Version classiqueVersion mobile
OpenEdition Books

Violence et télévision

 | 
Jean-Michel Lacroix

Violence on Television: a British Perspective

Barrie Gunter

Texte intégral

1Concern about violence on television can be traced back to the earliest days of the medium. Indeed, this concern has its roots in the unease that is expressed whenever a new entertainment medium appears on the scene that appeals to the masses. Such reactions were reported with the appearance of popular romantic and adventure novels in the nineteenth century and were repeated in response to the growing popularity of motion pictures in the early part of the present century.

2Politicians and professional lobbyists have derived considerable political mileage and publicity from attacking television. At the same time, the press, for long media rivals of television, have leapt at every opportunity to devote extensive coverage to controversy over the depiction of violent material on the small screen.

3In recent months in the United Kingdom, broadcasters have come under renewed pressure from government to be ever more vigilant in their presentation of violence. Of course, broadcasters and their regulators in the UK have held to a tradition of being vigilant in this respect. Codes of practice exist for programme makers which highlight areas of concern and give guidance on points with regard to which care is needed in the treatment of violence on screen and in the decision making about whether to show violence at all.

4The Broadcasting Act 1990 requires the Independent Television Commission (ITC), regulator of all commercial television channels in the UK, to draw up a general code governing the portrayal of violence. Both the ITC's immediate predecessors, the Independent Broadcasting Authority (IBA) and Cable Authority had well-established codes on violence. Adherence to the ITC's code rules is a licence condition for all ITC programme licensees whether their services are delivered terrestrially or by cable or satellite. The code on violence pays particular attention to the young and vulnerable and to the need to avoid portrayal of violent behaviour which can easily be imitated.

5Concern about television violence, of course, centres on the possible effects it might have on viewers. Whether or not there can be a demonstrated causal link between television viewing and crime or anti-social behaviour, the ITC believes that caution is required in the television portrayal of violence, given concern about the level of violence in society and the possibility of behaviour or attitudes being influenced by what is shown on television. Particular vigilance is necessary by broadcasters in terms both of the way violence is portrayed in individual programmes and of the possible accumulation of violence in successive programmes.

6There has been a tradition in the UK of conducting audience research about television violence. This has been particularly so in respect of the ITC and the IBA before it. This research has broadly fallen into two categories: 1) assessment of public perceptions and opinions about violence on television and 2) attempts to measure the influences on television violence on viewers.

7The debate about violence on television has many twists and turns. The topic is a more complex one than critics and political lobbyists often realise or are prepared to acknowledge. For the purposes of this paper, in dealing with the research done in Britain by the independent television regulator, the findings can be discussed under several broad headings:

  • Defining television violence
  • Attitudes towards television violence
  • Influences of television violence
    • General behavioural influences
    • Effects of specific incidents
    • Perceptions of crime
  • Crime reconstruction programmes

8The key questions addressed by social scientists have included: How much violence on television? What do people think about it? What influences does televised violence have on the viewing population? These are all important questions and all have been studied to varying degrees in the United Kingdom.

9Unfortunately in the public debate the fact that these questions address different matters is often confused. Thus, statements about the influences of television may be misleadingly and mistakenly derived front content analyses measuring the amount of violence in programmes or public opinion polls which assess viewers' attitudes towards television. It is essential that different types of evidence about television violence are distinguished and not misinterpreted if informed debate is to occur.

10When considering the influences of television violence, different kinds of effects can and need to be distinguished. There has been a great deal of public comment about the general effects television portrayals of violence might have on viewers' individual aggression levels or aggressive propensities and upon levels of crime and violence in society. Concern has also been expressed about the possible triggering effect specific incidents in programmes might have on individual viewers who identify closely with the characters, problems or situations depicted. There has also been concern about the role television might play in increasing levels of fear of crime.

11A particular feature of the drama and feature film output of broadcast television channels over the years has been a reliance upon crime-based storylines. On ITV, the UK's most watched commercial television channel, this has in recent years been supplemented by real-life crime series with titles such as Crimestoppers and Michael Winner's True Crimes reconstructions of recent murders or crimes of violence have often featured. The narrowness of the drama agenda in part reflects the enduring popular appeal of murder, crime and detection. Nevertheless, some public concern has been voiced about these programmes, and in some instances even by senior broadcasting executives.

Defining television violence

12Calls for control of violence on television necessitate some consideration of what is meant by violence in the first place. How should violent content be classified or defined in order that it may be monitored effectively, and, if necessary, limited in the extent and form of its occurrence. It is easy enough simply to condemn the depiction of violence on television. But the term 'television violence' is used almost indiscriminately. While most people may have some idea of what is meant by violence, the fact is that what one person sees in exactly the same way by somebody else. Viewers may have differing perceptions of television violence. Yet for all that has been debated and written about television violence, there have been few serious attempts among either critics or researchers to provide realistic and meaningful definitions of violence and to say precisely what it is on the screen that needs to be controlled.

13Traditional procedures for the measurement of violence on television do not normally take into account what viewers themselves perceive to be violent. Typically, the violence of a programme is assessed in terms of the number of incidents it contains which match what researchers themselves decide is violent. All incidents that match an a priori definition of violence are given equal weighting for intensity or seriousness, irrespective of the dramatic context in which they occur. Yet, viewers can distinguish between different kinds of violence, and take into account the nature of the programme context in which violence is depicted and the form of the violence where judging its seriousness or acceptability.

14Research sponsored by the IBA in the early 1980s explored an alternative method of assessing TV violence, relying on viewers' own opinions about the kinds of television portrayals that are considered to be violent (Gunter, 1985). A procedure was adopted that allowed viewers themselves to decide about the seriousness of violence contained in various fictional television portrayals. In all, a series of 12 experimental studies were carried out in which groups of people were shown scenes taken from British crime series, American crime series, westerns, science fiction series and cartoons.

15Viewers were invited to make a variety of personal judgements about each scene along a set of qualitative rating scales. Scenes were shown singly or in pairs for comparative judgements, in a small lecture theatre. Variations in perceptions of the scenes were examined in relation to a number of things: the types of programmes the scenes came from, the types of characters who were involved in the violence, the types of weapons or instruments of violence that were used, the physical setting of the action, and the degree of observable harm the violence caused to victims in each scene. The results indicated that viewers may be significantly influenced by many different attributes of television portrayals in their personal opinions about television violence.

16Familiarity of surroundings emerged as one of the most powerful factors influencing viewers' perceptions of television violence. The closer to home the violence was portrayed as being in terms of place and time, the more serious it was judged to be. Thus, violence in British crime series tended to be seen generally as more violent than similar portrayals on US series of the same type. Portrayals of 'violence' behaviour in cartoons or science fiction programmes, however, were seen as essentially non-violent.

17Perceptions of television violence were also found to vary significantly with a number of other characteristics of television portrayals. The kinds of fictional characters who inflict the violence, how the violence is inflicted, and how much harm is done to those on the receiving end, all emerged as important mediators of viewers' opinions about television violence.

18Some of the results, however, were paradoxical. Opinions about violence perpetrated by law enforcers and by criminals, or by men and by women, were markedly different, depending on the types of programmes from which they came. With extracts from British crime series, for instance, viewers were most concerned about violence inflicted by men upon women, while in scenes from American series they were concerned more by women being violent to men. In an American context, violence performed by criminals was perceived in more serious terms than that used by law enforcers (usually the police), while in British settings it was law-enforcer violence that viewers were more troubled by (Gunter, 1985).

19One can only speculate at this stage as to the reasons why these differences in opinions occur. One answer may lie in the societal norms regarding the use of violence. Traditionally, society approves of some forms of violence under certain circumstances, and disapproves of others. For example, violence used by police officers to uphold the law, or that used by private individuals in self-defence against an attacker are, within certain limits, permissible. Under these circumstances, however, the use of violence must not far outweigh the magnitude of the behaviour of the law-breaker. Where the force used to repel an attacker is much greater than that justified by initial provocation, violent retaliation will not be found so acceptable.

