Principles of multilevel modelling
International Journal of EpidemiologyPublished 1 February 2000
Sander Greenland
Citations479
SJR quartileQ1
SJR score2.69
SNIP2.48
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
The role of multilevel averaging ('shrinkage') in the reduction of estimation error, and the role of prior information in finding good averages are focused on.
Abstract
Multilevel modelling, also known as hierarchical regression, generalizes ordinary regression modelling to distinguish multiple levels of information in a model. Use of multiple levels gives rise to an enormous range of statistical benefits. To aid in understanding these benefits, this article provides an elementary introduction to the conceptual basis for multilevel modelling, beginning with classical frequentist, Bayes, and empirical-Bayes techniques as special cases. The article focuses on the role of multilevel averaging ('shrinkage') in the reduction of estimation error, and the role of prior information in finding good averages.
Keywords
MathematicsBiochemistry, Genetics and Molecular Biology
Bayesian Data Analysis
13,688 Citations2003Andrew Gelman, John B. Carlin +2 more
Journal of the American Statistical AssociationStatistics and Causal Inference
5,110 Citations1986Paul W. Holland
Introducing Multilevel Modeling
2,913 Citations1998Ita G. G. Kreft, Jan de Leeuw
Wiley series in probability and statisticsVariance Components
1,918 Citations1992S. R. Searle, George Casella +1 more
Zenodo (CERN European Organization for Nuclear Research)A Treatise on Probability
1,266 Citations1922Harold Jeffreys
This book is a searching analysis of the fundamental principles of the theory of probability and of the particular judgments involved in its application to concrete problems and is in agreement with the views expressed by Dr. Wrinch and the present reviewer.
Journal of the American Statistical AssociationStein's Estimation Rule and its Competitors—An Empirical Bayes Approach
875 Citations1973Bradley Efron, Carl N. Morris
Journal of the American Statistical AssociationData Analysis Using Stein's Estimator and its Generalizations
842 Citations1975Bradley Efron, Carl N. Morris
Journal of the Royal Statistical Society Series B (Statistical Methodology)Regression, Prediction and Shrinkage
621 Citations1983J. B. Copas
The American StatisticianAn Introduction to Empirical Bayes Data Analysis
453 Citations1985George Casella
Journal of the American Statistical AssociationThe Hierarchical Logistic Regression Model for Multilevel Analysis
434 Citations1985George Y. Wong, William M. Mason
EpidemiologyRandomization, Statistics, and Causal Inference
381 Citations1990Sander Greenland
It is concluded that probabilistic interpretations of conventional statistics are rarely justified, and that such interpretations may encourage misinterpretation of nonrandomized studies.
Journal of Educational StatisticsEmpirical Bayes Meta-Analysis
269 Citations1985Stephen W. Raudenbush, Anthony S. Bryk
Journal of the American Statistical AssociationBayes Empirical Bayes
235 Citations1981John Deely, D. V. Lindley
It is shown that empirical Bayes procedures are really non-Bayesian, asymptotically optimal, classical procedures for mixtures.
PubMedInsulin-like growth factor-I in relation to premenopausal ductal carcinoma in situ of the breast.
208 Citations1998Kari Bohlke, Daniel W. Cramer +2 more
American Journal of EpidemiologyTHE ROLE OF MODEL SELECTION IN CAUSAL INFERENCE FROM NONEXPERIMENTAL DATA
198 Citations1986James M. Robins, Sander Greenland
It is argued that all modelling strategies contain implicit prior beliefs about nature, and one can justify a modelling strategy only if the strategy is justified.
American Journal of EpidemiologyMulti-level Analysis in Epidemiologic Research on Health Behaviors and Outcomes
182 Citations1992Michael Von Korff, Thomas D. Koepsell +2 more
The authors review concepts and methods of multi-level analysis and their application to epidemiologic research on health behavior and health outcomes.
BiometricsWhen Should Epidemiologic Regressions Use Random Coefficients?
163 Citations2000Sander Greenland
It is argued that regression models with random coefficients offer a more scientifically defensible framework for epidemiologic analysis than the fixed‐effects models now prevalent in epidemiology.
Statistics in MedicineMethods for epidemiologic analyses of multiple exposures: A review and comparative study of maximum‐likelihood, preliminary‐testing, and empirical‐bayes regression
143 Citations1993Sander Greenland
This work reviews two hierarchical methods and compares their performance in a problem of predicting neonatal mortality rates and suggests that hierarchical methods should become part of the standard approaches to multiple-exposure studies.
EpidemiologyEmpirical-Bayes Adjustments for Multiple Comparisons Are Sometimes Useful
138 Citations1991Sander Greenland, James M. Robins
In situations in which an additional or even primary goal of analysis is to reach a set of decisions based on the data, Bayes and empirical-Bayes adjustments can provide a better basis for the decisions than conventional procedures.
Journal of the Royal Statistical Society Series A (Statistics in Society)Exchangeability and Data Analysis
127 Citations1993David Draper, James S. Hodges +2 more
The logic of judgments of exchangeability is analysed, to clarify the roles of context and data analysis in these judgments, and the nature of these judgments in description, inference and prediction is given.
EpidemiologyHierarchical Regression Analysis Applied to a Study of Multiple Dietary Exposures and Breast Cancer
114 Citations1994John S. Witte, Sander Greenland +2 more
This work uses hierarchical regression to analyze case-control data on diet and breast cancer to yield semi-Bayes relative risk estimates for dietary items by using a second-stage model to pull estimates toward each other when the corresponding variables have similar levels of nutrients.
