BACKGROUND

In social surveys, the answering performance of elderly respondents is more susceptible to external environmental interference, which may compromise the quality of survey data and the validity of research findings as well as policy decisions. This issue becomes particularly significant in the context of an aging population, where research and policy decisions on aging-related issues heavily rely on the quality of survey data. Existing research indicates that survey environment can influence respondents' answering performance, and this impact is more pronounced in the elderly group. The reasons may include the cognitive decline of the elderly, their increased social adaptability and loneliness, and their relatively lower education levels.

OBJECTIVE

This study aims to analyze the impact of survey environment factors, particularly the number and identity of bystanders, on the answering performance of the elderly in social surveys. It seeks to provide insights into improving survey data quality and better serving the national strategy of actively addressing population aging.

METHODS

This study utilizes data from the 2018 China Family Panel Studies (CFPS), a large-scale, national, and multidisciplinary social survey project conducted by the Institute of Social Sciences Survey at Peking University. The analysis focuses on respondents aged 60 and above, with a final sample size of 7,139 elderly individuals after excluding proxy response samples, telephone interview samples, and interrupted response samples. The study employs ordinary least squares (OLS) regression models to examine the relationship between bystander variables and the answering performance of elderly respondents. The answering performance is measured by the degree of missing information and the severity of reported issues. Control variables include individual characteristics of the respondents (age, gender, education, etc.), province fixed effects, and interviewer fixed effects.

RESULTS

The results indicate that the number of bystanders has a significant negative impact on the degree of missing information in elderly respondents' answers and a positive impact on the severity of reported issues. Both effects exhibit a marginally diminishing trend as the number of bystanders increases. Specifically, the probability and extent of missing information in responses decrease with an increasing number of bystanders, while the severity of reported issues increases. Regarding individual differences, female, younger, and less-educated elderly individuals are more susceptible to the influence of bystander numbers. Furthermore, the identity of bystanders also plays a role. When family members or relatives are present, the degree of missing information in responses is lower. When other individuals are present, the reported severity of issues is higher. Additionally, the "audience effect" is found to be significant only in the elderly group and not in younger respondents.

CONCLUSIONS

The presence of bystanders in the survey environment has a dual impact on the questionnaire survey quality of elderly respondents. On one hand, it can reduce the degree of missing information in responses, thereby improving the completeness of data collection. On the other hand, it may lead to an overestimation of the severity of subjective evaluation responses, compromising the objectivity of the collected information. Therefore, researchers should be aware of the potential biases in survey data caused by bystander factors and consider the characteristics of the survey environment when conducting analyses. In survey design, it is important to collect information about the survey environment, especially when sensitive or private questions are involved, to minimize external interference. Additionally, enhancing the training of interviewers to improve their communication skills can help establish trust with respondents, reduce their vigilance and resistance, and improve the completeness and authenticity of information collection.

CONTRIBUTION

This study enriches the understanding of factors influencing survey data quality by focusing on the elderly group. It provides new insights into improving the organization of elderly social surveys and enhancing data quality. The findings offer a reference for future research on survey methodology and contribute to the effective implementation of the national strategy of actively addressing population aging.