Journal of Interpersonal Violence
XX(X) xx –xx
© The Author(s) 2009
Reprints and permission: http://www.
A Survey of Police
About Crime Victim
A survey of police officers (n = 211) and prosecutors (n = 190) in Sweden
was conducted to assess law personnel’s beliefs about the behaviors and
reactions of victims of violent crimes. There were considerable differences
in the expected behavioral display of different types of crime victims, with
rape and domestic assault victims seen as particularly prone to expressive
self-presentation and self-blame. Despite empirical evidence showing
otherwise, most respondents thought that crime victims’ nonverbal and
emotional expression is to some extent related to the truthfulness of
their accounts. However, educational efforts appeared to have a corrective
influence on such beliefs. The perceived prevalence of false reports differed
across crime types, with rape and mugging receiving particularly high
estimates. Police officers believed false reports to be more common than
did prosecutors. Time constraints were seen, especially by prosecutors,
as an impediment to appropriate treatment of crime victims. Potential
1University of Gothenburg, Sweden
Karl Ask, Department of Psychology, University of Gothenburg, P.O. Box 500, SE 405 30,
J Interpers Violence OnlineFirst, published on August 28, 2009 as doi:10.1177/0886260509340535
2 Journal of Interpersonal Violence XX(X)
explanations for occupational differences and limitations associated with
the survey methodology are discussed.
crime victims, rape, self-presentation, false reports, beliefs
Given the traumatic nature of violent crimes, it is critical that professionals
involved in the legal process treat crime victims with consideration of their
special needs and concerns. The failure to do so may adversely affect the
victim’s well-being and long-term recovery from the trauma (Campbell
et al., 1999) and lessen the victim’s ability and willingness to cooperate in
the current and future criminal cases (Bard & Sangrey, 1986). Another
important consequence, which has received increased scholarly attention in
recent years, is that the failure to take into account the psychological reac-
tions of a victim may result in misinterpretations of the victim’s demeanor.
For instance, the composed and numbed behavior of a rape victim, lacking
signs of emotional distress, during a police interview may erroneously be
construed as a sign that the victim is not telling the truth about the event
(Kaufmann, Drevland, Wessel, Overskeid, & Magnussen, 2003; Winkel &
Koppelaar, 1991). This has serious implications for cases where physical
evidence is lacking and the legal outcome largely depends on the judged
credibility of the accuser and the accused, which is often true in cases of rape
and other sex-related crimes. Unfortunately, legal professionals, as well as
lay people, are very poor at detecting deception when merely observing a
statement, rarely performing better than what would be predicted by chance
alone (i.e., 50% in a dichotomous lie/truth judgment; Bond & DePaulo,
2006; Vrij, 2008). In addition, the cues that people report relying on when
assessing veracity have been found not to be consistently related to truthful-
ness or deception (DePaulo et al., 2003; Sporer & Schwandt, 2007). Hence,
an important measure to prevent misattribution of victims’ demeanor should
be to equip legal professionals with accurate knowledge. As a first step in
this process, the present research sought to document some of the existing
beliefs by means of a survey among police officers and prosecutors.
The psychological consequences of criminal victimization have been
studied systematically for more than three decades. In a pioneering study,
Burgess and Holmstrom (1974) examined the physical and psychological
reactions of rape victims who were admitted to the emergency department
of a major city hospital. The researchers found evidence for a so-called rape
trauma syndrome. That is, most of the female rape victims displayed a
similar sequence of reactions, characterized by an initial acute phase and
a long-term reorganization phase. In the acute phase, a number of somatic
manifestations were present (e.g., physical trauma, muscle tension) as
well as a range of emotional reactions (e.g., fear, self-blame). Two styles
of self-presentation were observed: the expressive style—in which nega-
tive emotions were evident through crying, sobbing, and tenseness—and
the controlled style—where any emotional reactions were hidden and a
calm, composed behavior was displayed. The two styles of expression were
equally common among the rape victims. The long-term reorganization
phase involved behaviors that helped victims cope with the trauma and
restore a functional lifestyle, but it was frequently accompanied by persis-
tent symptoms in the form of nightmares or phobias (Burgess & Holmstrom,
1974). The observed pattern was later found to be reliable across several
studies (for a review, see Frieze, Hymer, & Greenberg, 1987). However,
similar to the Burgess and Holmstrom study, most research in the area has
focused on the long-term psychological impact of the crime (for a review,
see Weaver & Clum, 1995) or the clinical treatment and recovery of the
victim (for a review, see Amstadter, McCart, & Ruggiero, 2007), rather than
the behaviors and reactions displayed toward legal professionals. Further-
more, the primary focus has been on rape victims, whereas relatively few
studies have investigated the differential impact of different types of crime.
It appears, however, that assaultive and nonassaultive offenses often lead to
the same types of long-term consequences, albeit of varying intensity (e.g.,
Lurigio, 1987; Norris & Kaniasty, 1994; Wirtz & Harrell, 1987). Recent
evidence indicates that the peritraumatic responses (i.e., at the time of the
offense) of rape victims tend to be more laden with emotions such as fear,
guilt, humiliation, numbness, and a sense of betrayal than the responses of
robbery and assault victims but that most of these emotions are quite
common also in the latter category of victims (Kaysen, Morris, Rizvi, &
Resick, 2005). However, across all types of crimes, there is a substantial
lack of documentation of the self-presentational styles that victims adopt
when communicating with others about the event.
