The Bifactor Model Fits Better Than the Higher-Order Model in More Than 90% of Comparisons for Mental Abilities Test Batteries
Open accessThe DOI matches its Crossref record; 12 of the 12 references it cites were found in a registry or an index.
Abstract
The factor structure of mental abilities has most often been depicted using a higher-order model. Under this model, general mental ability (g) is placed at the top of a pyramid, with “loading” arrows going from it to the other factors of intelligence, which in turn go to subtest scores. In contrast, under the bifactor model (also known as the nested factors/direct hierarchical model), each subtest score has its own direct loading on g; the non-g factors (e.g., the broad abilities) do not mediate the relationships of the subtest scores with g. Here we summarized past research that compared the fit of higher-order and bifactor models using confirmatory factor analysis (CFA). We also analyzed additional archival datasets to compare the fit of the two models. Using a total database consisting of 31 test batteries, 58 datasets, and 1,712,509 test takers, we found stronger support for a bifactor model of g than for the traditional higher-order model. Across 166 comparisons, the bifactor model had median increases of 0.076 for the Comparative Fit Index (CFI), 0.083 for the Tucker-Lewis Index (TLI), and 0.078 for the Normed Fit Index (NFI) and decreases of 0.028 for the root mean square error of approximation (RMSEA) and 1343 for the Akaike Information Criterion (AIC). Consequently, researchers should consider using bifactor models when conducting CFAs. The bifactor model also makes the unique contributions of g and the broad abilities to subtest scores more salient to test users.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
"General Intelligence," Objectively Determined and Measured
Spearman, C. (1904) · The American Journal of Psychology · doi:10.2307/1412107 · Cited by 6,207
Human Cognitive Abilities: A Survey of Factor-Analytic Studies
Carroll, J. B. (1993) · doi:10.1017/cbo9780511571312 · Cited by 4,121
Conventional wisdom on measurement: A structural equation perspective.
Bollen, K., & Lennox, R. (1991) · Psychological Bulletin · doi:10.1037/0033-2909.110.2.305 · Cited by 3,618
Wechsler Adult Intelligence Scale--Fourth Edition
Wechsler, D. (2012) · PsycTESTS Dataset · doi:10.1037/t15169-000 · Cited by 2,232
The Rediscovery of Bifactor Measurement Models
Reise, S. P. (2012) · Multivariate Behavioral Research · doi:10.1080/00273171.2012.715555 · Cited by 2,112 · cited 7×
“Thus, we turn to two additional analyses suggested by Mansolf and Reise [20]. a This set of columns pertains to Morgan et al.’s [43] simulations that used a bifactor model to analyze data that were generated to be truly bifactor. b This set of columns pertains to Morgan et al.’s [43] simulations that used a bifactor m…”
BackgroundInfluentialMcGrew, K. S. (2009) · Intelligence · doi:10.1016/j.intell.2008.08.004 · Cited by 1,354
The Development of Hierarchical Factor Solutions
Schmid, J., & Leiman, J. M. (1957) · Psychometrika · doi:10.1007/bf02289209 · Cited by 1,098 · cited 5×
“Although the Schmid-Leiman [64] decomposition2 provides a method for separating these variances and Brown [27] shows how this can be easily implemented using a spreadsheet (or even a hand calculator), we note that this is too rarely done in practice.”
BackgroundMethodInfluentialWorking memory is (almost) perfectly predicted by g.
COLOM, R. (2004) · Intelligence · doi:10.1016/j.intell.2003.12.002 · Cited by 436
Bifactor Modeling and the Estimation of Model-Based Reliability in the WAIS-IV
Gignac, G. E., & Watkins, M. W. (2013) · Multivariate Behavioral Research · doi:10.1080/00273171.2013.804398 · Cited by 200 · cited 4×
“Although they did not study unmodeled complexity, Morgan, Hodge, Wells, and Watkins [43] conducted a follow-up Monte Carlo simulation to Murray and Johnson [18] to examine bias in fit statistics.”
BackgroundInfluentialGlutting (2006) · doi:10.1177/00224669060400020101 · Cited by 122
Golay, P., Reverte, I., Rossier, J., Favez, N., & Lecerf, T. (2013) · Psychological Assessment · doi:10.1037/a0030676 · Cited by 94 · cited 3×
“Citation Battery/Notes Higher-Order Comparison Bifactor CFI TLI NFI RMSEA AIC χ2 df p ∆χ2 df p CFI TLI NFI RMSEA AIC χ2 df p Beaujean et al. [21] WISC-IV 0.981 0.976 0.960 0.039 226.74 150.74 82 <0.001 23.15 9 0.006 0.985 0.978 0.966 0.037 221.60 127.60 73 <0.001 Benson et al. [22] WISC-IV 0.956 0.950 0.925 0.042 1108…”
BackgroundMethodInfluentialBeaujean, A. A., Parkin, J., & Parker, S. (2014) · Psychological Assessment · doi:10.1037/a0036745 · Cited by 50 · cited 1×
“[21] did not provide TLI and NFI in their original paper; however, CFI, χ2, and df were provided for the hypothesized model.”
Background
Cited by
Canivez, G. L., McGill, R. J., Dombrowski, S. C., Watkins, M. W., Pritchard, A. E., & Jacobson, L. A. (2020) · Assessment · doi:10.1177/1073191118811609 · Cited by 53
Stephenson, K. G., Beck, J. S., South, M., Norris, M., & Butter, E. (2021) · Journal of Clinical Child & Adolescent Psychology · doi:10.1080/15374416.2020.1846543 · Cited by 35
Canivez, G. L., Dombrowski, S. C., & Watkins, M. W. (2018) · Psychology in the Schools · doi:10.1002/pits.22138 · Cited by 30
Kim, H., Li, N., Broyles, A., Musoka, L., & Correa-Fernández, V. (2023) · Journal of American College Health · doi:10.1080/07448481.2021.1892700 · Cited by 21
“Another factor model that is often used to represent both broad and more specific factors is a bi-factor model.(27) Both second-order models and bi-factor models assume that there is an overarching construct and at the same time, there are multiple domain-specific constructs (constructs that cover a subtle area).”
BackgroundFactor Structure of the Sussex-Oxford Compassion Scales
Halamová, J., & Kanovský, M. (2021) · Psihologijske teme · doi:10.31820/pt.30.3.5 · Cited by 21
Metadata
- Authors
- Cucina, J., & Byle, K.
- Venue
- Journal of Intelligence
- Publisher
- MDPI AG
- Published
- 2017-07-11
- Volume
- 5
- Issue
- 3
- Pages
- 27
- ISSN
- 2079-3200 (ISSN-L)
- OpenAlex
- W2735917811
- Cited by
- 98
- References
- 12
- FWCI
- 8.22
- Type
- journal article
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.