2316916код для вставки
Journal of Community & Applied Social Psychology, Vol. 1,275-289 (1991) Police Communication Programmes Aimed at Burglary Victims: a review of studies and an experimental evaluation FRANS WILLEM WINKEL* Department of Social Psychology, Free University De Boelelaan 1081, 1081 HV Amsterdam, The Netherlands ABSTRACT The effects and side-effects of a victim assistance project aimed at victims of burglary are experimentally evaluated. Key features of the programme included the provision of burglary prevention information by police officers during a personal interaction with the victim through a positive and limitative communication strategy. Results suggest that important programme goals (for example, stronger satisfaction with police performance, a more internal risk orientation concerning victimization, stronger preventive intentions) were met. One of the positive effects for which experimental support was found was an improvement in police-victim relations. As regards side-effects the programme led neither to response generalization, nor to increased fear of crime outside the home. However, the side-effects of increased fear in the home did occur in female victims and victims with an external risk-orientation. The implications for future victim assistance projects in this domain are finally discussed. Key words: Victim assistance, crime prevention, police communication, coping with victimization Crime prevention and victim assistance is sometimes depicted as a natural partnership. A familiar slogan of the ‘take a bite out of crime’ campaign in the USA for example is ‘victim assistance is crime prevention too’. A major portion of the Burglary leaflet, published by the National Association of Victims Support Schemes in Great Britain deals with questions of home security and crime prevention, also. In the Netherlands a central element of the recent governmental guidelines on the police role vis-u-vis crime victims is the provision of crime prevention information to victims who report their victimization. The provision of such information is seen as an important ingredient of a victim-oriented approach. However, scepticism is regularly voiced: police prevention publicity often does not work (Riley and Mayhew, 1980; Sacco and Silverman, 1981), or it might easily backfire as regards fear of crime and response generalization (Miransky and Langer, 1978; Scaglion and Condon, 1980; De Graaf, 1981; Holtom, 1985). In particular female victims and external *Portions ofthis article were presented at the4lst Annual Meeting of the American Society for Criminology (Reno, Nevada). 1052-928419 11040275- 1 5$07.50 0 1991 by John Wiley & Sons, Ltd. Received 18 February 1991 Accepted 15 October 1991 276 F. W. Winkel controllers appear to be vulnerable for these side-effects (Winkel, 1989c,d, 1990). This form of victim assistance is more closely scrutinized here by evaluating a police communication programme in which officers recontacted victims of a burglary several months after the incident was reported to them. A number of specific features were incorporated in this programme in an attempt to attain its intended effects and to avoid side-effects such as increased fear of crime and response generalization, which are likely to be stimulated among victims of burglary in particular. These features and the programme-goals will be considered in more detail in the ‘theoretical and empirical background’ section. After reviewing previous research relating to the present programme the experimental hypotheses are formulated, followed by a description of the experimental design, the dependent and independent variables. Next, the outcomes of the programme evaluation will be presented. Finally, we summarize the main results and their implications for future victim assistance programmes in this domain. THEORETICAL AND EMPIRICAL BACKGROUND Police communication will be defined here as the interpersonal provision of information on burglary prevention to victims by a pair of uniformed police officers who use a positive and limitative communication strategy in the form of a visit to the victim’s home some time after the incident was reported to the police. This definition refers to a number of features suggested in previous experiments on the effects and side-effects of preventive communication. In the first place reference is made to the importance of personal, face-to-face contacts with uniformed police officers. Winkel (1981, 1986, 1987b), for example, concludes that increasing police visibility is an efficient tool for controlling and reducing the public’s fear of crime. In these experiments high police visibility resulted in reduced fear both directly and indirectly, through strengthening police-community relations, diminishing subjective victimization risks and curtailing the perceived negative impact of victimization, indicating that less serious consequences were associated with victimization. In the second place the definition refers to the importance of using a so-called positive communication strategy (Ajzen and Fishbein, 1980; Jaccard, 1981), in which the positive personal consequences resulting from behavioural performance (complying with the preventive advice offered) are explicitly emphasized. Available evidence suggests that such a strategy is more effective (than a so-called ‘negative’ strategy) in persuading the general public to actually take preventive measures (Vrij, Winkel, Foppes and Volger, 1990; Zwanenburg and Winkel, 1990), and in changing their preventive intentions (Winkel, 1989b). Moreover, such a strategy does not backfire: increasedfearisnotalikelyresponseamongthegeneralpublic(Winkel,l981,1989a,1991). In the third place reference is made in the definition to using a limitative communication strategy, aimed at avoiding response generalization. This concept means that the communicator not only can induce the preventive reactions which he intends-the target set or set of intended preventive responses-but also unintended and undesired (buying weapons, watchdogs, and barricading one’s doors, for example) preventive reactions-the non-target set. Experimental outcomes, reported by Winkel (1987a), reveal response generalization due to a burglary prevention campaign to be particularly likely in specific subgroups among the public in general. Examples include: Police communication programmes 277 receivers characterized by an external risk orientation, for example persons who tend to perceive