Relatively little is known about the beliefs that people hold about crime
victims’ self-presentation styles. However, recent experimental evidence
suggests that people expect crime victims to act in a rather stereotypical
manner. Kaufmann and his colleagues (2003) showed that the emotional
display of a rape victim can influence judgments of the victim’s credibility.
When a victim’s testimony was accompanied by negative emotional dis-
plays, such as sobbing and signs of despair, her story was perceived as more
believable than when the same statement was given in a neutral (no
4 Journal of Interpersonal Violence XX(X)
emotions displayed) or incongruent (e.g., smiling) manner. The effect was
first demonstrated with lay people as observers but was later replicated in a
study of experienced police officers (Bollingmo, Wessel, Eilertsen, & Mag-
nussen, 2008). Interestingly, professional judges were found not to be
influenced by the victim’s emotional display (Wessel, Drevland, Eilertsen,
& Magnussen, 2006), perhaps indicating that their training in legal reason-
ing serves as a guard against behavioral misattribution—an explanation in
need of further testing. Winkel and Koppelaar (1991) showed experimen-
tally that a rape victim whose self-presentation was numbed not only was
perceived as less credible but also was blamed more for the rape than an
emotional victim. As further evidence for the existence of normative expec-
tations about crime victims’ emotional display, Rose, Nadler, and Clark
(2006) found that victims are expected to react in a way that is proportional
to the seriousness of the offense. An overly intense emotional display fol-
lowing a minor offense affected perceptions of a victim negatively, in the
same way as did the failure to display strong emotions in relation to a seri-
ous crime (Rose et al., 2006). Taken together, the empirical evidence
suggests that crime victims’ behaviors are gauged against culturally shared
stereotypes of normal reactions and that deviations from these stereotypes
tend to lower victims’ credibility. In addition, it appears that people are
attuned to the nonverbal behavior of crime victims when trying to assess
credibility, just as they have been found to do when judging criminal sus-
pects and witnesses (Vrij, 2008). This is problematic, given a large body of
research showing that nonverbal cues to deception are virtually nonexistent
(Vrij, 2008; see also DePaulo et al., 2003).
Although the above studies make a rather strong case that certain behav-
iors are expected from crime victims, the actual content of such expectations
was not directly addressed. Rather, normative beliefs were inferred from
participants’ responses to experimental manipulations. In addition, very few
studies on the issue have examined professionals in the criminal justice
system. To address these shortcomings, the present study surveyed the
beliefs held by police officers and prosecutors—two groups that frequently
meet crime victims and often make up a victim’s only contact with the crim-
inal justice system. The study also sought to compare the beliefs of these
two occupational groups, as they may differ for several reasons. First, police
officers meet a wider range of crime victims than do prosecutors, as the
latter predominantly meet the subset of victims whose case proceeds to
trial. Second, police officers engage in direct interaction with crime victims
(e.g., through investigative interviews, crime-scene visits) more often than
do prosecutors and hence may have a richer representation of crime victim
behaviors. Third, because the two groups serve different functions in the
legal system, police officers are motivated by different goals (e.g., solving
crimes) than are prosecutors (e.g., winning cases), which in turn may influ-
ence their attitudes toward crime victims.
A random 8 of the 21 regional police authorities in Sweden were approached.
Within these authorities, 17 senior officers responsible for teams of criminal
investigators and patrol officers were contacted via telephone and agreed to
distribute a questionnaire among their coworkers. A total of 304 paper ques-
tionnaires were then mailed for distribution. In addition, all 35 local public
prosecution offices in Sweden were approached. An e-mail invitation con-
taining a hyperlink to a Web-based, electronic version of the questionnaire
was sent to the registrar at each office, who was asked to forward the invita-
tion to prosecutors within the office. The distributors at both the police
authorities and the prosecutor offices were asked to invite only personnel
who encounter victims of violent crime as part of their work. The data col-
lection took place in 2007 during a period of 2 months.
In total, 401 professionals—211 police officers and 190 prosecutors—
responded to the questionnaire. Hence, the response rate among police
officers was 69.4%. It was not possible to establish a response rate among
prosecutors, as the actual number of invitations forwarded by administra-
tors at each prosecutor office could not be controlled. It should be noted,
however, that the number of prosecutors at the local offices in Sweden is
about 650. The demographic characteristics and professional experience of
the respondents are presented in Table 1.
The items in the questionnaire were arranged into sections corresponding
to different aspects of crime victim issues. Respondents were instructed
to answer the questions with adult (i.e., 15 years or older) crime victims
6 Journal of Interpersonal Violence XX(X)
Behavioral observations. Two multipart questions addressed behavioral
observations from participants’ work with crime victims. Respondents were
asked how often victims of rape, domestic assault, nondomestic battery, and
mugging, respectively, (a) show an expressive self-presentational style
(e.g., crying, despair, clear signs of distress) and (b) blame themselves for
what happened. Each rating was made on a 5-point scale (1 = very rarely,
5 = very often).
Beliefs. Nine statements about crime-victim behaviors (see Table 2) were
presented, and respondents were to indicate the perceived correctness of
each statement on 5-point scales (1 = totally incorrect, 5 = totally correct).
