FireDOC Search

Author
McDaniel, M. A. | Bickman, L. | Edelman, P. | Herz, E.
Title
Exploration of Non-Sampling Error in Fire Incident Surveys. Final Report.
Coporate
Loyola Univ., Chicago, IL
Sponsor
National Bureau of Standards, Gaithersburg, MD Department of Health, Education and Welfare, Washington, DC
Report
NBS GCR 78-135, December 1977, 24 p.
Distribution
AVAILABLE FROM National Technical Information Service (NTIS), Technology Administration, U.S. Department of Commerce, Springfield, VA 22161. Telephone: 1-800-553-6847 or 703-605-6000; Fax: 703-605-6900. Website: http://www.ntis.gov
Contract
GRANT-69015
Keywords
fire departments | fire incidence | fire reporting | fire fatalities | human behavior | sampling | fire statistics | surveys
Abstract
Non-sampling error in fire incident surveys is a potential problem for fire and human behavior research. The present report was prepared in an effort to examine fire victims' reporting biases concerning fire incidents. The telephones numbers of household fire victims in a midwestern city, during the months of September 1975 through August 1976, were obtained from the fire department. Also a sample of telephone numbers in the same area was radomly generated. All eligible and cooperative respondents were asked a series of questions concerning whether they experienced a fire during the target interval and, if they had a fire, they were questioned about it. Approximately one-half of the respondents who, according to fire department records, had a fire failed to report it to the interviewer. Results indicated that respondents who had more serious fires, reported the incident to the interviewer significantly more often than respondents who had less serious fires. Also fires which occurred later in the 12 month reference period (i.e., closer to the time of the interview) were more often reported to the interviewer than those which occurred earlier in the reference period. Several analyses indicated that, in most cases, the fire reporters' responses agreed with fire department records concerning the characteristics of the fire incident; however, total agreement was seldom found. In addition, it was found that fire reporters tend to be similar to fire-nonreporters in terms of demographic characteristics and most survey responses although some statistically significant differences were obtained.