Using Confirmatory Factor Analysis (CFA) to Assess Structure of SF-36 (Persian Version) questions for normal people and patients with chronic diseases

AUTHORS

Shideh Rafaati 1 , Zahra Bagheri 2 , Peyman Jafari 3 , * , Nahid Soltani 4

1 PhD Student in Biostatistical Social Departerment in Health Promotion Research Center, Hormozgan University of Medical Sciences, Bandar Abbas, Iran.

2 Assistant Professor Department of Biostatistics, Shiraz Universiuty of Medical Sciences, Shiraz, Iran.

3 Associate Professor Department of Biostatistics, Shiraz Universiuty of Medical Sciences, Shiraz, Iran.

4 MSc of Educational Management, Hormozgan University of Medical Sciences, Bandar Abbas, Iran.

How to Cite: Rafaati S , Bagheri Z , Jafari P , Soltani N . Using Confirmatory Factor Analysis (CFA) to Assess Structure of SF-36 (Persian Version) questions for normal people and patients with chronic diseases, Hormozgan Med J. 2015 ; 19(2):e87442.

ARTICLE INFORMATION

Hormozgan Medical Journal: 19 (2); e87442
Published Online: November 26, 2014
Article Type: Research Article
Received: March 03, 2014
Accepted: November 26, 2014

Crossmark

CHEKING

READ FULL TEXT
Abstract

Introduction: This study aims at analyzing factor structure of Short Form Healthy
Survey (SF-36), Persian Version, for both normal people and patients with chronic
diseases.
Methods: In this cross-sectional study, a total of 585 patients with chronic diseases and
642 over 18-years old normal people were selected using simple casual sampling method.
When the survey was completed by the participants, the Cronbach's alpha coefficient and
confirmatory factor analysis (CFA) were used to determine the survey's validity and
reliability. T-test also was used to compare the average score of the survey's domains in
both groups.
Results: For the group of patients with the chronic diseases, the highest mean belonged
to the social functioning domain's score (43.11±18.48). In contrast, for the normal
people's group, vitality found the highest mean score (50.83±9.95). T-test results showed
that the mean difference in all domains except the role limitations because of social
impediments for both normal people and patients was significant. Cronbach's alpha
coefficient for all eight domains of the survey for the normal people varied from 0.68 to
0.90, which it indicated an acceptable reliability for the survey; whereas the coefficient for
the eight domains of the survey for patients varied from 0.66 to 0.91. CFA results using
goodness-of-fit indicators showed that the 8-factor structure, suggested by the developer,
is not verified for the patients group and the CFA results in the normal people's group
showed that the 8-factor model has an acceptable fitness for the data.
Conclusion: CFA is an effective method to analyze structure of different domains of SF-
36 in various groups. For this reason, researchers need to be wary when they use the
survey and also they need to check initially its reliability and validity in different subgroups.

Keywords

Factor Analysis - Quality of Life - Iran

© 2015, Hormozgan Medical Journal. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/) which permits copy and redistribute the material just in noncommercial usages, provided the original work is properly cited.

Fulltext

References

  • 1.

    Refrences are included in the PDF.

  • COMMENTS

    LEAVE A COMMENT HERE: