ECPR

Install the app

Install this application on your home screen for quick and easy access when you’re on the go.

Just tap Share then “Add to Home Screen”

ECPR

Install the app

Install this application on your home screen for quick and easy access when you’re on the go.

Just tap Share then “Add to Home Screen”

How to Measure Interest Group Voice in the News Media? Addressing the Unitization Challenge

Interest Groups
Media
Methods
Lobbying
Darren Halpin
Australian National University
Max Grömping
Griffith University
Darren Halpin
Australian National University
Anne Sofie Cornelius Nielsen
Australian National University

Abstract

Appearing in the news media has long been considered an important avenue for interest groups to inform, shape or even frame the debate on issues that are of vital importance to their members and supporters. This has motivated a long-standing program of research which enumerates the scope of group voice in the media, leading to important findings of ‘bias’ in favour of economic interests, almost regardless of country context or time period. In this paper we address the challenge of unitization at the core of this quantitative research program: specifically, that it treats all instances of appearing in the news as equivalent. This, we argue, obscures important variation with potential consequences for relative and absolute assessments of group voice through the news media. The paper first documents varied practices of unitization within our field. Second, we propose a simple framework where our analytical attention is placed on the unit of group ‘statements’ – what linguists refer to as ‘reported speech’ – instead of simple appearances. Third, and taking Australian interest groups as a convenient context, we develop a new approach to programmatically extract ‘statements’ from a large corpus covering more than 2 million news articles from 2016 to 2024, and compare it to different empirical approaches to unitization already evident in the literature to assess implications for findings. This contributes to scholarship on interest groups and media agenda-setting, and to research on computational text analysis.