In addition, respondents were asked to estimate the frequency of false
reports (i.e., reports of crimes that in fact have not occurred) regarding rape,
domestic assault, nondomestic battery, and mugging, respectively. A range
Table 1. Demographic Characteristics of the Police and Prosecutor Samples
Female 85 (40.3%) 98 (51.6%)
Male 126 (59.7%) 92 (48.4%)
25-34 29 (13.7%) 28 (14.7%)
35-44 46 (21.8%) 64 (33.7%)
45-54 81 (38.4%) 67 (35.3%)
55-64 51 (24.2%) 31 (16.3%)
≥65 1 (0.5%) 0 (0.0%)
Missing 3 (1.4%) 0 (0.0%)
Length of service (years)
<5 27 (12.8%) 40 (21.1%)
5-9 14 (6.6%) 39 (20.5%)
10-19 35 (16.6%) 61 (32.1%)
20-29 63 (29.9%) 36 (18.9%)
≥30 71 (33.6%) 14 (7.4%)
Missing 1 (0.5%) 0 (0.0%)
Yes 81 (38.4%) 91 (47.9%)
No 129 (61.1%) 99 (52.1%)
Missing 1 (0.5%) 0 (0.0%)
Note: Percentages indicate proportions within each sample.
aSpecial training on crime victims’ psychological reactions and behaviors.
Table 2. Police Officers’ and Prosecutors’ Beliefs About Crime Victim Behaviors
Totally Rather correct/ Rather Totally
Occupational incorrect incorrect incorrect correct correct
group (%) (%) (%) (%) (%) M (SD) p
1. A crime victim’s display of emotions when telling about the crime is generally an
indicator of the veracity of his/her statement
Police officers 7.1 19.4 55.0 17.5 0.5 2.8 (0.8) .89
Prosecutors 7.9 17.9 57.4 16.3 0.5 2.8 (0.8)
2. A crime victim’s willingness or reluctance to spontaneously give a detailed account
of the crime is generally an indicator of the veracity of his/her statement
Police officers 12.8 32.7 40.3 13.7 0.0 2.6 (0.9) <.01
Prosecutors 14.7 46.3 31.6 7.4 0.0 2.3 (0.8)
3. The fact that a crime victim’s expressive style violates my expectations is generally
reason to examine the statement’s veracity extra carefully
Police officers 7.1 40.8 38.9 12.3 0.5 2.6 (0.8) .67
Prosecutors 11.6 35.3 40.0 12.1 0.5 2.5 (0.9)
4. A crime victim who displays negative emotions (e.g., crying, despair, clear signs
of distress) during his/her testimony is generally more likely to be believed in
Police officers 2.8 14.7 53.6 21.3 1.4 3.0 (0.8) <.001
Prosecutors 1.6 6.8 45.3 39.5 6.8 3.4 (0.8)
5. A crime victim who displays positive emotions (e.g., laughter, smiling) during his/her
testimony is generally less likely to be believed in court
Police officers 2.8 12.3 49.8 26.5 1.9 3.1 (0.8) <.05
Prosecutors 1.6 9.5 48.9 34.2 5.8 3.3 (0.8)
6. A crime victim’s inability to report details about the event shortly after the crime
(less than a day) is generally reason to question the veracity of the statement
Police officers 19.9 44.5 28.0 5.2 0.5 2.2 (0.8) <.05
Prosecutors 11.1 47.9 31.1 7.9 1.1 2.4 (0.8)
7. Details that appear in a crime victim’s memory after a period of time are generally
less reliable than those that the victim can report right from the start
Police officers 20.4 42.7 30.3 5.7 0.0 2.2 (0.8) .12
Prosecutors 12.6 47.4 32.1 5.8 1.1 2.3 (0.8)
8. The displayed reactions to a violent crime differ between crime victims with
different cultural backgrounds
Police officers 1.4 7.1 40.3 38.9 11.4 3.5 (0.8) <.001
Prosecutors 3.2 16.8 46.8 26.8 4.2 3.1 (0.9)
9. The type of relationship between the crime victim and the perpetrator generally
influences the victim’s expressive style and behavior
Police officers 0.5 2.4 33.6 48.3 14.2 3.7 (0.7) .08
Prosecutors 1.1 3.2 17.9 61.6 14.7 3.9 (0.7)
Note: Means and standard deviations are based on the rating scale from 1 (totally incorrect) to 5
8 Journal of Interpersonal Violence XX(X)
was to be estimated by indicating a minimum and a maximum percentage
of all reports that were believed to be false.
Treatment of crime victims. By rating their agreement with a set of state-
ments (1 = strongly disagree, 5 = strongly agree), respondents were to
indicate (a) whether they find it an important part of their work to make
crime victims feel well treated by the criminal justice system, (b) whether
they would be able to treat crime victims better if they had more time, and
(c) whether there is enough time for contacts with crime victims.
Self-rated knowledge. A final set of items were included to assess respon-
dents’ perceptions of their own knowledge about crime–victim behaviors,
both in absolute terms (1 = very little knowledge, 5 = very much knowledge)
and compared with their average colleague (1 = much less than average,
3 = average, 5 = much more than average). They were also asked whether
they had undergone special training concerning crime victims’ reactions
and behaviors, and whether they had tried to increase their knowledge on
the issue on their own initiative (e.g., by reading books, consulting research).
