Using mobile phone-based text message to recruit representative samples: Assessment of a cross-sectional survey about the COVID-19 vaccine hesitation

Highlights

• Using the mobile phone-based survey in Brazil, we achieved a representative sample of respondents.

• The Mobile phone-based survey is representative of all sociodemographic and health characteristics assessed.

• The Mobile phone-based survey is an essential tool during the Pandemic period for health purposes.

• The method of Mobile phone-based approach is a highly feasible and economical means of recruiting for survey research.

Abstract

Background

Limited research has examined mobile phone-based platforms for survey recruitment, especially during the COVID-19 pandemic in Brazil. Our objective was to investigate the feasibility and representativeness of mobile phone-based advertisement during a preliminary study about COVID-19 vaccine hesitation in Brazil. Moreover, we evaluate whether the older population can be reached through mobile phone-based platforms of the survey.

Methods

We conducted a study in December 2021 based on a preliminary survey about the COVID-19 vaccine hesitation in Assis, Brazil, Sao Paulo state. From a list of the adult population hesitant about the second dose of the COVID-19 vaccine, we sent a mobile phone-based advertisement inviting the participants to answer the survey for one week. The respondent's data were collected in a Google form platform. The comparison between the target population and the respondents was made using the Chi-squared test and the Welch's test, using a P-value of 0.05 as significative.

Results

The response rate was 9.99% after one week. The mean age of the respondent group was 33.97 (SD 14.99) and 35.05 (SD 14.19) of the population, with a P-value of 0.192 and a Cohen's d coefficient of 0.0754, corresponding to a small effect size between groups. We demonstrate that the mobile phone-based survey is a feasible and representative strategy during the pandemics in Brazil.

Moreover, the older population respondent was representative.

Conclusion

We achieved a representative sample of respondents using the mobile phone-based survey in Brazil. Furthermore, it was representative of all sociodemographic and health characteristics assessed. Finally, these findings suggest that the method is a highly feasible and economical means of recruiting for survey research.