Heterogeneity in victim participation: A new perspective on delivering a Victim Impact Statement

Abstract

A central question in the debate about victim participation in criminal justice procedures is which instrument available to victims ‘works’. The purpose of the present study was to examine which factors contribute to the likelihood of victims delivering a Victim Impact Statement (VIS). We extend previous research in two important regards. First, we examined victims’ perspectives on the purposes and function of the VIS. Consistent with previous research (for example, Roberts and Erez, 2004), we reveal a distinction between impact-related and expression-related use of VISs. However, this study adds a third component to the existing literature: the anticipation of negative consequences. Second, we examined which factors influence the likelihood of delivering a VIS and found three variables to be positively associated: posttraumatic stress symptoms, the type of crime, and the time of victimization. Against expectations, victims’ perspectives did not make a unique contribution to the model. Based on these findings, we argue that what is called for is a more heterogeneous approach to the study of procedural instruments available to victims.

Keywords

Crime severity Victim Impact Statement victim participation victim perspectives

Since the 1970s, the position of victims in the criminal justice system has been strengthened in many countries (Groenhuijsen and Letschert, 2008; Hall, 2010). Whereas 30 years ago it was correct to assert that the victim was the forgotten party of the criminal justice process, today such an assertion would be at odds with reality. The upsurge of the victim of crime has spurned a considerable amount of literature on both the advantages and disadvantages of granting victims participatory rights (for example, Ashworth, 2000; Erez, 1999; Groenhuijsen, 1999; Sarat, 1997) and the effectiveness of different victim-oriented measures (for example, Chalmers et al., 2007; Roberts and Erez, 2010; Roberts, 2009; Sanders et al., 2001; Sherman and Strang, 2007).

Research evaluating the contribution of victims’ rights to victims’ well-being tends to view victims of crime as a homogeneous group (Green, 2007; Pemberton et al., 2007). Where distinctions are made within the victim population, they primarily concern crime characteristics (for example, distinguishing victims of violent crime from victims of property offences) or demographic variables (for example, gender and age of the victim) (Pemberton and Reynaers, 2011). Often, references are made to ‘what victims want’ (for example, Strang, 2002; Braithwaite, 2002) and to whether measures ‘work’ or ‘don’t work’ for participating victims (compare Sanders et al., 2001, with Chalmers et al., 2007).

In contrast, the psychological and therapeutic literature widely acknowledges that individual differences in psychological characteristics, personality traits and victimization context have a large influence on victimization experience (for example, Bonanno et al., 2011), to the extent that two demographically matched victims of identical crimes may display different reactions and as a consequence may have different needs (Brewin et al., 2000; Ozer et al., 2003). In other words, in therapeutic approaches it makes obvious sense to match treatment to the individual circumstances of the victim. This article argues that this ‘matching’ is also relevant to the intersection between criminal justice and victims of crime. Rather than investigating whether an instrument ‘works’ for victims in general, it is often more fruitful to investigate for which victims or under which conditions an instrument works. The instrument of interest in the present article is the class of measures referred to as the Victim Impact Statement (VIS).

The Victim Impact Statement: Purposes and functions

Victim participation in criminal justice proceedings can take different forms. One of the instruments that enables victims to participate in the criminal justice procedure is the right to deliver a VIS. Although the precise form of the VIS can vary – from a written statement that primarily serves a function in awarding compensation to an oral statement that may influence the sentence given to the offender (also referred to as a Victim Statement of Opinion) – all have in common that they allow victims the right to express the harm they have experienced as a part of the court proceedings (Erez, 2004). In the Netherlands, the right to deliver an oral VIS was afforded to victims in 2005. The implementation of this right was accompanied by the possibility for victims to submit a written VIS, which is added to the file of the criminal case. In the Netherlands, the content of the VIS is restricted in the sense that victims can speak only about the consequences of the crime, and are not allowed to speak about the facts of the crime or a desired punishment.

The Dutch Explanatory Memorandum of the Oral Victim Impact Statement Act mentions four goals. First, a VIS may contribute to the information provision to the trial judge: the VIS may help judges in imposing a ‘just’ sentence (Roberts and Erez, 2004). Some even argue that a sentencing decision that does not take into account victim harm is incomplete and unfair (Edwards, 2004). The second and third goals refer to its preventative purposes, which can be, on the one hand, general (that is, establishing societal norms) and, on the other hand, specific (that is, decrease the relapse risk of the suspected offender). The fourth goal of the Act concerns the contribution of the VIS to the victim’s emotional recovery. In particular, it is assumed that the delivery of a VIS may facilitate recovery from the emotional harm that has been caused by the crime. However, the effectiveness of a VIS in facilitating recovery is widely debated at a theoretical level (Pemberton and Reynaers, 2011). Whereas some argue that the VIS is effective in helping victims to recover from the crime, others suggest that the VIS may even be counterproductive, in the sense that it may lead to secondary victimization. This is exemplified in contradictory statements such as ‘VIS can work, do work (for those who bother to make them)’ (Chalmers et al., 2007) and ‘VIS, don’t work, can’t work’ (Sanders et al., 2001).

