Are Households with Interrupted Phone Service Like Those with No Telephone Service?: Comparison Using Public-Use Microdata Set and Nationwide Personal Transportation Survey

Abstract

Research suggests that households with interrupted telephone service can be used as a proxy for households with no telephone service to help mitigate the inherent frame bias in a random digit-dialing (RDD) sample. If a sufficient quantity of households with a history of interrupted service can be surveyed, some researchers suggest that probability weights can be calculated to account for the households not covered by the telephone survey. For the weighting of households with interrupted telephone service to serve the purpose of representing households without any service, these two groups should be similar. The analysis presented compares the demographic and travel characteristics of people with a history of interrupted telephone service [from the 1995 Nationwide Personal Transportation Survey (NPTS)] with those who are tabulated as having no telephone service in the 1990 census. This research shows that there are significant differences between the two groups—for example, people who report interrupted telephone service are more likely to own their home, have more workers in the household, and have more vehicles available than are people in households without telephone service. And although both groups are poor, the people with interrupted service are slightly less poor than those with no service at all. For the NPTS, and perhaps other surveys, the large significant differences between these two groups suggests that simply adjusting the weights for the households with interrupted telephone service who complete the survey may not be a sufficient strategy to mitigate the noncoverage of households without telephone service for a nationwide RDD survey.