20It is worth considering more closely violence between the sexes. Among the kinds of violence which society most disapproves is that performed by men, the physically dominant and stronger sex, against women, who are generally regarded as physically weaker. Since violent television portrayals set in contemporary British locations are closer to home than those set in American locations, perceptions of violence in familiar British settings involving British characters may be strongly influenced by norms of conduct that prevail in the society in which British viewers live. Therefore a violent attack by a male character on a female character in a British programme will be judged according to the appropriate norms of conduct and rated as unacceptable. Similar types of portrayals in American crime series, however, are sufficiently distanced from the everyday reality of British viewers for the same rules of judgement not to apply. Instead, such factors as the frequency and familiarity of particular types of portrayal within these programmes, but not in relation to everyday life, may be more important mediators of opinions about them. Since most violence in American crime programmes is perpetrated by men, the rather unusual experience of seeing it used by women is relatively more disturbing.

21Of different physical forms of violence, shootings were seen as most violent, but stabbings, with their close-in ruthlessness, were perceived as most frightening and disturbing, particularly in British programmes. The amount of pain and suffering seen to be caused by violent acts was also important to the way viewers reacted. Those scenes in which observable harm to the victim of violence was clearly depicted were perceived to be significantly more violent and disturbing, especially in realistic settings.

22While the nature of violence on television and the settings in which it occurs appear to be important factors influencing viewers' opinions, viewers themselves do not always agree on what is or is not violent. Another important area to explore, therefore, is the variation among viewers in their perception of television violence, and whether these differences of opinion are related systematically to particular aspects of their psychological make-up.

23A detailed examination of viewers' perceptions of television violence and of their psychological profiles revealed an interesting series of links between the two. For example, it was found that viewers' perceptions of their own masculinity and femininity, rather than their actual sex, were related to their perceptions of certain kinds of television violence. Viewers (men and women alike) who gave themselves high femininity and low masculinity ratings tended to be especially sensitive to scenes depicting a man attacking a woman. Estimates of personal aggressiveness were related to opinions about certain kinds of physical violence on television. Those individuals with higher, self-reported propensities towards verbal rather than physical aggression tended to perceive fist-fights, particularly in British crime-series, as more violent and disturbing than people who had revealed tendencies towards physical aggression.

24There are important lessons to be learned from this kind of research. At the level of audience perception, fictional violence on television cannot be defined in simple terms as a single unitary entity. Violence is a complex feature of fictional programming and viewers' sensitivities to it may be highly discriminating.

Attitudes towards television violence

25There is a prevailing understanding, regularly fed by press sponsored opinion polls, that the public generally believe that there is too much violence on television. An accompanying "common sense" notion is that televised violence is potentially harmful, especially where young viewers are concerned. Public opinion about television violence, however, can be measured in different ways and opinion profiles can vary according to the way in which people are questioned. Viewers' opinions are important and in the UK, the 1990 Broadcasting Act requires the Independent Television Commission to make arrangements to carry out audience research to assess public opinion.

26To be useful, however, public opinion surveys must yield data which are relevant to policy and editorial decisions. There is frequently a tendency to oversimplify public attitudes in relation to an issue which has many perspectives and complexities. The public's views about television and violence can vary a great deal. Much depends on how questions about televised violence are asked, and whether or not they refer to general or specific areas of television output. It is easy to obtain an impression of strong concern, by using a particular line of probing public opinion. By focusing upon more specific aspects of the viewing experience, however, most notably the programmes which viewers watch, results are quite different (see Gunter and Stipp, 1993).

27How widespread are public concerns about violence on television? Opinion polls commissioned and published by major newspapers in the UK have frequently given the clear impression that there is widespread public concern.

28Research among viewers in Britain by the ITC and IBA which has investigated public attitudes towards a variety of taste and decency issues connected with television programming, however, has found that the prevalence of concern about violence can vary markedly with the form of the question (see Gunter and Stipp, 1993).

29One national survey conducted by the IBA in the late 1980s set out to examine the effects of different types of question on viewers' expressed opinions about controversial television content. To begin with, respondents were posed three different open-ended questions during the course of the survey interview. The first question asked them if they had personally seen or heard things on television which they had found offensive. Those who claimed that they had experienced something offensive on television (57 % of the sample in this particular survey) were next asked to give examples of the kinds of things that had caused the offence. A second question provided a different frame of reference in which to elicit viewers' perceptions of violence and other taste and decency matters on television. Respondents were asked to name any items they would not like to see on television. Finally, respondents in the same survey were asked which items in their opinion should never be shown on television.

30Comparisons of the results emerging from these three questions revealed that frame of reference can make some difference to the nature of viewers' expressed concerns. Although similar themes emerged throughout, the level of concern varied with the wording of the question (see Table 1).

31Focusing specifically on the violence item, mentions of violence occurred more frequently in response to questions about sources of offence and things viewers would not like to see, than in response to a question about things which should never be shown.

32The first pattern to emerge was that progressively fewer respondents mentioned any items at all as the frame of reference became more severe. In respect of violence on television, far fewer respondents took the view that it should never be shown compared with the number who said they either would not (personally) like to see violence on television or who claimed to have been offended by it. Responses mentioning sexual material were clearly less frequently occurring in the context of sources of offence than in connection with things respondents said they would not like to see or which should never be shown.

33The questions just discussed required respondents to supply their own answers. What happens to viewers' opinions when answers are already supplied and they have to choose from items presented to them? The same survey probed further for the kinds of things respondents would not want to see on television or which they believed should not be shown by using questions with ready-made response options. In the case of each question, a list of items was presented to respondents and they were asked to choose any which in the first instance represented items they personally would not want to see on television, and secondly, items which they believed should never be shown on television.

34The extent to which each listed item was selected by respondents as something they would not want to see was then compared with self-generated items for the same question (see Table 2). Rape, explicit sex, certain forms of violence and bad language were the items most often selected. Three points are of special interest here. First, more items were chosen from a prompt list than had earlier been self-generated by respondents. Second, items that had earlier been self-generated by survey respondents were selected by many more respondents when they occurred in a prompt list. Third, the rank order of items of concern from prompt list selections was different from that derived from self-generated choices. Thus, sexual material was more often chosen as the type of content respondents would not wish to see than violent items when a list of options was provided to respondents, while violence was more often mentioned than sex when respondents had to supply their own answers.

35A similar comparison was made between item mentions and item choices in response to a question asking respondents to select items which should never be shown on television (see Table 3). Again, more respondents made a response of some sort to a prompt listing, and items were mentioned more often when chosen from a prompt list than when self-generated answers were required. Further, the rank order of items common to prompted and self-generated questions varied. Rape was, once more, the most-often mentioned item from the prompt list, followed by explicit sex scenes. Concerns about violent material and bad language followed next, with the latter being mentioned a great deal more often when prompted.

General Opinions about TV Violence

36The controversy surrounding the portrayal of violence on television embodies many different worries and questions related to the possible social effects on audiences and to decisions governing what is permitted on screen and under what circumstances. When considering the nature of public opinion about violence on television, research in the UK has found that a number of distinctions need to be made on the basis of the form of question:

  1. General statements about violence on television can elicit apparently high levels of public concern.
  2. More focused questions about televised violence can yield varying levels of agreement, reflecting different apparent degrees of concern.
  3. Opinions about specific programmes which contain violence tend to yield much lower apparent levels of concern.

37One accusation often levelled against broadcasters is that there is simply too much violence on television. When issued as a bland, unspecified statement, a typical finding, as we have observed already, is that most people will agree that there is too much. Research published by the IBA in 1988, for instance, corroborated other press polls at around the same time, with six out of ten respondents stating their agreement (see Gunter and Wober, 1988).

