Do men and women respond differently to negative political communication? Only a limited collection of studies into the effects of negative campaigns have investigated this research question, and the conflicting results produced from such studies have prevented the development of a widely accepted answer. As campaigns transition to new media environments, further problems arise, as any potential gender gap may be magnified on the new political communication battlefield of social media. The present article contributes to this sparsely investigated area through an empirical study of men's and women's reactions on Facebook to US presidential candidate attacks during two general election campaigns (2012 and 2016) and two primaries (2016 Democratic and Republican). Across nearly 400 million reactions and 40 million unique users, women demonstrate lower receptivity to candidate attacks than men. Two potential explanatory factors for the gap are examined, but neither fully captures the magnitude of the differences observed. Conceptualizing the gender gap composition in terms of differential receptivity most accurately explains these findings and potentially resolves the competing explanations for the gap within the existing literature.

Beginning with the 1972 presidential election and for each election thereafter Harvard University's Institute of Politics in the John F. Kennedy School of Government has held a post-election symposium where all of the campaign managers, pollsters, political consultants and media advisors for all of the primary and general election candidates come together with leading journalists, electronic and print, and political commentators and pundits to discuss and dialogue about what occurred during the election among the candidates, nominees, and the winner and losers. The symposia have allowed campaign managers to describe what happened and forecast for the forthcoming presidential election. After the multi-day symposium a book length transcript is published. In 1992 the Annenberg School of Communication at the University of Pennsylvania launched its own symposia where the campaign managers of the Democratic and Republican nominees in the general election are invited to discuss what happened and forecast for the next presidential election. The ten books in the Harvard series and the three books in the Pennsylvania series are used as data sources in this article to determine if any of the campaign managers forecasted and/or predicted an African American presidential candidate, even when such candidates had appeared in previous years. While our findings uncover some of the interests and concerns of presidential campaign managers and advisors since 1972, the overriding focus of the symposia has been on the nature, scope, and significance of the African American electorate.