victimization as an outcome beyond their personal control, in persons reporting high ‘prior (that is: in the pre-campaign period) life stress’, and persons with a high communicative involvement with the theme of crime (heavy ‘crime news’ viewers, for example; Gerbner, Gross, Signorielli and Morgan, 1980). One might hypothesize that response generalization is thus likely to occur in crime victims exposed to preventive communication, because several of the conditions triggering generalization are simultaneously present in these individuals (Janoff-Bulman and Hanson-Frieze, 1983; Maguire, 1985). Response generalization might be avoided by utilizing a limitative strategy (Winkel, 1987a,b). This means that the communicator makes it clear to the receiver that the set of preventive measures given in the messagethe target set-is exhaustive and adequate. In the message, more far-reaching precautions should be presented as undesirable responses. Content of the preventive information provided is still another vehicle in achieving intended effects and avoiding harmful side-effects (Winchester and Jackson, 1983). Burglary is mainly typified as a form of opportunity crime, committed by amateurish perpetrators, using simple resources, like a screwdriver. As regards burglary prevention emphasis is put on the avoidance of ‘habits that have become natural’ but amount to asking for burglary (open doors/windows) and on the application of architectural provisions, like improving the fasteners on doors and windows. In short, stress is laid on simple techno-prevention, on target hardening. Such a focus might be of psychological importance to the receiver. Fischer (1984) and Scherer ( 1985) note that such a technical definition might positively influence the feelings of powerlessness and the more external orientation to victimization. Technical recommendations by the police are ‘more than professionalism and good public relations: they are also for victims regaining a sense of order, control and confidence in community’ (Fischer, 1984: 170). This aspect is nicely illustrated in Paap’s account of a victimization experience: ‘in some interactions the victim comes to define problems of burglary and home security not as moral or social problems, but as technological problems. This type of discussion helps the victim to define the situation as a technical not as an emotional matter and provides the feeling that increased control over one’s circumstances is available through technology. Defining the situation in terms of technical devices holds out some hope for personal power’ (Paap, 1981: 303). Achieving a more internal risk-orientation is not only positive in itself, but it might also contribute to realizing other goals, like reducing fear of crime. Tyler (1981), moreover, suggests that an internal risk-orientation increases the intention to take preventive measures, while Normoyle and Lavrakas (1984) report that such an orientation is accompanied by lower fear of crime. Drawing attention to the opportunity character of burglary might also be of psychological importance for the receiver. This implication is discussed by Waller and Okihiro (1978), Maguire (1980), Clarke and Hope (1984) and Hough (1984). Waller and Okihiro (1978: 110), for example, suggest that ‘we must do more to inform the public how they can help themselves to provide protection against the typical amateur offender’ (1978: 110). They stress the importance of de-dramatization in providing preventive messages. ‘Some reduction in fear might be achieved by more frequent publicity of the peaceful nature of residential burglary. The police called to the house could do more to provide reassurance to the victim by explaining the amateur and non-violent nature of most residential burglaries’ (1978: 105). Maguire 218 F. W. Winkel (1980: 270) moreover notes that for many victims the word burglary ‘conjures up pictures of masked intruders, ransacked rooms, and shadowy figures entering the bedroom while people sleep, all images perpetuated in fiction and in sensational media accounts of burglaries, but far from the reality of the mass of actual offences committed’. It thus seems plausible to interpret the initial symptoms of shock so frequently mentioned as a result of the combination of the unexpectedness of the event and the imagination of the victim.’ Maguire (1980: 271) therefore concludes that giving simple advice on security, reassuring victims that the burglar is probably a harmless teenager unlikely to return, although largely unproductive in thief-catching terms, are all valuable as aids both to the victim’s recovery and to police-public relations. Summarizing, we may conclude that increasing the acceptance of the point of view of the communicator that a fairly large number of burglaries is committed because an easy opportunity is offered is not only a goal positive in itself, but that increased acceptance can also contribute to lowering fear of crime and to counter the stimulation of extreme preventive reactions to burglary (cf. response generalization). Finally, we would like to note that the present intervention fits well into recommendations to limit the scope of police publicity campaigns (Riley and Mayhew, 1980; Winchester and Jackson, 1982; Heal, 1983; Mayhew, 1984),to address clearly identifiable social groups rather than employing ‘broad-spectrum’ techniques (Fattah, 1981; Fischer, 1984; and Kosberg, 1985), to utilize a more personal approach (Scaglion and Condon, 1980; Holtom, 1985), and that police officers should respond to victims’ informational needs, in particular to the very common need for preventive information (Shapland, 1982; Van Dijk, 1983; Yantzi and Brown, 1983; Reeves, 1985; Maguire, 1985). Victim recontacts One project that bears similarity to ours is Skogan’s (1985) victim recontact programme (Skogan and Wycoff, 1987) in which Houston police officers contacted victims of diverse offences a few weeks after the event. During the telephone conversation the officer asked about problems due to the victimization. Advice was offered on how to solve such problems and the officer mentioned various organizations able to give additional assistance. The telephone call was backed by a Crime Prevention brochure sent to the victims. Victims who had reported their victimization to the police were randomly assigned to two groups: one group was ‘recontacted’ while the other was not. Programme impact, however, was discouraging: ‘there were virtually