ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Discriminating Between Response Scores In A Diagnostic Test A Dummy Variable Regression Approach

Journal: International Journal of Scientific & Technology Research (Vol.3, No. 2)

Publication Date:

Authors : ; ; ; ;

Page : 18-24

Keywords : Keyword Dummy Variable; Multiple Regression; Classification; Discrimination; Probability Density Function; Parameter; Estimate.;

Source : Downloadexternal Find it from : Google Scholarexternal


Abstract This paper proposes a statistical method for the analysis of diagnostic screening test results reported as quantitative scores using dummy variable multiple regression techniques. The proposed method develops estimates of the expected test scores for subjects whose tests scores are critically below the minimum normal score those subjects whose test scores are critically above the maximum normal score those subjects whose test scores are either marginally below the minimum normal score or marginally above the maximum normal score and for those subjects whose test results or scores are normal. Test statistics are also developed for testing the existence of any significant difference between the expected scores by these various groups of subjects. The proposed method which may enable health practitioners in statistically discriminating screened subjects for specific health management programs by identified risk groups relative to the normal group is illustrated with some data.

Last modified: 2015-06-28 03:51:43