Gatekeeping in telephone surveys of female migrants in India
India has seen a large increase in the use of cell phones, with more than 90% of households having a mobile phone. The expanded growth of cell phones offers an alternative to collecting data in-person, and therefore allows social researchers to efficiently study the country’s large and diverse population. While this method of collecting data has been beneficial in efficiency, the sociocultural and economic environment in India presents challenges in implementing it. Cellphones act like a landline in many households, with several household members sharing a single phone number. Additionally, access to cellphones is gendered, as 50% of the adult male population has exclusive use of a cell phone, with the percentage dropping to 40% for women. Cellphones have been found to be a “risk to women’s reputation” due to fears of digital harassment combined with traditional gendered norms and roles. Due to this, the recruitment of women respondents for telephone surveys can be difficult. The “gatekeeping” of women responding to such interviews can present issues such as:
- Reducing response rates for surveys of women by reducing their participation
- Non-response bias can arise due to systematic differences between respondents and non-respondents
- Gatekeepers might refuse access to the woman respondent, but may offer to provide responses on their behalf, which can result in biased estimates.
In this paper, Faculty Associates Sharan Sharma (right) and Sonalde Desai, along with their colleagues, estimate the prevalence of gatekeeping, use call record data and examine different types of gatekeeping and their association with non-response and proxy interviews, and they then model the likelihood of the occurrence of a proxy interview as a function of household, respondent, and interviewer-level variables.
Desai and Sharma, along with their colleagues, found that in the India Human Development Survey (IHDS), gatekeeping arises in two stages, with the first one coming at the Computer Assisted Personal Interviewing (CAPI) stage when the root household does not share the migrant’s telephone number but instead that of the gatekeeper. This propagates through the process when the Computer-Assisted Telephone Interviewing (CATI) interviewer tries the telephone number and the gatekeeper either does not cooperate, or they insist on a proxy interview.
Overall, the team found that gatekeeping was at a prevalence estimate between 56% and 65%. When the gatekeeper is a husband or head of household, making contact with women is easier and the non-response rate is six percentage points lower than if the migrant could be directly contacted. Future research can look at if there are regional differences in gatekeeping prevalences and if survey responses differ between proxy interviews and interviews with the selected respondent. Additionally, more research could be done to uncover mechanisms governing gatekeeping, especially when looking at marital status.
Sharma, Sharan, Bijay Chouhan, O.P. Sharma, Jaya Koti, and Sonalde Desai. (2025). “Can I Interview Her? Gatekeeping in a Telephone Survey of Female Migrants in India.” Survey Practice 19 Special Issue (March). https://doi.org/10.29115/SP-2024-0029.
