The Impact of Fictional Television Suicide Stories on U.S. Fatalities: A Replication

Abstract
In a recent issue of this Journal, Phillips reported evidence from a time-series analysis of U.S. fatalities in 1977 that fictional suicide stories on daytime television serials, "soap operas," trigger subsequent real-life suicides and single-vehicle motor vehicle fatalities. This paper calls attention to a serious mistake in Phillips's data which invalidates his results. In addition, the paper describes a more precise approach which produces no evidence linking soap opera suicide stories to subsequent real-life fatalities. The error in Phillips's paper stems from the fact that he used newspaper summaries as sources for the soap opera suicide stories and, in eight out of 13 cases, misspecified the date of the event. This mistake invalidates his before-after analysis strategy. Analyses presented here corect this error and disaggregate the time series to daily information. Several stories that Phillips overlooked are included.