In the typical field-experiment, the investigator presents television programs in the normal viewing setting and observes behaviour where it naturally occurs. The investigator controls the television diet either by arranging a special series of programs or by choosing towns that in the natural course of events receive different television programs.
One early field-experiment was a study conducted by Stein and Friedrich (1972) for the Surgeon General's project. These investigators presented 97 pre-school children with a diet of either 'antisocial' 'prosocial', or 'neutral' television programs during a four-week viewing period. The antisocial diet consisted of twelve half-hour episodes of Batman and Superman cartoons. The prosocial diet was composed of twelve episodes of Mister Roger's Neighbourhood (a program that stresses such themes as sharing possessions and co-operative play). The neutral diet consisted of children's programming which was neither violent nor prosocial. The children were observed through a nine-week period, which consisted of three weeks of pre-viewing baseline, four weeks of television exposure, and two weeks of post-viewing follow-up. All observations were conducted in a naturalistic setting while the children were engaged in daily school activities. The observers recorded various forms of behaviour that could be regarded as prosocial (i.e. helping, sharing, co-operative play) or antisocial (i.e. pushing, arguing, breaking toys). The overall results indicated that children who were judged to be initially somewhat aggressive became significantly more so as a result of viewing the Batman and Superman cartoons. Moreover, the children who had viewed the prosocial diet of Mister Roger's Neighbourhood were less aggressive, more co-operative and more willing to share with other children.
In another field-experiment, Parke and his colleagues (Parke et al., 1977) found similar heightened aggression among both American and Belgian teenage boys following exposure to aggressive films. In the Belgian study-- which replicated the findings of two similar studies conducted in the United States--teenage boys residing in a minimum-security institution were presented with a diet of either aggressive or neutral films. This study included a one-week baseline observation period, followed by one week of film viewing, and a one-week post-viewing observation period. There were four cottages involved. Two cottages contained boys with high levels of aggressive behaviour; two contained boys with low levels of aggression. One of each pair of cottages was assigned to the aggressive film condition, while the other two viewed the neutral films. Only the two initially high-aggressive cottages were affected by the movies; those boys who saw the aggressive movies increased their level of aggression, while those who were exposed to the neutral films reduced their level of aggression.
Still, one might ask whether such results are found when the variation in television diets occurs naturally rather than by special arrangement. Williams and her colleagues (Joy, Kimball & Zabrack, 1986; Williams, 1986) had the opportunity to evaluate the impact of televised violence on the behaviour of children before and after the introduction of television in a Canadian community. They compared children living in the before/after television town with their peers in two other towns where television was well established. The three towns were called Notel (no television reception), Unitel (receiving only the government-owned commercial channel-CBC), and Multitel (receiving the CBC and three American commercial networks-ABC, CBS and NBC). Children in all three towns were evaluated at Time 1 when Notel did not receive a television signal and again at Time 2 when Notel had had television for two years (it had received the government channel-CBC). Results indicated that there were no differences across the three towns at Time 1, but at Time 2 the children from the former Notel town were significantly more aggressive, both physically and verbally, than the children in the Unitel or Multitel towns. Moreover, only children in the Notel town manifested any significant increase in physical and verbal aggression from Time 1 to Time 2.