Companion Website
About the Authors
Preface
1 RESEARCH AND STATISTICS
1.1 The methodology of statistical research
1.2 The statistical method
1.3 The logic behind statistical inference
1.3.1 Probability theory
1.3.2 Population size
1.3.3 Why do I need significance levels if I am investigating the whole
population?
1.4 General laws and theories
1.4.1 Objectivity and critical realism
1.5 Quantitative research papers
1.6 Concluding remarks
Questions
Further reading
References
2 INTRODUCTION TO STATA
2.1 What is Stata?
2.1.1 The Stata interface
2.1.2 How to use Stata
2.2 Entering and importing data into Stata
2.2.1 Entering data
2.2.2 Importing data
2.3 Data management
2.3.1 Opening data
2.3.2 Examining data
2.3.3 Making changes to variables
2.3.4 Generating variables
2.3.5 Subsetting data
2.3.6 Labelling variables
2.4 Descriptive statistics and graphs
2.4.1 Frequency distributions
2.4.2 Summary statistics
2.4.3 Appending data
2.4.4 Merging data
2.4.5 Reshaping data
2.5 Bivariate inferential statistics
2.5.1 Correlation
2.5.2 Independent t-test
2.5.3 Analysis of variance (ANOVA)
2.5.4 Chi-squared test
2.6 Conclusion
Questions
Further reading
3 SIMPLE (BIVARIATE) REGRESSION
3.1 What is regression analysis?
3.2 Simple linear regression analysis
3.2.1 Ordinary least squares
3.2.2 Goodness of fit
3.2.3 Hypothesis test for slope coefficient
3.2.4 Prediction in linear regression
3.3 Example in Stata
3.4 Conclusion
Questions
Further reading
References
4 MULTIPLE REGRESSION
4.1 Multiple linear regression analysis
4.1.1 Estimation
4.1.2 Goodness of fit and the F-test
4.1.3 Adjusted R²
4.1.4 Partial slope coefficients
4.1.5 Prediction in multiple regression
4.1.6 Standardization and relative importance
4.2 Example in Stata
4.3 Conclusion
Questions
Further reading
References
5 DUMMY-VARIABLE REGRESSION
5.1 Why dummy-variable regression?
5.1.1 Creating dummy variables
5.1.2 The logic behind dummy-variable regression
5.2 Regression with one dummy variable
5.2.1 Example in Stata
5.3 Regression with one dummy variable and a covariate
5.3.1 Example in Stata
5.4 Regression with more than one dummy variable
5.4.1 Example in Stata
5.4.2 Comparing the included groups
5.5 Regression with more than one dummy variable and a covariate
5.5.1 Example in Stata
5.6 Regression with two separate sets of dummy variables
5.6.1 Example in Stata
5.7 Conclusion
Questions
Further reading
References
6 INTERACTION/MODERATION EFFECTS USING REGRESSION
6.1 Interaction/moderation effect
6.2 Product-term approach
6.2.1 Interaction between a continuous predictor and a continuous
moderator
6.2.2 Interaction between a continuous predictor and a dummy moderator
6.2.3 Interaction between a dummy predictor and a dummy moderator
6.2.4 Interaction between a continuous predictor and a polytomous
moderator
6.3 Conclusion
Questions
Further reading
References
7 LINEAR REGRESSION ASSUMPTIONS AND DIAGNOSTICS
7.1 Correct specification of the model
7.1.1 All X-variables relevant, and none irrelevant
7.1.2 Linearity
7.1.3 Additivity
7.1.4 Absence of multicollinearity
7.2 Assumptions about residuals
7.2.1 That the error term has a conditional mean of zero
7.2.2 Homoskedasticity
7.2.3 Uncorrelated errors
7.2.4 Normally distributed errors
7.3 Influential observations
7.3.1 Leverage
7.3.2 DFBETA
7.3.3 Cook's distance
7.4 Conclusion
Questions
Further reading
References
8 LOGISTIC REGRESSION
8.1 What is logistic regression?
8.1.1 Tests of significance
8.2 Assumptions of logistic regression
8.2.1 Example in Stata
8.3 Conditional effects
8.4 Diagnostics
8.5 Multinomial logistic regression
8.6 Ordered logistic regression
8.7 Conclusion
Questions
Further reading
References
9 MULTILEVEL ANALYSIS
9.1 Multilevel data
9.1.1 Statistical reasons for using multilevel analysis
9.2 Empty or intercept-only model
9.2.1 Example in Stata
9.3 Variance partition or intraclass correlation
9.4 Random-intercept model
9.5 Level-2 explanatory variables
9.5.1 How much of the dependent variable is explained?
9.6 Logistic multilevel model
9.7 Random-coefficient (slope) model
9.8 Interaction effects
9.9 Three-level models
9.9.1 Cross-classified multilevel model
9.10 Weighting
9.11 Conclusion
Questions
Further reading
References
10 PANEL DATA ANALYSIS
10.1 Panel data
10.2 Pooled OLS
10.3 Between effects
10.4 Fixed effects (within estimator)
10.4.1 Explaining fixed effects
10.4.2 Summary of fixed effects
10.4.3 Time-fixed effects
10.5 Random effects
10.6 Time-series cross-section methods
10.6.1 Testing for non-stationarity
10.6.2 Lag selection
10.6.3 The TSCS model
10.7 Binary dependent variables
10.8 Conclusion
Questions
Further reading
References
11 EXPLORATORY FACTOR ANALYSIS
11.1 What is factor analysis?
11.1.1 What is factor analysis used for?
11.2 The factor analysis process
11.2.1 Extracting the factors
11.2.2 Determining the number of factors
11.2.3 Rotating the factors
11.2.4 Refining and interpreting the factors
11.3 Composite scores and reliability test
11.4 Example in Stata
11.5 Conclusion
Questions
Further reading
References
12 STRUCTURAL EQUATION MODELLING AND CONFIRMATORY
FACTOR ANALYSIS
12.1 What is structural equation modelling?
12.1.1 Types of structural equation modelling
12.2 Confirmatory factor analysis
12.2.1 Model specification
12.2.2 Model identification
12.2.3 Parameter estimation
12.2.4 Model assessment
12.2.5 Model modification
12.3 Latent path analysis
12.3.1 Specification of the LPA model
12.3.2 Measurement part
12.3.3 Structural part
12.4 Conclusion
Questions
Further reading
References
13 CRITICAL ISSUES
13.1 Transformation of variables
13.1.1 Skewness and kurtosis
13.1.2 Transformations
13.2 Weighting cases
13.3 Robust regression
13.4 Missing data
13.4.1 Traditional methods for handling missing data
13.4.2 Multiple imputation
13.5 Conclusion
Questions
Further reading
References
Index
)