38Turning from this standard and very general question to the responses given to more focused questions about televised violence, however, a different level of response can emerge. For example, when asked, in a nationwide IBA opinion survey, about the timing of violent programmes, an overwhelming proportion of UK viewers held the opinion that such programmes should be broadcast late al night (see Table 4). In addition to agreeing with the statement that "there is too much violence on television", more than half a national UK sample of viewers agreed that we would all be better off without violence on television. Relatively few respondents, however, wished to see violence in programmes become more realistic and exhibited mixed views about whether it was a sign of good sense and awareness to believe that television is harmful to people (see Gunter and Wober, 1988).

Perceived Harms of TV Violence

39Taking the question of the possible harms of television violence further, another national survey sample was presented with a group of items which dealt with purported harms of television violence (see Table 5). Nearly eight out of ten respondents felt that people are justified in being concerned about the impact of television violence on children today. At the same time, more than six out of ten believed that children will imitate violence seen on television and that television violence frightens children. Slightly smaller percentages of respondents believed that television violence can make children aggressive and that it is bad for children's sleep (see Gunter and Wober, 1988). In summary, when given direct prompts, most UK viewers have been found to believe that television violence may have potentially harmful effects on children, influencing their behaviour and disturbing them emotionally.

Violence in Programmes

40So far the spotlight has been shed on opinions given by viewers about violence on television in a broad sense, unrelated to any specific portrayals seen in programmes. Viewers, however, watch programmes and the style and content of the programmes they watch can vary significantly. Do viewers' opinions about violence seen in specific programmes reflect the level of concern apparently indicated by broader statements of opinion?

41Important insights into this last question were derived from a body of UK research during the mid-1980s which explored viewers' opinions about a range of UK-made and US-made crime drama series (Gunter and Wober, 1988). In these surveys, viewers of each week's episode of the selected series were questioned in detail about their perceptions of any violence it may have contained. While the UK series may not be familiar to non-UK readers, the important point to note is the degree of variation in opinions across programmes in terms of the perceived amounts and types of violence they contain.

Perceptions of Violence in British Crime Series

42This research assessed whether viewers felt that a programme happened to be the most violent series on television, or whether they felt that there are many other more violent series, whether the violence shown on screen was always necessary to the story, and whether in general a series contained too many extreme forms of violence for personal tastes (see Table 6). More than eight out of ten viewers of each series believed there were many more violent series on television than the one in question. This was felt to be the case most often for Bergerac and Juliet Bravo. More than six out of ten viewers of these series felt that from what they had seen the violence was often necessary to the story. More than seven out of ten thought this was true for Juliet Bravo and The Bill. Few respondents agreed that these series were far too violent for their liking. At least three times as many Dempsey and Makepeace viewers felt that the show was too violent as did viewers of Bergerac and Juliet Bravo.

Perceptions of Violence in US Crime Series

43Crime drama series imported from the United States have been a prominent and popular part of the television schedules in the UK for many years. Some of these programmes have been severely criticized, however, both in the USA and UK, for their violence content. Public anxieties in the UK have been sharpened particularly because these programmes tend to attract many young viewers. Continuing the polls about UK series, viewers' attitudes were compared on three US crime drama series which were broadcast on British television in the mid- to late- 1980s the time - Knight Rider, Magnum PI and Murder She Wrote (Gunter and Wober, 1988).

44Here, more than nine out of ten viewers of these series felt that there were many other more violent programmes on television than each of these series (see Table 7). There is a logical paradox in these results; for not only does each particular judgement imply that a particular series is not so violent; but when other series are investigated none of them is acknowledged to be the location of greatest violence. It is known from other research that viewers do not generally patronize just one series, but watch several series within a genre, so it is unlikely that viewers were saying that 'other series' were more violent just because they were unfamiliar with them. It is more likely that the question is being treated simply as an assessment of the level of violence the series was felt to show and the answers are reasonably interpreted as implying that viewers did not consider any of the series to be unduly violent.

45There was less widespread agreement over whether the violence shown in these series was always necessary to the story. Around six out of ten felt that this was true for Knight Rider and Magnum, while just under half felt that it was true of Murder She Wrote. Fewer than one in ten viewers of these programmes felt that they were too violent for personal tastes.

46Summing up, opinions regarding the portrayal of television violence have been found to vary quite a lot with the types of questions being asked and with the types of content being asked about. This in itself is a very important finding. It shows that the public's concerns about such TV content can be effectively measured only by taking into account a wide range of opinions about the many different ways in which violence and sex can be portrayed on TV. But it also suggests that some questions might lead to erroneous conclusions and that the public's opinions are very complex.

Influences of television violence

47A substantial body of research literature has accumulated over the years on the topic of the influences of television violence. Research carried out in the UK, sponsored by the IBA has explored three general types of effects: general behavioural influences of television violence, effects of specific incidents or portrayals, and influences on public perceptions of crime.

48One study investigated relationships between television viewing and aggression among school children aged 11 to 16 years (Lynn, Hampson and Agahi, (1989). Another explored the impact of a soap opera depiction of a suicide attempt upon real life suicides and parasuicides (Platt, 1987). A further group of surveys examined relationships between television viewing and public perceptions of crime (Gunter, 1987).

General Behavioural Influences

49Professor Richard Lynn, with his colleagues Susan Hampson and Edwina Agahi of the University of Ulster, studied sibling pairs to obtain data on the relationships between aggression, the viewing and enjoyment of television violence, and certain personality characteristics of these children. They invoked what they called a genotype-environment and interaction theory. This theory proposes that parents transmit characteristics to their children by both genetic and environmental mechanisms. One good example of genotype-environment influence occurs with intelligence, where intelligent parents transmit their intelligence to their children both genetically and by providing an environment where the intelligence of the child is enhanced.

50Genotype-environment interaction also embodies a process by which genetic predisposition determines personality and behaviour. This is the means by which genetically different children may react differently to the same environment in accordance with their genetic propensities. Bright children do well at school, are rewarded and become encouraged to work even harder. Less bright children struggle, become discouraged by their failure to learn and give up.

51Lynn and his colleagues argued that both genotype-environment correlation and interaction provides a promising perspective for studying the effects of television violence on aggression among young viewers. The two processes would act in different ways in accordance with whether the child is a passive or active viewer of television violence. A child would be a passive viewer of television violence in situations where the whole family views television. The thesis would be that some parents provide a sociopathic family environment in which a high level of aggression and the viewing of television violence are constituents. The high-level family-based aggression would be transmitted to children both genetically and environmentally. Within this theoretical framework, the viewing of television violence might play some part in augmenting aggression in children as an environmental component which forms a part of the total genetic - environmental mix.

52Lynn's research obtained data from pairs of siblings. The advantages of this approach were, first of all, that correlations between pairs of siblings for a trait such as aggression can provide evidence of the effect of family influences on that trait. Family effects are both genetic and environmental. These influences act on both siblings and hence will generate positive correlations between them in terms of their respective aggression scores. If the correlation between siblings on a personality variable such as aggression is zero, there can be no family influence effects, while if a positive correlation is present it may be explicable in terms of genetic similarity. A family environmental effect can be inferred where sibling correlations are higher than would be expected on the basis of genetic similarity.

53A second point is that sibling correlations can be calculated for pairs of variables. These correlations can be calculated both between families and within families. Traditional sociological theory of the effects of the viewing of television violence on aggression requires the existence of within family correlations for the two variables, i.e., the sibling who views more television violence should also be more aggressive.

54A genotype-environment correlation and interaction theory of the possible effects of viewing of television violence on aggression would start by positing the existence of some general sociopathic personality trait with a substantial heritability. Certain personality types might be expected to develop aggressive propensities and might also tend to become active viewers of television violence. The reason for individuals with this personality make-up viewing more television violence would be that they enjoy it more, and the enjoyment could be more important than the amount of viewing.

