Avoiding bias in mixed model inference for fixed effects
Open accessThe DOI matches its Crossref record; 6 of the 6 references it cites were found in a registry or an index.
Abstract
Analysis of a large longitudinal study of children motivated our work. The results illustrate how accurate inference for fixed effects in a general linear mixed model depends on the covariance model selected for the data. Simulation studies have revealed biased inference for the fixed effects with an underspecified covariance structure, at least in small samples. One underspecification common for longitudinal data assumes a simple random intercept and conditional independence of the within‐subject errors (i.e., compound symmetry). We prove that the underspecification creates bias in both small and large samples, indicating that recruiting more participants will not alleviate inflation of the Type I error rate associated with fixed effect inference. Enumerations and simulations help quantify the bias and evaluate strategies for avoiding it. When practical, backwards selection of the covariance model, starting with an unstructured pattern, provides the best protection. Tutorial papers can guide the reader in minimizing the chances of falling into the often spurious software trap of nonconvergence. In some cases, the logic of the study design and the scientific context may support a structured pattern, such as an autoregressive structure. The sandwich estimator provides a valid alternative in sufficiently large samples. Authors reporting mixed‐model analyses should note possible biases in fixed effects inference because of the following: (i) the covariance model selection process; (ii) the specific covariance model chosen; or (iii) the test approximation. Copyright © 2011 John Wiley & Sons, Ltd.
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
Small sample inference for fixed effects from restricted maximum likelihood
Kenward, M. G., & Roger, J. H. (1997) · Biometrics · doi:10.2307/2533558 · Cited by 4,552 · cited 1×
“We employed restricted maximum likelihood (REML) estimation with the Kenward–Roger approach [26] for all five cases.”
MethodLinear Mixed Models for Longitudinal Data
Verbeke, G. (1997) · Lecture Notes in Statistics · doi:10.1007/978-1-4612-2294-1_3 · Cited by 3,506
Power Calculations for General Linear Multivariate Models Including Repeated Measures Applications
Muller, K. E., Lavange, L. M., Ramey, S. L., & Ramey, C. T. (1992) · Journal of the American Statistical Association · doi:10.1080/01621459.1992.10476281 · Cited by 230 · cited 3×
“Methods in [29] allow exact computation and accurate approximation of the cumulative distribution function of Fm for finite samples.”
BackgroundMethodSelf-citationSelecting the Best Linear Mixed Model Under REML
Gurka, M. J. (2006) · The American Statistician · doi:10.1198/000313006x90396 · Cited by 200 · cited 1×
“There are no fully reliable ways to identify the best covariance model, at least in small samples [4].”
BackgroundSelf-citationImportance of Early Neglect for Childhood Aggression
Kotch, J. B., Lewis, T., Hussey, J. M., English, D., Thompson, R., Litrownik, A. J., Runyan, D. K., Bangdiwala, S. I., Margolis, B., & Dubowitz, H. (2008) · Pediatrics · doi:10.1542/peds.2006-3622 · Cited by 183
Approximate Power for Repeated-Measures ANOVA Lacking Sphericity
Muller, K. E., & Barton, C. N. (1989) · Journal of the American Statistical Association · doi:10.2307/2289941 · Cited by 177
Self-citation
Cited by
Robustness of linear mixed-effects models to violations of distributional assumptions
Schielzeth, H., Dingemanse, N. J., Nakagawa, S., Westneat, D. F., Allegue, H., Teplitsky, C., Réale, D., Dochtermann, N. A., Garamszegi, L. Z., & Araya‐Ajoy, Y. G. (2020) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.13434 · Cited by 1,393
Rights, J. D., & Sterba, S. K. (2019) · Psychological Methods · doi:10.1037/met0000184 · Cited by 413
An R2statistic for fixed effects in the generalized linear mixed model
Jaeger, B. C., Edwards, L. J., Das, K., & Sen, P. K. (2017) · Journal of Applied Statistics · doi:10.1080/02664763.2016.1193725 · Cited by 399
Self-citationSelecting a sample size for studies with repeated measures
Guo, Y., Logan, H. L., Glueck, D. H., & Muller, K. E. (2013) · BMC Medical Research Methodology · doi:10.1186/1471-2288-13-100 · Cited by 374
Self-citationWieling, M. (2018) · Journal of Phonetics · doi:10.1016/j.wocn.2018.03.002 · Cited by 362
“It is essential to assess which random intercepts and slopes need to be included, as failing to include a necessary random slope may yield p -values which are overcon fi dent (Gurka, Edwards, & Muller, 2011).”
Metadata
- Authors
- Gurka, M. J., Edwards, L. J., & Muller, K. E.
- Venue
- Statistics in Medicine
- Publisher
- Wiley
- Published
- 2011-07-12
- Volume
- 30
- Issue
- 22
- Pages
- 2696-2707
- ISSN
- 0277-6715 (ISSN-L)
- 1097-0258
- OpenAlex
- W2013051231
- Cited by
- 98
- References
- 6
- FWCI
- 5.56
- Type
- journal article
- Language
- en
- Open access
- green
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.