Bad News Travels Fastest: A Computational Approach to Predictors of Immediacy in Digital Journalism Ecosystems

Abstract

This paper studies the prevalence of immediacy in digital journalism from an ecosystem perspective. It combines insights into norms and routines in digital newswork with an analytical approach from news diffusion research and the power of computational methods to track story-based news flows at high granularity of time intervals in comprehensive media samples. We ask how attributes of news stories and situational preconditions of their production help explain whether the variety of newsrooms in a digital journalism ecosystem will converge on immediate story coverage so that wide-range bursts will emerge at the start of news diffusion processes. Based on the reconstruction of 95 news diffusion processes among 28 professional online news sites in Germany, we pool first reports from diffusion processes with the same attribute and compare the dynamics of the accumulation of issued first reports by attributes in event history analyses. We find that most news factors made no difference in a recurring pattern of basically fast diffusion dynamics. Only negative news and stories involving prominent personalities further accelerated diffusion processes and spread even faster. In contrast, events characterized by wide reach beyond small groups and the production break of many newsrooms during the night slowed down digital diffusion.