Reevaluating the Flynn effect, and the reversal: Temporal trends and measurement invariance in Norwegian armed forces intelligence scores

Since 1954, the Norwegian Armed Forces have annually administered an unchanged general mental ability test to male cohorts, comprising figure matrices, word similarities, and mathematical reasoning tests. These stable and representative data have supported various claims about shifts in general mental ability (GMA) levels, notably the Flynn effect and its reversal, influencing extensive research linking these scores with health and other outcomes. This study examines whether observed temporal trends in scores reflect changes in latent intelligence or are confounded by evolving test characteristics and specific test-taking abilities in numerical reasoning, word comprehension, and figure matrices reasoning. Our findings, using multiple-group factor analysis and multiple indicator multiple cause (MIMIC) models, indicate that while there was a general upward trend in observed scores until 1993, this was predominantly driven by enhancements in the fluid intelligence task, specifically figure matrices reasoning. Notably, these gains do not uniformly translate to a rise in underlying GMA, suggesting the presence of domain-specific improvements and test characteristic changes over time. Conversely, the observed decline is primarily due to decreases in word comprehension and numerical reasoning tests, also reflecting specific abilities not attributable to changes in the latent GMA factor. Our findings further challenge the validity of claims that changes in the general factor drive the Flynn effect and its reversal. Furthermore, they caution against using these scores for longitudinal studies without accounting for changes in test characteristics. • Unchanging Trends: Observed increases and subsequent decreases in intelligence scores do not reflect changes in latent intelligence among Norwegian conscripts. • Measurement Invariance: Fails to achieve scalar invariance across time, challenging the validity of using these scores to measure generational changes in intelligence. • Specific Subtests Influence: Significant score variations in figure matrices and word similarity tests largely account for reported IQ changes, not general intelligence. • Cultural and Educational Impacts: Changes in test scores over decades are likely influenced by educational reforms and cultural shifts, not by shifts in underlying cognitive abilities. • Implications for Research: Highlights the need for careful interpretation of intelligence scores over time and the importance of establishing measurement equivalence in longitudinal studies.