police, victims, and crime prevention

BRIT.J. CRIMINOL. VOL.31

NO. 3 SUMMER 1991

POLICE, VICTIMS, AND CRIME PREVENTION Some Research-based Recommendations on Victim-orientated Interventions FRANS WILLEM WINKEL*

Introduction

An integral component of victim assistance by the police or specialized victim support schemes is the provision of information on crime prevention and on how to avoid future victimization. Reference to its value is made, inter alia, in various recent international documents. Examples include the recommendations by the Council of Europe (Council of Europe 1988; Tsitsoura 1989) and the UN Declaration of Basic Principles ofjustice for Victims of Crime and Abuse of Power (United Nations 1986). Similar references can also be found in various national documents, guidelines, and legal reforms regarding crime victims in countries such as Great Britain (Maguire and Corbett 1987; Shapland et al. 1985), The Netherlands (Soetenhorst de Savornin Loman 1989), Israel (Geva 1988), Canada (Waller 1984), and the USA (Abell 1989). In view of its prominence in the literature, issues relating to the provision of crime prevention information deserve special consideration. This article will concentrate on four aspects: (1) how do we effectively stimulate victims or potential victims actually to take preventative measures—in particular, is communication an effective instrument? Which messages are more successful? Are some messages counterproductive? (2) Are increased fear of crime and response generalization possible side-effects of communication? Under what conditions do these effects occur, and how might we circumvent them? (3) What sort of skills do police officers need to provide information to victims in a cost-effective manner, and how might we train them in these skills? (4) What are the actual outcomes—both desired and undesired—of police—victim communication programmes directed at the prevention of crime? Discussion of these questions will be * Senior Lecturer, Department of Social Psychology, Free University Amsterdam, De Boelelaan 1081, 1081 HV Amsterdam, The Netherlands.

250

Downloaded from http://bjc.oxfordjournals.org/ at Universiteit van Tilburg on August 8, 2012

Documents produced by various international organizations, among them the United Nations and the Council of Europe, allot a central place in victim support efforts to the provision of information on how to prevent (re) victimization. This article presents an overview of some recent experimental studies conducted within the Dutch police; on the basis of these experiments, some recommendations are formulated on how victims may be effectively encouraged to cope with their experiences using strategies based on their emotions and on the problem of avoiding victimization. These recommendations concern the issues of enhancing the persuasive impact of preventative messages, of avoiding communication side-effects (such as increased fear of crime and response generalization) and the design of training programmes aimed at stimulating victim-orientated attitudes and behaviour in police officers and victim assistance workers. The outcomes of four evaluation studies, which examine the effects and side-effects of this kind of victim-orientated intervention using face-to-face re-contact procedures, are reported.

POLICE, VICTIMS, AND CRIME PREVENTION

mainly confined to reviewing a number of experiments conducted in The Netherlands over the past six years. Limitations on space prevent the inclusion here of details about statistical analyses, procedures, variables, and methods, but these can be found in the publications and reports cited. Issues in the Provision of Information to Victims Effectiveness

251


The issue of the effectiveness of giving information is dealt with in a number of communication models which specify the determinants of decisions whether or not to take preventative measures. The relevant models are (1) the reasoned action model (Fishbein and Ajzen, 1980); (2) the behavioural alternatives model (Jaccard 1981); and (3) models based on Kahneman and Tversky's (1984) prospect theory. All three models allow us to formulate specific predictions and hypotheses about effective preventative messages, namely, messages resulting in behavioural performance or behavioural change aimed at preventing crime. According to the first model, behavioural performance is primarily determined by behavioural intentions which, in their turn, are mainly influenced by two components: the attitude towards preventative behaviour and and the so-called subjective norms concerning the behaviour in question. As regards the latter component, behavioural performance is more likely to occur the stronger the perceived normative pressure to do so. This pressure can be increased by making it clear to the person addressed (the 'receiver') that others in the environment—for example, the police organization— expect that he or she should perform the behaviour in question and by strengthening the receiver's motivation to comply with that expectation. Attempts to encourage preventative behaviour in this way will be referred to as 'normative communication'. Evidence reported elsewhere (Winkel 1984, 1989) suggests that the perceived 'weight of the communicator' (Wc) is an essential condition for effective normative communication. Wc is a sort of collective term for various attributes perceived in the communicator by the receiver. It includes factors such as perceived expertise and the extent to which the communicator is considered to be honest, reliable, and objective. The greater this 'weight'—or, more particularly, the 'overweight' of the communicator over the receiver—the greater the desired effect produced by the normative message. The other main influence on behaviour specified by the reasoned action model is the attitudinal component; this will be referred to as 'attitudinal communication'. This concerns the receiver's general evaluative judgement about the proposed behaviour, in the sense that its performance is perceived as profitable or harmful, as (generally) good or bad. The more positive the evaluative judgement, the more likely it is that the behaviour will be performed. According to the model, performance will result if the receiver associates positive consequences with the behaviour and if he or she perceives those consequences as likely. Attitudes will become more negative, making performance less likely, if the receiver associates negatively evaluated consequences with the behaviour and considers those consequences likely. Together, these analyses suggest that an attitudinal campaign will be effective if the persuasive message explicitly stresses the relationships between the intended behaviour and its (positive) behavioural consequences, while a normative campaign will be successful if the message and its

