Big data meets survey science

Abstract

This article is part of the SSCR special issue on Big Data and Survey Science, guest edited by Adam Eck (Oberlin College), Ana Lucía Córdova Cazar (Universidad San Francisco de Quito), Mario Callegaro (Google Ltd.), and Paul Biemer (RTI International & UNC-CH).Surveys have long been the primary means of data collection about peoples’ attitudes, beliefs, and opinions and are useful for measuring specific characteristics of individuals as well as understanding public opinions and creating accurate and precise official statistics. Recently, artifacts of our increasingly digital lives have offered additional, broader information about our behaviors (e.g., purchase histories, personal interests captured through Internet browsing) in the form of “big data.” Big data and surveys have great potential to complement one another (Baker, 2017; Callegaro & Yang, 2018) to allow scientists to better understand people and the world in which we live—for example, by combining the low cost per data point of big data (offsetting the rising costs of survey-based data collection) with the ability to collect very specific information addressing research questions using survey data.