|
|
|
|
|
Andrew J. Houtenville, PhD and William A. Erickson, MS. Program on Employment and Disability, Cornell University, Rm 106, ILR Extension Building, Ithaca, NY 14853, (607)255-1540, wae1@cornell.edu
Preliminary analysis by the Census Bureau found a difference in the responses of people who mailed in the Census 2000 long form and those who were interviewed in person with regarded to the Go-Outside-Home Disability and Employment Disability questions. Approximately 25% of those who were mailed the long form did not return it and were subsequently interviewed (enumerated) by a Census employee (enumerator). When the long form was read aloud by the Census enumerator, questions 17c and 17d were separated by a page turn from the stem of question 17. The analysis suggests that people responded to the questions 17c and 17d as if they were asked whether they performed those activities, rather than if they had difficulty performing those activities. As a result, the estimated sizes of the Go-Outside-Home Disability and Employment Disability populations are upwardly biased, and this bias is systematically related to actually going outside the home and going to work. As a result of these errors, a portion of the Census 2000 (a major source of local-level disability statistics) has been eliminated. For instance, all statistics related to the poverty rate and educational attainment of people with disabilities are affected. The learning objectives of our presentation are to promote the understanding of (a) the Census Bureau’s analysis, (b) the portion of the Summary Files that are unaffected, (c) the abilities of the Census 2000 PUMS files and the American Community Survey to fill some of the void, and (d) other sources of local-level disability statistics.
Learning Objectives:
Keywords: Disability, Disability Policy
Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.