The data were analyzed using the SPSS (version 15) statistical software.
Parametric methods of analysis—analysis of variance (ANOVA), multi-
variate analysis of variance (MANOVA), and t tests—were used for
significance testing, based on the assumption that the rating scales used are
best treated on an interval, rather than ordinal, level. For all analyses, a
criterion level of α = .05 was used.
Respondents thought that an expressive style of self-presentation is rather
common among victims of all four crime types, as evidenced in an overall
mean rating of 3.6 (SD = 0.7) on the 5-point scale. However, a 4 (crime
type: rape vs. domestic assault vs. nondomestic battery vs. mugging) × 2
(occupation: police officer vs. prosecutor) mixed ANOVA revealed a sig-
nificant main effect of crime type, F(3, 1,143) = 245.63, p < .001, η2 = .39,
indicating that the perceived frequency of expressive behavior varied
between victims of different crimes. The highest rating was found for rape
(M = 4.1, SD = 0.9), followed by domestic assault (M = 3.9, SD = 0.8),
mugging (M = 3.2, SD = 1.0), and nondomestic battery (M = 3.1, SD = 0.9).
The ratings of all four crime types differed significantly from each other,
ps < .01 (Bonferroni post hoc tests). There was no main effect of occupa-
tion, F(1, 380) = 2.70, p = .10, η2 = .01. However, a significant Crime
Type × Occupation interaction emerged, F(3, 1,140) = 3.39, p < .05, η2 = .01.
Simple-effect analyses showed that police officers thought that an expres-
sive style was significantly more common among nondomestic battery
victims (M = 3.2, SD = 0.9) than did prosecutors (M = 2.9, SD = 0.9),
F(1, 1,140) = 23.27, p < .001. In addition, police officers found victims of
mugging to more often display an expressive style (M = 3.3, SD = 1.0) than
did prosecutors (M = 3.1, SD = 0.9), F(1, 1,140) = 5.41, p < .05. No occupa-
tional differences emerged for domestic assault or rape victims.
Across crime types, respondents rated victim self-blame as rather uncom-
mon (M = 2.5, SD = 0.7). However, a second mixed ANOVA showed that
there was significant variation between crimes, as indicated by a significant
main effect of crime type, F(3, 1,134) = 598.51, p < .001, η2 = .61. Domes-
tic assault victims were found to be most likely to put blame on themselves
(M = 3.4, SD = 1.0), followed by victims of rape (M = 3.0, SD = 1.1), non-
domestic battery (M = 1.9, SD = 0.8), and mugging (M = 1.6, SD = 0.8). The
ratings of all four crime types differed significantly from each other by post
hoc tests, p < .001 (Bonferroni). The main effect of occupation was also
significant, F(1, 378) = 4.46, p < .05, η2 = .01, indicating that prosecutors
saw victim self-blame as somewhat more common (M = 2.5, SE = 0.1) than
did police officers (M = 2.4, SE = 0.1). Finally, the Occupation × Crime
Type interaction was significant, F(3, 1,134) = 10.49, p < .001, η2 = .03.
Simple-effect analyses revealed that prosecutors saw self-blame as more
common among victims of rape (M = 3.1, SD = 1.1) than did police officers
(M = 2.9, SD = 1.2), F(1, 1,134) = 7.65, p < .01. Similarly, prosecutors rated
self-blame as more common among victims of domestic assault (M = 3.6,
SD = 0.9) than did police officers (M = 3.2, SD = 1.0), F(1, 1,134) = 37.69,
p < .001. No occupational differences were present regarding nondomestic
battery or mugging.
Table 2 presents respondents’ ratings of the nine statements about crime victim
behaviors. Although respondents tended not to endorse extreme standpoints,
frequently using the rating scale’s midpoint (3), trends toward agreement or
disagreement could be discerned. For instance, respondents tended to disagree
with the propositions that a victim’s inclination to give a spontaneous, detailed
report is indicative of truthfulness, that a victim’s inability to report details
10 Journal of Interpersonal Violence XX(X)
shortly after the crime is reason to question its veracity, and that details that
appear after some time in a victim’s memory are less reliable than those
remembered initially. In contrast, there was a tendency to believe that the
victim–offender relationship influences the victim’s style of expression and
behavior, and that the displayed reactions to a violent crime differ between
victims with different cultural backgrounds.
Respondents assigned some weight to the nonverbal behavior of crime
victims as an indicator of truthfulness. First, almost three quarters of the
respondents (73.8%) believed the proposition that a victim’s emotional
expression is indicative of veracity to be either partly correct/incorrect,
rather correct, or totally correct. Second, more than half the participants
(52.4%) reported similar levels of agreement with the statement that an
expectancy-violating style of expression is reason to question the veracity
of a victim’s statement extra carefully. As these beliefs are at odds with
research findings on nonverbal indicators of truthfulness, it is of interest to
know whether special training on crime victim issues has a corrective influ-
ence. A 2 (special training: yes vs. no) × 2 (occupation: police officer vs.
prosecutor) MANOVA was conducted on the two ratings. There was no main
effect of occupation, nor was there an Occupation × Training interaction.
