The Moral Foundations Dictionary for News (MFD-N): A Crowd-Sourced Moral Foundations Dictionary for the Automated Analysis of News Corpora

Presented During:  Mass Communication Division - Top Papers
Sponsor: Mass Communication Division
Fri, 11/9: 8:00 AM  - 9:15 AM 
Salt Palace Convention Center 
Room: 255C (Level 2) 
The Moral Foundations Dictionary (MFD; Graham, Haidt, & Nosek, 2009) is a widely-used tool to automatically extract moral information from textual corpora. Yet, the MFD inherits certain limitations such as ad hoc pre-selection and overlappings of word stems that limit its validity. In this paper, we introduce a crowd-sourced approach for developing a moral foundation dictionary derived from a total of 65,012 highlights of 1,009 online newspaper articles highlighted by a U.S. representative sample of 557 human coders. We cross-validate various methodologies and parameters for developing a wordcount-based Moral Foundations Dictionary for News (MFD-N) that reliably classifies news documents according to their latent moral profile and shows satisfactory convergent validity toward related content features such as topics and sentiment. We discuss further refinements and tunings of our dictionary and point towards future research directions for its implementation.

Author

Frederic Hopp, University of California, Santa Barbara  - Contact Me

Co-Author(s)

Devin Cornell, University of California, Santa Barbara  - Contact Me
Jacob Fisher, University of California, Santa Barbara  - Contact Me
Richard Huskey, Ohio State University  - Contact Me
Rene Weber, University of California, Santa Barbara  - Contact Me