ScholarWorks@중앙대학교

초록

As transitions in employment among public officials in South Korea emerge as a notable social concern, research on their turnover intentions is actively underway. Nevertheless, prior studies investigating the factors relating to public officials' turnover intentions have yielded varying and conflicting results despite utilizing similar datasets. This study seeks to illustrate that such inconsistencies may arise from biases in survey responses, particularly the inclination of public officials to provide socially desirable answers. To tackle this issue, we conducted an experiment where a sample of public officials was randomly divided into two groups: a treatment group utilizing the Bayesian Truth Serum to promote candid and thoughtful responses, and a control group employing conventional survey methods. The comparative examination of the survey data obtained from both groups yielded the following outcomes. At first, public officials displayed a tendency to exaggerate responses concerning job satisfaction and other socially desirable job aspects when lacking incentives for honest feedback. Secondly, the exaggeration of job-related factors resulted in either overestimation or underestimation of their correlation with turnover intentions. Lastly, the impact of social desirability was more pronounced among higher-ranking public officials (grade 5 and above) in comparison to those in lower ranks (grade 6 and below). These results indicate that social desirability bias significantly skews public official surveys, underscoring the necessity of employing appropriate research methodologies to mitigate such biases in surveys.

키워드