However, the main effect of training was significant, Wilks’ λ = 0.96, F(2,
393) = 7.59, p < .001, η2 = .04. Respondents who had received training
(43.0% of the total sample) believed significantly less strongly that emotional
expressions were indicative of truthfulness (M = 2.7, SD = 0.8) than did those
who had not received training (M = 3.0, SD = 0.8), F(1, 394) = 10.66, p < .01,
η2 = .03. Similarly, trained respondents were significantly less likely to see
expectancy-violating nonverbal behavior as an indicator of veracity (M = 2.4,
SD = 0.8) than did untrained respondents (M = 2.7, SD = 0.8), F(1, 394) =
10.35, p < .01, η2 = .03. Hence, special training appeared to have a positive
influence on respondents’ beliefs.
As evident in Table 2, the beliefs expressed by the two professional
groups differed in some respects. Police officers were more likely than
prosecutors to think that crime victims’ inclination toward spontaneously
giving a detailed account is indicative of veracity, t(398) = 2.77, p < .01,
d = .28, and that there are cultural differences in the ways victims react to
violent crimes, t(393) = 4.64, p < .001, d = .47. Prosecutors, however, were
more likely than police officers to think that believability in court depends on
a victim’s expression of negative emotions, t(386) = –5.01, p < .001, d = .51,
and positive emotions, t(385) = –2.50, p < .05, d = .25, and that an inability
to report details shortly after the crime is reason to question the veracity of
the victim’s statement, t(393) = –2.27, p < .05, d = .23.
Estimates of False Reports
Table 3 presents respondents’ estimates of the frequency of false reports.
The interpretation of these estimates is complicated by the fact that several
respondents chose not to answer these items. The proportion of respon-
dents who reported both a minimum and a maximum estimate was 80.5%
for rape, 80.3% for domestic assault, 79.3% for nondomestic battery, and
77.6% for mugging. Respondents with missing data for any of the eight
estimates were excluded from the following analysis, leaving data from a
total of 306 respondents (76.3% of the total sample). A 4 (crime type: rape
vs. domestic assault vs. nondomestic battery vs. mugging) × 2 (estimate:
minimum vs. maximum) × 2 (occupation: police officer vs. prosecutor)
mixed ANOVA, with the two first factors compared within participants,
was performed on the estimates. As maximum estimates are by definition
higher than minimum estimates, the main effect of estimate was highly
significant, F(1, 304) = 238.34, p < .001, η2 = .44. The analysis further
revealed a significant main effect of crime type, F(3, 912) = 60.62, p < .001,
η2 = .17. Pairwise comparisons showed that the average false-report esti-
mate for rape (M = 14.8, SE = 0.9) was significantly higher than for
mugging (M = 8.8, SE = 0.5), nondomestic battery (M = 7.5, SE = 0.5),
and domestic assault (M = 7.1, SE = 0.4), all ps < .001 (Bonferroni cor-
rected). Estimates for mugging were in turn higher than for domestic
assault (p < .01) but not quite significantly higher than for nondomestic
battery (p = .054). Estimates for domestic assault and nondomestic battery
did not differ from each other. The main effect of occupation was also sig-
nificant, F(1, 304) = 55.75, p < .001, η2 = .15. Across crime types, police
Table 3. Means of Police Officers’ and Prosecutors’ Maximum and Minimum
Estimates of the Frequency of False Reports by Crime Type
Police officers Prosecutors
Type of crime Minimum Maximum Minimum Maximum
Rape 15.5% (16.5) 25.9% (21.1) 5.0% (7.0) 12.2% (12.5)
Domestic assault 5.1% (6.8) 11.5% (11.0) 2.8% (3.8) 9.3% (10.6)
Nondomestic battery 6.2% (8.4) 13.2% (12.2) 2.3% (3.4) 8.1% (10.1)
Mugging 8.7% (9.5) 16.4% (13.2) 2.3% (3.3) 7.9% (10.0)
Note: Each respondent contributed with both a minimum and a maximum estimate.
Numbers in brackets represent standard deviations.
12 Journal of Interpersonal Violence XX(X)
officers made higher estimates of false reports (M = 12.9, SE = 0.6) than
did prosecutors (M = 6.2, SE = 0.7).
Three of the four possible interactions in the analysis were signifi-
cant. First, there was evidence for a Crime Type × Occupation interaction,
F(3, 912) = 22.26, p < .001, η2 = .07. Among police officers, estimates for
all crime types differed from each other, ps < .01, with the exception of
domestic assault and nondomestic battery, p = .29 (Bonferroni). Among
prosecutors, however, estimates for rape differed significantly from all
other crime types, ps < .001, whereas the estimates for the other crime types
did not differ from each other, ps > .19. Second, there was a significant
Crime Type × Estimate interaction, F(3, 912) = 31.91, p < .001, η2 = .09.
A wider range between respondents’ minimum and maximum estimates
was found for rape (M = 9.2, SD = 9.7) than for domestic assault (M = 6.5,
SD = 8.1), nondomestic battery (M = 6.4, SD = 8.4), and mugging (M = 6.8,
SD = 8.5), all ps < .001 (Bonferroni corrected), whereas the latter three did
not differ from each other. This suggests that, although the false-report rate
is believed to be higher for rape than for the other crime types, there is also
a greater uncertainty about the exact frequency. Third, the Crime Type ×
Estimate × Occupation interaction was found to be significant, F(3, 912) =
9.76, p < .001, η2 = .03. Tests of simple-simple effects revealed that police
officers chose a wider range than prosecutors between their minimum and
maximum estimates for rape, nondomestic battery, and mugging, (ps < .001)
but not for domestic assault (F < 1).
