When Does the Watchdog Bark? Conditions of Aggressive Questioning in Presidential News Conferences

In theories of the journalism-state relationship, the watchdog model of journalism competes with other models emphasizing either subservient or oppositional relations. Since actual journalistic practice is circumstantially variable, this study isolates the social conditions associated with aggressive journalism. Data are drawn from presidential news conferences from 1953 to 2000, and the focus is on the aggressiveness of the questions asked therein. Through multivariate models, four sets of explanatory conditions are explored: (1) the administration life cycle, (2) presidential popularity, (3) the state of the economy, and (4) foreign affairs. Results show (1) no evidence of a firstterm honeymoon period, but significantly more aggressive questions during second terms, (2) the president's Gallup job approval rating is not a significant independent predictor of aggressiveness, (3) both the unemployment rate and the prime interest rate are positively associated with aggressiveness, and (4) questions about foreign affairs are significantly less aggressive than questions about domestic affairs, and this differential has been stable for at least a half-century. We conclude by discussing the theoretical implications of these findings, which show that journalists modulate their conduct in complex ways that do not readily map onto any single model.