Mishima killed himself on Tuesday, November 24, 1970. In the experimental period of one week that followed, there were 117 motor vehicle fatalities. This week, then was compared to control weeks (taken from data generated between 1966 and 1973), that is weeks that did not experience any suicide. In sum, the number of motor vehicle deaths from the week of the Mishima case was significantly higher (117 compared to 98.88, or 9.12%).
Phillips extended his examination to the periods immediately before and after a suicide. Reasoning tht if a direct connection did exist between suicides and motor vehicle fatalities, then it would only be after the suicide hit the news that a motor vehicle suicide would occur. Therefore he studied data beginning two days before the suicide, the day of the suicide, and eleven days after the suicide. He then compared this data to his control years, focusing on exactly the same days. The pattern that he uncovered was rather surprising. Phillips found that there was only a slight difference between the week under consideration and the control week two days before the vent, and that the day of the event only a small increase followed. The same was true for the second day after the suicide, but then the third day brought with it an enormous increase in fatalities, some 31.29 % when compared to the control weeks. After the third day the rate of vehicular fatalities decreased precipitously and thus the key finding was that there was a “third day spike.” Further, the rate of subsequent fatalities was also linked to the amount of publicity the initial suicide garnered. If publicity of the event took place in one of the five largest of the California newspapers, then there was a 18.84 % increase in fatalities.
Phillips work was most revealing, but also would come under significant criticism in the years that followed.
To be continued….
David P. Phillips, “Motor Vehicle Fatalities Increase Just After Publicized Suicide Stories,” Science, (1977): 1464ff.
David P. Phillips, “Suicide, Motor Vehicle Fatalities, and the Mass Media: Evidence Toward a theory of Suggestion,” The American Journal of Sociology (1979): 1150ff.