The unfolding COVID-19 pandemic is resulting in unforeseen social and economic disruption, globally and locally. The rapid and unprecedented nature of the crisis requires the collection of contemporaneous research data. The disruptive nature of the pandemic produces practical challenges and methodological dilemmas for collecting suitable social science research data. In this chapter, we outline the problems, issues and opportunities associated with collecting research data using social surveys in this time of crisis.
There is a long history of using social surveys in social science research, dating back to Seebohm Rowntree's studies of York and Charles Booth's surveys of life and labour in London (Linsley and Linsley, 1993). The social survey is best considered as a methodological approach, rather than a single technique. Historically, questionnaires have been the main data collection instrument used in social surveys, but interviews are frequently used to collect data. Increasingly, a mixture of different modes of data collection are used, especially involving computers and new technologies (see De Leeuw, 2005).
Social surveys are designed to study statistical populations. Statistical populations are aggregates of specific entities or cases. A statistical population commonly contains too many cases to expediently study, and it is therefore more practicable to draw a sample (that is, a subset) of cases. Integral to the survey method is the collection and analysis of data from a sample of a larger statistical population. The social survey method is attractive because collecting data from a sample is more practical, and sample data have statistically efficient properties.
A social survey is a methodology that generates a matrix of research data. The usual components of the matrix are cases and variables. Variables are the result of collecting systematic measurements, and cases are the entities under investigation. Cases will commonly be individuals, but they could be households, families, businesses, schools, and so on. Integral to the survey method is the organized collection of systematic measures across a set of cases, to provide comparable data.
A fundamental aspect of constructing a survey sample from a statistical population is how well it represents, or reflects, the aspect of the statistical population under investigation.