Measuring Norms and Enumerator Effects: Survey Method Matters
通过比较标准面对面访谈和人类辅助自填问卷两种调查方法，发现方法显著影响回答，尤其在性别规范等敏感问题上，面对面访谈更易受社会期望偏差影响。
Abstract This paper examines how survey methods influence reported outcomes by comparing two survey methods: the standard face-to-face interview and an alternative method named the Human-Assisted Self-Administered (HASA) survey. In HASA, respondents are guided to self-answer on a tablet, while enumerators read the questions aloud and offer technical support. This approach provides continuous enumerator support while maintaining a higher level of confidentiality compared to face-to-face interviews. Leveraging a randomized controlled trial in Benin, with random assignment of the survey mode, the analysis reveals that survey methods significantly influence responses, especially for questions sensitive to enumerator presence. These differences are most salient for items related to gender norms and women’s agency. Respondents interviewed through HASA report fewer gender-equal attitudes and behaviors. Investigating the mechanisms, the findings suggest that face-to-face interviews may be more prone to social desirability bias.