Victims’ perspectives on criminal justice

The relationship between victims of crime and criminal justice instruments (such as the VIS) is complicated. This becomes evident when comparing the legal purposes and functions of a particular instrument available to victims with victims’ perceptions of that instrument. For example, victims may expect that their VIS will actually affect the outcome of the process, when this may not be so (Edwards, 2001; Lens et al., 2010). In a study by Lens et al. (2010), around 50 percent of the victims who delivered an oral VIS in court declared that one of their motivations was to influence the sentence. However, influencing the sentence given to the offender is not mentioned as one of the goals in the Explanatory Memorandum of the Oral Victim Impact Statement Act. Moreover, victims may differ in their views on the meaning and purposes of criminal justice procedures, which makes ‘matching’ an instrument to the victim’s needs complicated: a VIS may function as a means to award compensation, to reduce secondary victimization, to facilitate communication with the offender and/or allow the court to consider more closely the human costs of the crime at sentencing (Erez, 2004; Roberts, 2009; Roberts and Erez, 2010). Previous research has repeatedly shown that the impact of crime, often defined in terms of posttraumatic stress, anger and/or anxiety (for example, Orth and Maercker, 2009) may influence victims’ perspectives on criminal justice procedures and outcomes. For example, increased levels of posttraumatic stress and anxiety will reduce the ability of victims to engage directly with the offender in face-to-face mediation (Cheon and Regehr, 2006; Winkel, 2007). Moreover, the psychological impact of crime is associated with increased feelings of hostility (Orth and Wieland, 2006), revenge and retaliation (Cardozo et al., 2003; Orth et al., 2006) and punitiveness (Canetti-Nisim et al., 2009; Litvak et al., 2010; Pham et al., 2004), which suggests that the experienced severity of crime heightens the importance of retributive justice and consequently of the appropriateness/severity of the punishment meted out to the offender (see also Gromet and Darley, 2009; Tripp et al., 2007; Wenzel et al., 2008). Conversely, improved adjustment and coping are associated with forgiveness (Orth et al., 2008), which in turn is related to a more conciliatory stance towards the offender (Armour and Umbreit, 2005) and an increased emphasis on value-restorative outcomes (Wenzel et al., 2008).

In sum, the heterogeneity of victims’ crime experiences expresses itself both in the victims’ psychological characteristics and in their perception of instruments available to them in criminal justice procedures. In turn, both of these factors may determine whether or not victims choose to use the participatory rights offered to them in the criminal justice system. Both factors are explored in this article. Using a sample of victims of violent crime (N = 170) eligible to submit a Victim Impact Statement, this article examines the impact of the background characteristics of the victim and the crime and the impact of victims’ perspectives on the purpose and function of VISs on the likelihood of their submitting a VIS. Two hypotheses are formulated. First, we hypothesize that background characteristics of the victim and the crime influence the likelihood of delivering a VIS. More specifically, and in line with previous research (Lens et al., 2010; Leverick et al., 2007; Roberts, 2009), we predict that the impact of the crime on the victim will increase the likelihood of delivering a VIS (Hypothesis 1). Second, we expect that victims’ perceptions of the VIS make an additional contribution to the likelihood of delivering a VIS, above and beyond the influences of the crime’s impact on the victim (Hypothesis 2). More specifically, we predict that the anticipated negative consequences of delivering a VIS will negatively influence the likelihood of delivering one, whereas the anticipated positive consequences of delivering a VIS will increase the likelihood of delivering one.

Methods

Participants

Over a period of 12 months, 319 victims eligible to submit a VIS were invited to participate in the study. Eventually, 170 victims (53.3 percent) filled out the questionnaire. As no background information on the non-participants was available, possible non-response bias could not be estimated.

Procedure

Participants were recruited through all 19 district court offices of the Dutch Public Prosecution Service. To participate in the study, victims had to be eligible to submit a VIS and have sufficient mastery of the Dutch language. Furthermore, they had to face an upcoming trial. Potential participants received a letter with information about the survey from the Prosecution Service and the possibility that they would be further contacted by the investigators. A form was included through which they could opt out of receiving further information. If this form had not been returned within two weeks, victims were contacted by telephone and invited to participate. All participating respondents filled out a letter of informed consent. Participants were assured that there were no ‘right’ or ‘wrong’ answers and that the results of the study would be treated confidentially.

Design

The current study was part of a larger study into the effects of delivering a VIS on the emotional recovery of the victim. The original, longitudinal survey used a quasi-experimental, pre-test/post-test design. This design allowed for a comparison of the victim’s situation before and after the trial and between the two subgroups: (1) delivery of VIS: those who submitted an oral and/or written VIS; and (2) no delivery of VIS: those who declined to make a VIS, either oral or written. Respondents were asked to fill out two structured questionnaires: the first one two weeks before the trial in their case (pre-test), and the second one two weeks after (post-test). This article reports the results of the pre-test.

Measures

The questionnaire contained the following constructs: demographics, crime features, psychological characteristics of the victim, and victims’ perceptions of the VIS.

Demographics and crime features

Respondents were asked for the following demographics: gender, age, marital status, education level (lower/intermediate/higher education), employment (yes/no) and ethnic background (determined by the participant’s own and their parents’ birthplace). Furthermore, respondents had to indicate certain features of the crime that had been committed: the type of crime, the time that had elapsed since the commission of the crime (in months), their relationship (if any) to the offender, earlier victimization (if any), and victim vs. co-victim (of homicide). With the latter distinction (victim vs. co-victim), respondents were classified as either ‘directly’ or ‘indirectly’ harmed by the crime. According to the Declaration of the UN Commission on Human Rights, indirect victims are the family members of a direct victim. In the current study, this same classification was adopted.

Psychological characteristics

Following previous research (for example, Orth and Maercker, 2009), the psychological impact of the crime on the victim was measured by feelings of posttraumatic stress, anger and anxiety.

Posttraumatic stress

Using a forward and backward translation method, the Trauma Screening Questionnaire (TSQ; Brewin et al., 2002) was translated into Dutch and used to measure indications of posttraumatic stress disorder (PTSD). This validated, self-report screening tool has been adapted from the PTSD Symptom Scale-Self Report (Foa et al., 1993). The TSQ consists of 10 items that require straightforward ‘yes’ or ‘no’ responses. Five items concern re-experiencing of traumatic events, such as ‘Upsetting thoughts or memories about the event that have come into your mind against your will’. The remaining five items concern symptoms of arousal, such as ‘Heightened awareness of potential dangers to yourself and others’. A sum score of the TSQ is computed by adding the scores of the responses (yes = 1, no = 0), creating a continuous variable. The cut-off point of the TSQ is five, with six or more ‘yes’ responses indicating possible PTSD (Brewin et al., 2002). Cronbach’s alpha for the TSQ was found to be 0.88, indicating good reliability (Murphy and Davidshofer, 1998; Nunnally, 1978).

