An Evaluation of the Validity and Reliability of Survey Response Data On Household Electricity Conservation

Abstract

In this article, we evaluate the reliability and validity of responses to a survey question inquiring as to whether or not households reduced electricity consumption during the energy crisis by directly comparing survey responses with actual kwh electricity consump tion data before and after second energy crisis of 1979. A response bias arises to the extent that when respondents are confronted directly by an interviewer, they are inclined to provide the interviewer with a socially acceptable response, namely that they reduced their consumption of electricity. Our analysis using data from a large sample of Hawaii households revealed that a bias toward overreporting electricity conservation indeed existed. However, the larger source of error stemmed from response unreliability, as indicated by the weak correlation between reported conservation behavior and actual electricity conservation.