ESRA 2027 sessions by theme

Back to Overview of Sessions

Session Details

Measuring ideas accurately is the primary goal for survey research. A number of biases have been identified as reducing the accuracy of survey results, with biases typically divided into representation and measurement errors. Within the area of measurement errors, many biases involve cognitive processes related to how participants store and recall information or how they understand and react to survey tasks (some examples: primacy bias, set size bias, and telescoping effects).

However, other biases may result from social cues in the survey-taking experience itself. Social psychology is a field within psychology that studies how the real or imagined presence of others shapes our attitudes, intentions, and beliefs. The explicit presence of another person in the survey situation appears to alter a person’s response (as occurs with interviewer effects in in-person or telephone interviews), leading to socially-based biases such as socially desirable responding, positivity bias, and acquiescence bias. However, a question remains: could the mere thought that someone might see your responses and judge you (an ‘imagined’ presence) be enough to bias responses and increase inaccuracy? Most online surveys are typically completed anonymously and are self-administered, so the possibility of social biases would appear to be low.
Our session focuses on obtaining papers that discuss socially-based biases in online surveys and that can address three questions: First, can the biases be demonstrated? Are they social in nature? Do the biases affect the accuracy of survey response? Accuracy can be assessed by proximity to known benchmarks or degree of relatedness with behavior (showing that biases actually reduce predictive validity). Such papers can use cross-modal comparisons (with or without interviewers) or single-mode studies (e.g., online only).