We use a threshold model to drive our simulated network analysis testing public support for candidates in invisible primaries. We assign voter thresholds for candidates and vary number of voters, attachment to candidates and decay. Results of the algorithm show effects of size of lead, an unwavering base of support, and information loss.
Keywords: Network analysis; Threshold model, Invisible Primary, Campaigns and Elections, Presidency
Elizabeth A. Stiles, John Carroll University
Colin D. Swearingen, John Carroll University
Linda Seiter, John Carroll University
Brendan Foreman, KeyBank