Family Planning for Strangers: An Experiment on the Validity of Reported Contraceptive Use

Area of Science:

• Demography
• Sociology
• Public Health

Background:

• Sterilization rates in the Dominican Republic appear unusually high compared to regional trends.
• The standard practice of using unfamiliar interviewers in surveys may introduce bias into family planning data.
• Previous family planning data may be skewed due to interviewer-respondent dynamics.

Purpose of the Study:

• To investigate the impact of interviewer familiarity on reported sterilization rates.
• To examine potential biases in family planning data collection methods.
• To assess the influence of respondent self-presentation on survey outcomes.

Main Methods:

• A field experiment was conducted in a Dominican town in 2010.
• Interviewer assignment was randomized based on the degree of prior familiarity between interviewer and respondent.
• Sterilization use and respondent self-presentation tendencies were analyzed in relation to interviewer familiarity.

Main Results:

• Reported sterilization use was significantly higher when interviewers were perceived as outsiders.
• Conversely, sterilization use was lower when interviewers were known to the respondent or from the same community.
• A positive correlation was observed between high sterilization use and respondents' tendency towards positive self-presentation.

Conclusions:

• The use of unfamiliar interviewers in demographic and health surveys may lead to biased reporting of sterilization.
• Findings challenge the routine reliance on strangers for data collection in family planning research.
• Methodological adjustments in survey design are needed to mitigate bias and ensure data validity.