Short abstract:

Surveys on social perception of science are a tool with several limitations. To address them, we present an innovative approach to analyse the social perception of science with a questionnaire developed according to Item Response Theory Models, based on the representational model of measurement.

Long abstract:

Public Understanding of Science studies are a branch of Science and Technology Studies that methodologically rely on surveys measuring the social perception of science. Several issues limit the information provided by these surveys. In particular, these surveys have been developed without regard to measurement theory and this makes it difficult to know what information the data actually provide. Furthermore, surveys in general have become such a ubiquitous tool for research, marketing and even business management that social reluctance to answer them is now widespread. In addition to the response bias generated by this reluctance, there is a generalized aversion to answering long questionnaires. We approached these difficulties from an innovative perspective with the aim of designing a questionnaire to analyse the social perception of science based on Item Response Theory (IRT) models. IRT models are supported by the representational model of measurement. This model defines measurement as the process of assigning numbers to objects based on rules, so that they reflect empirical relationships between objects. From this perspective, a questionnaire is a procedure for the objective and standardized measurement of a sample of behaviours to identify a latent trait and measuring would be an action comparable to finding the positions of people and items in a line. The mathematical expression of the IRT models ensures that people's response patterns are consistent with the ordering of the item locations along the variable and, therefore, measurement is independent of the items and the sample the data are obtained from.