# Statistical Analysis of Longitudinal Categorical Data in the Social and Behavioral Sciences: An Introduction with Computer Illustrations

## Synopsis

A comprehensive resource for analyzing a variety of categorical data, this book emphasizes the application of many recent advances of longitudinal categorical statistical methods. Each chapter provides basic methodology, helpful applications, examples using data from all fields of the social sciences, computer tutorials, and exercises. Written for social scientists and students, no advanced mathematical training is required. Step-by-step command files are given for both the CDAS and the SPSS software programs.

## Excerpt

Statistical analysis of longitudinal categorical data in the social and behavioral sciences - An introduction with computer illustrations

A large amount of data in the social sciences comes in the form of categorical variables. Diagnostic units such as schizophrenic and psychotic, verdicts such as guilty and not guilty, and simple preferences such as like and dislike are all examples of categorical data. Furthermore, many variables that are commonly thought of as continuous can be functionally defined in categorical terms. For instance, body temperature can be meaningfully divided up into hypothermic, normal, and fevered.

Many social scientists, however, analyze this type of data using methods intended for continuous data or use only χ2 and simple hierarchical log-linear models. Even now, almost 25 years after the publication of the classic Discrete Multivariate Analysis by Bishop, Fienberg, and Holland (1975), the use and analysis of categorical variables is quite lacking. This stagnation is most apparent in the analysis of longitudinal and developmental data.

Although many social scientists have been limited in their use and analysis of categorical data, there have been many interesting statistical developments for the analysis of such data (see, for instance, Hagenaars, 1990; von Eye &Clogg, 1996). With these developments, any statistical goal that one could pursue using continuous data can be pursued using categorical data. Examples include the analysis of longitudinal data, the analysis of causal assumptions, the prediction of dependent events, the formation of groups and clusters, and the modeling of specific theory driven models.

This text is intended to provide a means for applying the many advances that have been made in longitudinal categorical statistical methods. This volume is designed to be accessible to the average social scientist and statistics student. No advanced mathematical or statistical training is necessary. Additionally, this book has been . . .

Search by...
Show...

### Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.