no significant differences between contacted victims and victims in the control group on any of the outcome measures’ (Skogan, 1984: 3). The effect variables included willingness to take preventive action, satisfaction with police performance, attitude towards the police, perceived seriousness of crime in the neighbourhood and fear of crime. Results moreover indicated the occurrence of side-effects: ‘those who were contacted thought there was more crime in their area and victims with a linguistic heritage other than English were negatively rather than positively affected by the program’ (Skogan, 1984: 3). According to Skogan, victim recontact programmes would lead to positive results if the following criteria were considered: the victim is approached sooner; (2) contact is personal rather than by telephone; (1) Police communicationprogrammes 279 ( 3 ) attention is concentrated on specific groups of victims, for instance on those who are known to have problems; (4) active assistance is offered, rather than referring victims to other organizations; (5) programme personnel participate in training aimed at improving the skills needed in police-victim interactions. Our project conforms to four of these five criteria. In our case victims are visited in person by two uniformed police officers. The visits are limited to burglary victims, to whom direct assistance is offered in the form of preventive information, while a security check was made of vulnerable areas in the dwelling. Finally, project personnel participated in a training programme on burglary prevention communication which lasted a number of days, incorporating practical exercises through role-pays. HYPOTHESES The previous notes may be summarized by listing the most important (side-)effect measurements conducted in the experiment. In addition to fear of crime and response generalization the following seven measures are relevant: (1) satisfaction with police performance; (2) attitude towards the police; (3) agreement with the sender’s view on the opportunity-character of crime; (4) perceived police protection against burglary; (5) perceived police protection against crimes of violence; (6) risk-orientation regarding victimization; and (7) willingness to take burglary-preventive measures. With respect to these variables we hypothesize that police communication will result in: greater satisfaction with police performance, a more positive attitude towards the police, greater agreement regarding the opportunity-character of crime, a more favourable assessment of police protection against burglary, a more internal risk-orientation, and increased willingness to take preventive measures. Regarding fear of crime it was assessed whether increased fear as a side-effect is avoided in victims generally, and more specifically among female victims and among victims with an external risk-orientation. In line with the remarks above we hypothesize that especially women and victims with an external risk-orientation are likely to react to police publicity with increased fear. We also hypothesize that police publicity will achieve a reduction in fear among male victims and victims with an internal risk-orientation. Finally, it was assessed whether response generalization is avoided generally, and whether it does not occur among victims with an external risk-orientation, a high communicative involvement with the theme of crime, or among victims who already applied preventive measures. Parallel to a distinction made between rather extreme reactions to crime ‘inside the home’ and more extreme reactions ‘outside’ (cf. Winkel, 1987b), measures differentiate between fear of indoor crime and fear of outdoor crime. METHOD Procedure Assisted by the state police all persons who had reported a burglary in the preceding 18 months were listed. Victims were divided among three municipalities (Weesp, 280 F. W. Winkel Muiden and Muiderberg). Victims in Weesp were further divided according to five neighbourhoods. Next, the victims from the distinct geographical areas were assigned at random to either the control group or the experimental group. Since confidential personal data were at issue, all victims in the control group were informed of the survey (by telephone or in writing) and requested to participate. Willingness to cooperate was extremely high: only two victims refused, one by reason of illness, another on account of a death. A Free University pollster handed the victims in the control group a questionnaire, subsequently retrieved by the same person. Victims in the experimental group were approached by two uniformed state police officers, who had participated in the training programme referred to above. In the training attention was given to burglary as an opportunity crime, common methods by which burglars gain entry to a building, technical, constructional and organizational means of burglary prevention and so on. The officers also participated in extensive role-playing sessions, learning to establish and maintain contact with a victim by way of preventive communication. These exercises were discussed and evaluated per individual by the section Crime Prevention. One of the goals of both training and role-playing sessions was to teach officers how to use a positive and limitative communication strategy. As noted, all victims in the experimental group were visited by two officers in uniform. A visit contains three phases: (1) oral communication on burglary prevention is given; (2) a security check is made in and around the house; ( 3 ) a brochure on burglary prevention is distributed. Oral information was structured in conformity with the brochure Burglary Prevention. In this phase the officers spoke of the opportunity character of burglary and, based on the principles mentioned above, possible ways of preventing break-ins were suggested. Following the oral message a safety check was carried out in and around the home, inspection taking place in terms of the checklist ‘House Break-ins’. Vulnerable points around the house were pointed out and ‘tailored’ advice was given to reduce such vulnerability (for example, ladders and tools stored out of sight; the garden landscaped such that a burglar cannot work ‘invisibly’, etc.). The house itself, too, was checked out. Front doors, rear entrances, (sliding) doors opening to the garden, and garage doors were looked over. Are the doors equipped with security locks? Is it possible to insert a screwdriver between door and