55From an initial sample of more than 2000 schoolchildren aged 11 to 16 years, 99 pairs of brothers, 113 pairs of sisters and 174 brother-sister pairs were extracted. Measures were taken of aggression, personality, intelligence overall television viewing, television violence viewing and enjoyment of television violence. Personality measures comprised the Eysenck personality factors of Extraversion, Neuroticism and Psychoticism. Of particular relevance here was the Psychoticism factor which represents a broad sociopathic personality trait which might underlie both aggressiveness and viewing of television violence.

56Across families, a multiple regression analysis which explored the strengths of various possible predictors of aggression found that the personality variable, Psychoticism, was the major determinant of aggression together with a minor effect of Neuroticism. There was also a substantial effect for the enjoyment of television violence. The amount of viewing of television violence, however, exhibited no effect on aggression (see Table 8).

57Within families, this research found no evidence that the sibling who watched more violence on TV was more aggressive. Correlations were computed between pairs of siblings (both same-sex pairs and different-sex pairs) in terms of IQ, personality variables, aggression and television variables. While siblings from the same family household showed significant degrees of association in the amount of television they reportedly watched and in the amount of violent programming they reportedly watched, there were no significant correlations between them in terms of level of aggressiveness (see Table 9).

58Since pairs of siblings share the same family environment including family viewing of television violence, if the viewing of such content has an effect on aggression, it should act on both siblings. Low correlations between siblings was interpreted as evidence for a genetic influence on aggression, rather than of environmental effects (including television violence viewing). There was no tendency for the correlations between the viewing of television violence and aggression to increase over the age range 11 to 15, as would be expected from a sociological theory of the cumulative effects of viewing TV violence.

59The significant television variable which was related to aggression was the enjoyment of television violence rather than the amount of viewing. Siblings raised in the same family are exposed to the same family viewing of television violence, but they react to it differently. The sibling who enjoyed television violence was also the more aggressive.

60According to Lynn, there are genetic differences between children in their tendencies to develop sociopathic personalities. Of particular relevance here is the Eysenckian measure of Psychoticism. Children who score highly on this dimension develop aggressive personalities. This has been established elsewhere. Lynn's study also revealed that high Psycoticism scorers viewed and enjoyed television violence more in the case of males only. For Lynn, the enjoyment of television violence should be regarded as causal to the amount of viewing. Siblings who enjoyed television violence more were also more aggressive.

61Summing up, males scored higher on Psychoticism than did females, viewed more television violence, derived more enjoyment from television violence, and were more aggressive. Since boys and girls experienced the same family environment for television viewing, Lynn argued that it was difficult to explain these differences in terms of traditional sociological theory. A genotype-environment theory was credited as a better explanation by positing largely genetically-derived sex differences in Psychoticism leading to sex differences in the enjoyment of television violence and in aggression.

Specific Incident Effects

62Effects of specific unusual fictional media portrayals in triggering rare acts of serious criminal aggression, an important facet of the issue of media effects under natural conditions, have been reported, but they have not been established in scientific analysis. Press reports of such incidents usually suggest a simple causal connection.

63One often cited example of such an incident is a Russian roulette scene with a handgun in the film, The Deer Hunter. This was identified as giving rise to a number of real life "imitations". Psychologists who have analysed each of those and a number of other reports of supposedly movie-inspired violence, while not totally ruling out the possibility of imitation in these incidents, have not been totally convinced that these were indeed examples of imitation (Wilson and Hunter, 1983). Detailed scientific examination of causal connections between specific television portrayals of crimes and subsequent apparently imitative criminal acts by individuals has shown that establishing such links faces formidable obstacles and that such apparently clearcut connections are more complex and tenuous than they seem at first glance.

64One other example of this type of media portrayal is factual suicide portrayal on television and subsequent shifts in rates of suicides and attempted suicides. Such an incident occurred on British television in a popular BBC soap opera called East Enders in 1986. In an episode broadcast on Thursday 27 February 1986, a leading female character, Angie Watts, was portrayed taking a drug overdose. The trigger for this behaviour was her discovery that her husband Den Watts, had been having an affair. Angie was a long-suffering wife to an unfaithful husband and this incident represented the latest in long line of adulterous affairs. Angie had already developed a drink problem. At the end of this particular episode, however, she was pictured alone, drunk, and pouring out several boules of pills which she was witnessed as starting to consume along with more alcohol. The episode was watched by 14.4 million viewers aged 4 and over, 28 % of the potential audience.

65The episode was criticised for providing an ill-advised role model and for encouraging copycat behaviour. The IBA commissioned a study from the Medical Research Council's Unit for Epidemiological Studies in Psychiatry, Edinburgh to provide systematic, scientific evidence on the impact of this programme (Platt, 1987). The researcher conducted an objective and thorough evaluation of the aftermath of this fictional event by measuring any subsequent fluctuations in overdose cases treated in hospitals throughout Great Britain.

66The research obtained data on the number of cases of deliberate overdosage treated in accident and emergency departments in the week following the episode (the experimental period). This was compared to the number in the week prior to the episode (control period). A period of one week was chosen because of earlier evidence that the effect of a suicide story is detectable 0-7 days after its publication, (Bollen and Phillips, 1982; Phillips and Bollen, 1985). Data were obtained from 63 hospitals out of an original contact sample of 154 hospitals which comprised all hospitals with major accident and emergency departments in Great Britain.

67Since this before and after analysis may be invalidated by a temporal or seasonal trends in parasuicide, two further sets of data were sought. First, information about hospital-related overdosage obtained on a daily basis over a five-week span in 1986, which included the experimental and control periods. Secondly, baseline data for 1985 were obtained covering the two weeks equivalent to the control and experimental weeks in 1986. It was assumed that 1985 figures would present a 'normal' picture of parasuicide under 'normal' circumstances. Any increase in hospital - treated drug overdosage in the experimental period in 1986 compared to the corresponding control period, having controlled for trends in 1985, would be a consequence of the screened episode.

68The findings to emerge were somewhat conflicting and ambiguous. There was a 14.6 % rise in the total number of episodes of self-poisoning during the day after the broadcast, and an even larger increase (31 %) three days later compared with the day before the episode. Across the week following the episode, drug overdosage rates increased 13.7 % compared with one full week prior to the episode. This was a bigger increase than had been observed over the same period one year earlier. This was not a statistically significant difference however.

69Among different demographic groups, the biggest increase in deliberate drug overdose rates in 1986 following the episode of EastEnders occurred among males aged 45+, although changes in the same direction also occurred among all females and, in particular, among females aged 30 to 44 years. Only the result for all females achieved statistically significance, however. For the rate of change among females in the same age-group as the fictional character, the results did not achieve significance (see Table 10).

70A further analysis compared deliberate drug overdosage rates across different regions. Here, it needs to be borne in mind that EastEnders is set in London. Would viewers who live nearer to the scene of this drama be more likely to identify with it and hence exhibit higher levels of similar behaviour? Results showed that there was no tendency for people living in the same part of the world as the drama was set to exhibit greater rates of 'imitative' behaviour. Indeed, the London region actually experienced a decline in deliberate drug overdosage rates after the key Eastenders episode had been televised. At the same time, London exhibited the highest viewing levels for this programme (see Table 11).

71The researchers concluded that the case for imitative effects following this television portrayal of a suicide attempt was not proven.

Perceptions of Crime

72There has been growing concern in the UK about the role of television in contributing towards increased levels of fear of crime. During the 1980s the IBA conducted a number of studies which explored relationships between television viewing and public perceptions and fears of crime. This research investigated the possibility that relationships obs(erved by previous authors between amount and type of television viewing and social anxiety, mistrust and alienation might be explained in terms of an influence of television or in terms of characteristics other than demographic ones which relate to individuals' personalities.

73One early study of these relationships indicated that initially significant degrees of association between diary measures of television viewing and measures of fear of crime and mistrust of others disappeared in the presence of statistical controls for demographic and personality variables. At that time, it was reasoned that relationships observed between fearfulness and television viewing may be little more than epiphenomena of deeper seated personal dispositions or more general social conceptions (see Wober and Gunter, 1982).

