Abstract
Surveys are widely used to measure attitudes, perceptions, experiences and other constructs that cannot be observed directly. However, developing a questionnaire that produces meaningful and interpretable data involves considerably more than writing a set of Likert-scale questions. This session will explore the process of designing and validating quantitative survey instruments, drawing on practical examples from research developing measures of students' attitudes towards astronomy and science, and experiences of inclusion in science engagement.

This webinar will consider how to move from defining a construct to developing appropriate survey items, including the use of existing literature and instruments, expert review, cognitive testing, and piloting. Key quantitative approaches to evaluating instruments will be reviewed, including item analysis, internal consistency and reliability, and exploratory and confirmatory factor analysis. The session will then discuss complimentary qualitative approaches to evaluating the appropriateness and effectiveness of survey instruments.

Rather than treating validation as a single statistical test, the session will emphasise validation as an ongoing process of accumulating evidence about whether an instrument supports the interpretations researchers wish to make from its scores. Participants will leave with a foundation for validating their own surveys, with practical examples, and an overview of common pitfalls and limitations.