Is Accepting Proxies in the 2009 National Household Travel Survey Beneficial?

• Activity Number: 612
• Type: Topic Contributed
• Date/Time: Thursday, August 5, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Social Statistics Section
• Author(s): Promod Chandhok*+
• Companies: Bureau of Transportation Statistics
• Keywords: response error

Abstract

Proxy provides additional observations on the population at a lower marginal cost than would be obtained from additional self-reports. Accepting proxies reduces survey costs, but the additional proxy responses may have substantive response errors questioning the added utility of proxies. For example, in travel surveys it is seen that proxies report a lesser number of trips than self-respondents. By not accepting proxies, proxy response bias is absent but we end up with fewer observations, leading to a larger variance in estimates. There is thus a trade-off between higher variance and no proxy response bias and a lower variance with potentially large proxy response bias. This presentation reports results from the 2009 National Household Travel Survey where proxies were accepted from 18 years and older.