The evolution of mobile-phone surveys in low- and middle-income countries

Abstract

Mobile-phone interviews are becoming an increasingly appealing mode of data collection in low- and middle-income countries. Therefore, we used data from 36 countries to examine mobile phone coverage of adults aged 15-49 years. We paid special attention to the (1) coverage rate (percentage of individuals who own at least one mobile phone) versus penetration rate (mobile-phone subscriptions per 100 inhabitants), and (2) proxy-covered population (i.e., persons who don’t own mobile phones but are potentially accessible through devices owned by other household members). Study findings suggest that (1) penetration rates notably overestimate mobile-phone coverage of the population; (2) mobile-phone coverage structure varies substantially; (3) mobile-phone owners are more likely to be males, urban residents, literate, married, and relatively wealthy; and (4) proxy-covered adults are distinct with regard to demographic characteristics and survey outcomes in comparison to directly covered and not-covered adults. This study informs practitioners about the current coverage structure in several low- and middle-income countries and promotes proper procedures for studying coverage structure before implementing mobile-phone interviews in those countries. Finally this study introduces the term proxy-coverage and advocates for approaching the proxy-covered population to reduce noncoverage bias, especially in settings where mobile-phone sharing among family members is common.