Related Experiment Videos
Reliability and validity of disability questions for US Census 2000.
E M Andresen1, C A Fitch, P M McLendon
1Department of Community Health, Saint Louis University School of Public Health, MO 63108, USA. andresen@slu.edu
American Journal of Public Health
|August 11, 2000
Summary
New 2000 US census disability questions showed low agreement between individuals with disabilities and their proxies. These census disability questions may not accurately reflect disability in America.
Area of Science:
- Public Health
- Sociology
- Health Services Research
Background:
- The 2000 US census introduced new questions to assess disability.
- Accurate disability data is crucial for policy and resource allocation.
Purpose of the Study:
- To evaluate the validity and proxy reliability of the 7 new disability questions from the 2000 US census.
- To compare responses with established instruments like BRFSS and ADL.
Main Methods:
- Interviewed 131 individuals with disabilities and their proxies in St. Louis, MO, and Massachusetts.
- Assessed concordance between self-reports and proxy-reports.
- Compared census question responses with the Behavioral Risk Factor Surveillance System (BRFSS) and Activities of Daily Living (ADL) instruments.
Main Results:
- Proxies reported higher levels of impairment compared to individuals with disabilities.
- Low agreement (kappa = 0.24-0.55) was observed between self- and proxy-reports.
- Moderate concordance was found between the census questions and their BRFSS and ADL counterparts.
Conclusions:
- The 2000 US census disability questions demonstrated limited reliability when answered by proxies.
- Findings suggest potential inaccuracies in the disability profile generated by the Census 2000 questions.
- Further research may be needed to refine disability measurement in large-scale surveys.