NCJ Number
53251
Date Published
1978
Length
168 pages
Annotation
INTENDED FOR RESEARCHERS LACKING A THOROUGH KNOWLEDGE OF MATRIX ALGEBRA, THIS BOOK DISCUSSES FORMS FOR DATA, APPROPRIATE MULTIVARIATE ANALYTIC TECHNIQUES, THE USE OF THESE AND THE INTERPRETATION OF RESULTS OBTAINED.
Abstract
THIS BOOK INTRODUCES A SELECTION OF GENERALLY USEFUL MULTIVARIATE ANALYSIS TECHNIQUES. A LARGE PART OF THE BOOK DEALS WITH A GROUP OF MULTIVARIATE TECHNIQUES KNOWN GENERICALLY AS FACTOR ANALYSIS. FACTOR ANALYSIS IS CONSIDERED IMPORTANT IN HELPING A BEHAVIORAL SCIENCE RESEARCHER REPLACE A LARGE SET OF OBSERVABLE VARIABLES BY FEWER UNOBSERVED CONSTRUCTS, AND IT IS USED IN THE ANALYSIS OF DATA OBTAINED IN SPECIAL, RELATIVELY UNFAMILIAR WAYS. ANOTHER CENTRAL PURPOSE OF SOME MULTIVARIATE TECHNIQUES IS TO PROVIDE WAYS OF DISCRIMINATING BETWEEN SEVERAL GROUPS (OR SAMPLES) ON THE BASIS OF DIFFERENCES IN THEIR SCORES ON SEVERAL VARIABLES. ALTHOUGH THE WELL-KNOWN ANALYSIS OF VARIANCE CAN BE USED FOR THIS PURPOSE IN THE CASE WHERE ONLY ONE VARIABLE IS USED FOR DISTINGUISHING BETWEEN SEVERAL GROUPS, WHEN MORE THAN ONE VARIABLE IS USED, SPECIAL MULTIVARIATE TECHNIQUES ARE REQUIRED. THESE TECHNIQUES ALSO DEAL WITH THE PROBLEMS OF CLASSIFICATION, SUCH AS THE USE OF MEASURES ON SEVERAL VARIABLES AS A BASIS FOR DECISIONS ABOUT THE ALLOCATION OF SUBJECTS TO GROUPS. MOST OF THE TECHNIQUES IDEALLY REQUIRE VARIABLES TO BE MEASURABLE QUANTITATIVELY AND, IN SOME CASES, TO BE AT LEAST APPROXIMATELY NORMALLY DISTRIBUTED. MANY VARIABLES INVOLVED IN SOCIAL AND BEHAVORIAL SCIENCE, HOWEVER, ARE QUALITATIVE. A SELECTION OF USEFUL MULTIVARIATE TECHNIQUES APPLICABLE TO SUCH DATA IS DISCUSSED. THE RELATIONSHIPS BETWEEN THE VARIOUS TECHNIQUES, SOME OF WHICH CAN BE USED IN COMBINATION WITH OTHERS, ARE ILLUSTRATED. EXTENSIVE USE IS MADE OF NUMERICAL EXAMPLES.