scholarly journals Errors of Measurement: Regression Toward the Mean

2017 ◽  
Keyword(s):  
The Mean ◽  
Regression Toward The Mean

Reversion Toward the Mean Independently of Regression Toward the Mean

Methodology ◽  
2009 ◽  
Vol 5 (1) ◽  
pp. 3-6 ◽  
Author(s):  
Merton S. Krause
Keyword(s):  
Random Error ◽  
Evaluation Studies ◽  
Internal Validity ◽  
Intervention Evaluation ◽  
Population Mean ◽  
Experimental Intervention ◽  
Sample Data ◽  
The Mean ◽  
Regression Toward The Mean

There is another important artifactual contributor to the apparent improvement of persons subjected to an experimental intervention which may be mistaken for regression toward the mean. This is the phenomenon of random error and extreme selection, which does not at all involve the population regression of posttest on pretest scores but involves a quite different and independent reversion of subjects’ scores toward the population mean. These two independent threats to the internal validity of intervention evaluation studies, however, can be detected and differentiated on the sample data of such studies.

Comment on “Three Aspects of Cognitive Development”

1978 ◽  
Vol 46 (1) ◽  
pp. 129-130 ◽  
Author(s):  
William Asher
Keyword(s):  
Cognitive Development ◽  
Experimental Design ◽  
Age Groups ◽  
Quasi Experimental ◽  
The Mean ◽  
Regression Toward The Mean ◽  
Two Age Groups

In Exp. 2 of Siegler's 1976 study a faulty quasi-experimental design was used. The stated results, that older and younger children with apparently equal initial performance derived different benefits from identical experience, can also be explained by a confounding with regression toward the mean. These effects are a result of a selective matching of subjects from two age groups on a fallible variable correlated with age.

Markov Chains and Regression toward the Mean

1991 ◽  
Vol 26 (1) ◽  
pp. 115-125 ◽  
Author(s):  
Robert W. Kolb ◽  
Ricardo J. Rodriguez
Keyword(s):  
Markov Chains ◽  
The Mean ◽  
Regression Toward The Mean

Physical Attractiveness and Occupational Prestige among Male College Graduates

1980 ◽  
Vol 47 (3_suppl) ◽  
pp. 1275-1280 ◽  
Author(s):  
Jack Sparacino
Keyword(s):  
Socioeconomic Status ◽  
Physical Attractiveness ◽  
College Graduates ◽  
Archival Data ◽  
Occupational Prestige ◽  
Graduate Degree ◽  
Male College ◽  
The Mean ◽  
Regression Toward The Mean

A set of archival data was used to test the hypothesis that physical attractiveness is positively associated with socioeconomic status among men. Consistent with earlier research, more attractive men earned equivalent grades while undergraduates. They were also no more likely to have subsequently earned a graduate degree or to have held jobs characterized by higher status as indexed by Duncan (1961) prestige scores. Attractiveness itself was stable over the 25 yr. for those initially judged to be intermediate in attractiveness. Those who were initially most and least attractive were judged to have declined and increased, respectively, over the period in question, providing evidence for a regression toward the mean.

scholarly journals Intelligence, consistency, and Emerson's dilemma

2019 ◽  
Author(s):  
Kimmo Sorjonen ◽  
Tomas Hemmingsson ◽  
Bo Melin
Keyword(s):  
Reading Ability ◽  
Statistical Power ◽  
Positive Association ◽  
Response Style ◽  
Measurement Instruments ◽  
Negative Effect ◽  
The Mean ◽  
Regression Toward The Mean ◽  
Response Consistency ◽  
Large Representative

In a large representative sample of Swedish male conscripts (N = 49,246), we demonstrate a positive association between intelligence and response consistency on items measuring endorsement of a strong military defense. This association is accentuated, to some degree, by an apparent difficulty to handle reversed items among those with low intelligence. Consequently, due to regression toward the mean, reversed items in measurement instruments would tend to have a negative effect on statistical power when analyzing the association between whatever the instrument is measuring and intelligence. On the other hand, the presence of reversed items could mitigate bias in the measurement due to an acquiescent response style among those with low intelligence. Lack of motivation/carelessness and inadequate reading ability among some subjects are two possible confounders for the found association in the present study.

Toward a General Theory of Diary Panels

2009 ◽  
Vol 105 (3_suppl) ◽  
pp. 1131-1153 ◽  
Author(s):  
Rex S. Toh ◽  
Michael Y. Hu
Keyword(s):  
General Theory ◽  
Conceptual Framework ◽  
Item Nonresponse ◽  
Estimation Errors ◽  
Level Of Involvement ◽  
Record Keeping ◽  
Participation Level ◽  
Conditioning Effect ◽  
The Mean ◽  
Regression Toward The Mean

This article represents an integration of findings reported in seven articles on diary panels that the authors have published, based on the AT&T database. There are four major issues involved with diary panels, to wit: respondent noncooperation involving item nonresponse and attrition, bias leading to estimation errors, mathematical artifacts involving regression toward the mean, and the conditioning effect of being observed. An integrated conceptual framework for diaries is advanced, consisting of five independent variables (questionnaire design, length of participation, level of aggregation, duration of usage, and defining usage rate with length used), four mediating variables (degree of difficulty, participation fatigue, natural mortality, and level of involvement), and six dependent variables (item nonresponse, attrition, accuracy, regression toward the mean, the conditioning effect, and estimation regression). This attempt at a general theory of diary panels is admittedly incomplete, but is meant to serve as a useful conceptual framework for further research on longitudinal studies involving record keeping and reporting.

Sign in / Sign up

Export Citation Format

Share Document