Part 1:
1. Choose a paper describing the development or validation of a measure of relevance in health disparities research (please give the full citation and/or upload the paper if that's possible).
Schilling, L. S., Dixon, J. K., Knafl, K. A., Lynn, M. R., Murphy, K., Dumser, S., & Grey, M. (2009). A new self-report measure of self-management of type 1 diabetes for adolescents. Nursing Research, 58(4), 228-236. doi:10.1097/NNR.0b013e3181ac142a.
2. What was the definition of the construct?
Schilling et al. developed a measure to assess self-management of type 1 diabetes in adolescents (SMOD-A). The authors define this construct as “a multidimensional concept that includes activities that youth and their parents perform to care for the disease, as well as processes of collaboration between youth and their parents and between youth and health care providers as youth move toward the goal of assuming full responsibility for managing their diabetes” (p. 229).
3. How did the authors provide evidence on the validity of the measure? Could you think of additional approaches to validating the measure?
Schilling et al. first incorporated content validity into the process of developing the SMOD-A, using an initial qualitative study with adolescents with diabetes and their parents to inform item content and develop measure items. This was followed with expert review by “three panels of expert judges” (p. 230) to either approve, eliminate, or re-write measure items.
After these two phases, a field-study was used to test the psychometric properties of the SMOD-A. The authors measured construct validity, specifically convergent validity, by assessing the correlation of the five subscales of the SMOD-A with the Diabetes Self-Management Profile (DSMP; n = 16), the Self-Efficacy for Diabetes Scale (SEDS), the Diabetes Quality of Life for Youth Questionnaire (DQOL-Y), and the Self-Care Inventory (SCI). The SEDS, DQOL-Y, and SCI were all administered to the entire sample. The authors state that these measures were chosen for convergent validity because they were linked theoretically or empirically to self-management, or because they measured the same concepts. They also discuss in detail previously established psychometric properties of all the comparison measures, as well as the hypothesized direction of correlation between the measures and the SMOD-A subscales. The correlation of the five SMOD-A subscales with hemoglobin A1c was also assessed as a measure of criterion validity. Finally, the authors conducted exploratory factor analyses, separately for Part I and Part II of the SMOD-A.
Schilling et al. addressed content, criterion-related, and convergent validity, making this a thorough psychometric assessment of the SMOD-A. One limitation, as the authors note, is the sample that the SMOD-A was tested with: 80% white, from one geographic region (two study sites), and with “reasonably good metabolic control” (p. 10). This limits the study’s generalizability. The SMOD-A would need to be further tested among diverse populations before making conclusions about its broader generalizability and applicability.
4. How did the authors provide evidence on the reliability of the measure? Could you think of additional approaches to evaluating the reliability of the measure?
The researchers employed test-retest reliability (Pearson’s correlation), re-administering the SMOD-A to randomly selected participants from both study sites. Re-administration at 2 weeks (+/- 3 days; n = 74) was used to assess reliability, while re-administration at 3 months (+/- 1 week; n = 113) was used to assess the stability of self-management over time. Cronbach’s alpha was computed for subscales of the SMOD-A to test internal consistency.
5. Describe the implications of a lack of measurement validity or reliability for future research applications.
The validity of a measure is foundational to building a rigorous study: a perfectly designed study will be invalidated if the instruments used do not accurately measure the concept under study, or if they measure some other concept instead. Furthermore, because validity is compromised by systematic error, a measure that lacks validity may systematically over- or under-represent the concept under study and could lead to reporting of incorrect findings.
Part 2:
1. Find a paper describing a health disparity (please give the full citation or, even better, upload the paper so everyone else can look at it too)
Kim, H. J., Jen, S., & Fredriksen-Goldsen, K. I. (2017). Race/ethnicity and health-related quality of life among LGBT older adults. The Gerontologist, 57(suppl_1), S30-S39. doi:10.1093/geront/gnw172
2. Summarize the construct and measurement of the dimension of disparity (e.g., racial inequalities?, SES inequalities?) and the outcome measured (e.g., self-rated health).
This study assessed both health promoting and health risk factors related to racial/ethnic differences in self-rated health-related quality of life (HRQOL) between African American, Hispanic, and non-Hispanic White LGBT older adults. The health promoting and health risk factors addressed included lifetime LGBT-related discrimination and victimization, day-to-day discrimination, and socioeconomic, identity-related, spiritual, and social resources. Physical HRQOL components included vitality, mobility, pain, dependence on medical treatment, satisfaction with sleep, daily living activities, and work capacity, while psychological HRQOL components included positive and negative affect, body image acceptance, self-esteem, concentration, and personal beliefs.
3. What is the evidence for the validity and reliability of the measures?
The authors only report Cronbach’s alpha, separately for both physical HRQOL (a = .85) and psychological HRQOL (a = .85). Cronbach’s alpha is a measure of internal consistency reliability, or how consistent the results are for different items for the same construct within the instrument. The authors do cite the study that validated the measure of HRQOL that they used, but do not further discuss its psychometric properties. This would have been particularly important in this context, where a measure developed for use among the general U.S. population is applied within a particular subpopulation, namely, older LGBT individuals across racial/ethnic categories.
4. What is the reference category used for the disparity measure (ie, who is the comparison group)? Why does this reference category make sense (or not) for this research question?
Kim et al. use older LGBT Whites as the reference category for differences in HRQOL, compared to African American and Hispanic LGBT older adults. As the framework for this design, the authors use the Health Equity Promotion Model (HEPM), which addresses the intersectional nature of discriminatory and supportive experiences across axes of race/ethnicity for LGBT individuals.