PubMedCardiovascular diseases and diabetes mellitus in different ethnic groups: the Finnmark study.
107 Citations1998J S Witte, S Greenland +1 more
EpidemiologyProbability Logic and Probabilistic Induction
100 Citations1998Sander Greenland
How probability logic can give precise meanings to the concepts of inductive support, corroboration, refutation, and related notions, as well as provide a foundation for logically sound statistical inference, is described.
The American StatisticianReview of Five Multilevel Analysis Programs: BMDP-5V, GENMOD, HLM, ML3, VARCL
92 Citations1994Ita G. G. Kreft, Jan de Leeuw +1 more
BiostatisticsSmall-sample bias and corrections for conditional maximum-likelihood odds-ratio estimators
87 Citations2000Sander Greenland
The rationale and performance of several corrections for the conditional maximum-likelihood estimator of the odds ratio for matched pairs with a dichotomous exposure are contrasted.
Statistics in MedicineA semi‐bayes approach to the analysis of correlated multiple associations, with an application to an occupational cancer‐mortality study
80 Citations1992Sander Greenland
The application of empirical-Bayes methods to the problem of multiple inference in epidemiologic studies is generalized to allow for non-exchangeable parameters and non-independent estimates, and the resulting method is Bayesian in so far as some feature of the prior distribution are specified from prior information.
Statistical ScienceStein Estimation: The Spherically Symmetric Case
75 Citations1990Ann Cohen Brandwein, William E. Strawderman
Environmental Health PerspectivesHierarchical regression for epidemiologic analyses of multiple exposures.
74 Citations1994Sander Greenland
Based on the literature to date, it is suggested that hierarchical methods should become part of the standard approaches to multiple-exposure studies and are compared to two hierarchical methods, empirical-Bayes regression and a variant here called "semi-Baye regression.
Archives of Environmental Health An International JournalEmpirical-Bayes and Semi-Bayes Approaches to Occupational and Environmental Hazard Surveillance
70 Citations1994Sander Greenland, Charles Poole
This work provides an elementary introduction to empirical-Bayes analysis of occupational and environmental hazard surveillance data, using a study of occupational exposures and cancer mortality to illustrate key concepts.
Statistics in MedicineSIMULATION STUDY OF HIERARCHICAL REGRESSION
55 Citations1996John S. Witte, Sander Greenland
A simulation study of logistic regression in which hierarchical regression fitted by a two-stage procedure to ordinary maximum likelihood is compared, indicating that hierarchical modelling of continuous covariates offers worthwhile improvement over ordinary maximum-likelihood.
The Annals of StatisticsUsing Empirical Partially Bayes Inference for Increased Efficiency
51 Citations1985Bruce G. Lindsay
Statistics in MedicineSECOND-STAGE LEAST SQUARES VERSUS PENALIZED QUASI-LIKELIHOOD FOR FITTING HIERARCHICAL MODELS IN EPIDEMIOLOGIC ANALYSES
50 Citations1997Sander Greenland
Three relatively simple choices for estimation of the regression coefficients are compared: maximum-likelihood first stage combined with a weighted-least-squares second stage (MLLS); joint iteratively reweighted least squares fitting of first andsecond stage (JILS); and empirically penalized quasi- likelihood (EPQL).
EpidemiologySoftware for Hierarchical Modeling of Epidemiologic Data
44 Citations1998John S. Witte, Sander Greenland +1 more
This work describes the existing software packages that one can use to fit hierarchical models and presents SAS code for analyzing epidemiologic data with hierarchical models, allowing epidemiologists tofit hierarchical models with readily available software.
BiometrikaA general framework for model-based statistics
35 Citations1990Joe R. Hill
A general framework for model-based statistics based on extended concepts of sufficiency and ancillarity is presented, which exhibits in a single equation the reference distributions for decision theory, model checking, and conditional inference for a general statistical model.
Scandinavian Journal of Social MedicineSummarization, smoothing, and inference in epidemiologic analysis
26 Citations1993Sander Greenland
This paper proposes that some resolution can be achieved by separating data analysis into summarization, smoothing, and inferential phases, and suggests improvements could be made to all three phases simply by keeping the phases distinct.
PubMedMultilevel modeling and model averaging.
22 Citations1999Sander Greenland
This article provides an elementary introduction to multilevel modeling as a model-averaging technique that provides an alternative to model selection, and it emphasizes the role of prior information in finding good models.
BiometricsReducing Mean Squared Error in the Analysis of Stratified Epidemiologic Studies
12 Citations1991Sander Greenland, Leslie A. Kalish
It is shown that for unmatched stratified studies, the form of the approximately optimal estimator is very simple, and may often correspond to no more than a small correction to the ordinary stratified estimator.
BiometricsReducing Mean Squared Error in the Analysis of Pair-Matched Case-Control Studies
11 Citations1990Leslie A. Kalish
The possibility of optimizing the trade-off between bias and precision is explored and the family of weighted averages of the stratified and pooled estimates is considered, and the weight that minimizes an asymptotic approximation of mean squared error is derived.
The American StatisticianEstimation When Using a Statistic That is Not Sufficient
8 Citations1987Glen Meeden