Abstract
INTRODUCTION: Generating unbiased and accurate assessments of sexual behavior among unmarried individuals is a major challenge, as it is governed by many social, moral, legal conventions and restrictions. These estimates are particularly susceptible to Fieldworker-induced social desirability bias. Accurate estimation of pre-marital sexual behavior is crucial for evaluating many aspects of sexual and reproductive health. The National Family Health Survey (NFHS) is the only reliable source of information on pre-marital sexual behavior in India, especially in the absence of any other comparative large-scale survey. This study provides us with an opportunity to understand the Fieldworker-level bias in the reporting of pre-marital sex among unmarried men and women. DATA AND METHODS: This study uses nationally representative data from the fifth round of the National Family Health Survey (NFHS-5). The analysis included 2950 female and 2073 male fieldworkers, as well as all unmarried women and men of reproductive age. Bayesian cross-classified multilevel modelling was performed to examine the Fieldworker's effect on the reporting of premarital sex among unmarried respondents. Additionally, Multivariate decomposition analysis was used to quantify the effect of various fieldworker characteristics in reporting pre-marital sex. RESULTS: The fieldworker effect accounted for around 31.7% of total variation in the reporting of intimate partner violence for females and 24.6% for IPV reporting among males. Results from the decomposition analysis indicated that age and residence of fieldworker were associated with a 2.4% increase and 1.3% decrease in the likelihood of reporting pre-marital sexual activity, respectively. After adjusting fieldworker and individual-level characteristics, an association between an age difference of greater than ten years and higher reporting of ever having had sexual intercourse was observed for both sexes. CONCLUSION: Although we believe that NFHS surveys are a potentially valuable source of information on pre-marital sexual behavior, especially in the absence of proven alternatives, the evidence from this study suggests that fieldworker effects may bias reporting of ever-had sexual intercourse. Therefore, estimates from these surveys should be evaluated carefully and interpreted in the context of other available information, given their wide use in policy formulation and monitoring in India.