PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 12034011

Fitting structural equation models using estimating equations: a model segregation approach.

Ke-Hai Yuan·Wai Chan
View original source ↗Export metadata ↓

Abstract

Problems such as improper solution, non-convergence, subsets of variables having different distribution, and latent variables with single indicators are common in the practice of structural equation modelling. In such cases, it may be feasible to fix some model parameters at prespecified values while concentrating on estimating some other parameters. This paper formulates such a model fitting process through a model segregation approach. The statistical properties of this procedure are studied using the theory of estimating equations and optimal estimating functions. The dependency of the new parameter estimates on those of the prespecified parameter estimates is characterized for several commonly used estimating equations. A rescaled model fit statistic is proposed. Examples illustrate various applications of this procedure.

Explore related subjects

Behavioral SciencesHumansModels, PsychologicalJournal ArticleResearch Support, Non-U.S. Gov't

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Ke-Hai Yuan, Wai Chan. 2002. Fitting structural equation models using estimating equations: a model segregation approach.. https://doi.org/10.1348/000711002159699

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2003

PhD faculty in clinical departments of U.S. medical schools, 1981-1999: their widening presence and roles in research.

PURPOSE: To analyze the growth, research activities, and academic status of PhD faculty in U. S. medical school clinical departments between 1981 and 1999. METHOD: Full-time U.S. medical school faculty who had PhDs and appointments in clinical departments in 1981 and 1999 and junior PhD faculty who became assistant professors between 1981 and 1990 were selected from the Faculty Roster System of the Association of American Medical Colleges. Their research activities and academic statuses were compared with those of MD or MD/PhD faculty in clinical departments or PhD faculty in basic science departments in the same years. RESULTS: The number of PhD faculty in clinical departments now exceeds the number of PhD faculty in basic science departments. PhD faculty in clinical departments come from diverse research backgrounds, contribute substantially to the research intensity of their institutions, and are more likely than their counterparts in basic science departments to become involved in research involving human participants or human tissues. PhD faculty in clinical departments are less likely than their counterparts in basic science departments, but are as likely as physicians in clinical departments, to be rewarded with academic promotion. They are less likely than their physician colleagues to be promoted in research-intensive departments such as departments of medicine and at top 20 research-intensive schools. CONCLUSIONS: The burgeoning career opportunities for PhD faculty in clinical departments should be reflected in the course work, mentorship, and potential thesis topics of PhD training programs. In lieu of tenure, research-intensive medical schools should develop alternative career tracks providing somewhat greater job stability for these faculty.

Behavioral Sciences↗
PubMed2003

Overcoming barriers to teaching the behavioral and social sciences to medical students.

Most U.S. medical schools offer courses in the behavioral and social sciences (BSS), but their implementation is frequently impeded by problems. First, medical students often fail to perceive the relevance of the BSS for clinical practice. Second, the BSS are vaguely defined and the multiplicity of the topics that they include creates confusion about teaching priorities. Third, there is a lack of qualified teachers, because physicians may have received little or no instruction in the BSS, while behavioral and social scientists lack experience in clinical medicine. The authors propose an approach that may be useful in overcoming these problems and in shaping a BSS curriculum according to the institutional values of various medical schools. This approach originates from insights gathered during their attempts to teach various BSS topics at four Israeli medical schools. They suggest that medical faculties (1) adopt an integrative approach to learning the biomedical, behavioral, and social sciences using Engel's "biopsychosocial model" as a link between the BSS and clinical practice, (2) define a hierarchy of learning objectives and assign the highest priority to acquisition of clinically relevant skills, and (3) develop clinical role models through teacher training programs. This approach emphasizes the clinical relevance of the BSS, defines learning priorities, and promotes cooperation between clinical faculty and behavioral scientists.

Behavioral Sciences↗
PubMed2002

Toward a racial abyss: eugenics, Wickliffe Draper, and the origins of The Pioneer Fund.

The Pioneer Fund was created in 1937 "to conduct or aid in conducting study and research into problems of heredity and eugenics . . . and problems of race betterment with special reference to the people of the United States." The Fund was endowed by Colonel Wickliffe Preston Draper, a New England textile heir, and perpetuates his legacy through an active program of grants, some of the more controversial in aid of research on racial group differences. Those presently associated with the Fund maintain that it has made a substantial contribution to the behavioral and social sciences, but insider accounts of Pioneer's history oversimplify its past and smooth over its more tendentious elements. This article examines the social context and intellectual background to Pioneer's origins, with a focus on Col. Draper himself, his concerns about racial degeneration, and his relation to the eugenics movement. In conclusion, it evaluates the official history of the fund.

Behavioral Sciences↗