Treatment of Crime Victims
Respondents found it an important part of their work to make crime vic-
tims feel well treated by the criminal justice system (M = 4.7, SD = 0.5).
There was, however, a slight difference between the occupational groups,
with police officers agreeing more strongly with the statement (M = 4.8,
SD = 0.5) than did prosecutors (M = 4.7, SD = 0.5), t(399) = 2.22, p < .05,
d = .22. Respondents also tended to think that they would be able to treat
crime victims better if they could spend more time with them (M = 3.9,
SD = 1.0). Prosecutors (M = 4.2, SD = 0.9) agreed more strongly with this
than did police officers (M = 3.7, SD = 1.1), t(398) = –5.11, p < .001, d = .51.
Opinions were more mixed in terms of whether there is enough time for con-
tacts with crime victims (M = 2.6, SD = 1.0). Again, prosecutors (M = 2.2,
SD = 0.9) perceived the lack of time to be greater than did police officers
(M = 2.9, SD = 1.0), t(399) = 7.15, p < .001, d = .72.
Respondents tended to rate their own degree of knowledge about crime vic-
tims’ reactions and behaviors as rather high (M = 3.5, SD = 0.8), and police
officers were more so inclined (M = 3.6, SD = 0.7) than were prosecutors
(M = 3.4, SD = 0.8), t(399) = 3.15, p < .01, d = .32. As a further indicator of
respondents’ confidence, they tended to think that they had more knowledge
than their average colleague (M = 3.4, SD = 0.6), t(399) = 13.59, p < .001,
d = 0.68 (one-sample t test; reference value = 3). Police officers and prose-
cutors did not differ in this respect. Most participants (65.1%) claimed to
have tried to increase their knowledge on the issue on their own initiative.
Again, no significant difference was found between the proportions of
police officers and prosecutors who had done so.
The results of the present research give insights into the beliefs of profes-
sionals within the criminal justice system regarding issues related to crime
victims. The professionals surveyed expressed a perception that victims’
reactions and behaviors differ as a function of crime type. For instance,
victims of rape and domestic assault were thought to more often display
an expressive self-presentational style and blame themselves than victims
of other assaultive crimes. Although the correspondence of these beliefs
to the actual frequency of victim behaviors is difficult to assess, they do
indicate that a certain type of behavior is generally expected from victims
of rape and domestic assault. In contrast, research indicates that the
expressive style is about as common among rape victims as is the con-
trolled style (Burgess & Holmstrom, 1974; Frieze et al., 1987). The
discrepancy between research findings and professionals’ beliefs may, of
course, stem from actual differences in the behaviors that crime victims
display in research settings and when in contact with the legal system.
Another possibility, worthy of further investigation, is that professionals
hold erroneous expectations about normal crime victim behavior.
The results further indicated that respondents found nonverbal behavior to
be somewhat useful when determining the veracity of a victim’s statement.
This belief conflicts with a large body of research showing that the nonverbal
cues people rely on when trying to detect deception are unrelated to actual
truthfulness (DePaulo et al., 2003; Sporer & Schwandt, 2007; Vrij, 2008).
The present finding adds to the literature by showing that professionals’
14 Journal of Interpersonal Violence XX(X)
beliefs about nonverbal reliability cues extend beyond judgments of crimi-
nal suspects and witnesses, which have been the focus of most previous
research (Granhag & Strömwall, 2004). The weight placed on nonverbal
behavior is particularly alarming given the fact that respondents tended to
expect an expressive self-presentational style from rape victims. Hence,
there is an apparent risk of misjudgment when rape victims behave in a
controlled manner. On a more positive note, however, it appears that educa-
tional efforts may be a useful step in coming to terms with misconceptions.
Participants who had undergone special training placed less weight on non-
verbal behavior compared with untrained participants. The data give no
answer as to what particular kind of training participants had received or
which educational efforts are the most effective but nonetheless imply that
targeting professionals’ knowledge is an important task for researchers and
others with the aim to improve the criminal justice system.
Another notable set of findings concerns respondents’ estimates of false
reports. Allegations of rape and mugging were seen as particularly likely to
be fabricated. On average, police officers thought that at least one out of six,
and possibly as many as one out of four, reports of rape is false and that
potentially one out of six claims of mugging is false. Although prosecutors
generally made more cautious estimates, false reports of rapes were believed
to be considerably more common than false reports of the other crime types.
The issue of false crime allegations, especially false claims about rape, is
highly controversial and has spurred a lively discussion in the media in
recent years (e.g., Krantz & Wahlgren, 2007; Ream, 2007). Some of the
proposed motives behind false reports are the prospect of economic com-
pensation, the intention to harm the accused, and the need for sympathy and
attention (Kanin, 1994). One fact that is all too obvious to legal profession-
als is that alleged rapes are notoriously difficult to investigate, primarily
because there is rarely substantial evidence other than the testimonies of the
alleged victim and the suspect (Hazelwood & Burgess, 2001). As a conse-
quence, many rape allegations will not be proven either true or false, and the
actual frequency of false allegations is therefore extremely difficult to
assess. Whether correct or incorrect, the estimates reported in the present
study indicate a considerable degree of suspiciousness toward rape victims.
