepts – was significantly correlated with general intelligence (r ≈ 0.62). However, this correlation may be underestimated, as few studies employed robust measures of both intelligence (more than one subtest) and financial literacy (>12 items). Additionally, cognitive reflection has not been investigated for its unique capacity at predicting financial literacy, an important consideration, as cognitive reflection has been suggested to be a measure of financial cognition. Beyond cognitive factors, non-cognitive predictors like need for cognition and attitude toward numbers (comfort with numerical concepts) may also influence financial literacy. In our study of 688 young adults (17–40 years), we administered four intelligence subtests (including a measure of cognitive reflection), a financial literacy test, and questionnaires assessing need for cognition and attitude toward numbers - the first to investigate all of these dimensions simultaneously. Results revealed a stronger latent correlation (0.76) between general intelligence and financial literacy than previously reported. Cognitive reflection was found to yield some incremental predictive variance in predicting financial literacy beyond general intelligence. Finally, attitude toward numbers and need for cognition both correlated positively with both general intelligence and financial literacy, however, only attitude toward numbers uniquely predicted financial literacy in a structural equation model. These findings suggest that financial literacy shares closer to 50–60% of its variance with general intelligence. They also highlight the potential value of fostering a positive numerical attitude alongside financial knowledge in educational interventions.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 20 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.