frame near the lock? and so on. Attention was given to such windows as featured by the house: casement windows, sliding windows, cantilevered windows, skylights, dome windows and what have you (a sample checkpoint: are retaining slats affixed such that the glass is not easily removed?). During the third phase of the visit a map containing information was presented, in which the two previous phases were expanded on and explained. This map was composed of a 20-page information brochure dealing with burglary prevention, a checklist ‘House Break-ins’, a brochure Mark Your Possessions on marking and identifying one’s goods, a registration form ‘Valuable Possessions’ and a folder ‘32 Tips for Carefree Holidays’. To illustrate we reproduce some fragments of this material. Informa tion fragmen ts The brochure Burglary Prevention was subdivided into a number of chapters dealing with matters such as the burglar and his ‘favourite working conditions’, vulnerability to break-ins, three lines of defence: around the house, locks and keys, doors and windows. Each chapter features extensive graphic illustrations. The first chapter contains the following fragment: Police communicationprogrammes 28 1 Can you protect yourself against these risks? Yes, you can. You can make use of ‘technology’: you can install better locks. The first and most important rule is: make the burglar’s working conditions unattractive. You do this by following a few basics. They are rather obvious if you look at them from the burglar’s point of view. These basics are: A BURGLAR IS ALWAYS IN A HURRY Time is his greatest enemy. Every minute counts. The more time he needs, the more risks he runs. A ‘job’ that will take too long frightens him off. And if he does get in, he has less opportunity to collect his loot. A BURGLAR LIKES SIMPLE TOOLS A professional safecracker does not hesitate to attack a bank’s safe with a cutting torch and ultra-modern equipment. The (opportunistic) thief staking homes prefers to work with simple tools. They do not hamper his movements. This is why proper security measures make things very difficult for this kind of criminal, and he will abandon his project. NOTHING IS IMPREGNABLE Complete security does not exist. Sooner or later a burglar will find a way to get around the problem. These basic rules underlie reasonable protection. Reasonable, because making a house completely secure would mean changing it into a fortress or a bunker. Safe, certainly, but not very pleasant to live in. And that is not the intention. That’s why we would rather speak of ‘burglary-discouraging’. That is to say: the situation is made as difficult and as unattractive as possible for the burglar. This brochure explains ways to do this. A second fragment is taken from the Crime Prevention folder entitled ‘32 Tips for Carefree Holidays’: What should be remembered before you go on holidays? And what should be taken care of whilst you are gone? A number of things. So much, in fact, that it seemed a good idea to put things down on paper. Of course, also during holiday seasons the police try to prevent offences as much as they can. But you, too, can do a great deal. Crime prevention needs both the police and you. When you know that most criminal offences are ‘petty crimes’, and when you know that these are often committed simply because the opportunity was there, then you understand immediately what you can do to prevent them. THE LESS OPPORTUNITY OFFERED THE FEWER CRIMES COMMITTED This folder explains how you can reduce opportunity. A convenient checklist tells you what to remember, what you should do, what you should certainly not do. Mark the items that apply to your situation and try to stick to them. We wish you carefree holidays! Variables The questionnaire assessed a number of background characteristics, including age, educational level and possible other victimizations experienced, such as theft outside the house, assault, vandalism. In addition to gender, the following receiver characteristics were recorded: (1) communicative involvement with crime as theme; (2) riskorientation regarding burglary victimization; ( 3 ) preventive behaviour displayed earlier, i.e. prior to the publicity campaign. Communicative involvement was determined by way of two questions answerable with ‘never’, ‘sometimes’, ‘regularly’ and ‘often’. The questions were: Do you talk with others about crime? How often do you read or hear about crime in the paper, on the radio or on television? Risk-orientation 282 F. W. Winkel was assessed utilizing three propositions: (a) one can do something about the likelihood of a burglary oneself, (b) if one takes a number of simple preventive actions one can avoid becoming a victim of a break-in, (c) one can reduce burglary risks oneself by installing some simple devices. Answers were in terms of seven-point scales ranging from ‘agree’ to ‘disagree’. Previous preventive action was assessed through asking ‘did you take burglary-preventive action in the past’? Regarding the dependent variables the attitude towards the police was assessed in terms of agreement with five statements: the police take people seriously; the police feel involved with the public; the police here are usually helpful; police here are open to suggestions; the relation between police and public is pleasant. Answers were in terms of seven-point scales, varying from ‘very strong agreement’ to ‘very strong disagreement’. Satisfaction with the police was measured through the question: ‘Are you satisfied so far with police performance on your behalf?’ The corresponding seven-point scale ran from ‘satisfied’ to ‘dissatisfied’. Agreement with the communicator’s view that burglary is a crime with an opportunity character was measured via statements such as: a good number of break-ins takes place simply because people themselves provide the opportunity; a fairly important part of crime is a matter of offered opportunity; crimes are perpetrated because the chance is provided; things are made too easy for burglars because people leave themselves wide open. Agreement with these statements could be expressed on seven-point scales from ‘agree’ to ‘disagree’. Perceived police protection against crime was determined through the questions: ‘To what extent do you consider yourself police-protected against burglaries’, and ‘To what extent. . . against violent crime?’ Both questions had seven-point scales, from ‘adequately’ to ‘inadequately’. Willingness to take preventive action was assessed by asking: ‘Do you intend to take action to reduce the likelihood that your house is broken into?’ One could answer ‘yes’ or ‘no’. Fear of crime indoors was assessed through seven items, including: How safe or unsafe do you feel when home alone in the evening? How worried or unworried are you that a burglary will take place during the day, when you are not at home? Would you indicate to what degree the chance that your house will be broken into evokes at this moment the feelings listed below? Would you indicate to what degree the chance that, indoors, you will be the victim of a crime against your person evokes at this moment the feelings listed below? All of these items carried a seven-point scale anchored with terms like ‘calm’and ‘tense’, ‘safe’and ‘unsafe’, ‘quite unworried’ and ‘quite worried’. Subjective burglary risk was assessed by asking ‘How likely or unlikely do you personally think it is for you to become a victim of a burglary?’ The answer, on a seven-point scale, ranges from ‘very unlikely’ to ‘very likely’. Perceived negative impact of burglary was measured through asking ‘How serious do you expect the consequences to be if you become the victim of a break-in?’ The seven-point scale varied from ‘not serious’ to ‘very serious’. Fear of crime outdoors was determined by questions such as: How safe or unsafe do you feel when walking the street alone at night in this neighbourhood? How worried or unworried are you that you may be bothered, threatened or assaulted in this neighbourhood in daytime? How worried or unworried are you that this may happen in the evening? Please indicate the degree to which the chance that you will be the victim of violent crime in these streets evokes at this moment the feelings listed below. Corresponding seven-point answering scales terminated in terms like ‘calm’ and ‘tense’, ‘safe’ and ‘unsafe’, ‘quite unworried’ and ‘quite worried’. Police communication programmes 283 Subjective victimization risk was found through four questions: How great or how small do you estimate the chance of your becoming the victim of vandalism _ . . of theft on the street .. . of maltreatment . .. of threats? The relevant seven-point scales varied from ‘very slight’ to ‘very great’. Expected negative impact of victimization was determined through the question: ‘How serious do you expect the consequences to be if you become the victim of violence?’ Here again, a seven-point answering scale was utilised. Possible occurrence of response generalization was checked by offering 23 statements regarding relatively extreme preventive reactions to crime. Twelve of these related to indoor prevention, the others to prevention outside the house. These statements were categorized in terms of five indices: moderate mobilization, extreme mobilization, police intensification, avoidance, and outdoor mobilization. These statements include: There is a great deal to be said for getting a watchdog; It is really worth the money to be included in a private surveillance service’s route; It is sensible to keep a steel pipe or a wrench at ready under your bed; You really should tell the police when you go away, even for just one day; There isn’t a street you can walk safely at night; It is a good thing to have something at hand to defend yourself with outside; You cannot really go out at night without carrying some means to defend yourself. Responses to these statements were entered on seven-point scales ranging from ‘disagree’to ‘agree’. RESULTS Sample The total sample consists of 250 subjects. The average age was 46. The experimental group consisted of 92 subjects, while 158 victims belonged to the control group of persons not exposed to police communication. More than half of the sample (52 per cent) consists of women, 48 per cent is male. With respect to level of education 32 per cent had elementary schooling, 33 per cent intermediate and 35 per cent higher education. The majority of the victims live in the municipality of Weesp (70 per cent); 30 per cent comes from Muiden/Muiderberg. The greater part by far (86 per cent) does not live alone in the house; 14 per cent does. Exposure to mass media news on crime was reported by 37 per cent of the subjects; 55 per cent reported regular exposure; 36% regularly discuss crime with others; 52% says they do so sometimes. A clear majority is positive about police performance while the crime was reported: 89 per cent describes treatment as friendly (4 per cent as unfriendly), 82 per cent as expert (5 per cent as bungling), 85 per cent as interested (7 per cent as uninterested). A negative comment often expressed in this connection is that once the crime was reported the police did not contact the victims again. In addition to burglaries, part of the subjects reported that they were victimized this past year: theft outdoors (19 per cent), vandalism (17 per cent), maltreatment (1.6 per cent), and a fight (1.2 per cent). A very large majority of subjects in the experimental group expressed themselves positively regarding the police communication offered through the present project: 89% of them characterized the visit as useful (1 per cent not useful) and 85 per cent as worthwhile (3 per cent as not worthwhile). Experimental outcomes Prior to the analyses three indices were constructed: attitude towards the police, degree of acceptance of the sender’s view that crime has an opportunity character, 284 F. W. Winkel and risk-orientation. The attitude index is built up of five variables, reliability is alpha = 0.87. The index for the sender’s view consists of five variables as well. Its reliability is alpha = 0.89. Three variables are included in the index for riskorientation; reliability is alpha = 0.78. The score per person is in every case his or her average for the constituent variables. Table 1 represents the first part of the results obtained. Table 1. Means (m) and standard deviations (SD) of the effect variables surveyed and the willingness to take preventive measures (distribution in percentages) of control subjects (no communication) and experimental subjects (communication)with relevant test statistics(Fix2) and levels of statisticaldeDarture ( Q ) Control group Satisfaction with police action Attitude towards the police Agreement with sender’s view Perceived burglary protection Perceived violence protection Risk orientation Preventive willingness Yes no Experimental group m SD m SD FIX’ n 4.90 2.76 3.79 2.90 2.95 4.66 1.82 1.08 1.60 1.60 