As concluded in earlier research, an overly suspicious attitude toward crime
victims may result in a secondary victimization, which may increase their
suffering and impede their long-term recovery (Campbell et al., 1999).
Hence, it is obvious that police officers and prosecutors face a challenge in
maintaining a fine balance between suspicion and acceptance when in con-
tact with crime victims.
The fact that police officers consistently estimated false reports to be
more frequent than did prosecutors calls for an explanation. It may be that
prosecutors are less suspicious toward crime victims because the two pro-
fessional groups are exposed to different samples of victims. The police
come in contact with all crime victims who file a report—even those who
retract their allegations after a brief period of time. Prosecutors, however,
come in contact only with those victims who maintain their allegations long
enough for the case to be handed over to the prosecutor’s office. It is pos-
sible that the lower exposure to retractors causes prosecutors to view false
reports as less common. This account is plausible given that people often
rely on availability as a cue to probability judgments (Tversky & Kahne-
man, 1973); the more easily they come to think of examples from an event
category, the more frequent or probable the type of event is thought to be.
A second interpretation is that the observed results simply reflect different
degrees of cautiousness when responding to the survey. In most legal sys-
tems, including the Swedish, the decisions made by prosecutors are more
pivotal and consequential than those made by individual police officers.
Hence, police officers may feel less accountable and freer to express their
subjective opinions on controversial issues such as false crime reports.
Prosecutors, however, may be motivated to portray themselves as unpreju-
diced, as a suspicious and speculative attitude toward crime victims may
damage the public’s trust in the authority. It remains for future research to
establish which, if any, of the above explanations that best accounts for the
Respondents expressed concern that the time available for contact with
crime victims is too scarce. This concern was particularly strong among
prosecutors. Unfortunately, the insufficiency of time appears to negatively
affect the manner in which crime victims are treated, as a majority of the
respondents claimed that more time available would lead to improved treat-
ment. This is an important aspect to consider when policy makers seek to
improve the performance of the legal system. Ultimately, the lack of time
may be conducive to misjudgments and maltreatment of crime victims. It is
a well-documented social-psychological finding that time pressure, and
other circumstances that limit people’s information processing capacity,
increase the reliance on stereotypes as a basis for social judgments (Fiske,
Lin, & Neuberg, 1999; Kaplan, Wanshula, & Zanna, 1993). Hence, present
working conditions for practitioners in the criminal justice system may
function as an effective perpetuator of erroneous expectations about crime
victim behaviors. This may, for instance, further discredit rape victims who
do not display the expected expressive self-presentation style.
16 Journal of Interpersonal Violence XX(X)
Although the present research provided insight into police officers’ and
prosecutors’ beliefs about crime victims, future studies should seek to address
the shortcomings that are associated with this type of questionnaire research.
First, there may be strong self-presentational concerns among actors in the
criminal justice system. This is mainly because these agencies are highly
scrutinized by the media and the public, and the expression of controversial
attitudes and beliefs may elicit strong critique and loss of public support. To
investigate whether the present results underestimate the extremeness of pro-
fessionals’ beliefs, studies should be carried out where self-presentational
concerns are kept to a minimum. Second, many of the items in the question-
naire concerned perceptions of victims of violence in general. This has the
disadvantage of making respondents picture an average victim, rather than a
victim under clearly defined and specific circumstances, which, of course,
is more akin to what respondents encounter in actual cases. One way to
come to terms with this problem would be to present respondents with
detailed, scenario-based descriptions and ask what behaviors would be
expected from an individual victim in that situation. Third, in the areas
where respondents’ beliefs were found not to correspond with scientific
knowledge (e.g., the value of nonverbal behavior when judging credibil-
ity), it still remains unclear what practical consequences follow from such
beliefs. Equally unclear is whether the beneficial effects of education
extend beyond practitioners’ beliefs and influence the actual treatment of
crime victims. Such questions are best addressed through the use of experi-
In conclusion, criminal justice professionals’ perceptions of crime vic-
tim behaviors are yet a relatively uncharted research area. The present
research provides a first glance at the contents of these perceptions and
points out some directions for future research efforts. The primary targets
of such studies should be to draw a more complete and nuanced picture of
dominant beliefs and to investigate their practical consequences on judg-
ments of crime victim credibility. Ultimately, the main beneficiaries of
this research will be crime victims as a group, whose chances of receiving
an appropriate treatment are enhanced as scientific knowledge on the
The author would like to thank Eva Bloch, Lisbeth Johansson, Frida Pedersson,
Kajsa Rapp, Gunn Sjögren, and Leif Strömwall for their generous help at various
stages of the research process.
Declaration of Conflicting Interests
The author declared that he had no conflicts of interests with respect to his author-
ship or the publication of this article.
The author disclosed receipt of the following financial support for the research and/
or authorship of this article: This research was supported by the Crime Victim
pensation and Support Authority (grant nr. 8914/2006).