Anger

A Dutch translation of the seven-item Dimensions of Anger Reactions Scale (DAR; Novaco, 1975) was used to measure feelings of anger. This validated, self-report scale consists of seven items that are answered on a nine-point Likert scale, ranging from 0 (‘not at all’) to 8 (‘exactly so’). Sample items include ‘When I get angry I stay angry’ and ‘My anger has had a bad effect on my health’. Test-retest reliability of the Dutch translation of the DAR was examined by Nederlof et al. (2009): they found a correlation coefficient of .84, supporting the reliability of the scale. The DAR provides an indicator of key aspects of anger dysregulation, including frequency, intensity, duration, violent expression and problematic consequences for psychosocial functioning and well-being (Forbes et al., 2004). The instruction preceding the DAR was altered to ensure that respondents would report anger post victimization, rather than anger per se. The author of the DAR approved the appropriateness of this adaption (for example, Kunst et al., 2011). Respondents were asked to indicate whether or not they had experienced feelings of anger in the previous two weeks. A sum score of the DAR is computed by adding up the scores of all responses. Cronbach’s alpha was found to be 0.89, indicating good reliability (Murphy and Davidshofer, 1998).

Anxiety

Feelings of anxiety were measured with the anxiety subscale of the Hospital Anxiety and Depression Scale-Dutch version (HADS-NL). The HADS-NL was translated from English (Zigmond and Snaith, 1983) into Dutch by Pouwer, Snoek and Van der Ploeg in 1983 by permission of the publishers and validated with Dutch respondents. Test-retest reliability of the total HADS and both subscales were found to be good in different groups of Dutch subjects (Spinhoven et al., 1997). The anxiety subscale consists of seven self-report items that are answered on a four-point Likert scale. A sample item is ‘I can sit at ease and feel relaxed’. Scores range from 0 to 3, with a total score of 9 or more indicating a psychiatric state of anxiety. Cronbach’s alpha for the subscale anxiety was found to be 0.91, indicating good reliability (Murphy and Davidshofer, 1998).

Victims’ perceptions of the VIS

Respondents were asked to score 16 perceptions regarding the purpose or the consequences of submitting an oral or written VIS on a five-point Likert scale, ranging from 1 (‘totally disagree’) to 5 (‘totally agree’). The items were partially derived from the distinction Roberts and Erez (2004) made between impact-related and communicative/expression-related use of VIS. In their first model, the instrumental model, the key issue is the effect of the statement on the sentence imposed on the offender. In the second model, VISs are viewed as vehicles of expression, and the central issue is the victims’ communication with other participants in the criminal justice process, whether that is the judge, the prosecutor or the defendant (Roberts and Erez, 2004). Sample items of impact-related use of the VIS included: ‘I expect the VIS to influence the sentence given to the offender’, ‘I expect the VIS to receive a lot of attention during the process’, ‘I expect the VIS to help me get more understanding from the judge(s)/public prosecutor’, and ‘I expect the VIS to positively influence my entitlement to compensation’. Expression-related use of VIS was measured with items related to both emotional recovery and (positive influences of) creating understanding. These items included: ‘I expect the VIS to positively influence my emotional recovery’, ‘I expect the VIS to help me get more understanding from the offender’, ‘I expect the VIS to be an emotional burden’ (reversed in analyses), ‘I expect the VIS to prevent recidivism’, ‘I expect the VIS to have a positive influence over the expressed emotions/sorrow of the offender’, ‘I expect the VIS to help me get more understanding from “others” present in the courtroom (for example, family members, press)’, and ‘I expect the VIS to have no influence over my emotional recovery’ (reversed in analyses). Additionally, to measure anticipated negative consequences of participating in the criminal justice procedure (for example, Herman, 2003), several items were added to the distinction made by Roberts and Erez (2004): ‘I expect the offender to get angry at me (after delivering a VIS)’, ‘I expect negative reactions from third parties (after delivering a VIS)’, ‘I expect the offender to take revenge (after my delivery of a VIS)’, ‘I expect the VIS to negatively influence the process’, and ‘I expect the VIS to misrepresent the case’.

Statistical analyses

As a first step in our analyses, characteristics of the victim and the crime were examined. Means and standard deviations (SD) were computed for continuous variables, and percentages were presented for categorical variables. Second, a principal component analysis (PCA) with orthogonal rotation was conducted on victims’ perceptions of the VIS. Third, a binary logistic regression analysis was conducted to analyse the relative contribution of each variable to the likelihood of delivering a VIS. Odds ratios were calculated from the logistic regression coefficients to provide an estimate of the likelihood of submitting a VIS. In preparation for the logistic regression analysis, and to simplify and clarify the interpretation of the coefficients in this analysis, the continuous independent variable ‘type of crime’ was recoded into six categories: threat, stalking, sexual offences, homicide (surviving relative), violent crimes (for example, attempted murder, robbery, grievous bodily harm, hostage taking) and traffic offences (being guilty of a serious and/or fatal accident). Furthermore, extreme outliers in the variable ‘time elapsed since the commission of the crime’ were removed for reasons of clarification. Removal of these outliers did not have a significant effect on the model. Before conducting the multivariate logistic regression analysis, Pearson’s product-moment correlation coefficients between the selected variables and the delivery of the VIS were calculated. Variables with significant univariate correlations with the target dependent variable (delivery of VIS; p < .05) were retained for logistic regression analysis and entered into the equation simultaneously.