FRANS WILLEM WINKEL

252


delivery make the receiver aware of the presence of normative expectations and strengthen motivation to comply with those expectations. With respect to attitudinal communication, the message should make it plausible that performing the suggested form of behaviour will be profitable for the receiver. To obtain this goal the communicator can formulate the message in three different ways. He or she can (1) emphasize that performance of the desired behaviour is to the receiver's advantage; or (2) emphasize that not performing it is disadvantageous; or (3) fuse both strategies into one message simultaneously conveying the positive consequences of carrying out the advice and the negative consequences of disregarding it. I will refer to these three strategies as the 'positive', the 'negative', and the 'mixed' strategies of persuasive communication. Each of these meets the criterion of effectiveness, since in all three cases the message links behaviour to consequences. In place of 'simple' attitudes and intentions, the second, behavioural alternatives model introduces differential attitudes and intentions. These concepts refer to differences between attitudes or intentions respecting different behavioural alternatives. On the simplest level, one can always speak of two behavioural alternatives, for example (1) performing preventative behaviour and (2) not performing it. In this case the differential attitude simply refers to the difference between the attitude towards performance and the attitude towards non-performance. The same reasoning applies to differential intentions. This model is based on the assumptions that (1) behavioural performance will occur if the intention to perform the behaviour is stronger than the intention not to perform, and (2) that the likelihood of behavioural performance increases with the magnitude of the discrepancy between these two intentions: the greater the difference between the two intentions, the higher the likelihood of performance. Given the fact that differential intentions are, according to the model, determined by differential attitudes, effective messages are those messages that succeed in effecting a substantial difference between the attitude towards performing preventative behaviour and the attitude towards not performing it. According to this model, the hypothesis follows that a mixed message will be the most effective, while the positive and the negative styles of communicating will be about equally effective. Basically, the mixed message can be characterized as a two-sided message: not only does it stress the positive consequences of performing preventative behaviour, thereby influencing the receiver's attitude towards performance in a positive direction, it also stresses the negative consequences (increased risk of victimization) of not performing that behaviour, thereby influencing the receiver's attitude towards non-performance in a negative direction. From this perspective the positive and the negative messages can be regarded as one-sided, stressing either positive or negative behavioural consequences in isolation. Using Kahneman and Tversky's prospect theory, however, an alternative hypothesis might be formulated. This alternative states that a mixed message will be completely unsuccessful, a positive one will be successful, and a negative message will be counterproductive and result in boomerang effects. According to this third model, an individual's behavioural decisions are determined by the way in which a choice situation is presented to him or her (the 'framing effect'). If a choice situation is presented in terms of potential gains—a positive message—individuals tend to respond with risk-avoiding behaviour; if that same situation is described negatively, for example in terms of potential losses—a negative message—individuals tend to exhibit


Side-effects

Two classes of communication side-effects prevail in victimological research: response generalization and increased fear of crime. Response generalization Generalization across preventative responses refers to the production of an effect on non-stimulated preventative responses (the non-target set), alongside a change in stimulated ones (the target set). Transfer from the target to the non-target set constitutes an undesired side-effect when the non-target set consists of extreme preventative behaviour, such as buying watchdogs, or keeping iron bars or guns at hand. One example of such an effect is when a media campaign persuading victims to lock their doors when out of the house produces not only the desired locking effect (opportunity reduction) but also a fortification mentality. Building on models involving exposure mechanisms, learning theory, and individual differences, Winkel (1987a) devised two hypotheses: (1) 'carry-over' (generalization) will automatically emerge in all receivers of preventative information; and (2) the phenomenon is confined to specific types of receivers. Experimental outcomes generally supported the second hypothesis, suggesting that exposure to relatively innocuous preventative 253


risk-seeking behaviour. The performance of behaviour aimed at preventing crime can be characterized as risk-avoidance: by taking these measures the person's objective risk of victimization is reduced. In a similar vein, not performing that behaviour could be characterized as risk-seeking: by not taking these measures the individual's risk of victimization is increased. Taken together, these analyses suggest that the mixed message will not be successful in bringing about preventative behaviour, because it combines the hypothetically productive elements (risk-avoidance) of the positive message with the counter-productive elements (risk-seeking) of the negative message. All three hypotheses, based on the three models described, were simultaneously tested in an experimental set-up, described in detail elsewhere (Winkel 1989c). The experimental outcomes were quite clear-cut and provide substantial support for the hypothesis based on model (1) and that based on model (2). These outcomes provide us with the following indications concerning effective design and communication of measures to prevent burglary. Effectiveness appears to be enhanced by (1) explicitly linking the performance of preventative behaviour to desirable or positive consequences, or by linking negative behavioural consequences to non-performance, using a positive, a negative, or a mixed frame; and (2) focusing attention on the normative expectations held by the police regarding the performance of that behaviour, for example, by stressing the common responsibility of the police and victims in preventing crime. In attitudinal campaigns, effectiveness is particularly enhanced by using twosided rather than one-sided crime prevention messages. Simultaneously emphasizing opportunity reduction, decreased victimization risks, and rendering a target unattractive to burglars as positive outcomes of performance, together with their negative counterparts emanating from non-performance, in one and the same preventative message is the most promising strategy. As regards normative communication, effectiveness is enhanced if the 'weight' of the communicator as perceived by the receiver is great. Performance is facilitated if the victim more strongly perceives the communicator to be an expert, and that he or she is reliable, objective, and honest.

