Conclusions, Extensions, and Applications
As stated in the Preface, this book has been written with quantitatively oriented social and behavioral science researchers in mind. Its main purposes have been to thoroughly explain the methodological issues and principles involved in marginal modeling and to indicate the types of substantive research questions for which marginal modeling is very useful or even the only means of obtaining the appropriate answers. Therefore, a large variety of real-world examples have been presented to help the reader understand the main issues involved. The book does provide a solid statistical basis for marginal modeling, but mathematical statisticians might prefer a more theoretical mathematical approach. Especially for our intended audience, it is important to have good and accessible computer programs and routines to carry out themarginal analyses. In the last section of this chapter, several computer programs, including our own Mathematica and R routines will be introduced. The reader can find all data sets used in this book and the Mathematica and R routines to carry out the book’s analyses on these data sets on the book’s website, which will be given in the same section; these routines entail the often complicated matrices with the generalized exp-log notation needed to define the models.
KeywordsGeneralize Estimate Equation Weight Little Square Loglinear Model Marginal Model Partner Effect
Unable to display preview. Download preview PDF.