Results

Demographics

The total sample consisted of 170 victims of serious crimes (73 men and 97 women) who were eligible for delivering a VIS. The respondents’ age varied from 14 to 91, with a mean age of 37.1 years (SD = 14.2). Minors who wanted to participate in the study were asked to fill out the questionnaire with the help of one or both of their parents. Forty-four percent of the respondents reported not having a partner, and 24.1 percent were married. Of the remaining 31.8 percent, 15.3 percent reported living with a partner, 14.1 percent reported being divorced, and 2.4 percent reported being a widow/widower. About half (58.2 percent) of the participants completed intermediate education, 20.6 percent completed lower education and 21.2 percent completed higher education. A majority of the participants held a paid job (66.5 percent), and the other 33.5 percent of the respondents did not. Seventeen percent of the respondents were of non-Dutch origin (for example, Moroccan, Chinese or Polish).

Crime features

Our total sample (N = 170) consisted of 159 victims, 4 co-victims of homicide, and 3 family members of a victim who was killed in a car accident; in the remaining four cases, an underage victim was represented during the trial by one of his parents. Crime types were grouped into six categories: threat (n = 36, 21.2 percent), stalking (n = 29, 17.1 percent), serious violent crimes (n = 72, 42.4 percent) (grievous bodily harm, robbery, hostage taking, attempted murder, and a combination of crime types: e.g. threat and assault), sexual offences (n = 20, 11.8 percent), traffic offences (n = 9, 5.3 percent) and homicide (n = 4, 2.4 percent). If a victim indicated that he/she had experienced more than one crime (for example, threat and sexual offence), he/she was assigned to the most severe category. On average, the crime took place 12.2 months (SD = 24.7) before completion of the pre-test. After removing eight extreme outliers, ranging from 40 to 192 months, the mean time elapsed since victimization was 7.5 months (SD = 6.8). A majority (60.6 percent) of the respondents knew the perpetrator before the crime was committed; most of them were acquaintances, friends or family members. Furthermore, 57 victims (33.5 percent) had previous victimization experiences, either of the same type of crime (n = 34, 20.0 percent) or of a different type of crime (n = 39, 22.9 percent).

Psychological characteristics

The mean score on the TSQ was 5.3 (SD = 3.4) and the mean score on the HADS was 9.3 (SD = 4.9). Following the psychometric characteristics of both scales, about half of the respondents (51.2 percent) showed posttraumatic stress symptoms and 54.7 percent showed signs of severe, clinically relevant anxiety. Furthermore, participants displayed moderately high levels of anger concerning their victimization: the mean score on the DAR scale was 20.0 (SD = 14.4). For comparison, Kunst et al. (2011) recently found mean posttraumatic anger scores for victims of violent crimes without probable PTSD of 17.5 (SD = 12.3) and for victims with probable PTSD of 30.6 (SD = 14.3).

Victims’ perceptions of the VIS

A PCA with varimax rotation was conducted on victims’ perceptions of the VIS. Prior to performing the PCA, the suitability of data for factor analysis was assessed. First, the sample size was examined. Although there is little agreement among authors concerning how large a sample should be (Pallant, 2001), our sample size of 170 was suitable to meet the standard of a 10 to 1 ratio, that is 10 cases for each item to be analysed (for example, Nunnally, 1978). Second, the strength of the relationship between the items was examined. Inspection of the correlation matrix revealed the presence of many coefficients of .3 and above, which indicates medium to large correlation effects (Cohen, 1988). Furthermore, the Kaiser–Meyer–Olkin value was 0.76. This exceeds the recommended value of 0.5, which means the sample is sufficiently large to conduct a PCA (Kaiser, 1974). Finally, Bartlett’s Test of Sphericity (Bartlett, 1950) reached statistical significance, supporting the factorability of the correlation matrix. The PCA initially revealed the presence of five components with eigenvalues exceeding 1, explaining 28.5 percent, 16.8 percent, 8.6 percent, 7.7 percent and 6.8 percent of the variance, respectively. Using Cattell’s (1966) scree test, it was decided to retain three components for further investigation. To aid in the interpretation of these components, varimax rotation was performed. The rotated solution (presented in Table 1) revealed the presence of a simple structure (Thurstone, 1947), with all components showing a number of strong loadings and all variables loading substantially on only one component.

Table 1.

Factor loadings for factor analysis with varimax rotation.

	Expression	Anticipation of negative consequences	Impact
Emotional recovery	.772	.105	.045
Understanding offender	.753	−.041	.211
Emotional burden (reverse scored)	.701	.253	−.005
Prevent recidivism	.697	.004	.232
Emotions/sorrow offender	.678	.200	.276
Understanding others	.583	−.047	.270
No emotional recovery (reverse scored)	−.460	.334	.100
Misrepresenting case	−.032	.837	−.011
Negative influence process	−.040	.753	.009
Revenge perpetrator	.082	.746	.162
Angry perpetrator	.168	.710	.247
Negative reactions third parties	.068	.544	.097
Influence sentence	.022	.031	.837
Attention during process	.206	.073	.704
Understanding judges / public prosecutor	.368	.167	.624
Compensation	.139	.322	.541

Note: Factor loadings greater than .40 are shown in bold.

The three-factor solution explained a total of 53.8 percent of the variance, with component 1 contributing 28.5 percent, component 2 contributing 16.8 percent and component 3 contributing 8.6 percent. Inspection of the three components revealed coherent underlying dimensions. In line with the distinction made by Roberts and Erez (2004) between expressive and impact-related VIS functions, component 1 consists of items relating to the former function, while component 3 concerns items connected to the desire to influence the outcome of the criminal trial. In addition to these components, our analysis adds a dimension that could be described as the anticipation of negative consequences: component 2 contains items that concern a negative influence of participation on the course or objectivity of the trial, retaliatory responses by the perpetrator, or negative reactions from the victims’ social surroundings.

