A Step Toward Formalising Visual Data Analysis Practices in Human Computer Interaction

Abstract Visual data is an integral part of qualitative research, yet it is not always clear to researchers how to use or interpret it once gathered. Existing methods for qualitative data analysis largely rely on textual approaches such as thematic analysis, or grounded theory. Open coding is a term that is frequently used to describe a analysis that follows a logical—yet undocumented—process, but these can often be applied ad hoc, and are necessarily created after the fact of data collection. Researchers often develop their own ways of interrogating visual data, but can struggle to prove the legitimacy of not employing an existing approach. This article outlines a proven and replicable process for the analysis of static visual imagery that has been developed by the author and utilised together with research collaborators over a seven year period. The approach, which I call Annotated Visual Analysis (AVA), can be used for sketched or artistic data from participants or in first person, subjective research, in order to generate insight, gather requirements or confirm hypotheses. Here, I outline five case studies using the approach in differing contexts, in order to demonstrate its applicability, and outline the process and guidelines so that other researchers might employ the method. I also discuss the need for other researchers using visual methods to share and consolidate their forms of analysis to develop knowledge for use by others in the field.