Applied logistic regression analysis / Scott Menard.

Format:

Book

Author/Creator:

Menard, Scott W.

Series:

Quantitative applications in the social sciences ; no. 07-106.

Sage university papers. Quantitative applications in the social sciences ; no. 07-106

Language:

English

Subjects (All):

Regression analysis.

Logistic distribution.

Physical Description:

viii, 111 pages : illustrations ; 22 cm.

Edition:

Second edition.

Place of Publication:

Thousand Oaks, Calif. : Sage Publications, [2002]

Summary:

The focus in this Second Edition is again on logistic regression models for individual level data, but aggregate or grouped data are also considered. The book includes detailed discussions of goodness of fit, indices of predictive efficiency, and standardized logistic regression coefficients, and examples using SAS and SPSS are included. -- More detailed consideration of grouped as opposed to case-wise data throughout the book -- Updated discussion of the properties and appropriate use of goodness of fit measures, R-square analogues, and indices of predictive efficiency -- Discussion of the misuse of odds ratios to represent risk ratios, and of over-dispersion and under-dispersion for grouped data

Updated coverage of unordered and ordered polytomous logistic regression models.

Contents:

1. Linear Regression and the Logistic Regression Model 1

1.1 Regression Assumptions 4

1.2 Nonlinear Relationships and Variable Transformations 11

1.3 Probabilities, Odds, Odds Ratios, and the Logit Transformation for Dichotomous Dependent Variables 12

1.4 Logistic Regression: A First Look 14

2. Summary Statistics for Evaluating the Logistic Regression Model 17

2.1 R[superscript 2], F, and Sums of Squared Errors 18

2.2 Goodness of Fit: G[subscript M], R[superscript 2 subscript L], and the Log Likelihood 20

2.3 Predictive Efficiency: [lambda subscript p], [tau subscript p], [phi subscript p], and the Binomial Test 27

2.4 Examples: Assessing the Adequacy of Logistic Regression Models 36

2.5 Conclusion: Summary Measures for Evaluating the Logistic Regression Model 41

3. Interpreting the Logistic Regression Coefficients 41

3.1 Statistical Significance in Logistic Regression Analysis 43

3.2 Interpreting Unstandardized Logistic Regression Coefficients 48

3.3 Substantive Significance and Standardized Coefficients 51

3.4 Exponentiated Coefficients or Odds Ratios 56

3.5 More on Categorical Predictors: Contrasts and Interpretation 57

3.6 Interaction Effects 61

3.7 Stepwise Logistic Regression 63

4. An Introduction to Logistic Regression Diagnostics 67

4.1 Specification Error 67

4.2 Collinearity 75

4.3 Numerical Problems: Zero Cells and Complete Separation 78

4.4 Analysis of Residuals 80

4.5 Overdispersion and Underdispersion 89

4.6 A Suggested Protocol for Logistic Regression Diagnostics 90

5. Polytomous Logistic Regression and Alternatives to Logistic Regression 91

5.1 Polytomous Nominal Dependent Variables 94

5.2 Polytomous or Multinomial Ordinal Dependent Variables 97

Appendix Probabilities 107.

Notes:

Includes bibliographical references (pages 108-110).

ISBN:

0761922083

OCLC:

46729126