A COMPARISON OF THREE METHODS OF ANALYZING ATTRIBUTE DATA IN A TWO-FACTOR CLASSIFICATION,

Abstract

The report describes a comparison of three methods of analyzing attribute data in a two-factor classification. Sets of proportions (where each proportion is based on a sample of ten) were developed within the constraints of various combinations of seven influencing factors. Each set of proportions was subjected to factorial analysis by the (0, 1) analysis of variance, factorial chi-square, and arcsine transformation of proportions methods. Using the percentage points of the F or chi-square distribution which coincide with each application of the three methods as response data, an analysis of factors influencing these percentage points was undertaken. The results of this analysis were used to conclude that the arcsine transformation of proportions method is the best of the three for use within the levels of the factors studied in this report. (Author)

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 1969
Accession Number
AD0691237

Entities

People

  • Donald L. Jerger

Tags

DTIC Thesaurus Topics

  • Analysis Of Variance
  • Classification
  • Computing-Related Activities
  • Data Science
  • Information Science
  • Interdisciplinary Science
  • Mathematical Analysis

Readers

  • Regression Analysis.