Citation

Sen A, Soot S, Yang L, Christopher E. Transp. Res. Rec. 1995; 1493: 170-177.

Copyright

(Copyright © 1995, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

unavailable

PMID

unavailable

Abstract

Because response rates vary by household type and by neighborhood, certain groups can be underrepresented. Factoring can rectify this situation somewhat for data used for descriptive purposes, but assumptions underlying many model estimation procedures are violated if factored data are used. Perhaps the only practical solution is to increase the sampling rate of underreporting groups. Because sampling rates should be proportional to reciprocals of response rates, a model to estimate response rates is presented. Such a model could be of value for implementing future surveys. A logit regression model was constructed with the demographic data as independent variables. Observations on the dependent variable, response rate, were obtained from a large-scale household travel survey conducted in the Chicago metropolitan area by the Chicago Area Transportation Study.

Record URL:
http://onlinepubs.trb.org/Onlinepubs/trr/1995/1493/1493-020.pdf

Language: en

Keywords

Mathematical models; Regression analysis; Optimization; Surveys; Sampling; Estimation; Motor transportation