FRANS WILLEM WINKEL

Perceived risk, negative impact, and fear According to the 'expectancy-value' model of fear of crime (Winkel, 1981, I989d), emotional responses to communication have a cognitive basis. These responses are rooted in two so-called 'cognitive products': subjective victimization risks (SVR) and perceived negative impact of victimization (NI). This model predicts increased fear of crime in response to preventative communication if, and only if, that information results in inflated estimates of victimization risk and/or in strengthening perceived negative impact of victimization. On the basis of this model, and using findings from the 'fear appeal' literature together with victimological ideas and studies dealing with persuasion and describing fear of crime, the following three hypotheses about the relationship between fear and preventative communication were formulated and experimentally tested: (1) the message hypothesis (specifying conditions relating to the message that would cause fear); (2) the effectiveness hypothesis; and (3) the 'interaction' hypothesis (the latter two specifying conditions relating to the receiver that would cause fear). According to the message hypothesis, the likelihood of fear occurring is dependent on the communication strategy used. This hypothesis implies that reported fear will be highest with a negative communication strategy, lower if no information is provided (control group), lower still with a mixed strategy, and lowest with a positive strategy. Such an effect was observed both in reports of fear of crime and in the relevant cognitive products SVR and NI. According to the effectiveness hypothesis, fear of crime in the receiver varies with his or her perception of the effectiveness of the recommended preventative measures. This hypothesis implies that negative communication leads to more fear among receivers, who will experience the proffered advice as of little use; conversely, less fear is reported among those who hold the recommended measures to be effective. As part of the interaction hypothesis, it was predicted that negative communication will elicit fear particularly in female receivers, and in receivers with an external risk-orientation, but not among male receivers and 'internal controllers'. All of the above predictions extended to the fear-related cognitive products SVR and NI as well. The hypotheses were tested in an experiment (Winkel 1989?) which confirmed all 254


information might be more likely to stimulate extreme preventative responses, like installing complex alarm devices, or distrusting all strangers, in receivers who are emotionally or communicatively highly involved with the idea of crime, and/or who perceive the protection offered by the police as inadequate, the risk of victimization as high, or victimization as an outcome beyond their control. Given the empirical fact (Hanson-Frieze el al. 1987) that several of these characteristics can easily coincide in crime victims, these data suggest that victims are particularly likely candidates for carry-over effects to occur in response to the provision of preventative information. Winkel (1987a, b) and Winkel and Huismans (1986) provide some empirical evidence that generalization can be inhibited (and even that a reverse process of response block can be triggered) by means of what they termed a limitative communication strategy. In such a strategy, the communicator attempts to make clear to the receiver in one way or another that the preventative advice contained in the message is all that should be acted upon. The non-target set needs to be explicitly discussed with the receiver, making him or her aware that extreme preventative responses are not desired.


Skills and training

The processing of suspects and perpetrators is considered to be a traditional focus of police organizations. Moreover, it is regularly noted that 'thief-catching' is a dominant value in the locker-room culture, which takes it as constituting 'real police work' (Dunham and Alpert 1989; Toch 1986). From this perspective it is no great surprise to come across numerous victimological recommendations pointing to a fundamental need for special training programmes aimed at stimulating victim-orientated attitudes and behaviour in police personnel. Examples include Finn and Lee (1987), Skogan (1984), the APA Task Force on crime victims (1984), Rosenbaum (1987), Davis (1987), and the UN/Council of Europe documents mentioned above. A recurrent basic assumption in this literature is that current police behaviour is bad in that it often merely expresses a general tendency to blame victims, thereby creating a 'second wounding' (Van Dijk 1989; Steinmetz 1990; Rosenbaum 1987). However, some caveats are in order. First, the evidence for /w/ic^-induced 'secondary victimization', emanating from a psychological need to believe that people 'get what they deserve, or conversely, deserve what they get' (Lerner 1980) is mainly anecdotal (Winkel \988b). Moreover, empirical evidence suggests something different (Burgess and Holmstrom 1974). Second, the psychological reasoning underlying this reproach is often inaccurate, and at least incomplete. Stokols and Schopler (1973: 199) cogently argue that 'central to Lerner's hypothesis is the assumption that observers unable to find a behavioural link between a victim and a misfortune, will be motivated to supply a characterological justification for his suffering'. Thus, the assignment of responsibility to the victim encompasses either behavioural or characterological attributions. More recent psychological evidence (Winkel and Steinmetz, 1990; Janoff-Bulman and Hanson-Frieze 1983) suggests that characterological attribution has adverse psychological consequences on the victim, while the behavioural attribution may have a negative impact, depending upon such factors as the time at which it is done (e.g. immediately during reporting to the police) and the tone (cynical or not) in which the attribution is made. Thus, behavioural attributions by those around the victim, or even 255


three and suggested that the relationship between fear and communication is conditional rather than automatic. Both message-based characteristics and receiver-based characteristics (like risk-orientation, gender, and perceived effectiveness of the recommended measures) constituted relevant conditions for the (non-)occurrence of fear responses and related cognitive products. Together, these findings provide the following suggestions with regard to controlling fear of crime in response to the provision of crime-prevention information: (1) avoid using a negative communication strategy— that is, the communicator should not stress negative consequences (e.g. increased opportunity and victimization risks) emanating from non-compliance with the preventative advice; (2) use a mixed or positive communication strategy, so that side-effects are avoided and fear of crime reduced; (3) as all three hypotheses, in one way or another, touch upon the notion of the 'controllability of crime', this element should be expressly referred to by the communicator, for example, by stressing the relationship between adopting the suggested preventative tactics and positive behavioural consequences, the efficacy of the recommended measures, and the victim's own capacity to control risks of victimization (fostering self-efficacy and perceived internal control).