Binary logistic regression analysis

As a first step in the preparation for the logistic regression analysis, bivariate associations between the independent variables and the dependent variable (likelihood of delivering a VIS) were calculated. Table 2 shows eight factors that had significant bivariate associations with the likelihood of delivering a VIS. The following background characteristics increased the likelihood of delivering a VIS: gender (woman) and marital status (divorced). Furthermore, the type of crime (stalking, violent crimes, sexual offences and traffic offences) was bivariately associated with the likelihood of delivering a VIS, as were the time elapsed since victimization and symptoms of posttraumatic stress, anger and anxiety. From victims’ perceptions regarding the VIS, only the ‘expression’ component had a significant bivariate association with the likelihood of delivering a VIS. Because the total sample contained only four co-victims of homicide, the assumption of the minimum expected cell frequency was violated. Therefore, bivariate associations with the likelihood of delivering a VIS could not be computed for this group.

Table 2.

Bivariate associations with the likelihood of delivering a VIS.

	Delivery of VIS, N = 170OR (95% CI)
Background characteristics
Gender (women)	3.61 (1.75–7.47)***
Age	1.02 (1.00–1.05)
Level of education
Lower (reference)	1.00
Intermediate	0.49 0(.18–1.36)
Higher	0.91 (0.27–3.08)
Marital status
Single (reference)	1.00
Living together	1.37 (0.48–3.92)
Married	1.59 (0.68–3.75)
Divorced	4.66 (1.24–17.48)*
Widow/widower	2.21 (0.22–22.48)
Employment (no)	1.23 0(.58–2.62)
Ethnic background
Native (reference)	1.00
First-generation	2.33 (0.48–11.29)
Second-generation	0.65 (0.17–2.54)
Crime characteristics
Type of crime
Threat (reference)	1.00
Stalking	6.33 (1.97–20.33)**
Violent crimes	6.53 (2.46–17.38)***
Sexual offences	8.17 (2.10–31.76)**
Traffic offences	16.33 (1.75–152.82)*
Homicide	–
Time elapsed (in months)	1.09 (1.02–1.18)*
Victim (vs. co-victim)	–
Relationship with offender (yes)	1.08 (0.52–2.21)
Earlier victimization (yes)	0.93 (0.45–1.92)
Psychological characteristics
Posttraumatic stress	1.45 (1.27–1.66)***
Anger	1.03 (1.01–1.06)*
Anxiety	1.22 (1.11–1.33)***
Victims’ perceptions of the VIS
Expression	1.13 (1.04–1.23)**
Fear negative consequences	0.96 (0.88–1.04)
Impact	1.11 (0.99–1.24)

Note: OR = odds ratio; CI = confidence interval.

p < .05; **p < .01; ***p < .001.

As a second step in the preparation for the logistic regression analysis, Pearson’s product-moment correlation coefficients were calculated for the remaining eight variables. Whereas correlations ranged from small (r = .025) to large (r = .783), correlations between the psychological characteristics were very high: very strong (see Cohen, 1988) positive mutual correlations were found between indications of posttraumatic stress symptoms, anger and anxiety, ranging from r = .619 to r = .783, with significance values less than .01. The magnitude and sign of the associations of these variables are similar to those previously reported (Orth and Wieland, 2006). Because the bivariate correlation between posttraumatic stress symptoms and anxiety exceeded the recommended maximum of .7 (r = .783, p < .01: see Tabachnick and Fidell, 2007), the latter was omitted from the logistic regression analysis. Posttraumatic stress and anxiety have a correlation of .783 and so the value of R² will be (.783)² = .61: posttraumatic stress symptoms share 61 percent of the variability in anxiety scores.

A logistic regression analysis was performed to assess the impact of a number of factors on the likelihood of delivering a VIS. The remaining seven independent variables with bivariate correlations with the likelihood of delivering a VIS were included in the model: gender, marital status, type of crime, time elapsed since victimization, indications of posttraumatic stress, anger and expression. The full model containing all predictors was statistically significant, χ² (14, N = 134) = 69.91, p < .001, indicating that the model was able to distinguish between respondents who wanted to deliver a VIS and those who did not. The model as a whole explained between 40.6 percent (Cox and Schnell) and 56.2 percent (Nagelkerke R²) of the variance in likelihood of delivering a VIS, and correctly classified 80.6 percent of cases. As shown in Table 3, only three variables made a statistically significant unique contribution to the model: type of crime, time elapsed since victimization and symptoms of posttraumatic stress. These three variables increased the likelihood of a VIS being delivered, controlling for all other factors in the model. The Hosmer–Lemeshow goodness-of-fit statistic (Hosmer and Lemeshow, 1989) revealed the model to adequately fit the data (4.23, p = .836).

Table 3.

Logistic regression predicting the likelihood of delivering a VIS.

	Delivery of VIS, N = 170OR (95% CI)
Background characteristics
Gender (women)	2.86 (0.89–9.22)
Marital status
Single (reference)	1.00
Living together	2.72 (0.59–12.62)
Married	1.23 (0.34–4.53)
Divorced	4.36 (0.66–28.83)
Widow/widower	3.21 (0.05–200.36)
Crime characteristics
Type of crime
Threat (reference)	1.00
Stalking	1.45 (0.28–7.48)
Violent crimes	7.22 (1.78–29.25)**
Sexual offences	0.67 (0.07–6.39)
Traffic offences	6.99 (0.44–110.31)
Homicide	–
Time elapsed (in months)	1.15 (1.02–1.29)*
Psychological characteristics
Posttraumatic stress	1.58 (1.22–2.05)***
Anger	0.98 (0.93–1.03)
Victims’ perceptions of the VIS
Expression	0.98 (0.85–1.14)

Note: OR = odds ratio; CI = confidence interval.

p < .05; **p < .01; ***p < .001.