1 .58 1.18 5.41 2.51 4.42 3.28 3.05 5.12 1.48 1.03 1.44 1.71 1.61 1.28 5.02 2.55 9.64 2.91 0.21 8.21 0.01 0.05 0.00 0.04 0.30 0.00 8.13 0.00 66% 34% 84%) 16% Higher score = more satisfied; more negative attitude; more agreement; more protection; more internal orientation. According to the p-values listed, on almost every dependent variable surveyed here a significant difference emerges between the control group and the experimental group. The one exception is perceived police protection against violent crime. There is no difference in the assessment between those exposed to police communication and those not so exposed: in both groups this protection is characterized as inadequate. The other variables do reveal differences. It turns out, first, that subjects exposed to police communication report greater satisfaction with police activity on their behalf so far than control subjects. Secondly, police communication positively affects the attitude towards the police. Exposed victims are typically more positive than control victims. Differences also emerge regarding acceptance of the sender’s view on opportunity crime: agreement is stronger among exposed subjects. Police communication also effectuates a shift in perceived police protection against burglary: exposed victims are more positive. Risk-orientation, too, is influenced positively by police communication: compared to non-exposed victims, those who received information shift to a more internal orientation to victimization. A final significant difference concerns reported willingness to take preventive action against burglary; police information increases willingness to do so. In sum, results suggest that police communication has a favourable impact on the dependent variables surveyed here. At any rate, compared to the control group, the experimental group more strongly confirms to a number of preconditions (cf. risk-orientation, agreement with sender’s view, perceived burglary protection) relevant to the avoidance of side effects. The question whether the present campaign actually does circumvent undesired effects, especially relative to fear of crime, is answered in Table 2. Police communication programmes 285 Table 2. Means (m) and standard deviations (SD) of fear of crime indoors for control and experimental subjects in general and for victims with either an internal or an external riskorientation and for male and female victims, with test statistics ( f ) and levels of departure b)* Control group Fear of crime ExternaI risk-orientation Internal risk-orientation Male Female ~ Experimental group rn SD rn SD F P 3.84 3.87 3.98 3.75 4.03 1.37 1.48 1.28 1.35 1.47 3.90 4.12 3.49 3.43 4.23 1.42 1.45 1.21 1.10 1.49 0.12 4.91 0.72 0.0 1 2.01 0.07 ~~ * In addition to analyses of variance, analyses of co-variance were always applied (also for the results to be reported later). Testing of the interaction effect communication x risk orientation is checked for gender influence. The calculated statistic is F = 5.86 (p = 0.00). Testing of the interaction with gender IS checked for risk orientation influence. The calculated statistic is F = 1.99 (p = 0.07). Since the analyses of co-variance yielded no different interpretation of the data they are not reproduced here. Prior to determining the influence of police communication on fear, two indices were constructed: fear of crime indoors (cf. Table 2) and fear of crime outdoors. The first index is composed of seven variables (alpha = 0.87). The second index contains five variables (alpha = 0.83). Table 2 suggests that communication, among victims in general, does not result in side-effects in the form of increased fear. However, a reduction of fear of crime indoors is not visible either. Table 2 further reveals an exposure by risk-orientation interaction and a trend for male and female victims to react to police communication in a different manner. Reported means suggest that victims with an external risk-orientation react to communication with increased fear, while among ‘internals’ fear is reduced. The means also reveal a tendency for women to react to police communication with increased fear. Among males the opposite effect is visible. Police communication, then, appears to result in side-effects among female victims and among victims with an external risk-orientation. Table 3 offers insight with regard to the subjective perception of burglary risk and the expected negative impact of burglary. Analyses of variance performed with respect to the variables listed in Table 3 suggest that police communication does not result in main effects on subjective risk of burglary ( F = 0.007; p = 0.78) or on the expected negative impact of burglary ( F = 0.00; p = 0.95), nor does it give rise (in interaction with the receiver’s gender) Table 3. Means (rn) and standard deviations (SD) on the subjective burglary risk and the perceived negative impact of burglary of experimental and control victims with an internal or external risk-orientation, with test statistics (F) for the interaction term communication x risk-orientation and levels of statistical departure (p) Control group Risk-orientation External burglary risk Internal External negative impact Internal Experimental group rn SD rn SD F P 4.34 4.25 5.21 5.31 1.74 1.39 1.64 1.51 4.79 3.95 5.54 5.00 1.62 1.61 1.67 1.70 2.74 0.05 2.10 0.07 286 F, W. Winkel to differential effects on subjective burglary risk ( F = 0 . 0 5 ; = ~ 0.81) or on the expected negative impact of burglary ( F = 0.26; p = 0.60). According to Table 3, differences do occur for subjective burglary risk in interaction with the risk-orientation, while for expected negative impact of burglary a tendency to differential effects becomes visible. Given Table 3, police communication to the externally oriented proves to occasion a higher estimate of subjective burglary risk, while among internally oriented victims a lower subjective risk assessment obtains. Further, there is a tendency that among the externally oriented, police communication reinforces the expected negative impact of burglary, while among the internally oriented it is weakened. Police communication has no influence at all on fear of crime outdoors. Neither the main effect of information ( F = 0.37; p = 0.54), nor the effects of interaction with gender ( F = 0.31; p = 0.57), nor the interaction with risk-orientation ( F = 0.23; p = 0.62) are significant. In short, police communication appears to have no sideeffectsin the area of fear of crime outdoors. With respect to expected negative impact of violent crime the situation is identical. Police communication has no effect on this. Neither the main effect of information ( F = 1.18; p =0.27), nor the effects of interaction with gender ( F = 0.15; p = 0.70), nor the interaction with risk-orientation ( F = 0.95: p = 0.33) are significant. For the most part this situation is repeated in the case of subjective victimization risk regarding offences other than burglary (alpha = 0.84). Neither the effect of communication in interaction with the risk-orientation ( F = 0.24; p = 0.86) nor the effect of communication in interaction with gender ( F = 0.03; p = 0.86) is significant. A practically significant main effect does occur (F=2 . 