FRANS WILLEM WINKEL

256


behavioural self-blame by victims themselves, do not inherently trigger adverse psychological consequences. On the contrary, behavioural attributions may sometimes facilitate the victim's recovery, by providing an opportunity to re-establish control over the situation (Hanson-Frieze et al. 1987). These caveats, of course, do not refute the need for training programmes. Indeed, our own research, involving role-playing sessions with hundreds of police officers, suggests that officers often say or do the wrong thing to victims, or do the right thing at the wrong time. However, these findings (Koppelaar et al. 1990) do suggest that failures or errors in police officers' dealing with victims are generally of a much more subtle nature than a blatant blaming of the victim. Frequently encountered 'errors' include, inter alia, not providing (or sometimes even exaggerating, thus over-dramatizing) emotional support, not providing enough non-verbal communication, ordering the (psychologically necessary) consecutive phases of the interview process wrongly, and pressing a victim to tell his or her story in a strict (chrono)logical order at an early stage of the interview. In providing crime prevention information to victims I would identify skills in the following areas as particularly relevant: (1) interaction; (2) communication; and (3) prevention. Training programmes for police officers should therefore, ideally consist of at least three modules, one for each of these skill areas. This selection of modules mirrors the basic contention that the provision of crime prevention information should be conceived of as a process in which distinct phases have to be traversed correctly. The model approach with regard to victims of burglary, elaborated elsewhere (Winkel and Koppelaar 1988; Winkel 1989a), consists of seven phases: (1) establishing contact; (2) assessing the victim's psychological condition; (3) recognizing the victim's emotions; (4) correcting unwarranted notions, putting things in perspective; (5) presenting preventative recommendations; (6) conducting a security survey of the victim's property; and (7) terminating the conversation. A basic typology, available in the scholarly literature on social support (House 1981), is interwoven within this model: (a) the provision of emotional support (phases 2, 3, and 4); (b) the provision of informational support (phase 5) and (c) the provision of tangible or practical support (phase 6). Its relevance is exemplified injacobson's (1986) review of studies of support timing, implying that different kinds of support appear to be appropriate in different phases. One of his major conclusions is that 'out-of-sequence support will neither be effective, nor be recognized as helpful by the distressed individual' (Jacobson 1986: 254). The same behaviour, offered by others and intended to be supportive, may be seen as helpful by the recipient if provided at the right time, and as unhelpful if provided at the wrong time. Jacobson notes, for example, that it is unusual for informational support to be 'needed or appreciated prior to reducing the emotional arousal provoked by' a victimization (1986: 255). The typical temporal patterning of helpful interventions appears to be 'emotional support in the early stages and informational and practical support later on'. Within this framework, the capabilities required typically involve the psychological skills of providing emotional support, establishing rapport, and creating an empathetic atmosphere, and the techniques of 'cognitive restructuring' and stimulating existing processes of selective evaluation in victims. In this first module police officers (or victim assistance workers) are made aware of victims' most common emotional responses, such as fear and anger, and learn how to deal with these through conversational techniques


' A recent Dutch study (Korthals AJtej 1989) suggests that a majority of burglars (64 per cent) will return to the same neighbourhood after having committed a burglary. However, only a minority (about 30 per cent of those returning to the same neighbourhood) will return to the tame address, mainly because ofthc loot expected or the 'easy opportunity' offered. Other studies tuggest a temporal pattern in rc-viciimization risks (Polvi tl al. 1990). Because re-victimization studies (e.g. Forrester el al. 1988) arc only now beginning to emerge, some caution is needed here.

257


that provide an opportunity for the victim to ventilate thoughts and feelings about the crime experience. During the first three phases five points (NIMH training manual 1987; Koppelaar et al. 1990) are of crucial importance. First, look at the victim while listening. Research (Winkel and Vrij 1990) indicates that individuals who frequently look at their interlocutors are considered to be more credible, attentive, and competent, and more skilful socially. Second, maintain eye contact 'horizontally': for example, avoid conversations while you are standing and the victim is seated. Third, let the victim control the degree of eye contact; recognize that 'forced' eye contact might be experienced as threatening (Exline and Messick 1967). Therefore, avoid sitting directly opposite the victim, choosing instead a position at an oblique angle. Fourth, provide (non-)verbal responses while listening: nod and change facial expression, reflect ('you are saying that. . . am I correct?'), and paraphrase and summarize what has been said. Finally, 'follow' the victim's story and avoid interrupting, unless there is real confusion; be open and tolerant, that is, do not prejudge, moralize, or condemn the victim. Be willing to accept another person's feelings as important to him or her. The officer should not try to tell the victim how he or she should feel. However, particularly during phase 4, a role transition should take place, with the initiative gradually shifting from the victim to the officer. Typically, this phase brings cognitive restructuring techniques into play (see Folkman and Lazarus 1986 for a review). Basically, these techniques come down to attempts by the police officer to uncover and challenge victims' 'irrational, distorted or faulty' beliefs about themselves, others, the world, and the future, that are assumed to give rise to adjustment problems. For example, strong fear responses in burglary victims are sometimes rooted in the faulty idea that the perpetrator will soon return (statistical data 1 suggest that this likelihood is over-estimated) or in attributing a 'giant-like' status to the perpetrator, whereas in most cases offenders are relatively unsophisticated teenagers, taking advantage of an 'opportunity' offered. The officer's task here is to get the victim to give up these maladaptive modes of thought on which emotional distress is based in favour of more accurate ones. Another promising technique is for officers to try to stimulate and enhance existing tendencies for 'selective evaluation' (Winkel and Steinmetz 1990). Taylor, Wood, and Lichtman (1983) suggest that victims regularly try to get rid of adversely experienced victim status by selectively evaluating themselves and their situation in ways that are self-enhancing. In particular, five mechanisms of selective evaluation that minimize the perceived negative impact associated with being a victim of crime are postulated, such as making downward comparisons with less fortunate others, focusing on attributes that make one appear advantaged, creating hypothetical worse worlds, and construing benefit from the event. Phases 1-4 mainly centre on stimulating 'emotion-focused coping' (see Figure 1 and Folkman et al. 1986). Phase 5 signals the beginning of problem-focused coping, where the victim is prompted to take preventative action to avoid future victimization. Communication skills are particularly relevant to these latter phases. Key features of this module are to train officers to use a mixed or positive communication strategy,

