skip navigation

PUBLICATIONS

Register for Latest Research

Stay Informed
Register with NCJRS to receive NCJRS's biweekly e-newsletter JUSTINFO and additional periodic emails from NCJRS and the NCJRS federal sponsors that highlight the latest research published or sponsored by the Office of Justice Programs.

NCJRS Abstract

The document referenced below is part of the NCJRS Virtual Library collection. To conduct further searches of the collection, visit the Virtual Library. See the Obtain Documents page for direction on how to access resources online, via mail, through interlibrary loans, or in a local library.

 

NCJ Number: 68156 Find in a Library
Title: ANALYSIS OF NONMETRIC DATA - THE RELATION OF DUMMY DEPENDENT VARIABLE REGRESSION USING AN ADDITIVE-SATURATED GRIZZLE-STARMER-KOCH MODEL
Journal: SOCIOLOGICAL METHODS AND RESEARCH  Volume:8  Issue:4  Dated:(MAY 1980)  Pages:369-388
Author(s): M KUECHLER
Date Published: 1980
Page Count: 20
Type: Statistics
Format: Article
Language: English
Country: United States of America
Annotation: STATISTICAL PROPERTIES AND THE INTERPRETATION OF COEFFICIENTS DETERMINED BY USUAL REGRESSION PROCEDURES WHEN THE DEPENDENT VARIABLE IS A DICHOTOMY ARE EXPLORED AND COMPARED WITH A MORE REFINED APPROACH.
Abstract: DURING THE LAST 10 YEARS, A NUMBER OF NEW TECHNIQUES FOR THE ANALYSIS OF NONMETRIC DATA HAVE BEEN DEVELOPED THAT ARE EQUIVALENT TO CLASSIC METRIC APPROACHES IN STATISTICAL RIGOR BUT ARE BASED ON ASSUMPTIONS THAT WILL GENERALLY BE MET BY MOST DATA THAT ARISE IN SURVEYS. LITTLE DISAGREEMENT EXISTS AMONG METHODOLOGISTS THAT ORDINARY REGRESSION SHOULD NOT BE USED WITH DUMMY DEPENDENT VARIABLES. WHEN THE ASSUMPTIONS OF HOMOSCEDASTICITY AND NORMAL DISTRIBUTION OF THE DEPENDENT VARIABLE ARE NOT MET, INVALID INFERENCES OFTEN RESULT; NEVERTHELESS, THIS APPROACH IS STILL VERY POPULAR WITH RESEARCHERS MORE INTERESTED IN THE SUBSTANTIVE ISSUES THAN THE METHODOLOGY. EXAMINING DIFFERENCES AND SIMILARITIES BETWEEN DUMMY DEPENDENT VARIABLE REGRESSION AND THE NEW TECHNIQUES IS AN URGENT TASK AIMED AT IMPROVING THE DATA ANALYSES DONE BY THE SUBSTANTIVELY ORIENTED RESEARCHER. IN THIS ARTICLE, ANOTHER TECHNIQUE IS DISCUSSED: A SIMPLE VERSION OF THE WEIGHTED LEAST SQUARES APPROACH DEVELOPED BY GRIZZLE, STARMER, AND KOCH (GSK). THIS SIMPLE VERSION OF THE GSK APPROACH USES A SATURATED MODEL WHERE ALL INTERACTIONS, AS WELL AS DIRECT EFFECTS, ARE POSTED. THE NEW TECHNIQUE IS COMPARED TO THE USUAL DUMMY REGRESSION APPROACH. MATHEMATICAL EQUATIONS AND AN ILLUSTRATIVE APPLICATION OF THE TECHNIQUE ARE PROVIDED, ALONG WITH NOTES.
Index Term(s): Evaluation; Mathematical modeling; Sociology; Statistical analysis
To cite this abstract, use the following link:
http://www.ncjrs.gov/App/publications/abstract.aspx?ID=68156

*A link to the full-text document is provided whenever possible. For documents not available online, a link to the publisher's website is provided. Tell us how you use the NCJRS Library and Abstracts Database - send us your feedback.