☆ 4.4 Article

Inconvenient truths about logistic regression and the remedy of marginal effects

PUBLIC ADMINISTRATION REVIEW (2023)

Related references

Note: Only part of the references are listed.

Editorial Material Health Care Sciences & Services

Controversy and Debate: Questionable utility of the relative risk in clinical research: Paper 2: Is the Odds Ratio portable in meta-analysis? Time to consider bivariate generalized linear mixed model COMMENT

Mengli Xiao et al.

Summary: Studies in the Cochrane Database of Systematic Reviews have shown that study-specific OR tends to be higher in studies with lower baseline risks and there is a strong negative correlation between OR (RR or RD) and baseline risk, with conditional effects notably varying with baseline risks.

JOURNAL OF CLINICAL EPIDEMIOLOGY (2022)

Add to Collection

Article Social Sciences, Mathematical Methods

Interpreting logit models

Luca J. Uberti

Summary: The article reviews a range of methods for correctly interpreting and effectively applying logit models, emphasizing the importance of using Stata and highlighting that interaction terms are typically easier to interpret in practice than implied by the literature.

STATA JOURNAL (2022)

Add to Collection

Article Psychology, Multidisciplinary

The Partial Derivative Framework for Substantive Regression Effects

Dale S. Kim et al.

Summary: Regression models are widely used in psychological sciences, but interpreting coefficient estimates in nonlinear modeling scenarios can be problematic. We address this issue by developing a framework that explicitly models substantive variables and uses partial derivatives to summarize the relationship between predictors and outcomes.

PSYCHOLOGICAL METHODS (2022)

Add to Collection

Editorial Material Health Care Sciences & Services

Noncollapsibility, confounding, and sparse-data bias. Part 1: The oddities of odds COMMENT

Sander Greenland

Summary: To prevent statistical misinterpretations, it is advised to focus on estimation instead of statistical testing. Measures based on odds or their logarithms are often promoted for their statistical properties, but they have the undesirable property of noncollapsibility. This problem is illustrated with a basic numeric example in this note, and is amplified in odds ratios and logistic regression.

JOURNAL OF CLINICAL EPIDEMIOLOGY (2021)

Add to Collection

Article Health Care Sciences & Services

Noncollapsibility, confounding, and sparse-data bias. Part 2: What should researchers make of persistent controversies about the odds ratio?

Sander Greenland

Summary: The article discusses the issue of noncollapsibility of odds ratios and provides a basic numerical example illustrating the difference between noncollapsibility and confounding effects, as well as its connection to sparse-data bias in logistic regression.

JOURNAL OF CLINICAL EPIDEMIOLOGY (2021)

Add to Collection

Article Social Sciences, Mathematical Methods

Using Predictions and Marginal Effects to Compare Groups in Regression Models for Binary Outcomes

J. Scott Long et al.

Summary: This study introduces methods for group comparisons using predicted probabilities and marginal effects on probabilities in regression models for binary outcomes. Unlike traditional approaches, these methods are not affected by the scalar identification of regression coefficients and are expressed in the natural metric of the outcome probability. The interpretive framework developed in this study can be applied to a wide range of regression models and can be extended to any number of groups.

SOCIOLOGICAL METHODS & RESEARCH (2021)

Add to Collection

Article Social Sciences, Mathematical Methods

On Group Comparisons With Logistic Regression Models

Jouni Kuha et al.

SOCIOLOGICAL METHODS & RESEARCH (2020)

Add to Collection

Review Education & Educational Research

A review of the application of logistic regression in educational research: common issues, implications, and suggestions

Lian Niu

EDUCATIONAL REVIEW (2020)

Add to Collection

Article Health Care Sciences & Services

Log Odds and the Interpretation of Logit Models

Edward C. Norton et al.

HEALTH SERVICES RESEARCH (2018)

Add to Collection

Editorial Material Medicine, General & Internal

Odds Ratios-Current Best Practice and Use

Edward C. Norton et al.

JAMA-JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION (2018)

Add to Collection

Article Health Care Sciences & Services

The alarming problems of confounding equivalence using logistic regression models in the perspective of causal diagrams

Yuanyuan Yu et al.

BMC MEDICAL RESEARCH METHODOLOGY (2017)

Add to Collection

Article Political Science

Compression and Conditional Effects: A Product Term Is Essential When Using Logistic Regression to Test for Interaction

Carlisle Rainey

POLITICAL SCIENCE RESEARCH AND METHODS (2016)

Add to Collection

Article Economics

A Primer on Marginal Effects-Part II: Health Services Research Applications

E. Onukwugha et al.

PHARMACOECONOMICS (2015)

Add to Collection

Article Public, Environmental & Occupational Health

Estimating predicted probabilities from logistic regression: different methods correspond to different target populations

Clemma J. Muller et al.

INTERNATIONAL JOURNAL OF EPIDEMIOLOGY (2014)

Add to Collection

Article Political Science

Behind the Curve: Clarifying the Best Approach to Calculating Predicted Probabilities and Marginal Effects from Limited Dependent Variable Models

Michael J. Hanmer et al.

AMERICAN JOURNAL OF POLITICAL SCIENCE (2013)

Add to Collection

Article Social Sciences, Mathematical Methods

Using the margins command to estimate and interpret adjusted predictions and marginal effects

Richard Williams

STATA JOURNAL (2012)

Add to Collection

Article Sociology

Logistic Regression: Why We Cannot Do What We Think We Can Do, and What We Can Do About It

Carina Mood

EUROPEAN SOCIOLOGICAL REVIEW (2010)

Add to Collection

Article Economics

Robustness of Logit analysis: Unobserved heterogeneity and mis-specified disturbances

J. S. Cramer

OXFORD BULLETIN OF ECONOMICS AND STATISTICS (2007)

Add to Collection

Article Political Science

Understanding interaction models: Improving empirical analyses

T Brambor et al.

POLITICAL ANALYSIS (2006)

Add to Collection

© Peeref 2019-2024. All rights reserved.