Conclusion and discussion

In recent decades, the attention afforded to victims of crime has increased considerably (for example, Groenhuijsen and Pemberton, 2009). This upsurge in the victim of crime has spawned a considerable amount of literature on both the advantages and disadvantages of procedural instruments available to victims, such as the VIS. However, instead of taking into account differences in victims’ characteristics and needs, references are made to ‘what victims want’ (for example, Braithwaite, 2002; Strang, 2002) and to whether measures ‘work’ or ‘don’t work’ for victims (compare, for example, Sanders et al., 2001, with Chalmers et al., 2007). In other words, individual characteristics and perspectives that may determine whether or not a victim decides to participate in the criminal justice procedure are neglected. The present study was designed to fill this theoretical and empirical gap by examining which factors contribute to the likelihood of delivering a VIS.

The purpose of this study was twofold. First, victims’ perspectives on the purposes and function of the VIS were explored. Factor analysis showed that people’s perceptions of the VIS could be divided into three different components, with clear underlying topics: expression, impact and anticipation of negative consequences. This finding was partially consistent with previous research on motives for delivering a VIS: earlier research revealed a distinction between impact-related and expression-related use of VISs (Roberts and Erez, 2004; Sanders et al., 2001). Our study, however, was the first to reveal a third component: the anticipation of negative consequences. Second, we examined which variables contribute to the likelihood of a VIS being delivered. More specifically, we considered the impact of both background characteristics of the victim and the crime and of the victims’ perspectives on the purposes and function of VISs on the likelihood of delivering one. A binary logistic regression analysis revealed three variables to be positively associated with the likelihood of delivering a VIS, when controlling for all other variables in the equation: posttraumatic stress symptoms, the type of crime committed and the time elapsed since victimization. More specifically, and in line with Hypothesis 1, this study showed that the impact of the crime on the victim is positively correlated with the likelihood of delivering a VIS; in particular, victims displaying signs of poorer psychological functioning as a consequence of their victimization are likely to opt for the delivery of a VIS. This also applies to victims of serious violent crimes (for example, grievous bodily harm, robbery, hostage taking and attempted murder). Moreover, time since victimization is positively correlated with the likelihood of delivering a VIS. This means that victims of more complex, and often more serious cases are more likely to opt for delivering a VIS. These findings are in line with earlier research, which stated that the seriousness of the crime may influence the likelihood of delivering a VIS (Lens et al., 2010; Leverick et al., 2007; Pemberton, 2010; Roberts, 2009).

Although victims’ perceptions regarding the VIS were bivariately correlated with the likelihood of delivering a VIS, they did not make unique contributions to the model when controlling for the background variables of the victim and the committed crime (as suggested in Hypothesis 2). In sum, although one might presume that victims’ perceptions regarding the purpose and function of victim instruments determine whether or not a victim chooses to use these participatory rights on offer, this study showed that in fact victims’ choices are strongly influenced by the impact of the crime on the victim. More specifically, this study showed that the higher the impact of the crime on the victim, the more likely the victim is to deliver a VIS.

Practical implications

Based on these results, we argue that a more heterogeneous approach to the study of procedural instruments available to victims is needed: instead of looking at what victims want, we should focus on which instrument works for whom and under which conditions. The results of this research suggest that one must consider the reality that participants may differ on relevant characteristics from non-participants and that accounting for heterogeneity is an important element of incorporating the victim’s perspective in criminal justice. Weighing the advantages and disadvantages of modes of victims’ participation should not neglect the heterogeneity in the victim experiences, perspectives and needs we investigated. This study has some important implications. First, and at the most abstract level, incorporation of psychological constructs and concepts into the study of victims in the criminal justice system allows victimological research to do justice to individual differences between victims (see also Pemberton, 2009). We argue that the first steps in matching victim instruments in the criminal justice procedure with victims’ needs are to examine which victims feel the need to use the participatory rights on offer and why they do so. In the Netherlands, for example, only victims of serious violent crimes are allowed to deliver a VIS in court. This ‘restriction’ of the circle of rights-bearers of the VIS has long been debated: both professionals and lay-people have asked themselves whether victims of less serious crimes would have the same need to participate in the criminal justice procedure. This study is the first to show that victims who opt for the delivery of a VIS differ in their perspectives regarding the VIS from victims who decline their use of a VIS. However, this study revealed that it is not victim perspectives of the VIS but the impact of the crime on the victim that determines the likelihood of its use. Second, we argue that this important predictor of delivering a VIS (that is, crime severity) should be taken into account when determining the legal content of the VIS. Besides the circle of rights-bearers, another important debate in the Netherlands focuses on the content of the VIS. In the Netherlands, victims are allowed to speak only about the consequences of the crime and are not allowed to speak about the facts of the crime or a desired punishment. We argue that victims who are severely affected by the crime probably feel the strongest need to ‘stretch’ this legal content of the VIS. Research has repeatedly shown that the psychological impact of crime is associated with increased feelings of hostility (Orth and Wieland, 2006), revenge and retaliation (Cardozo et al., 2003; Orth et al., 2006) and punitiveness (Canetti-Nisim et al, 2009; Litvak et al., 2010; Pham et al., 2004). Therefore, we argue that victims of serious violent crimes (for example rape) are more inclined to give way to these feelings by heaping abuse upon the defendant or uttering a wish for a severe punishment. When debating the above-mentioned limitations of the VIS it is important to take into account these characteristics of both the victim and the crime.