74Later British research attempted to look in more detail at different levels of judgement about the prevalence of crime and of concerns about personal involvement, and how in each case they are related to viewing programmes with and without crime-related themes (Gunter and Wakshlag, 9). This research investigated the ways in which respondents' television viewing patterns (measured in terms of proportion of viewing time devoted to different categories of programming in addition to overall time devoted to overall amount of viewing) were related to societal-level and personal-level judgements about crime in a variety of locations, urban and rural, both close to home and distant from it. This work explored: a) whether societal-level judgements were more closely related to television viewing than were personal-level judgements; b) whether perceptions of crime in some settings were especially closely related to television viewing; and c) whether viewing of specific categories of programming, particularly those with crime-related content, predicted perceptions of crime better than did television viewing per se.

75A television viewing panel, based in London, completed a one-week viewing diary and attached a questionnaire to provide information about the kinds of programmes they watched and their perceptions of the likelihood of crime and fears of personal victimization in different locations. Some locations were geographically close to home, while others were more distant. Some perceptions were at the level of concerns for society and others were pitched at the level of personal concerns.

76Thus, respondents were first asked to estimate along a five-point scale (ranging from "not at all likely" (1) to "very likely" (5)) the probability that a person living in any of five locations would be assaulted in their lifetime (societal-level judgements). The five locations given were London, Glasgow, the Cotswolds (a rural location in England), Los Angeles or on a farm in the United States. Secondly, estimates were requested from respondents concerning the likelihood that they might themselves fall victim to violent assault (Personal level judgements) if they were to walk alone at night for a month around the area where they live, in a local park, through the streets of London's West End, through the streets of Glasgow, or through the streets of New York. They were also asked to say how likely they thought it was that they would become "the victim of some type of violent behaviour sometime in your lifetime" and that "you will have your home broken into during the next year".

77Finally, respondents were asked to say how concerned they would be for their personal safety (along a five-point scale ranging from "not at all concerned" (1) to "very concerned" (5)) if their car broke down at night in the English countryside, if they had to walk home alone late at night from a local pub, or if they found themselves having to walk through several streets in Los Angeles at night to reach their car.

78In order to find out if television viewing or viewing of specific programme types were related to risk perceptions independently of these other variables, a series of multiple regression analyses was run in which ten television viewing variables, demographics, personal experience with violence (direct and indirect), belief in ability for self-defence, and belief in a just world, were related to each risk perception. Each regression procedure was executed with all independent variables entered equally.

79The results for perceptions of risk for others indicated that, in the presence of multiple statistical controls for other variables, there was only one instance where a television viewing variable exhibited a significant relationship with a victimisation-likelihood perceptions (see Table 12).

80Viewing of television news was negatively related to perceived likelihood of victimisation for someone who lives in Los Angeles. Heavier viewing of the news predicted the perception of less danger in Los Angeles for others. None of the serious drama or crime-related programme categories (e.g., action-adventure, US crime-drama, UK crime-drama) was significantly related to any perceptions of risk for others. More significant to these perceptions were whether respondents knew a victim of an assault. Respondents who knew a victim perceived greater danger for others who live in Los Angeles, Glasgow and rural USA.

81Further, similar analyses were computed for perceptions of likely risk to self in different locations. Six significant relationships emerged between these perceptions and television viewing variables (see Table 13). Viewing of soap operas and of UK crime-drama predicted perceived risk in own neighbourhood. Heavier viewing of both programme types predicted the perception of greater danger to self in this setting. Total television viewing was significantly related to perception of potential danger in a local park and in London's West End at night. In both instances, heavier viewing predicted perception of greater risk. Finally, soap operas emerged as a significant predictor of perceived personal danger if walking alone at night in the streets of New York and perceived likelihood of having one's home burgled in the next year. Heavier soap opera viewing predicted greater perceived danger in New York; but less perceived danger of being burgled.

82Self-defence capability emerged most consistently as a significant predictor of perceived likelihood of self-victimisation across settings. Greater confidence in being able to defend oneself was associated with a reduction in perceived likelihood of being assaulted.

83Finally, the results for fear of victimisation showed that heavier total television viewing was a significant predictor of level of concern in all three scenarios (see Table 14). Throughout, heavier television viewing predicted greater concern for personal safety. With regard to the scenario closest to home, (i.e., walking home alone at night from a local pub) concern for safety was also predicted by amount of viewing of action-adventure, US crime-drama and sport. Heavier viewing of each of these programme categories predicted greater concern for personal safety.

84In summary, this survey among London residents which investigated their perceptions of crime at home and abroad, found that perceived likelihood of victimisation for others and for self, and concern about victimisation for self, varied with the situation, demographic characteristics of respondents, their direct experience with crime, and confidence in personal ability for self-defence in the face of an assault. Television viewing patterns, however, were relatively weak and inconsistent indicators of judgements about crime.

Crime reconstruction programnes

85Recent concerns about television's depiction of crime and violence have focused upon a specific programme genre, namely programmes which include dramatic reconstructions of real crimes. Some of these programmes, such as the BBC's Crimewatch UK and ITV's Crimestoppers and Crime Monthly carry reports of crimes, interviews with police and victims, and reconstructions of crimes, and invite viewers to join the fight against crime by telephoning the programme or the police (anonymously if they wish) with any information they may have about particular featured crimes.

86These programmes are popular with audiences and valued highly by the police, not simply as a public relations exercise, but also because the programmes have encouraged witnesses or informants to come forward with information which has resulted in the apprehension of perpetrators of crimes.

87Generally distinguished from these programmes are another collection of crime reconstruction programmes which dramatically depict crimes which occurred in the past (sometimes the recent past) and show how they were solved. A recent example of such a television series was produced by London Weekend Television for broadcast on the ITV network, called Michael Winner's True Crimes.

88Critics of these programmes, which depict both violent and non-violent crimes, argue that their main effect is to create an exaggerated fear of crime among viewers. Criticisms of television crime reconstruction programmes have frequently been voiced in the absence of systematic evidence concerning what people feel about such programmes, at the level of taste, or how they respond to or are influenced by them.

89In the past four years, the ITC has conducted survey research to tap into public opinion about crime reconstruction programmes. Although this research was not designed to test the effects of such programmes, it did, in one instance, examine the degree of association between reported viewing of a crime reconstruction programme and separately reported fear of crime.

90In 1990, a set of three surveys were carried out with regional television viewing panels in three ITV regions, London, Central Scotland and the East of England, to explore viewers' experience of and opinions towards crime reconstruction programmes on television. One focus of the research was to examine attitudes towards one particular programme — ITV's Crimestoppers. Additional background information was collected from respondents concerning their perceptions and worries about crime, and general views about media coverage of crime. Response rates of well over eighty per cent were obtained in each regional survey, although these varied somewhat across individual questions (see Wober and Gunter, 1990).

91The great majority of respondents indicated a willingness to telephone the police if they suspected a crime had happened, but only a minority (about one in three) felt able to defend themselves against an unarmed attacker. Well over half reported being more careful about going out alone after dark, but less than half were worried about being burgled and fewer than one in three claimed often to talk about crime. Under half of respondents felt that most crimes happened in big cities rather than in rural or suburban areas, while fewer still reported an increase in burglaries in their own area. Under one in five felt that, of late, violent attacks in their own locality had become more frequent.

92Turning to the media and crime, the main source of information about crime mentioned by respondents across all regions was a local newspaper. Regional TV news was placed second in London and the East of England, while "other people" were named as the second source of crime news among respondents in Central Scotland (see Table 15).

93Most respondents claimed to enjoy watching factual programmes about crime and police work, and over half observed that local newspapers these days seem to be full of crime news. Local television news, on the other hand, was far less widely seen to be preoccupied with crime, and few respondents believed that television news concentrates too much on crime (see Table 16).