2 8 ; ~ = 0.06). According to the means, police communication results in a reduction of subjective victimization risk. For exposed victims mean subjective victimization risk is rn = 2.72 (SD = 1.18). Table 4 gives insight into the occurrence of response generalization. Relevant prevention indices were subjected to a reliability analysis once again. The reliability of the index for moderate mobilization indoors is alpha = 0.73. Extreme mobilization indoors corresponds with an alpha of 0.78. The alpha of the index for avoidance reactions is 0.90; of the index for mobilization outdoors 0.73. The police intensification index has a reliability of alpha = 0.55 (which is too low). Only when the third item is removed is a reliability achieved of alpha = 0.61. Hence, it was decided to deal with this item separately in the analyses. At issue is the statement: ‘At night money and bonds are best kept under your pillow’. Table 4 suggests that police communication does not lead to response generalization. There are no significant differences in agreement with moderate mobilization, extreme mobilization and police intensification between those who were exposed to police communication and those who were not. Nor do significant differences emerge in relation to agreement with avoidance reactions. Significant differences do occur in the amount of agreement with mobilization reactions outdoors and with the statement just cited. The means, however, suggest that in both cases agreement is less in the experimental group than in the control group. In other words, these significant differences indicate that police communication results in reaction blockade: agreement with extreme prevention is reduced. In addition to the analyses of variance reported in Table 4, such analyses were also performed for victims with a high risk of response generalization. These are victims with a high communicative involvement with the theme of crime, victims with an external risk-orientation and victims who displayed preventive behaviour 287 Police cominunicationprogrammes Table 4. Response generalization: means (m) and standard deviations (SD) of the degree of agreement with extreme preventive reactions, grouped according to various prevention indices of victims in the control group and the experimental group, with test statistics (4 and levels of statistical departure (p) Control group Prevention index Moderate mobilization Extreme mobilization Intensifying police surveillance Money under your pillow Avoidance reactions Mobilization outdoors Experimental group m SD m SD F P 3.53 2.89 1.97 2.50 3.22 3.07 1.36 I .76 1.35 2.04 1.73 1.60 3.56 2.76 1.98 2.04 3.18 2.67 1.36 1.70 1.34 1.68 1.51 1.47 0.03 0.27 0.00 3.30 0.03 3.80 0.85 0.60 0.97 0.03 0.85 0.03 Higher score = more agreement. before. None of these within-groups analyses led to significant differences between exposed and control victims. Response generalization, then, does not occur in these groups. To avoid an endless row of figures we will not list the statistics and so on here. We should, however, mention mobilization reactions outdoors in the group with a high communicative involvement ( F = 8.90; p = 0.00) and in victims who previously displayed preventive behaviour ( F = 7 . 0 9 ; = ~ 0.00). Inspection of the relevant means suggests the occurrence of response blockade. Mean agreement with mobilization reactions outdoors of highly communicative involved controls ( m = 3.29; SD = 1.73) is stronger than that of comparable experimental subjects ( m = 2.50; SD = 1.44). When preventive behaviour was present beforehand these means are m = 3.16 (SD = 1.60) and m = 2.57 (SD = 1.49) respectively. Thus, within these risk groups of victims police communication results in reaction blockade as well: the amount of agreement with mobilization reactions outdoors is thus reduced. DISCUSSION The impact of the victim assistance programme studied here is mostly positive. Results suggest that it is possible to achieve important goals through personal contact of uniformed police officers with burglary victims, during which both oral and written preventive information is presented and a security inspection is conducted. Apparently, this procedure of police communication holds more promise of success than ‘victim recontact’ programmes conducted in the United States. Those achieved either zero-effects (in terms of the goals sought) or side-effects. The results reported here, however, suggest that police communication results in improved relations between police and victims and will lead to greater satisfaction with the police. Moreover, police communication contributes to accepting the idea that burglary is an ‘opportunity-induced’ type of crime. Due to the information offered, police protection as perceived by victims is strengthened and a more internal risk-orientation of victimization established. Willingness to take preventive action increases as well. With regard to side-effectspolice communication does not result in response generalization. There are some indications that the programme results in reaction blockade, even within groups of victims ‘vulnerable’ to an increased risk of response generalization. These outcomes underscore the desira- 288 F. W. Winkel bility of applying a limitative communication strategy in future assistance programmes as well. Police communication does not lead to greater fear of crime outdoors. The more ambitious goal of reducing fear of crime indoors is realized in part only, among male victims and victims with an internal risk-orientation. But side-effects emerge in the form of increased fear of crime indoors among female victims and among victims with an external risk-orientation. It appears that utilization of a positive