Who Did Not Respond to the Population Census of Japan? Analysis of Nonrespondents Using Online Survey Data

Abstract

Recent concerns have been voiced about the increases in nonresponses or “not reported” cases in Japan’s population censuses. The purpose of this study was to explore the factors related to nonresponse to obtain information that might contribute to correction, imputation, and/or interpretation of existing census data and to improve the quality of future censuses. Analysis of the individual-level data collected through online surveys found that nonrespondents were more likely than respondents to be young, unmarried, living alone, and short-term residents. Age was considered the most important factor related to nonresponse. Further, urban residents tended to be nonrespondents, privacy-consciousness was unexpectedly unrelated to nonresponse, and understanding the census was negatively associated with nonresponse regardless of age. Therefore, scholars should be aware that spurious geographical patterns and correlations might be observed when using population census data for regional analyses. Finally, we suggest that the government provide more information to the public regarding the usefulness and availability of the census data, regardless of age-group, before and after performing future censuses.