FRANS WILLEM WINKEL

• conversational techniques - nonverbal communication - cognitive restructuring • stimulating selective evaluation

Emotion-focused coping (1) emotional support

- positive or mixed communication strategy • limitative strategy control-perspective (self-efficacy)

Problem-focused coping (2) informational support

— knowledge of modus operandi, opportunity character of crime; security terminology F I G . 1. Victim assistance: goal, type of support offered, and skills needed

and also to use a limitative communication strategy so that victims do not carry prevention to extremes. Negative strategies should be avoided, while the efficacy of the recommended measures and the victim's self-efficacy (internal control) should be stressed. These strategies have been discussed in the previous sections on effectiveness and side-effects. Officers are instructed to back up their oral preventative messages by simultaneously providing written information material, such as leaflets and brochures, and by trying to engage the victim's interest in this material. Previous research (Winkel 1990) suggests that communication impact is enhanced by imparting information through more than one channel, auditory as well as visual. A distinct skill relevant to phase 6 concerns the adoption of a standardized step-wise approach in giving preventative advice during a security check of premises. An example is to start with the perpetrator's place of entry and the surrounding 'wall openings'. Then the survey should move from that point to the other side of the house and upwards. Each wall opening should be dealt with systematically, for example, working from the upper to the lower side of windows, doors, etc. Here, too, officers are instructed to involve the victim by referring to written documents to illustrate what they are saying. The prevention module mainly involves the officers gaining a knowledge of home security. The programme attempts to broaden their insight into what prevention actually means and what opportunities there are for victims themselves to shape crime prevention. Attention is given to the opportunity character of petty crimes, to the modus operandi used by perpetrators, and to the technical, behavioural, and architectural preventative measures that might be considered. Officers should have a thorough knowledge of the preventative devices available, how they are technically described, how they can be implemented, where victims can get them, and how much they cost. A 258


— see (2); also — illustrating/referring to written documents — step-wise approach

(3) practical support


The outcomes: programme evaluations

Four evaluation studies have been carried out to examine the effects (and side-effects) of police communication programmes aimed at victims of burglary (Winkel and Koppelaar, 1987; Winkel 1989a, b,J). As half of these were literal replications, only two evaluations will be discussed in detail here. In the first, victims of burglary who had reported their victimization to the police were randomly assigned to a re-contact programme. Experimental subjects were approached by a couple of uniformed police officers, some time after reporting the crime. These victims were provided with oral and written information about preventing future burglaries, and a security check was made in and around their houses. All officers had participated in the training programme. However, trainees participating in programme 1 were exposed particularly to the interaction and prevention modules, while programme 2 participants were mainly exposed to the communication and prevention modules. A random sample of burglary victims who were not re-contacted served as a control group. Except for the noted training differences, both programmes had important common features, including re-contacting, the police officers' wearing uniforms, and personal, face-to-face contacts, while outcome measures encompassed both desired and sideeffects. Deviating from the 'standard procedure' of the Dutch police guidelines for dealing with victims, it was decided to provide preventative information through a personal re-contact in the victim's home, sometime after the first contact (reporting the crime to the police) had taken place (the guidelines assume this to happen during the first contact). Given its complexity, the combination of reporting and security survey was considered undesirable: not only should all sorts of technical details of the case have been raised at the first contact, but officers were assumed also to be providing the victim with information about compensation, restitution, how the police (or even the criminal justice system in general) would proceed with the case, and what victim support schemes were available. The risk of overburdening the victim was very real. Providing crime prevention information during the first contact might therefore easily have resulted in committing the 'neglected mediator fallacy' (McGuire 1981). This states that persuasive impact is seriously hampered if the victim is not given a real opportunity to be exposed to the preventative message, or is stopped from attending to 259


thorough knowledge of prevention issues is a sine qua non in making an expert and therefore credible impression on victims. Skills training programmes, incorporating those modules, entail more than simply addressing some basic victimological and social-psychological notions through lectures. Essential ingredients are exposing trainees to specially constructed videos depicting different types of approach and victim-officer interactions, actively discussing those fragments and evaluating the officer's own and others' styles of approaching and informing victims, and performing role-playing sessions with professional actors and actresses as victims. A substantial set of programmes have been designed, during the last decade, by the Circon group for training development. A recent evaluation in approximately 150 Dutch police forces, performed by an independent research team (Driessen 1989), suggests that both the content and the format of these programmes are generally appreciated and positively assessed by a large majority of police officers.

FRANS WILLEM WINKEL

Desired outcomes

Results of re-contacts

• Satisfaction with the police

• Stronger satisfaction

• Attitude towards the police

• More positive attitude: police perceived as being more helpful, caring, and committed

