An Analysis of Response Errors: A Case Study
Rikuma Ito · The Journal of Business · 1963
Not cached yet. This record was fetched live for this request and is stored after the lookup. Reopening the page should serve the cached copy.
The Journal of Business · University of Chicago Press · article · 1963
Abstract
CONSUMER surveys provide an important basis for economic analyses. With the increasing use of consumer surveys, the question of the accuracy of data collected by this method has become an important practical problem. The of sampling error enables one to measure or control sampling error in sample-derived estimates. However, this elaborate loses its practical importance once the basic assumptions of the are not realized in practice. For instance, with the presence of bias, the conventional formula of standard error of the mean fails to measure properly the reliability of an estimate.
Numerous sources of can be cited.' Because of the numerous sources of and lack of generality in the nature of biases from different sources, it is impossible to develop an acceptable theory of bias by abstract mathematics or logic. The problem of can be attacked only by extensive empirical studies. The importance of empirical research is confirmed by several investigations in the area of non-sampling error.2 This study uses data from a Federal Reserve Board study of consumer finances, in which respondents were asked, inter alia, to state the amount of their monthly payments and the gross and net amount of their loan on a new car.
The survey itself obtained other data on the respondents, and additional data were available on some interviewer characteristics and, from the lenders, on the true values of the loan and the monthly payment. Thus, it was possible to measure the errors in each respondent's reply and to test for statistical relationships between the errors and some characteristics of the respondent, the interviewer, and the interview itself.
Open the resolved locator · 10.1086/294635
Metadata licence: OpenAlex metadata (CC0)