IBM SPSS Advanced Statistics 22 - University of Sussex

1 IBM spss Advanced Statistics 22 NoteBefore using this information and the product it supports, read the information in Notices on page InformationThis edition applies to version 22, release 0, modification 0 of IBM spss Statistics and to all subsequent releases andmodifications until otherwise indicated in new 1. Introduction to 2. GLM Multivariate Analysis .. 3 GLM Multivariate of Multivariate Multivariate Profile Multivariate Post Hoc Multivariate Command Additional 3. GLM Repeated Measures .. 11 GLM Repeated Measures Define Repeated Measures of Repeated Measures Repeated Measures Profile Repeated Measures Post Hoc Comparisons.

2 16 GLM Repeated Measures Repeated Measures Command Additional 4. Variance Components Components of Squares (Variance Components)..23 Variance Components Save to New Command Additional 5. Linear Mixed Mixed Models Select Mixed Models Fixed Non-Nested Nested of Mixed Models Random Mixed Models Mixed Models Mixed Models EM Mixed Models Command Additional 6. Generalized Linear Models 33 Generalized Linear Models Linear Models Reference Category36 Generalized Linear Models Linear Models Linear Models Linear Models Linear Models Initial 38 Generalized Linear Models Linear Models EM Linear Models Linear Models Command Additional 7.

3 Generalized Estimating Equations Type of Model .. 47 Generalized Estimating Equations 49 Generalized Estimating Equations Estimating Equations Predictors .. 49 Generalized Estimating Equations Options .. 50 Generalized Estimating Equations Estimating Equations Estimation .. 51 Generalized Estimating Equations Initial Values52 Generalized Estimating Equations Estimating Equations EM Means .. 53 Generalized Estimating Equations Estimating Equations Command Additional 8. Generalized linear a generalized linear mixed a Custom Effect and Build by Effect Means: Significant Means: Custom 9.

4 Model Selection Analysis Define Analysis Selection Loglinear Analysis 72 HILOGLINEAR Command Additional Features .. 72 Chapter 10. General Loglinear Analysis 73 General Loglinear Analysis Loglinear Analysis Loglinear Analysis Command Additional 11. Logit Loglinear Analysis 77 Logit Loglinear Analysis Loglinear Analysis Loglinear Analysis Command Additional 12. Life Tables Define Events for Status Variables .. 82 Life Tables Define Tables Command Additional 13. Kaplan-Meier Define Event for Status 86 Kaplan-Meier Compare Factor Save New Command Additional 14.

5 Cox regression Analysis 89 Cox regression Define Categorical Variables .. 89 Cox regression regression Save New regression regression Define Event for Status Variable .. 91 COXREG Command Additional 15. ComputingTime-Dependent a Time-Dependent regression with Time-Dependent CovariatesAdditional 16. Categorical VariableCoding 17. Covariance Structures.. spss Advanced Statistics 22 Chapter 1. Introduction to Advanced StatisticsThe Advanced Statistics option provides procedures that offer more Advanced modeling options than areavailable through the Statistics Base Multivariate extends the general linear model provided by GLM Univariate to allow multipledependent variables.

6 A further extension, GLM Repeated Measures, allows repeated measurements ofmultiple dependent Components Analysis is a specific tool for decomposing the variability in a dependentvariable into fixed and random Mixed Models expands the general linear model so that the data are permitted to exhibitcorrelated and nonconstant variability. The mixed linear model, therefore, provides the flexibility ofmodeling not only the means of the data but the variances and covariances as Linear Models (GZLM) relaxes the assumption of normality for the error term andrequires only that the dependent variable be linearly related to the predictors through a transformation,or link function.

7 Generalized Estimating Equations (GEE) extends GZLM to allow Loglinear Analysis allows you to fit models for cross-classified count data, and ModelSelection Loglinear Analysis can help you to choose between Loglinear Analysis allows you to fit loglinear models for analyzing the relationship between acategorical dependent and one or more categorical analysis is available through Life Tables for examining the distribution of time-to-eventvariables, possibly by levels of a factor variable; Kaplan-Meier Survival Analysis for examining thedistribution of time-to-event variables, possibly by levels of a factor variable or producing separateanalyses by levels of a stratification variable; and Cox regression for modeling the time to a specifiedevent, based upon the values of given covariates.

8 Copyright IBM Corporation 1989, 201312 IBM spss Advanced Statistics 22 Chapter 2. GLM Multivariate AnalysisThe GLM Multivariate procedure provides regression analysis and analysis of variance for multipledependent variables by one or more factor variables or covariates. The factor variables divide thepopulation into groups. Using this general linear model procedure, you can test null hypotheses aboutthe effects of factor variables on the means of various groupings of a joint distribution of dependentvariables. You can investigate interactions between factors as well as the effects of individual factors.

9 Inaddition, the effects of covariates and covariate interactions with factors can be included. For regressionanalysis, the independent (predictor) variables are specified as balanced and unbalanced models can be tested. A design is balanced if each cell in the modelcontains the same number of cases. In a multivariate model, the sums of squares due to the effects in themodel and error sums of squares are in matrix form rather than the scalar form found in univariateanalysis. These matrices are called SSCP (sums-of-squares and cross-products) matrices.

10 If more than onedependent variable is specified, the multivariate analysis of variance using Pillai's trace, Wilks' lambda,Hotelling's trace, and Roy's largest root criterion with approximateFstatistic are provided as well as theunivariate analysis of variance for each dependent variable. In addition to testing hypotheses, GLMM ultivariate produces estimates of useda prioricontrasts are available to perform hypothesis testing. Additionally, after anoverallFtest has shown significance, you can use post hoc tests to evaluate differences among specificmeans.