• Agreement with the 'crime as opportunity' perspective

• Stronger acceptance

•

• Stronger protection

Perceived police protection against burglary

• Risk-orientation

• Shift to 'internal' locus of control

•

• Stronger willingness to take preventative measures

Behavioural intentions

F I G . 2. Desired and achieved effects of programmes 1 and 2

In comparing the programmes with regard to side-effects, a different picture emerges: outcomes clearly diverged. In programme 1 both types of side-effects emerged. There was response generalization: re-contacted victims reported relatively more attitudes (and intentions) indicative of extreme forms of preventative behaviour. And although, with this programme, no significant differences between experimental and control subjects emerged on the scale of overall fear of crime, side-effects showed up in particular sub-scales: re-contacted victims reported a higher fear of property 260


it, or from comprehending and accepting it (forming basic mediators of the ultimate desired behavioural impact). Although crime prevention officials, who form part of the specialized units within the Dutch police organization, do not usually wear uniforms, it was decided to deploy uniformed officers on the basis of several previous studies (Winkel 1986, 19876, 1988), which suggested that, at least with regard to the general public, uniforms represent reassurance: the visible presence of the police tended to have a fear-reducing effect. Our preference in the evaluation for personal, face-to-face contacts was a response to the rather disappointing results obtained by Skogan and Wycoff (1987) in their evaluation of re-contacting victims by telephone, while our emphasis on crime prevention as an element of victim assistance came from suggestions about effective services for victims recently formulated by Davis (1987). As regards desired effects, the two programmes revealed identical pictures. The victims, are summarized in Figure 2. In both evaluations re-contacts were generally very positive. In programme 1 an overwhelming majority felt the re-contact to be useful (94 per cent) or worthwhile (91 per cent). About 88 per cent of the victims re-contacted were of the opinion that re-contacting should have top priority within the major outcomes, reflecting significant differences between re-contacted and control police organization. The same pattern was visible in programme 2. Here, 89 per cent of the re-contacted victims characterized the visit as useful and 85 per cent as worthwhile. Moreover, as shown in Figure 2, both programmes resulted in a number of positive outcomes.


Discussion

The above analyses suggest that the provision of crime prevention information may be considered an important component of victim support if that information is supplied during a re-contact, on the basis of a personal, face-to-face meeting between a couple of uniformed officers and a victim. Police officers (and victim assistance workers in general) can enhance the persuasive impact of their preventative messages by using a mixed communication strategy, by stressing the victims' own ability to take measures and the efficacy of the measures recommended, and, as far as normative communication is concerned, by making an expert, reliable, objective, and honest impression. Communication side-effects, such as increased fear of crime and response generalization, can be controlled by using a mixed or positive communication strategy. A negative strategy of communication should be avoided. Proper assistance needs fully to take into account the distinction between emotion-focused and problem-focused coping processes. Officers should be aware that out-of-sequence support is not perceived as helpful: on the contrary, it merely confuses the victim. Skills are needed to stimulate these coping processes effectively. Such skill may be learned and strengthened through active participation in training programmes, through exposing trainees to lectures or 261


crime (burglary, vandalism, car theft, and theft from cars) and stronger fears of burglary and vandalism. In programme 2, however, all of these side-effects were absent. Together these outcomes suggest that running an assistance programme for victims of burglary is not a risk-free operation: under certain conditions programmes may backfire. Programme 2 outcomes clearly suggest, however, that this need not happen; side-effects can be avoided. A comparison of the two programmes suggests that avoiding undesirable side-effects depends on giving the participating police officers the appropriate training. In programme 1 the communication module was relatively under-exposed, while communication skills were strongly emphasized in programme 2, which avoided side-effects. It seems advisable, therefore, to strive for a sophisticated balance between interaction, communication, and prevention modules in future training programmes. In both programmes some interesting features emerged. In programme 1, fear reduction and the opposite of response generalization occurred in those victims who were characterized by 'high prior life stress'—that is, victims who had been confronted with 'negative life events' before their present victimization—and also in victims who reported having had financial, emotional, physical, or relationship problems during the previous year. These types of victims apparently obtained 'emotional benefit' from the programme (see also Cook et al. 1987). Increased fear and response generalization were visible pre-eminently in the opposite segment of victims: those who did not report having had prior life problems. To these people, their present victimization was something completely new. In programme 2 a slight increase in fear of crime was reported in response to the re-contact by victims who had an external risk-orientation, and also in female victims (statistical trend). Together these features underlie the need to broaden our knowledge of relevant psychological individual differences between victims. Awareness of these differences will enable us to provide better answers to the questions who needs what (sort of assistance), from whom, and at what time.

FRANS WILLEM WINKEL

REFERENCES

R. B. (1989), 'A Federal Perspective on Victim Assistance in the United States of America', in Viano, Crime and its Victims: 213-27. APA (1984), American Psychological Association Task Force on the Victims of Crime and Violence, Final Report. Washington, DC: APA. BUROESS, A. W., and HOLMSTROM, L. L. (1974), 'Rape Trauma Syndrome', American Journal of ABELL,

Psychiatry, 131:981-5. COOK, R. F., SMITH, B. E., and HARRELL, B. E. (1987), Helping Crime Victims: Levels of Trauma

and Effectiveness of Services. Washington, DC: National Institute of Justice. COUNCIL OF EUROPE (1988), Assistance to Victims and Prevention of Victimisation, Recommendation

R (87) 21 and Explanatory Memorandum. Strasbourg: Council of Europe. DAVIS, R. C. (1987), 'Studying the Effects of Services for Victims in Crisis', Crime and Delinquency, 33/4: 520-31. DRIESSEN, F. (1989), Gebruik van de training conflicthantering door korpsen gemeenttpolitie, districten

rijkspolitie en politiescholen, Rapport t.b.v. de Directie Politie, Ministerie van Binnenlandse Zaken, Utrecht. DUNHAM, R. G., and ALPERT, G. P., eds. (1989), Critical Issues in Policing: Contemporary Readings. Prospect Heights, 111.: Waveland. 262


