##
**Multiple regression in behavioral research: explanation and prediction.
3rd ed.**
*(English)*
Zbl 0920.62137

Fort Worth, TX: Harcourt Brace College Publishers. v, 1058 p. (1997).

This is the third revised version of the book (the first edition appeared 1973 and the second 1982). The author updated the second edition according to the present thinking and practice trying to retain the overall objectives and nonmathematical approach of the former editions. The book is divided into four parts and two appendicies:

I. Foundations of multiple regression analysis; II. Multiple regression analysis: explanation; III. Structural equation models; IV. Multivariate analysis; Appendix A: Matrix algebra, introduction; Appendix B: Tables.

Part I is devoted to theoretical foundations of multiple regression. Part II explores the relation between multiple regression analysis and analysis of variance. Also basic elements of categorical data models, continuous and categorical independent variables and logistic regression are explained. Part III concerns various structural equation models, indirect effects in structural equation models, path analysis, the use of LISREL and EQS. Part IV is devoted to the elements of multivariate methods: discriminant analysis, canonical correlation, multivariate analysis of variance, factor analysis. Newly included topics or substantially rewritten parts are: Regression diagnostics, logistic regression, multilevel analysis, research examples.

The book assumes knowledge on the level of a basic course in statistics, inluding elements of inferential statistics. The emphasis is on a non-mathematical approach, applications of suitable statistical methods for particular problems, interpretation of results, and decsription of the most often made erroneous decisions. The author discusses the assumptions under which a particular method can be applied and makes a large number of useful remarks and comments. He also shows on examples how to use the four widely used statistical packages BMDP, MINITAB, SAS and SPSS. The book can be used as a textbook, but it is certainly a useful source of information for those who have to analyse data from the area of behavioral sciences.

I. Foundations of multiple regression analysis; II. Multiple regression analysis: explanation; III. Structural equation models; IV. Multivariate analysis; Appendix A: Matrix algebra, introduction; Appendix B: Tables.

Part I is devoted to theoretical foundations of multiple regression. Part II explores the relation between multiple regression analysis and analysis of variance. Also basic elements of categorical data models, continuous and categorical independent variables and logistic regression are explained. Part III concerns various structural equation models, indirect effects in structural equation models, path analysis, the use of LISREL and EQS. Part IV is devoted to the elements of multivariate methods: discriminant analysis, canonical correlation, multivariate analysis of variance, factor analysis. Newly included topics or substantially rewritten parts are: Regression diagnostics, logistic regression, multilevel analysis, research examples.

The book assumes knowledge on the level of a basic course in statistics, inluding elements of inferential statistics. The emphasis is on a non-mathematical approach, applications of suitable statistical methods for particular problems, interpretation of results, and decsription of the most often made erroneous decisions. The author discusses the assumptions under which a particular method can be applied and makes a large number of useful remarks and comments. He also shows on examples how to use the four widely used statistical packages BMDP, MINITAB, SAS and SPSS. The book can be used as a textbook, but it is certainly a useful source of information for those who have to analyse data from the area of behavioral sciences.

Reviewer: M.Huškova (Praha)

### MSC:

62P15 | Applications of statistics to psychology |

62-01 | Introductory exposition (textbooks, tutorial papers, etc.) pertaining to statistics |

62P25 | Applications of statistics to social sciences |

62H20 | Measures of association (correlation, canonical correlation, etc.) |

62H25 | Factor analysis and principal components; correspondence analysis |