94Crime reconstructions on TV were widely thought to be a good idea, and well over half of respondents in all regions felt that more programmes were needed which help the police and appeal for help from the public. However, there was less widespread support for programmes which concentrate on violent crimes or attacks on people (see Table 17).

95Respondents thought that there were many ways in which television can help the police. The great majority of respondents believed that television could help by widely reporting crime on local and national news (especially in London), and through special reports and items such as Police 5, Crimewatch and Crimestoppers, which promote the availability of free phone lines which the public can use to give information about crimes to the police (see Table 18).

96Focusing on "Crimestoppers", respondents in London were far less likely to have watched any editions of this over the previous month than were respondents from the other two regions. Fewer than half of London respondents compared with two-thirds of Central Scotland respondents and three-quarters of East of England respondents reported having watched any editions of the programme.

97Among respondents who reported viewing Crimestoppers, over half agreed that although they themselves had not been made afraid by the programme, viewers should be warned beforehand when it is due to show a reconstruction of a violent crime. Just one in three respondents indicated they thought that Crimstoppers had made them feel more cautious about going out alone in the dark or that other people had probably also become more afraid of crime as a result of watching the programme. At the same time, a similar proportion said they were reassured by Crimestoppers that the police solve a great many crimes. Other results suggested a positive function of the programme format. Whether people see more or less of Crimestoppers episodes makes no effective difference to their perceptions of amounts of crime, or of precautions against or fears evoked by crime. This is in distinction to evidence of viewing in general — and heavier viewers do tend to be more careful or worried about crime.

98Although there was no widespread belief that Crimestoppers should be shown only after 9pm, equally few viewers felt it was appropriate for children to watch the programme. Crimestoppers did not generally make respondents talk more about crime, and few respondents reported that editions they had seen had been, in their opinion, unnecessarily violent (see Table 19).

99Hardly any respondents reported making a telephone call to the police about a crime featured in Crimestoppers or indeed any other crime. A small number claimed they had had to reassure someone in their family not to worry, however, after watching the programme (see Table 20).

100Finally, a number of relationships were explored in each region between age and crime-related perceptions; overall reported viewing of television and crime-related perceptions with age effects statistically controlled; and claimed viewing of Crimestoppers and crime-related perceptions, with age and general television viewing effects statistically controlled.

101These correlations reveal that older people were, in general, more anxious about crime than were younger people. They had less confidence in being able to defend themselves, and were more preoccupied with crime, as evidenced by being more likely than younger people to say that they talked about crime with others. In London and the East of England, older people were more likely to phone the police about crime (see Table 21).

102Heavier watching of television, regardless of age, was also linked with greater anxieties about crime, less confidence in being able to defend oneself, and greater preoccupation with crime. Heavier viewers across the country were not consistently more or less likely to call the police.

103Claimed viewing of Crimestoppers, in particular, was rarely linked to crime-related perceptions, and then only weakly and inconsistently across different parts of the UK. In London only, there was a weak positive correlation between reported viewing of Crimestoppers and confidence to define oneself. Crimestoppers viewers in London felt better able to cope with an attacker than non-viewers. Only in the East of England was reported viewing of Crimestoppers associated with greater fear of crime. Generally, regular viewers of Crimestoppers were people who enjoyed watching factual programmes about crime and police work.

104Further opinions about television's role in helping the police were examined in relation to claimed viewing of Crimestoppers. Although findings were not consistent across regions, there emerged a general impression that people who watched Crimestoppers held positive opinions about this type of television programme. Heavier claimed viewing of this programme was associated with stronger agreement that television can help the police via reconstructions of crimes, arranging confidential telephone lines and wider reporting of crime. Such people believed that full length and shorter programmes with reconstructions have value. These people were also more likely to reject negative views about television's presentation of crime (see Tables 22 to 24).

Summary and discussion

105The early 1990s have seen the re-emergence of concern about levels of violence on television and its impact upon the audience in the United Kingdom. The broadcast regulators have kept their codes of practice relating to the depiction of violence in programmes under constant review, and have conducted regular audience research to assess the nature and source of public anxieties about screen violence.

106The independent television regulator in the UK has carried out research since the early 1980s on a variety of issues including the definition of violence, attitudes towards television violence, and the influences of television violence on general aggression, specific incidents of violence, and perceptions and fear of crime. More recently, close attention has been paid to the content and impact of crime reconstruction programmes.

107The research has revealed that viewers' perceptions of violence are complex and multi-faceted, and vary with the type of programme, form of violence, setting, depicted consequences, and nature of characters involved. Viewers attach greatest seriousness to close-in violence featuring sharp instruments which are shown in contemporary, domestic settings (Gunter, 1985).

108Attitudes towards television violence can vary significantly with the way they are measured. Different types of question —prompted versus unprompted, general or programme-specific— can yield dramatically different profiles of apparent public concern about violence (Gunter and Wober, 1988). Viewers are more likely to indicate widespread concern when asked about violence on television in general than when their attention is focused upon particular programmes which contain violent scenes. It is important to distinguish between these methodological artifacts when judging the validity of public opinion polls. It is possible that some types of question may lead viewer to express worries which they do not in fact experience when watching television programmes (Gunter and Stipp, 1992).

109Quite apart from matters of taste, the question as to whether television violence can shape the behaviour of viewers, making them more aggressive as personalities or increasing the likelihood that they will overtly display preexisting latent aggressive tendencies, has been a subject of debate since the earliest days of this mass medium. There is growing concern also that even if television does not act directly on viewers' behaviours, it may cultivate certain exaggerated perceptions of the real world relating to the occurrence of crime and violence.

110Research from the UK on the first of these issues, obtained in the late 1980s, did not yield evidence indicative of an influence of television violence viewing upon aggressive dispositions of children and adolescents. There was, however, some evidence to suggest that individuals with particularly tough-minded personality profiles were both more prone to aggression and exhibited greater enjoyment of programmes containing violence (Lynn et al, 1990).

111Concern about the potential influence of highly specific or unusual portrayals in popular programmes to trigger similar, imitative behaviour among susceptible viewers has been fed by isolated but often widely publicised cases. In the UK, there was a suspicion that an attempted suicide portrayed by a prominent character in a leading peak-time soap opera had resulted in a spate of copycat suicide attempts among members of the public during the week following this crucial episode. Closer investigation of suicide and parasuicide trends linked to drug overdoses across the UK failed to produce conclusive evidence of a causal role for television (Platt, 1987).

112Research into public perceptions of crime has reportedly shown widespread concerns about crime across the UK. Television has been held responsible by some critics. However, research has revealed that not all crime-related perceptions are invariably linked with television viewing. The perceived likelihood of other people falling victim to crime was principally predicted by whether members of the public know victims of crime themselves. There was little evidence that television played much of a role here (Gunter, 1987; Gunter and Wakshlag, 1986).

113Perceptions of likelihood that people would themselves fall victim to crime was, once again, linked most powerfully and consistently to personal experience. However, there were also some indications of a link with television viewing. Thus, overall television viewing significantly predicted perceived likelihood of self-victimization in London and New York (among London residents). Viewing of soap operas was also linked to perceptions of greater likelihood of falling victim to crime in one's own neighbourhood or in Glasgow (again, among London residents). Finally, viewing of UK-produced crime-drama was linked to perceived greater likelihood of self-victimization in one's own neighbourhood (Gunter and Wakshlag, 1986).

114Television viewing predictors were most prominent in relation to concern about crime. Heavier television viewing, heavier viewing of action-adventure, sport, and US crime drama predicted greater concern about walking home at night from a public house or bar. Women and those individuals with personal experience of crime and a lack of confidence in self-defence capability were also more concerned in this situation.

115Turning to crime-reconstruction programmes, the evidence emerged that viewers of these programmes were made more afraid by them. Indeed, this genre was held in generally high regard, with viewers believing that they performed a useful function (Wober and Gunter, 1990). Viewers appear to welcome the assumption of a degree of responsibility by television in helping the police tackle crime, and encouragement through that medium of their own involvement in this process.