video-taped interactions, and in role-playing sessions with victims. Ideally, these training programmes should encompass prevention, communication, and interaction modules. The analyses also suggest that decisions regarding the referral of victims to re-contact programmes, or to other forms of victim assistance, should not be based only on the type of crime experienced, nor just on relevant biographical or sociographical data. Psychological data such as the victim's perceived locus of control for victimization risks, prior life stress, and past year life problems indicate the amount of psychological distress induced by the victimization, and should also be taken into consideration. Finally, it must be noted that in order to replicate these programmes elsewhere, the police should be in a position to monitor carefully local crime trends and the modus operandi used by perpetrators. To this end, many police forces in The Netherlands have now appointed 'crime analysts', who belong either to special 'crime investigation units' or to one of the twenty-three regionally organized crime prevention units. Together with one of these units we recently finished a study (Vrij et al. 1990) which examines more closely the impact of these programmes on local burglary rates. A one-year prevention programme consisting of a package of burglary prevention measures (distributing burglary prevention leaflets, security visits to citizens, victim re-contacts, and increased police surveillance on request during holidays), targeted at two high burglary-rate areas of Grootebroek and Bovenkarspel, resulted in a 50 per cent drop in burglaries reported to the police, compared to a 19 per cent drop in the control (no special programme) neighbourhoods. These outcomes are in line with studies suggesting that improving home security will result in reducing re-victimization risks (Forrester et al. 1988) and Hough and Mo's (1986) analysis suggesting that targethardening prevents attempted burglaries from becoming completed burglaries. However, one of the issues that urgently needs empirical clarification in future studies concerns the psychological impact of re-victimization on re-contacted victims.


R., and MESSICK, D. (1967), 'The Effects of Dependency and Social Reinforcement upon Visual Behavior during an Interview', British Journal of Social and Clinical Psychology, 6: 256-66. FINN, P., and LEE, B. N. W. (1987), Serving Crime Victims and Witnesses. Washington, DC: US Department ofjustice, National Institute ofjustice, Office of Communication and Research Utilization. FISHBEIN, M., and AJZEN, J. (1980), Understanding Attitudes and Predicting Social Behavior. Englewood Cliffs, NJ: Prentice-Hall. FOLKMAN, S., LAZARUS, R. S., DUNKEL-SCHETTER, C , DELONGIS, A., and GRUEN, R. J. (1986), 'Dynamics of a Stressful Encounter: Cognitive Appraisal, Coping, and Encounter Outcomes', Journal of Personality and Social Psychology, 50/5: 992-1003. FORRESTER, D., CHATTERTON, M., and PEASE, K. (1988), The Kirkkolt Burglary Prevention Project, Rochdale, Home Office Crime Prevention Unit paper no. 13. London: Home Office. GEVA, R. (1988), 'Feelings of Insecurity and Fear and Co-operation with the Police: The Israeli Experience', paper presented at the 6th International Symposium on Victimology, Jerusalem. HANSON-FRIEZE, I., HYMER, S., and GREENBERO, M. S. (1987), 'Describing the Crime Victim: Psychological Reactions to Victimization', Professional Psychology; Research and Practice, 18/4: 299-315. HOUGH, J. M., and Mo, J. (1986), 'If At First You Don't Succeed', Home Office Research Bulletin 21: 10-13. HOUSE, J. S. (1981), Work Stress and Social Support. Reading, Mass.: Addison-Wesley. JACCARD, J. (1981), 'Attitudes and Behavior: Implications of Attitudes toward Behavioral Alternatives', Journal of Experimental Psychology, 17: 286-307 JACOBSON, D. A. (1986), 'Types and Timing of Social Support', Journal of Health and Social Behavior, 27: 250-64. JANOFF-BULMAN, R., and HANSON-FRIEZE, I. (1983), 'A Theoretical Perspective for Understanding Reactions to Victimization', Journal of Social Issues, 39/2: 1-17. KAHNEMAN, D., and TVERSKY, A. (1984), 'Choices, Values, and Frames', American Psychologist, 39/4: 341-50. KOPPELAAR, L., VAN DER STEEN, J. C , VRIJ, A., and WINKEL, F. W. (1990), 'Angst na verkrachting: slachtoffergeorienteerde vaaridgheden bij de politie', in Winkel and Wurff, Angst voor Criminalileit: 112-27. KORTHALS ALTES, H. J. (1989), Eindrapportage Modus operandi-onderzoek Woninginbraken, deel A: onderzoeksresultaten. 's Gravenhage, Bureau Criminaliteitspreventie, Landelijk Bureau Voorkoming Misdrijven. LAZARUS, R. S., and FOLKMAN, S. (1984), Stress, Appraisal and Coping. New York: Springer. LERNER, M. J. (1980), The Belief in a Just World: A Fundamental Delusion. New York: Plenum. MCGUIRE, W. J. (1981), 'Theoretical Foundations of Campaigns', in R. E. Rice and W. J. Paisley, eds., Public Communication Campaigns: 15—41. London: Sage. MAOUIRE, M., and CORBETT, C , eds. (1987), The Effects of Crime and the Work of Victim Support Schemes. Aldershot: Gower. NIMA (National Institute of Mental Health) (1986), Training Manualfor Human Service Workers in Major Disasters, no. 86-853. Rockville, Md.: US Dept. of Health and Human Services. EXUNE,

Victimisation: Time Course and Crime Prevention Opportunity', Journal of Police Science and Administration, 17/1: 8-11. 263


POLVI, N., LOOMAN, T., HUMPHRIES, C , and PEASE, K. (1990), 'Repeat Break-and-Enter

