Access

You are not currently logged in.

Access your personal account or get JSTOR access through your library or other institution:

login

Log in to your personal account or through your institution.

If you need an accessible version of this item please contact JSTOR User Support

On Stratification, Grouping and Matching

Dallas W. Anderson, Leslie Kish and Richard G. Cornell
Scandinavian Journal of Statistics
Vol. 7, No. 2 (1980), pp. 61-66
Stable URL: http://www.jstor.org/stable/4615774
Page Count: 6
  • Get Access
  • Read Online (Free)
  • Download ($12.00)
  • Cite this Item
If you need an accessible version of this item please contact JSTOR User Support
On Stratification, Grouping and Matching
Preview not available

Abstract

The effects of coarseness of classification are given a joint multivariate treatment for three contexts of application: proportionate stratified sampling, regression with grouped data, and matching for controls in data analysis. The link among the three contexts is found in the "relative explanatory power", $REP_{p}^{2}$, assuming a linear regression model between the predictand and the multivariate predictor variables. The high efficacy of a few, coarse classes for each predictor variable is favorable to multivariate designs in all three contexts, as shown in tables.

Page Thumbnails

  • Thumbnail: Page 
[61]
    [61]
  • Thumbnail: Page 
62
    62
  • Thumbnail: Page 
63
    63
  • Thumbnail: Page 
64
    64
  • Thumbnail: Page 
65
    65
  • Thumbnail: Page 
66
    66