Although this study has important practical implications, some limitations need to be addressed. First, this article represents cross-sectional data, which prevents us from determining causality or the exact nature of the relationships between our variables. Second, the psychological state of the victim was measured with the help of self-report questionnaires. Therefore, these results must be interpreted with care. For example, the TSQ has been developed to measure indications of possible PTSD: we cannot conclude that respondents who scored above the cut-off point of 5 really are suffering from PTSD. Moreover, the authors of the TSQ recommend that screening be conducted three to four weeks post-trauma to allow for normal recovery processes to take place (Brewin et al., 2002). Given the specific nature of this research, we were not able to meet this criterion. However, speaking against this argument, participants were never asked to fill in a questionnaire before this period of three to four weeks post-trauma. Third, although victims’ perceptions of the VIS reveal consistent trends with earlier research (Roberts and Erez, 2004; Sanders et al., 2001), they are generally solicited after victims have been given information by victim services. It can therefore not be ruled out that their views in part reflect official views.

Footnotes

Acknowledgements

We would like to thank the Dutch Ministry of Security and Justice for its financial support of the data collection. Moreover, we would like to thank Daphne C. Oldenhof and Janne van Doorn, who assisted in conducting the study.

References

Armour

Umbreit

(2005) The paradox of forgiveness in restorative justice. In: Worthington

(ed.) Handbook of Forgiveness. New York: Routledge.

Ashworth

(2000) Victims’ rights, defendants’ rights and criminal procedure. In: Crawford

Goodey

(eds) Integrating a Victim Perspective within Criminal Justice. Aldershot: Dartmouth Publishing.

Bartlett

(1950) Tests of significance in factor analysis. British Journal of Psychology 3: 77–85.

Bonanno

Westphal

Mancini

(2011) Resilience to loss and potential trauma. Annual Review of Clinical Psychology 7: 1–25.

Braithwaite

(2002) Restorative Justice and Responsive Regulation. Oxford: Oxford University Press.

Brewin

Andrews

Valentine

(2000) Meta analysis of risk factors for posttraumatic stress disorder in trauma-exposed adults. Journal of Consulting and Clinical Psychology 68(5): 748–766.

Brewin

Rose

Andrews

(2002) Screening for posttraumatic stress disorder in civilian populations. In: Orner

Schnyde

(eds) Reconstructing Early Intervention after Trauma. Oxford: Oxford University Press.

Canetti-Nisim

Halperin

Sharvit

Hobfoll

(2009) A new stress-based model of political extremism. Personal exposure to terrorism, psychological distress and exclusionist political attitudes. Journal of Conflict Resolution 53(3): 363–389.

Cardozo

Kaiser

Gotway

Agani

(2003) Mental health, social functioning, and feelings of hatred and revenge of Kosovar Albanians one year after the war in Kosovo. Journal of Traumatic Stress 16(4): 351–360.

10.

Cattell

(1966) The scree test for the number of factors. Multivariate Behavioral Research 1: 245–276.

11.

Chalmers

Duff

Leverick

(2007) Victim impact statements: Can work, do work (for those who bother to make them). Criminal Law Review, 360–379.

12.

Cheon

Regehr

(2006) Restorative justice models in cases of intimate partner violence: Reviewing the evidence. Victims and Offenders 1(4): 369–394.

13.

Cohen

(1988) Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Erlbaum.

14.

Edwards

(2001) Victim participation in sentencing: The problems of incoherence. The Howard Journal 40: 39–54.

15.

Edwards

(2004) An ambiguous participant: The crime victim and criminal justice decision-making. British Journal of Criminology 44: 967–982.

16.

Erez

(1999) Who’s afraid of the big bad victim? Victim impact statements as victim empowerment and enhancement of justice. Criminal Law Review 1: 345–356.

17.

Erez

(2004) Integrating restorative justice principles in adversarial proceedings through victim impact statements. In: Cape

(ed.) Reconcilable Rights? Analysing the Tension between Victims and Defendants. London: Legal Action Group.

18.

Foa

Riggs

Dancu

Rothbaum

(1993) Reliability and validity of a brief instrument for assessing posttraumatic stress disorder. Journal of Traumatic Stress 6: 459–473.

19.

Forbes

Hawthorne

Elliot

McHugh

Biddle

Craemer

Novaco

(2004) A concise measure of anger in combat-related posttraumatic stress disorder. Journal of Traumatic Stress 17: 249–256.

20.

Green

(2007) The victims movement and restorative justice. In: Johnstone

Van Ness

(eds) Handbook of Restorative Justice. Cullompton, Devon: Willan Publishing.

21.

Groenhuijsen

(1999) Victims’ rights in the criminal justice system: A call for more comprehensive implementation theory. In: Van Dijk

JJM

Van Kaam

RGH

Wemmers

J-AM

(eds) Caring for Crime Victims: Selected Proceedings of the 9th International Symposium of Victimology. Monsey, NY: Criminal Justice Press.

22.

Groenhuijsen

Letschert

(2008) Reflections on the development and legal status of victims’ rights instruments. In: Groenhuijsen

Letschert

(eds) Compilation of International Victims’ Rights Instruments, 2nd edn. Nijmegen: Wolf Legal Publishers.

23.

Groenhuijsen

Pemberton

(2009) The EU framework decision for victims of crime: Does hard law make a difference? European Journal of Crime, Criminal Law and Criminal Justice 17: 43–59.

24.

Gromet

Darley

(2009) Punishment and beyond: Achieving justice through the satisfaction of multiple goals. Law & Society Review 43(1): 1–38.

25.

Hall

(2010) Victims and Policy Making: A Comparative Perspective. Cullompton, Devon: Willan Publishing.

26.

Herman

(2003) The mental health of crime victims: Impact of legal intervention. Journal of Traumatic Stress 16(2): 159–166.

27.

Hosmer

Lemeshow

(1989) Applied Logistic Regression. New York: Wiley.

28.

Kaiser

(1974) An index of factorial simplicity. Psychometrika 39: 31–36.

29.

