Document Detail


Model modifications in covariance structure analysis: the problem of capitalization on chance.
MedLine Citation:
PMID:  16250105     Owner:  NLM     Status:  MEDLINE    
Abstract/OtherAbstract:
In applications of covariance structure modeling in which an initial model does not fit sample data well, it has become common practice to modify that model to improve its fit. Because this process is data driven, it is inherently susceptible to capitalization on chance characteristics of the data, thus raising the question of whether model modifications generalize to other samples or to the population. This issue is discussed in detail and is explored empirically through sampling studies using 2 large sets of data. Results demonstrate that over repeated samples, model modifications may be very inconsistent and cross-validation results may behave erratically. These findings lead to skepticism about generalizability of models resulting from data-driven modifications of an initial model. The use of alternative a priori models is recommended as a preferred strategy.
Authors:
R C MacCallum; M Roznowski; L B Necowitz
Related Documents :
721365 - The estimation of disease frequency using a population sample.
7272415 - Matching in epidemiologic studies: validity and efficiency considerations.
20216775 - Estimation of particle size distribution parameters with forward-scattering techniques.
7723655 - The effect of trial size and variability on statistical power.
18208605 - Signs: a parallelized, open-source, freely available, web-based tool for gene selection...
22249595 - Improved surface laplacian estimates of cortical potential using realistic models of he...
Publication Detail:
Type:  Journal Article    
Journal Detail:
Title:  Psychological bulletin     Volume:  111     ISSN:  0033-2909     ISO Abbreviation:  Psychol Bull     Publication Date:  1992 May 
Date Detail:
Created Date:  2005-10-26     Completed Date:  2005-10-27     Revised Date:  -    
Medline Journal Info:
Nlm Unique ID:  0376473     Medline TA:  Psychol Bull     Country:  United States    
Other Details:
Languages:  eng     Pagination:  490-504     Citation Subset:  IM    
Affiliation:
Ohio State Univeristy, USA.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:
Adolescent
Algorithms
Data Interpretation, Statistical*
Empirical Research
Factor Analysis, Statistical
Humans
Intelligence Tests / statistics & numerical data
Job Satisfaction
Models, Psychological*
Models, Statistical*
Questionnaires
Reproducibility of Results
Sample Size
Sampling Studies
Software

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  Effect of Chinese herbal medicine on patients with primary hepatic carcinoma in III stage during per...
Next Document:  Reflections on public bioethics: a view from the trenches.