Analysis Methods

ANALYSIS CORE
Our mission is to foster the improvements in the analysis of health related phenomena among older minority; encourage the development of methods and measures that better capture the health and determinants of health of diverse elders; promote collaboration between RCMAR sites on analysis issues; and disseminate new knowledge in this area.

Analysis Methods


Methods for Developing, Adapting, and Testing Measures for Minority Populations
Qualitative Methods in Minority Aging Research (UCLA, under development)
Quantitative Analytic Methods
  • Estimating Ordinal Reliability Using SAS The following paper on the SAS site provides a method for calculating reliability for ordinal data.
    (from the paper) The typical method of computing Cronbach’s alpha in SAS® is using the CORR procedure and the ALPHA option. This default method calculates alpha based on the Pearson correlation matrix (or standard covariance matrix) and has the underlying assumption that the data are continuous. When the data are ordinal this method may give biased results (Zumbo, Gadermann and Zeisser,2007). Gadderman, Guhn and Zumbo (2012) provide code to estimate reliability for ordinal data using R, however it was the desire of the author to be able to more easily automate reporting and do the whole analysis use SAS® solutions. In this paper we discuss and give SAS® code to estimate reliability for ordinal data. (updated 12/2016)
  • Strengthening Causal Inference in Nonrandomized Health Disparity Designs (2013)
  • Moderator and Mediator Analyses for Investigating Determinants of Health Disparities (WSU/UMich-MCUAAAR under development)
Use of Internet Panels to Conduct Health Disparities Research
Methodological Issues Conducting Interventions to Reduce Disparities
Methods of Conducting Meta-Analyses
RCMAR ANALYSIS CORE HOMEPAGES

CADC

CADC (University of California, San Francisco) Measurement Core Website Recommends and provides information on measures (many in English and Spanish) of several determinants of health and health disparities including socioeconomic status, interpersonal processes of care, perceived discrimination, and culture in the medical encounter. It additionally provides annotated bibliographies on methodological topics including recruitment and measurement issues in diverse populations, developing and/or translating surveys, and locating appropriate measures for use in health disparities research.

CHIME

CHIME (University of California, Los Angeles) (University of California, Los Angeles) Analysis Core Website has English and Spanish language survey instruments (e.g., SF-36, Kidney Disease Quality of Life, patient evaluations of health care), resource materials (e.g., presentations)about methods of assessment of older individuals, and links to analytic resources.

MCUAAAR

MCUAAAR (University of Michigan & Wayne State University) Measurement Core Website contains information about upcoming seminars and workshops. It additionally provides research training to junior investigators through partnerships and expert consultations on measurement and design issues.

NERC

NERC (University of Colorado) American Indian and Alaska Native Methods and Measurement Resources contain Quality of Life and Health-related Quality of Life Measures, Mental Health Measures for depression English and non-English, reviews and summaries of measures, racial and ethnic identity, religiosity, epidemiological, measure development issues, methods of scale construction and other references.