Developing a Computational Ontology from Mixed-Methods Research: A Workflow and Its Challenges

Hanchard, M. and Merrington, P. (2019) Developing a Computational Ontology from Mixed-Methods Research: A Workflow and Its Challenges. In: Digital Humanities Congress 2018, Sheffield, UK, 6-8 Sep 2018,

Abstract

In this paper, we will discuss some of the challenges faced when developing a data ontology from mixed methods research. Beyond the Multiplex is a three year project that seeks to understand how to enable a wider range of audiences to participate in a more diverse film culture. A key part of the project is exploring regional patterns of film audience experience. In technical terms, this involves inductively generating an ontological data model to formally describe film audiences by drawing on primary mixed methods research. Ultimately, we are seeking to develop a relational understanding of film audiences which will culminate in a searchable triplestore database. This paper will cover the challenges of developing the data ontology. For example, how we maintained coherence across qualitative and quantitative datasets whilst incorporating taxonomies for controlling factors such as socio-cultural indicators, film type, and venue/platform. Overall, our data ontology draws on: a socio-cultural index of audience engagement with film; 200 semi-structured interviews, including 30 elite interviews with film industry and policy professionals; a longitudinal survey of 2000 respondents across three sample points; 16 focus groups using film elicitation to understand how audiences interpret specialised film and experience stories; and a discourse analysis of industry and policy documents. In combination, our mixed-methods approach generates various data types. Drawing on thematic qualitative analysis alongside cluster and latent class analyses, we are integrating our findings by drawing on literature at the intersection of audience studies and theories of cultural consumption. In this paper, we explain how social theory and data analysis combined have enabled us to iteratively generate a taxonomy. We then explain how that taxonomy has become the basis of both a triplestore database and allows an analysis of film audience engagement.

• Item Type: Conference Proceedings
• Status: Published
• Refereed: Yes
• Glasgow Author(s) Enlighten ID: Merrington, Dr Peter and Hanchard, Dr Matthew
• Authors: Hanchard, M., and Merrington, P.
• College/School: College of Social Sciences > School of Social and Political Sciences > Sociology Anthropology and Applied Social Sciences
• ISSN: 2050-7224
• Copyright Holders: Copyright © 2018 The Authors
• First Published: First published in Proceedings of the Digital Humanities Congress 2018. Studies in the Digital Humanities.
• Publisher Policy: Reproduced under a Creative Commons License