Release the Hypotheses! Tools & Techniques for Experimental Studies “in the Wild”

Background

Psychology and allied behavioural science disciplines increasingly want to explore the human world beyond the laboratory in order to improve the resolution, significance, validity and robustness of studies. As research ‘in the wild’ becomes more commonplace, we are seeing increasing reliance on an array of technologies and techniques to record, analyse and assign structure and meaning to data captured to study real-world human behaviour. While mobile devices, sensors and networks – the ‘Internet of Things’ (IoT) – are providing increased access to data collection, the ability to control or influence the human setting in which the data collection occurs remains underexplored. Furthermore, traditional approaches to pattern matching and intelligent algorithms are often designed in ways which ignore scientific hypotheses or experimental methods, with concerns arising around the validity and reliability of findings generated in processes that are often described as ‘dredging’ for statistical significances in massive datasets. Research in the Wild is therefore often characterised by qualitative or mixed-methods techniques which can better cope with the subjectivity of real-world settings, making them highly sensitive to context in comparison with traditional laboratory experiments.

Project summary

The community held a number of workshops to: create and consolidate connections across GW4 academics and scope the knowledge base of existing research; build an understanding of methods and epistemologies across the community, and to develop collaborations leading to outline proposals and further projects.