Bibliographie

REFERENCES

Bollen K.A. and Phillips D.P., 1982, "Imitative suicides: A national study of the effects of television news stories", American Sociological Review, 47, 502-809.

Gunter B., 1985, Dimensions of Television Violence, Aldershot, UK: Gower Publishing Company.

Gunter B., 1987, Television and the Fear of Crime. London, UK: John Libbey and Company.

Gunter B. and Stipp H., 1993, "Attitudes and sex and violence on television in the United States and Great Britain: A comparison of research findings". Medien Psychologie, December, 267-286.

Gunter B. and Wakshlag J., 1986, "Television viewing and perceptions of crime among London residents". In P. Drummond and R. Paterson (Eds.). Television and its Audience, London: British Film Institute, 191-209

Gunter B. and Wober M., 1988, Violence on Television: What the Viewers Think. London, UK: John Libbey and Company.

Lynn R., Hampson S. and Agali E., 1989, "Television violence and aggression: A genotype-environment, correlation and interaction theory". Social Behaviour and Personality, 17(2), 143-164.

Platt, S., 1987, "The aftermath of Angie's overdose: Is soap (opera) damaging to your health?" British Medical Journal, 294, 954-957.

Phillips D.P. and Bollen K.A., 1985, "Same time, last year: Selective data dredging for negative findings". American Sociological Review, 50, 364-371.

Wilson, W. and Hunter R., 1983, "Movie-inspired violence". Psychological Reports, 53 435-441

Wober M. and Gunter B., 1982, "Television and personal threat: Fact or artifact? A British survey". British Journal of Social Psychology, 21, 239-247.

Wober M. and Gunter B., 1990, Crime reconstruction programmes: Viewing experience in three regions linked with perceptions of and reactions to crime. London: Independent Broadcasting Authority, Research Paper, August 1990.

Annexes

TABLE 1: Viewers' perceptions of controversial content

Base:

Items should never be shown
1195
%

Would not like to see
1195
%

Found offensive
1195
%

Violence/brutality/cruelty

16

27

25

Explicit scenes of love-making/sex scenes

14

18

11

Factual scenes of people being killed/dying

5

7

3

Child abuse

5

-

-

Bad language

4

9

22

Nudity

1

2

-

None/not stated

52

34

11

Source: Gunter and Stipp, 1993

TABLE 2: Items would not want to see on television

Base: All respondents

Prompted
1195
%

Unprompted
1195
%

A woman raped by a man/rape/unspec.

55

-

Explicit scenes of love-making

31

18

The killing of an innocent victim

29

27

Animals killing/ attacking each other/ cruelty to animals

28

3

Bad language

26

9

Frontal male nudity

16

-

The killing of a criminal

14

27

Frontal female nudity

13

-

Close-up of child-birth

9

-

Nudity (unspec)

5

2

Source: Gunter and Stipp, 1993

TABLE 3: Items which should never be shown on television

Base: All respondents

Prompted
1195
%

Unprompted
1195
%

A woman raped by a man

60

-

Explicit scenes of love-making

36

14

The killing of an innocent victim

32

16

Animals killing/ attacking each other/

27

-

Bad language

25

4

Frontal male nudity

23

1

Frontal female nudity

21

1

The killing of a criminal

20

16

Close-up of child-birth

12

-

Source: Gunter and Stipp, 1993

TABLE 4: General opinions about violence on TV

Agree
%

Neither
%

Disagree
%

TV programmes containing violence should be broadcast late at night

80

12

8

There are too many programmes on television that contain violence

60

21

19

We would be better off without violence on television; there is already enough in real life

55

24

21

I sometimes wish that violence programmes (like the Sweeney) was more realistic

15

22

63

People who say that television is harmful to people don't know what they're talking about

32

33

35

Source: Gunter and Wober, 1988

TABLE 5: Perceived harms of television violence

Agree
%

Neither
%

Disagree
%

People are justified in being concerned about the impact of TV violence on children today

78

17

5

Children will imitate the violence they see on TV

64

22

14

Watching TV programmes containing violence often frightens children unnecessarily

63

20

17

If children often watch TV programmes containing violence, they get accustomed to violence

61

25

14

By watching TV programmes containing violence, children become aggressive

58

27

15

Watching programmes containing violence is bad for children's sleep

57

27

16

Source: Gunter and Wober, 1988

TABLE 6: Perceptions of violence in UK crime series

TABLE 6: Perceptions of violence in UK crime series

Source: Gunter and Wober, 1988

TABLE 7: Perceptions of violence in US crime series

Knight
Rider
%


Magnum
%

Murder
She Wrote
%

There are many more violent programmes than… on TV at the moment

91

93

93

From what I have seen, the violence on… is often necessary to the story

58

60

49

… features too many extreme forms of violence for my liking

8

8

7

Source: Gunter and Wober, 1988

TABLE 8: Television and adolescent aggressiveness

TABLE 8: Television and adolescent aggressiveness

Notes: 1. Significance levels: *** p<.001; **p<.01; *p<.05
2. Predictor strengths given as beta weights
Source: Lynn, Hampson and Agahi, 1989

TABLE 9: Television viewing and aggressiveness patterns among sibling pairs

TABLE 9: Television viewing and aggressiveness patterns among sibling pairs

Notes: 1. Significance levels: *** p<.001; **p<.01; *p<.05
2. Degrees of association shown as correlation coefficients

TABLE 10: Episodes of deliberate drug overdosage treated in 63 emergency and accident departments in Great Britain, 1985 and 1986

TABLE 10: Episodes of deliberate drug overdosage treated in 63 emergency and accident departments in Great Britain, 1985 and 1986

Note: 1. Significance level - * p<.O5 (one-tail test)
Source: Platt, 1987

TABLE 11: Change in drug overdosage rates, viewing of Eastenders and geographical location

TABLE 11: Change in drug overdosage rates, viewing of Eastenders and geographical location

Notes: 1. Five hospitals omitted from the analysis
2. TVR (television rating) is the percentage of the potential audience (aged 4 plus years) who watched the EastEnders episode
Source: Platt, 1987

TABLEAU 12: Factors related to perceived likelihood of victimisation for others

TABLEAU 12: Factors related to perceived likelihood of victimisation for others

Levels of statistical significance: ***P<0.001, **P<0.01, *P<0.05

TABLE 13: Factors related to perceived likelihood of self-victimisation

TABLE 13: Factors related to perceived likelihood of self-victimisation

Levels of statistical significance: ***P<0.001, **P<0.01, *P<0.05
Source: Gunter, 1987

TABLE 14: Factors related to fear of crime

TABLE 14: Factors related to fear of crime

Levels of statistical significance: ***P<0.001, **P<0.01, *P<0.05
Source: Gunter, 1987

TABLE 15: Most likely source of crime news in own area or region

Base:

London
544
%

Central Scottland
448
%

East of England
485
%

In a local newspaper

60

65

59

On your BBC regional news magazine

11

4

13

From other people

10

13

4

On programmes like Police 5

7

1

1

On your ITV regional news magazine

6

7

13

On 'Crimestoppers'

1

6

5

Other

4

3

4

Q: Where would you be MOST likely to hear about crimes in your area or region?
Source: Gunter and Wober, 1990

TABLE 16: Opinions about media coverage of crime

Base:

London
452
%

Central Scottland
379
%

East of England
419
%

I enjoy watching factual programmes about crime and police work

67

72

64

The local newspapers these days are full of news about crimes

55

60

54

Our local television news programme is full of news about crimes these days

32

41

41

Television news concentrates too much on crime

15

14

18

Note: Percent agreeing with each statement
Source: Gunter and Wober, 1990

TABLE 17: Opinions about crime reconstruction programmes

Base:

London
452
%

Central Scottland
379
%

East of England
419
%

Crime reconstruction programmes like 'Crimewatch' are a good idea

81

87

80

We need more programmes that help police by reconstructing crimes and appealing for help from the public