Amstadter, A. B., McCart, M. R., & Ruggiero, K. J. (2007). Psychosocial interven-
tions for adults with crime-related PTSD. Professional Psychology: Research
and Practice, 38, 640-651.
Bard, M., & Sangrey, D. (1986). The crime victim’s book (2nd ed.). New York:
Bollingmo, G. C., Wessel, E. O., Eilertsen, D. E., & Magnussen, S. (2008). Credibil-
ity of the emotional witness: A study of ratings by police investigators. Psychol-
ogy, Crime & Law, 14, 29-40.
Bond, C. F., Jr., & DePaulo, B. M. (2006). Accuracy of deception judgments.
Personality and Social Psychology Review, 10, 214-234.
Burgess, A. W., & Holmstrom, L. L. (1974). Rape trauma syndrome. American
Journal of Psychiatry, 131, 981-986.
Campbell, R., Sefl, T., Barnes, H. E., Ahrens, C. E., Wasco, S. M., & Zaragoza-
Diesfeld, Y. (1999). Community services for rape survivors: Enhancing psy-
chological well-being or increasing trauma? Journal of Consulting and Clinical
Psychology, 67, 847-858.
DePaulo, B. M., Lindsay, J. J., Malone, B. E., Muhlenbruck, L., Charlton, K., &
Cooper, H. (2003). Cues to deception. Psychological Bulletin, 129, 74-118.
Fiske, S. T., Lin, M., & Neuberg, S. L. (1999). The continuum model: Ten years
later. In S. Chaiken & Y. Trope (Eds.), Dual-process theories in social psychol-
ogy (pp. 231-254). New York: Guilford.
Frieze, I. H., Hymer, S., & Greenberg, M. S. (1987). Describing the crime victim:
Psychological reactions to victimization. Professional Psychology: Research
and Practice, 18, 299-315.
Granhag, P. A., & Strömwall, L. A. (Eds.). (2004). The detection of deception in
forensic contexts. Cambridge, UK: Cambridge University Press.
Hazelwood, R. R., & Burgess, A. W. (Eds.). (2001). Practical aspects of rape inves-
tigation: A multidisciplinary approach (3rd ed.). New York: CRC Press.
Kanin, E. J. (1994). False rape allegations. Archives of Sexual Behavior, 23, 81-92.
Kaplan, M. F., Wanshula, L. T., & Zanna, M. P. (1993). Time pressure and
information integration in social judgment: The effect of need for structure.
18 Journal of Interpersonal Violence XX(X)
In O. Svenson & A. J. Maule (Eds.), Time pressure and stress in human judgment
and decision making (pp. 255-267). New York: Plenum.
Kaufmann, G., Drevland, G. C. B., Wessel, E., Overskeid, G., & Magnussen, S.
(2003). The importance of being earnest: Displayed emotions and witness cred-
ibility. Applied Cognitive Psychology, 17, 21-34.
Kaysen, D., Morris, M. K., Rizvi, S. L., & Resick, P. A. (2005). Peritraumatic
responses and their relationship to perceptions of threat in female crime victims.
Violence Against Women, 11, 1515-1535.
Krantz, K., & Wahlgren, A. (2007, March 24). Oskyldigt anklagade [Falsely
accused]. Expressen, p. 18.
Lurigio, A. J. (1987). Are all victims alike? The adverse, generalized, and differen-
tial impact of crime. Crime & Delinquency, 33, 452-467.
Norris, F. H., & Kaniasty, K. (1994). Psychological distress following criminal vic-
timization in the general population: Cross-sectional, longitudinal, and prospective
analyses. Journal of Consulting and Clinical Psychology, 62, 111-123.
Ream, A. K. (2007, June 17). Cheers can’t drown out painful truths. Chicago
Tribune, p. 25.
Rose, M. R., Nadler, J., & Clark, J. (2006). Appropriately upset? Emotion norms and
perceptions of crime victims. Law and Human Behavior, 30, 203-219.
Sporer, S. L., & Schwandt, B. (2007). Moderators of nonverbal indicators of decep-
tion: A meta-analytic synthesis. Psychology, Public Policy, and Law, 13, 1-34.
Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency
and probability. Cognitive Psychology, 5, 207-232.
Vrij, A. (2008). Detecting lies and deceit: Pitfalls and opportunities (2nd ed.).
Chichester, UK: Wiley.
Weaver, T. L., & Clum, G. A. (1995). Psychological distress associated with inter-
personal violence: A meta-analysis. Clinical Psychology Review, 15, 115-140.
Wessel, E., Drevland, G. C. B., Eilertsen, D. E., & Magnussen, S. (2006). Credibility
of the emotional witness: A study of ratings by court judges. Law and Human
Behavior, 30, 221-230.
Winkel, F. W., & Koppelaar, L. (1991). Rape victims’ style of self-presentation and
secondary victimization by the environment: An experiment. Journal of Inter-
personal Violence, 6, 29-40.
Wirtz, P. W., & Harrell, A. V. (1987). Assaultive versus nonassaultive victimization: A
profile analysis of psychological response. Journal of Interpersonal Violence, 2, 264-277.
Karl Ask is a researcher and lecturer at the department of psychology, University of
Gothenburg, Sweden. He received his PhD from the University of Gothenburg in
2006. His research interests include investigative psychology, witness psychology,
and motivation and emotion in legal settings.