Measurement invariance of the SF-12 across European-American, Latina, and African-American postpartum women
- 336 Downloads
The purpose of this study was to determine whether a postpartum-specific version of the SF-12 was invariant across three ethnic groups. Specifically, we examined the presence of differential item functioning (DIF) among European-American, Latina, and African-American mothers. DIF refers to differential endorsement of item responses that are not due to the construct being measured. DIF can result in biased group comparisons.
We analyzed cross-sectional data of postpartum women (n = 655) who delivered at an urban hospital in the northeast region of the USA. Multiple indicators multiple causes (MIMIC) model was used to examine differential item functioning.
The analyses revealed the presence of DIF for three items: Item 1 “self-assessed general health,” item 8 “bodily pain,” and item 9 “calm and peaceful.” Only two DIF effects were meaningful based on odds ratios and on the percentage of the total effect accounted for by the DIF effect. Specifically, African-American women differentially endorsed item 8 “bodily pain” when compared to European-American women (OR = 2.11, CI95 = 1.20, 3.71) and Latinas were more likely to endorse item 9 “calm and peaceful” when compared to European-American women (OR = 2.62, CI95 = 1.64, 4.17).
The results of this study indicate that the SF-12 is to a great degree an invariant measure for the assessment of HRQoL among postpartum ethnically diverse women. More research is needed to examine other aspects of invariance (e.g., configural and metric) and longitudinal invariance in ethnically diverse samples. To better understand ethnic differences in health, future studies need to examine the factors that may underlie DIF effects in quality of life.
KeywordsSF-12 Structural equation modeling (SEM) Multiple indicators multiple causes modeling (MIMIC) Differential item functioning (DIF) Postpartum Quality of life
Short-form health survey
Multiple indicator multiple causes
Differential item functioning
Health-related quality of life
Maternity outcomes project
Physical component summary
Mental health component summary
Structural equation modeling
Confirmatory factor analysis
Weighted least squares mean and variance adjusted estimator
Maximum likelihood estimator with robust standard errors estimator
Missing completely at random
Comparative fit index
Root mean square error of approximation
Weighted root mean square residual
This study was based on the first author’s master’s thesis. This research was supported by Agency for Healthcare Research and Quality (RO1 HS09698-3) and Robert Wood Johnson Foundation Grant (42680).
- 15.Sword, W., Watt, S., Krueger, P., Thabane, L., Landy, C. K., Farine, D., et al. (2009). The ontario mother and infant study (TOMIS) III: A multi-site cohort study of the impact of delivery method on health, service use, and costs of care in the first postpartum year. BMC Pregnancy and Childbirth, 9, 16.PubMedCrossRefGoogle Scholar
- 18.Gallo, J. J., Anthony, J. C., & Muthén, B. O. (1994). Age differences in the symptoms of depression: A latent trait analysis. Journal of Gerontology: Psychological Sciences, 49, 251–264.Google Scholar
- 19.Hagell, P., & Westergren, A. (2011). Measurement properties of the SF-12 health survey in Parkinson’s disease. Journal of Parkinson’s Disease, 1(2), 185–196.Google Scholar
- 20.Fleishman, J. A., & Lawrence, W. F. (2003). Demographic variation in SF-12 scores: True differences or differential item functioning? Medical Care, 41, 75–86.Google Scholar
- 21.Teresi, J. A., Ocepek-Welikson, K., Kleinman, M., Cook, K. F., Crane, P. K., Gibbons, L. E., et al. (2007). Evaluating measurement equivalence using the item response theory log-likelihood ratio (IRTLR) method to assess differential item functioning (DIF): Applications (with illustrations) to measures of physical functioning ability and general distress. Quality of Life Research, 16(Supplement 1), 43–68.PubMedCrossRefGoogle Scholar
- 28.Brown, T. A. (2006). Confirmatory factor analysis for applied research. New York: Guilford Press.Google Scholar
- 38.Little, R. J. A., & Rubin, D. B. (2002). Statistical analysis with missing data. Hoboken, NJ: Wiley.Google Scholar
- 39.Marsh, H. W., Hau, K.-, & Wen, Z. (2004). In search of golden rules: Comment on hypothesis-testing approaches to setting cutoff values for fit indexes and dangers in overgeneralizing hu and bentler’s (1999) findings. Structural Equation Modeling: A Multidisciplinary Journal, 11, 320–341.CrossRefGoogle Scholar
- 42.Muthén, B. O. (1998–2004). Mplus technical appendices. Los Angeles, CA: Muthén & Muthén.Google Scholar
- 43.Yu, C. Y. (2002). Evaluating cutoff criteria of model fit indices for latent variable models with binary and continuous outcomes. Unpublished doctoral dissertation.Google Scholar
- 44.Yang, F. M., Tommet, D., & Jones, R. N. (2009). Disparities in self-reported geriatric depressive symptoms due to sociodemographic differences: An extension of the bi-factor item response theory model for use in differential item functioning. Journal of Psychiatric Research, 43, 1025–1035.PubMedCrossRefGoogle Scholar
- 48.Leventhal, H., Halm, E., Horowitz, C., Leventhal, E., & Ozakinci, G. (2004). Living with chronic illness: A contextualized, self-regulation approach. In S. Sutton, A. Baum, & M. Johnston (Eds.), The sage handbook of health psychology (pp. 197–240). London: Sage.Google Scholar