59

66

56

Programmes don't need to concentrate on violence crimes/ attacks on people in order to help police in their work

43

39

39

Note: Percent agreeing with each statement
Source: Gunter and Wober, 1990

TABLE 18: Opinions about how television can help the police

Base:

London
488
%

Central Scottland
406
%

East of England
452
%

By widely reporting crime, especially violent crime, on local/national news

97

79

73

Regular promotion/free confidential phone lines for the public to ring the police with information about crimes

86

88

81

By local programmes like Police 5 which report a range of 'everyday' crimes like burglaries

82

73

77

By full-length programmes (like "Crimewatch") that reconstruct major crimes

80

83

77

By short spots like "Crimestoppers" in peak viewing time that reconstruct violent crimes and appeal for help

72

81

74

Note: Percent agreeing with each statement
Source: Gunter and Wober, 1990

TABLE 19: Opinions about crimestoppers

Base:

London
209
%

Central Scottland
254
%

East of England
322
%

Viewers should be warned beforehand when Crimestoppers shows a dramatic reconstruction of a violent crime

57

52

54

Nothing I have seen on Crimestoppers has made me more afraid of crime

51

52

52

Crimestoppers has made me feel more cautious about going out alone in the dark

34

33

37

Crimestoppers has made me feel more confident that the police solve a great many crimes

32

31

33

From what I have seen, I expect many viewers will be made afraid of being victims of crime from watching Crimestoppers

39

27

23

Crimestoppers should be shown only after 9pm

30

20

24

Crimestoppers is generally suitable for children to watch

25

37

28

Crimestoppers has made me talk more to other people about crlme

13

18

12

Some Crimestoppers editions I have seen were unnecessarily violent

14

8

9

Note: Percent agreeing with each statement
Source: Gunter and Wober, 1990

TABLE 20: Action contingent on watching crimestoppers

TABLE 20: Action contingent on watching crimestoppers

Q: As a result of seeing "Crimestoppers" have you ever done any of these things?
Source: Gunter and Wober, 1990

TABLE 21: Relationships between age, viewing and crime-related perceptions

TABLE 21: Relationships between age, viewing and crime-related perceptions

Notes: 1. L - London; CS - Central Scotland; EE - East of England
2. All correlation coefficients significant at p<.05 level
Source: Gunter and Wober, 1990

TABLE 22: Relationships between viewing television or crimestoppers and perceptions of television's role

TABLE 22: Relationships between viewing television or crimestoppers and perceptions of television's role

Notes: 1. L - London; CS - Central Scotland; EE - East of England
2. All correlation coefficients significant at p<.05 level
Source: Gunter and Wober, 1990

TABLE 23: Relationships between viewing television or crime stoppers and opinions about crime reconstruction programmes

TABLE 23: Relationships between viewing television or crime stoppers and opinions about crime reconstruction programmes

Notes: 1. L - London; CS - Central Scotland; EE - East of England
2. All correlation coefficients significant at p<.05 level
Source: Gunter and Wober, 1990

TABLE 24: Relationships between viewing television or crimestoppers and opinions about television's coverage of crime

TABLE 24: Relationships between viewing television or crimestoppers and opinions about television's coverage of crime

Notes: 1. L - London; CS - Central Scotland; EE - East of England
2. All correlation coefficients significant at p<.05 level
Source: Gunter and Wober, 1990

Table des illustrations

Titre TABLE 6: Perceptions of violence in UK crime series
Légende Source: Gunter and Wober, 1988
URL http://books.openedition.org/psn/docannexe/image/4869/img-1.jpg
Fichier image/jpeg, 119k
Titre TABLE 8: Television and adolescent aggressiveness
Légende Notes: 1. Significance levels: *** p<.001; **p<.01; *p<.052. Predictor strengths given as beta weightsSource: Lynn, Hampson and Agahi, 1989
URL http://books.openedition.org/psn/docannexe/image/4869/img-2.jpg
Fichier image/jpeg, 87k
Titre TABLE 9: Television viewing and aggressiveness patterns among sibling pairs
Légende Notes: 1. Significance levels: *** p<.001; **p<.01; *p<.052. Degrees of association shown as correlation coefficients
URL http://books.openedition.org/psn/docannexe/image/4869/img-3.jpg
Fichier image/jpeg, 112k
Titre TABLE 10: Episodes of deliberate drug overdosage treated in 63 emergency and accident departments in Great Britain, 1985 and 1986
Légende Note: 1. Significance level - * p<.O5 (one-tail test)Source: Platt, 1987
URL http://books.openedition.org/psn/docannexe/image/4869/img-4.jpg
Fichier image/jpeg, 200k
Titre TABLE 11: Change in drug overdosage rates, viewing of Eastenders and geographical location
Légende Notes: 1. Five hospitals omitted from the analysis2. TVR (television rating) is the percentage of the potential audience (aged 4 plus years) who watched the EastEnders episodeSource: Platt, 1987
URL http://books.openedition.org/psn/docannexe/image/4869/img-5.jpg
Fichier image/jpeg, 132k
Titre TABLEAU 12: Factors related to perceived likelihood of victimisation for others
Légende Levels of statistical significance: ***P<0.001, **P<0.01, *P<0.05
URL http://books.openedition.org/psn/docannexe/image/4869/img-6.jpg
Fichier image/jpeg, 294k
Titre TABLE 13: Factors related to perceived likelihood of self-victimisation
Légende Levels of statistical significance: ***P<0.001, **P<0.01, *P<0.05Source: Gunter, 1987
URL http://books.openedition.org/psn/docannexe/image/4869/img-7.jpg
Fichier image/jpeg, 315k
Titre TABLE 14: Factors related to fear of crime
Légende Levels of statistical significance: ***P<0.001, **P<0.01, *P<0.05Source: Gunter, 1987
URL http://books.openedition.org/psn/docannexe/image/4869/img-8.jpg
Fichier image/jpeg, 255k
Titre TABLE 20: Action contingent on watching crimestoppers
Légende Q: As a result of seeing "Crimestoppers" have you ever done any of these things?Source: Gunter and Wober, 1990
URL http://books.openedition.org/psn/docannexe/image/4869/img-9.jpg
Fichier image/jpeg, 154k
Titre TABLE 21: Relationships between age, viewing and crime-related perceptions
Légende Notes: 1. L - London; CS - Central Scotland; EE - East of England2. All correlation coefficients significant at p<.05 levelSource: Gunter and Wober, 1990
URL http://books.openedition.org/psn/docannexe/image/4869/img-10.jpg
Fichier image/jpeg, 199k
Titre TABLE 22: Relationships between viewing television or crimestoppers and perceptions of television's role
Légende Notes: 1. L - London; CS - Central Scotland; EE - East of England2. All correlation coefficients significant at p<.05 levelSource: Gunter and Wober, 1990
URL http://books.openedition.org/psn/docannexe/image/4869/img-11.jpg
Fichier image/jpeg, 134k
Titre TABLE 23: Relationships between viewing television or crime stoppers and opinions about crime reconstruction programmes
Légende Notes: 1. L - London; CS - Central Scotland; EE - East of England2. All correlation coefficients significant at p<.05 levelSource: Gunter and Wober, 1990
URL http://books.openedition.org/psn/docannexe/image/4869/img-12.jpg
Fichier image/jpeg, 117k
Titre TABLE 24: Relationships between viewing television or crimestoppers and opinions about television's coverage of crime
Légende Notes: 1. L - London; CS - Central Scotland; EE - East of England2. All correlation coefficients significant at p<.05 levelSource: Gunter and Wober, 1990
URL http://books.openedition.org/psn/docannexe/image/4869/img-13.jpg
Fichier image/jpeg, 117k

Auteur

Head of Research, Independent Television Commission, Londres.

© Presses Sorbonne Nouvelle, 1997

Conditions d’utilisation : http://www.openedition.org/6540