A rapid review of research methodologies for remote research among young people insub-Saharan Africa

Working paper number: 463

Author: Yasmin Dunkley, Nontokozo Langwenya, Samantha Malunga, Jane Kelly, Elona Toska
Unit: AARHub


The COVID-19 pandemic represents a new opportunity for survey research teams to design, pilot and implement remote survey modalities. The Helping Empower Youth Brought Up in Adversity with their Babies and Young Children (HEY BABY) study planned to collect follow-up data in 2020 to follow up on 1,046 adolescent mothers and their children. While baseline data collection involved a combination of face-to-face interviews supported by audio computer-assisted self-interviewing (ACASI), follow-up data collection needed to happen remotely due to mobility restrictions and pandemic control regulations. To inform this methodological shift, a rigorous literature review was conducted in 2020. Despite a dearth of literature assessing remote survey modalities with young peoplein South Africa, this review–based on general concept papers, case studies across low-and middle-incomecountries and think-pieces–demonstrates that remote survey methodologies may be both feasible and acceptable in Africa. Hybrid models, based oncomputer assisted technology interviewing(CATI), withShort Message Service(SMS)follow up and higher frequency data collection,may beacceptable alternativesfor field-research compared to face-to-face models that are currently unavailable. Question topicswill need to be screened to ensure they do not distress participants. CATI models of remote data collection should average 30 minutes;however,frequent check-ins are acceptable in a South African context. Incentivisation and the use of different survey modes will encourage survey response uptake, but care must be taken to ensure the new remote data methodology questions are as alike as possible to the face-to-face baseline. Intensive tracking and subgroup weighting can address the effect of not including individuals without mobile phones in the study. Data analysis and results interpretation will have to factor in mode effects of a mixed modality research study.

Publication file: DunkleyLangwenyaMalungaKellyToska.pdf