strategy of communication is not sufficient. Nevertheless, in light of the preponderance of positive effects continuation of police communication programmes seems justified. REFERENCES Ajzen, J. and Fishbein, J. M. (1980). Understanding Attitudes and Predicting Social Behavior, Prentice Hall, Englewood Cliffs, New Jersey. Clarke, R. and Hope, T. (1984). Introduction, in: Clarke, R. and Hope, T. (Eds) Coping with Burglary, Kluwer-Nijhoff, Boston, pp. 1-15. Dijk, J. J. M. van, (1983). Victimologie in theorie en praktijk: een kritische reflectie op de bestaande en nog te creen voorzieningen voor slachtoffers van delicten, Justitiele Verkenningen, W.0.D. C., 6,5-37. Fattah, E. A. (1981). Becoming a victim: the victimization experience and its aftermath, Victimology, 6,2947. Fischer, C. T. (1984). A phenomenological study of being criminally victimized: contributions and constraints of qualitative research, Journal of Social Issues, 40, 161-1 78. Gerbner, G., Gross, L., Signorelli, N. and Morgan, M. (1980). Aging with television: images on television drama and conceptions of social reality, Journal of Communication, 30,3141. Graaf, R. J. M. de, (I98 1). Geintegreerde aanpak van inbraakbestrijding, Algemeen Politie Blad, 22, 508-51 3. Heal, K. (1983). The police, the public and the prevention of crime, Howard Journal, 22, 91-1 00. Holtom, C. (1985). Slachtofferprojecten in Groot-Brittannie, Justitiele Verkenningen, W.O.D.C.,2,18-31. Hough, M. (1984). Residential burglary: a profile from the British Crime Survey, In Clarke, R. and Hope, T. (Eds) Coping with Burglary, Kluwer-Nijhoff, Boston, pp. 15-27. Jaccard, J. (1981). Attitudes and behavior: implications of attitudes toward behavioral alternatives, Journal of E,xperimental Social Psychology, 17,286-307. Janoff-Bulman, R. and Hanson-Frieze, J. (1983). A theoretical perspective for understanding reactions to victimization, Journal of Social Issues, 39, 1-17. Kosberg, J. I. (1985). Victimization of the elderly: causation and prevention, Victimology, 10,376396. Maguire, M. (1980). Buglary as opportunity, Research Bulletin, Home Office Research Unit, 10,6-9. Maguire, M. (1985). Victims' needs and victim services: indications from research, Victimology 10,539-560. Mayhew, P. (1984). Target hardening: how much of an answer? In Clarke, R . and Hope, T. (Eds) Coping with Burglary, Kluwer-Nijhoff, Boston, pp. 29-45. Miransky, J. and Langer, E. J. (1978). Burglary (n0n)prevention: an instance of relinquishing control, Personality and Social Psychology Bulletin, 4, 399405. Normoyle, J. and Lavrdkas, P. J. (1984). Fear of crime in elderly women: perceptions of control, predictability, and territoriality, Personality and Social Psychology Bulletin, 10, 191-202. Paap, W. R. (1981). Being burglarized: an account of victimization, Victimology,6, 297-305. Reeves, H. (1985). Victim support schemes: the United Kingdom model, Victimology, 10, 679-686. Riley, D. and Mayhew, P. (1980). Crime prevention publicity; an assessment, Home Office Research Study 63, HMSO, London. Police communicationprogrammes 289 Sacco, V . F. and Silverman, R. A. (1981). Selling crime prevention: the evaluation of a massmedia-campaign, Canadian Journal of Criminology, 23, 191-203. Scaglion, R. and Condon, R. G. (1980). Determinants of attitudes toward city police, Criminology, 17,485494. Scherer, J. (1985). Themes in social vulnerability: research avenues in victimology, Victimology, 10,2633. Shapland, J. (1982). Ervaringen van slachtoffers met polities en justitie in Engeland, Justitiele Verkenningen, W.0.D. C., 6,39-52. Skogan, W. G. (1984). Recent experiments in fear reduction, A summary of research conducted by the Police Foundation 1983-1984, W.O.D.C.,’s-Gravenhage. Skogan, W. G. (1985). Einige nieuwe politie-experimenten in de V.S., Justitiele Verkenningen, W.O.D.C.,l, 101-116. Skogan, W. G. and Wycoff, M. A. (1987). Some unexpected effects of a police service for victims, Crime and Delinquency, 33,490-501. Tyler, T. R. (1981). Perceived control and behavioral reactions to crime, Personality and Social Psychology Bulletin, 7,212-217. Vrij, A., Winkel, F. W., Foppes, J. H. and Volger, D. M. (1990). Politiele preventie-programma’s en angst voor criminaliteit: het rendement van een planmatige aanpak. In: Winkel, F. W. and A. van der Wurff, A. (Eds) Angst voor Criminaliteit: Theorie, Onderzoek en Interventie, Swerts en Zeitlinger, Amsterdam, 179-1 87. Waller, J., and Okihiro, N. (1978). Burglary: the Victim and the Public, University of Toronto Press, Toronto. Winchester, S., and Jackson, H. (1983). Residential Burglary: the limits of prevention, Home Office Research Study 74, HMSO, London. Winkel, F. W. (198 1). Angst voor criminaliteit: verklarende modellen, Tijdschrift voor Criminologie, 23,265-278. Winkel, F. W. (1986). Reducing fear of crime through police visibility: a field experiment, Criminal Justice Policy Review, 1, 381-398. Winkel, F. W. (1987a). Response generalisation in crime prevention campaigns: an experiment, British Journal of Criminology, 27, 155-174. Winkel, F. W. (1987b). Politie en Voorkoming Misdrijven; effecten en neveneffecten van voorlichting, Mens en Recht, Amsterdam. Winkel, F. W. (1989a). Increased fear of crime and related side-effects of persuasive communication: the price-tag of burglary prevention campaigns? In: Viano, E. C. (Ed.) Crime and its Victims: international research and public policy issues, Hemisphere, Washington, DC, chapter 26, pp. 273-297. Winkel, F. W. (1989b). Police crime prevention campaigns: a social-psychological perspective on effective preventive communication, Journal of Police and Criminal Psychology, 5, 11-21. Winkel, F. W. (1989~).Cognitieve en affectieve reacties op preventieve voorlichting: de rol van informatie- en ontvangerkenmerken, Gedrag en Organisatie, 2, 149-1 64. Winkel, F. W. (1989d). Responses to criminal victimisation: evaluating the impact of a police assistance program and some social psychological correlates, Police Studies, 12, 59-73. Winkel, F. W. (1990). Psychologische reacties bij slachtoffers van misdrijven: een toetsing van enkele kern-concepten uit een sociaal-psychologisch model, Gedrtag en Organisatie, 3,48-70. Winkel. F. W. (1991). Police, victims and crime prevention: some research-based recommendations on victim-orientated interventions, British Journal of Criminology, 31, 250-265. Yantzi, M. and Brown, S. D. (1983). Behoeften van slachtoffers van misdrijven, Justitiele Verkenningen, W.O.D.C., 6 , 8 6 9 3 .