Measuring Differences among Non-Randomized Groups: An Epidemiological Model for Identifying Successful School Programs

Random assignment of students to innovative educational programs is not always possible in establishing proper and adequate control groups; therefore, it is recommended that the biomedical model of adjusted statistics designed to overcome the well-known difficulty educational investigators face when...

Full description

Saved in:
Bibliographic Details
Published in:The Journal of experimental education 1979-10, Vol.48 (1), p.50-59
Main Authors: Marascuilo, Leonard A., Dagenais, Fred
Format: Article
Language:English
Subjects:
Age
Critical values
Demography
Modeling
Mortality
Pretests
Proportions
Sample mean
Sample size
School districts
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Random assignment of students to innovative educational programs is not always possible in establishing proper and adequate control groups; therefore, it is recommended that the biomedical model of adjusted statistics designed to overcome the well-known difficulty educational investigators face when attempting to randomize subjects be adopted. The model's utility is demonstrated by using hypothetical data for fifth grade students tested by criterion and norm referenced tests. Results indicate that the adjusted discrepancies between group statistics are considerably smaller than is indicated by inspection of raw, or unadjusted, sample values. The educational analogue of the biomedical model of adjusted rates provides a way to obtain a more accurate estimate of program success or failure when comparisons across classrooms or other units are desirable.
ISSN:0022-0973
1940-0683
DOI:10.1080/00220973.1979.11011715