FRANS WILLEM WINKEL

D. P. (1987), 'Coping with Victimization: The Effects of Police Intervention on Victims' Psychological Readjustment', Crime and Delinquency, 33/4: 502-19. SHAPLAND, J., WILLMORE, J., and DUFF, P. (1985), Victims in the Criminal Justice System (Cambridge Studies in Criminology). Aldershot: Gower. SKOGAN, W. G. (1984), Recent Experiments in Fear Reduction: A Summary of Research Conducted by the Police Foundation 1983-1984. The Hague: R.D.C. and WYCOFF, M. A. (1987), 'Some Unexpected Effects of a Police Service for Victims', Crime and Delinquency, 33/4: 490-502. SOETENHORST DE SAVORNIN LOHMAN, J., ed. (1989), Slachtoffers van misdrijven: ontwikkelingen in hulpverlening, recht en beleid. Arnhem: Gouda Quint. STEINMETZ, C. H. D. (1990), De hulpverlening aan slachtoffers van ernslige misdrijven, dissertatie. Arnhem: Gouda Quint. STOKOLS, D., and SCHOPLER, J. (1973), 'Reactions to Victims under Conditions of Situational Detachment', Journal of Personality and Social Psychology, 25: 199-209. TAYLOR, S. E., WOOD, J. V., and LICHTMAN, R. R. (1983), 'It Could Be Worse: Selective Evaluation as a Response to Victimization', Journal of Social Issues, 39/2: 19—40. TOCH, H., ed., Psychology of Crime and Criminal Justice. Prospect Heights, 111.: Waveland. TSITSOURA, A., 'Assistance to Victims and Prevention of Victimisation: Recent Activities of the Council of Europe', in Viano, Crime and its Victims: 205-13. UNITED NATIONS (1986), Declaration of Basic Principles of Justice for Victims of Crime and Abuse of Power. New York: UN Dept. of Public Information, August. VAN DIJK, J. J. M. (1989), 'Eigen Schuld, dikke bult', in Soetenhorst de Savornin Lohman, Slachloffers van misdrijven: 11-25. VIANO, E. C , ed. (1989), Crime and its Victims: International Research and Public Policy Issues. New York: Hemisphere. VRIJ, A., WINKEL, F. W., FOPPES, J. H., and VOLOER, D. M. (1990), 'Politiele preventieprogramma's en angst voor criminaliteit: het rendement van een planmatige aanpak', in Winkel and Wurff, Angst voor Criminaliteit: 179-86. WALLER, I. (1984), 'Assistance to Victims of Burglary', in R. Clarke and T. Hope, eds., Coping with Burglary: Research Perspectives on Policy: 233—49. Boston: Kluwer. WINKEL, F. W. (1981), 'Angst voor criminaliteit; verklarende modelen', Tijdschrift voor Criminologie, 23/2: 53-73. (1984), 'Changing Misconceptions about Rape through Informational Campaigns', Viclimology, 9/2: 262-72. (1986), 'Reducing Fear of Crime through Police Visibility: A Field Experiment', Criminal Justice Policy Review, 1/4: 381-98. (1987a), 'Response Generalization in Crime Prevention Campaigns: An Experiment', British Journal of Criminology 27/2: 155-74. (1987A), Politie en Voorkoming Misdrijven: effecten en neveneffecten van voorlichting. Amsterdam: Mens & Recht. (1988a), 'The Police and Reducing Fear of Crime: A Comparison of the Crime-centred and Quality of Life Approaches', Police Studies, 11/4: 183-90. (1988A), 'Rape Reporting to the Police: Testing the Social-psychological Impact of a Persuasive Campaign on Cognitions, Attitudes, Normative Expectations and Intentions', paper presented at the 6th International Symposium on Victimology, Jerusalem. (1989a) 'Preventieve voorlichting als vorm van slachtofferhulp: de kunst van het voorlichten', in Soetenhorst de Savornin Lohman, Slachtoffers van misdrijven: 69-97. ROSENBAUM,


264


Setting', Social Behaviour: An International Journal of Applied Social Psychology, 5/4: 335—50.

and VAN DER WURFF, A., eds. (1990), Angst voor Criminaliteit: Theorie, Onderzoek en Interventie. Amsterdam: Swets and Zeitlinger.

265


(1989A), 'Responses to Criminal Victimisation: Evaluating the Impact of a Police Assistance Programme and some Social-psychological Characteristics', Police Studies, 12/2: 59-74. (1989c) 'The Effectiveness of Crime Prevention Campaigns: Testing some Socialpsychological Guidelines on Message Design', Journal of Police and Criminal Psychology, 5/2: 11-21. (\989d), 'Cognitieve en affecticve readies op preventieve voorlichting: de rol van informatie- en ontvangerkenmerken', Gedrag en Organisatie, 2/3: 149—64. (1989«J, 'Increased Fear of Crime and Related Side Effects of Persuasive Communications: The Price Tag of Burglary Prevention Campaigns?', in Viano, Crime and its Victims: 273-99. (1989/), 'Crime Prevention and Victim Assistance: An Experiment on Police Communication to Burglary Victims', paper presented at the 41st Annual Meeting of the American Society of Criminology, on 'Crime in Social and Moral Contexts', Reno, Nevada. (1990), 'Massamedia, voorlichting en angstgevoelens voor criminaliteit: een informatieverwerkingsmodel', in Winkel and Wurff, Angst voor Criminaliteit: 47-75. and HUISMANS, S. E. (1986), 'Refutational Messages on Donor Cards: A Test of Boomerang Effects', Psychological Reports 59: 899-910. and KOPPELAAR, L. (1988), 'Police Information for Victims of Crime: A Research and Training Perspective from The Netherlands', Police Studies, 11/2: 73-81. and STEINMETZ, C. H. D. (1990), 'Angst voor criminaliteit en slachtofferschap: een schokverwerkingsmodel', in Winkel and Wurff, Angst voor Criminaliteit: 22—47'. and VRIJ, A. (1990), 'Interaction and Impression Formation in a Cross-cultural Dyad: Frequency and Meaning of Culturally Determined Gaze Behaviour in a Police Interview