Kunst

MJJ

Winkel

Bogaerts

(2011) Posttraumatic anger, recalled peritraumatic emotions, and PTSD in victims of violent crime. Journal of Interpersonal Violence 5: 1–19.

30.

Lens

KME

Pemberton

Groenhuijsen

(2010) Het spreekrecht in Nederland: Een bijdrage aan het emotioneel herstel van slachtoffers? Tilburg: INTERVICT/PrismaPrint.

31.

Leverick

Chalmers

Duff

(2007) An Evaluation of the Pilot Victim Statement Schemes in Scotland. Edinburgh: Scottish Executive Social Research.

32.

Litvak

Lerner

Tiedens

Shonk

(2010) Fuel in the fire: How anger impacts judgement and decision making. In: Potegal

Stemmler

Spielberger

(eds) International Handbook of Anger: Constituent and Concomitant Biological, Psychological, and Social Processes. Dordrecht: Springer.

33.

Murphy

Davidshofer

(1998) Psychological Testing: Principles and Applications, 4th edn. Hillsdale, NJ: Prentice Hall.

34.

Nederlof

Hovens

Muris

Novaco

(2009) Psychometric evaluation of a Dutch version of the Dimensions of Anger Reactions. Psychological Reports 105: 585–592.

35.

Novaco

(1975) Anger control: The development and evaluation of an experimental treatment. Contemporary Psychology 21: 397–398.

36.

Nunnally

(1978) Psychometric Theory. New York: McGraw-Hill.

37.

Orth

Maercker

(2009) Posttraumatic anger in crime victims: Directed at the perpetrator and the self. Journal of Traumatic Stress 22(2): 158–161.

38.

Orth

Wieland

(2006) Anger, hostility, and posttraumatic stress disorder in trauma-exposed adults: A meta-analysis. Journal of Consulting and Clinical Psychology 74: 698–706.

39.

Orth

Montada

Maercker

(2006) Feelings of revenge, retaliation motive, and posttraumatic stress reactions in crime victims. Journal of Interpersonal Violence 21: 229–243.

40.

Orth

Berking

Walker

Meijer

Znoj

(2008) Forgiveness and psychological adjustment following interpersonal transgressions: A longitudinal analysis. Journal of Research in Personality 42(2): 365–385.

41.

Ozer

Best

Lipsey

Weiss

(2003) Predictors of posttraumatic stress disorder and symptoms in adults: A meta-analysis. Psychological Bulletin 129: 52–73.

42.

Pallant

(2001) SPSS Survival Manual. Buckingham: Open University Press.

43.

Pemberton

(2009) Victim movements: From varying needs to diversified criminal justice agendas. Acta Criminologica 22: 1–23.

44.

Pemberton

(2010) The Cross-over: An Interdisciplinary Approach to the Study of Victims of Crime. Antwerp: Maklu Publishers.

45.

Pemberton

Reynaers

(2011) The controversial nature of victim participation: Therapeutic benefits in victim impact statements. In: Erez

Kilchling

Wemmers

J-AM

(eds) Therapeutic Jurisprudence and Procedural Justice for Victims of Crime: International Perspective. Durham, NC: Carolina Academic Publishing.

46.

Pemberton

Winkel

Groenhuijsen

(2007) Taking victims seriously in restorative justice. International Perspectives in Victimology 3(1): 4–14.

47.

Pham

Weinstein

Longman

(2004) Trauma and PTSD symptoms in Rwanda: Implications for attitudes towards justice and reconciliation. Journal of the American Medical Association 292(5): 602–612.

48.

Roberts

(2009) Listening to the crime victim: Evaluating victim input at sentencing and parole. In: Tonry

(ed.) Crime, Punishment, and Politics in Comparative Perspective. Crime and Justice: A Review of Research, vol. 38. Chicago: Chicago University Press.

49.

Roberts

Erez

(2004) Expression in sentencing: Exploring the expressive function of victim impact statements. International Review of Victimology 10: 223–244.

50.

Roberts

Erez

(2010) Communication at sentencing: The expressive function of Victim Impact Statements. In: Bottoms

Roberts

(eds) Hearing the Victim: Adversarial Justice, Crime Victims and the State. Cullompton, Devon: Willan Publishing.

51.

Sanders

Hoyle

Morgan

Cape

(2001) Victim Impact Statements: Don’t work, can’t work. Criminal Law Review 6: 447–458.

52.

Sarat

(1997) Vengeance, victims and the identities of law. Social and Legal Studies 6(2): 163–190.

53.

Sherman

Strang

(2007) Restorative Justice: The Evidence. London: The Smith Institute.

54.

Spinhoven

Ormel

Sloekers

PPA

van Kempen

GMJ

Speckens

AEM

van Hemert

(1997) A validation study of the Hospital Anxiety and Depression Scale (HADS) in different groups of Dutch subjects. Psychological Medicine 27: 363–370.

55.

Strang

(2002) Repair or Revenge: Victims and Restorative Justice. Oxford: Oxford University Press.

56.

Tabachnick

Fidell

(2007) Using Multivariate Statistics, 5th edn. Boston: Allyn & Bacon.

57.

Thurstone

(1947) Multiple Factor Analysis. Chicago: University of Chicago Press.

58.

Tripp

Bies

Aquino

(2007) A vigilante model of justice: Revenge, reconciliation, forgiveness and avoidance. Social Justice Research 20(1): 10–34.

59.

Wenzel

Okimoto

Feather

Platow

(2008) Retributive and restorative justice. Law and Human Behavior 32: 375–389.

60.

Winkel

(2007) Posttraumatic Anger: Missing Link in the Wheel of Misfortune. Inaugural lecture, Tilburg University. Nijmegen: Wolf Legal Publishers.

61.

Zigmond

Snaith

(1983) The Hospital Anxiety and Depression Scale. Actua Psychiatrics Scandinavia 67(6): 361–370.