IBM SPSS Statistics 19 Statistical Procedures Companion: Chapters
- Introduction. An overview of the statistical procedures described in the book.
- Getting to Know IBM SPSS Statistics. Tutorials; windows; dialog boxes; the online help system; Pivot Table Editor; Chart Editor; using command syntax.
- Introducing Data. Planning the data file; getting data into IBM SPSS Statistics; the Text Wizard; creating new variables; selecting cases.
- Preparing Your Data. Checking variable definitions; case counts; data values.
- Transforming Your Data. Computing new variables; changing coding schemes; ranking.
- Describing Your Data. Taking a first look at your data with tables, charts, and descriptive statistics.
- Testing Hypotheses. Samples and populations; missing values; steps in testing a hypothesis; calculating confidence intervals; reporting your results correctly; commonly used tests for popular hypotheses.
- T-tests. One-sample, paired-samples, and independent-samples t-tests; data setups for the different t-tests; interpreting the output.
- One-way Analysis of Variance. Preparing the data file; examining the data; checking assumptions; interpreting the output; atoning for multiple comparisons; setting up contrasts.
- Crosstabulation. Chi-square tests; McNemar's test; measures of association and agreement; measures of risk; testing hypotheses about odds ratios.
- Correlation. Plotting the data; scatterplot matrices; correlation coefficients based on ranks; partial correlation coefficients; identifying points in a scatterplot.
- Bivariate Linear Regression. Least squares regression line; measures of fit; assumptions and transformations; looking for unusual points.
- Multiple Linear Regression. Formulating the problems; interpreting the coefficients; including categorical variables; comparing models; automated model building; checking for violations of assumptions; residuals; unusual observations.
- Automatic Linear Modeling. All possible subsets regression models; specifying the model; model building summary; comparing observed and expected values; examining the coefficients; examining assumptions; looking for influential points; ensemble models; automatic data preparation; model summary; evaluating the predictors.
- Discriminant Analysis. Calculating the functions; testing hypotheses; classifying cases into groups; automated model building; analyzing more than two groups; classification function coefficients.
- Logistic Regression Analysis. Basics of the model; predicted probabilities; coefficients; testing hypotheses; categorical variables; interaction terms; evaluating linearity; automated model building; diagnostics; model calibration; model discrimination; diagnostics for individual cases.
- Cluster Analysis. Hierarchical clustering; k-means clustering; two-step clustering; distance and similarity measures; interpreting the results.
- Factor Analysis. Basics of the model; determining the number of factors; goodness-of-fit tests; methods for factor extraction and rotation; computing factor scores.
- Reliability Analysis. Reliability coefficients: Cronbach's alpha, split-half reliability, Guttman's lower bounds; testing hypotheses about scales: parallel and strictly parallel models; Cochran's Q; intraclass correlation coefficients.
- Nonparametric Tests. One-sample tests: chi-square, binomial, runs; two related groups: sign test, Wilcoxon test; two independent groups: Wilcoxon, Wald-Wolfowitz runs test; three or more groups: Kruskal-Wallis, median, Friedman, Kendall's W, Cochran's Q.
- General Loglinear Analysis. Basic model; fitting a saturated model; fitting an unsaturated model; goodness-of-fit tests; models for ordinal data; incomplete tables; tests for square tables; Poisson regression; standardizing tables.
- Univariate General Linear Model. Regression; two-way ANOVA; randomized complete block design; randomized complete block design with empty cells; analysis of covariance; mixed effects nested designs; split-plot designs.
- Multivariate General Linear Model. Multivariate two-way fixed-effects model with interaction; profile analysis; setting up custom linear hypotheses.
- Repeated Measures Designs. Checking assumptions; testing hypotheses; doubly multivariate repeated measures analysis of variance.