post 2 questions pertaining to articles

699

Violence Against Women Volume 15 Number 6

June 2009 699-719 © 2009 Sage Publications

10.1177/1077801209332182 http://vaw.sagepub.com

hosted at http://online.sagepub.com

Trauma Victim: Yes or No? Why It May Be Difficult to Answer Questions Regarding Violence, Sexual Abuse, and Other Traumatic Events Siri Thoresen Carolina Øverlien Norwegian Centre for Violence and Traumatic Stress Studies

The aim of this study is to explore the reasons why it may be difficult to answer questions regarding violence, sexual abuse, and other potentially traumatic events (PTes). Qualitative and quantitative methods are used with a nonrepresentative community sample of 628 women who respond to a Web survey. altogether, 65% do not find it difficult to answer any PTe questions. Difficulties in answering occur more frequently with violence and sexual abuse items and are associated with exposure. Reasons for difficulties in answering include “event fit” (severity, frequency, force), “me as a victim” (responsibility, reactions, protec- tion, memory), and “you as abuser” (intention, protection).

Keywords: measurement; potentially traumatic events; sexual abuse; violence

a large body of research has documented negative social and psychological con-sequences of experiencing violence, sexual abuse, and other potentially trau- matic events (PTes; Dube et al., 2005; graham-Bermann & Seng, 2005; Jarvis, gordon, & Novaco, 2005; Kessler, Sonnega, Bromet, & Nelson, 1995). It is of great importance to society to identify the prevalence of traumatic events in different populations and reveal their causes and effects. However, exposure to PTes may be hard to measure, and we know little about how informants comprehend our ques- tions and what kinds of events informants perceive to qualify for a “yes” response. The aim of this study was to investigate why it may be difficult to answer questions regarding violence, sexual abuse, and other PTes.

A Wide Range of PTEs

When investigating the relationship between exposure to traumatic events and sub- sequent social and mental health problems, much research has focused on specific trauma populations, such as victims of sexual assault, victims of violence, combat veterans, or terror and disaster victims.1 However, there are several well-documented

Article

700 Violence against Women

reasons for the necessity of measuring lifetime exposure to a variety of traumatic events. For example, the number of exposures seems to have a cumulative effect and experiencing more than one traumatic event seems to be quite common (Follette, Polusny, Bechtle, & Naugle, 1996; Frans, Rimmo, aberg, & Fredrikson, 2005; green et. al., 2000; Resnick, Kilpatrick, Dansky, Saunders, & Best, 1993). In large population studies, there is often a requirement that screening instruments for PTes be brief. Several brief measures of lifetime exposure to traumatic events, including violence and sexual assault, have been developed (for a review, see Norris & Hamblen, 2004).

Traumas Are Gendered

PTes are not evenly distributed in the population, but follow specific gender and age patterns. Women and girls are more exposed to sexual abuse than men and boys; men are more exposed to nondomestic violence and combat, and women are more exposed to domestic violence (Nemeroff et al., 2006; Norris, 1992). On a general level, women and girls are more often victims of intrafamilial violence, and men and boys are more often victims of extrafamilial violence. These gender patterns are related to other gender-specific behavior patterns in societies. Interpersonal traumas, especially rape and sexual molestation, are thought to have more devastating effects on social and mental functions than other PTes, for both men and women (Creamer, Burgess, & McFarlane, 2001; Dube et al., 2005; green et al., 2000).

Stigma and Taboo: Can We Ask These Questions and How?

Violence and sexual abuse are issues that are loaded with taboos, social stigma, and moral preconceptions. Hence, researchers may be reluctant to include PTe measures and wonder if it is acceptable for informants to be asked such questions, or whether informants will react negatively, be unwilling to answer, be bothered by reminders, be put in physical danger because their violent partner is bothered by such questions, or drop out of the study altogether. Because of such concerns, PTe measures may be omitted from large epidemiological studies, resulting in a loss of important information. In a recent article, Becker-Blease and Freyd (2006) argued that most such objections are not supported by research. empirical investigations seem to contradict the hypothetical unwillingness to respond to PTe questions. For exam- ple, it has been found that violence questions are well accepted, and that nonresponses to single items occurred much more frequently with income questions than with sexual abuse questions (Black, Kresnow, Simon, arias, & Shelley, 2006). Research on childhood victimization has been found to be well tolerated by participating women,

Thoresen, Øverlien / Trauma Victim: Yes or No? 701

adverse reactions are less common than usually expected (Newman, Walker, & gefland, 1999), and most individuals make favorable cost–benefit appraisals regard- ing their participation (Newman & Kaloupek, 2004).

One of the most important causes of measurement error in this research area is underreporting of violence and sexual abuse, while false positives are probably rare (Fergusson, Horwood, & Woodward, 2000; Hardt & Rutter, 2004). Willingness to disclose (give a “yes” answer) seems to be dependent on how the questions have been formulated, in the sense that the use of “labeling” terms, such as “rape,” will reduce the number of respondents who answer yes, while using behaviorally specific questions (such as asking directly about forced vaginal, oral, or anal penetration) will increase disclosure (Harned, 2004; Resnick et al., 1993). Two recent studies using both types of question formulations supported this hypothesis (Harned, 2004; Thombs et al., 2006).

Are Informants Consistent in Their Answers to PTE Questions?

Several studies have empirically investigated consistency in responses to trau- matic event screening questionnaires. Two studies found the number of reported events to increase (Roemer, Litz, Orsillo, ehlich, & Friedman, 1998; Southwick, Morgan, Nicolaou, & Charney, 1997), whereas another study found the number of reported events to remain unchanged 1 year later (Bramsen, Dirkzwager, van esch, & van der Ploeg, 2001). Research indicates that it is common (>80%) to change at least one response at follow-up (Bramsen et al., 2001; Spinhoven, Bean, & eurelings-Bontekoe, 2006). The overall frequency of inconsistent reporting of PTes in a study with a 6-year follow-up was 64% (Hepp et al., 2006). Reports of child- hood sexual abuse and physical punishment have also been found to be relatively unstable over time (Fergusson et al., 2000).

Inconsistencies in self-reports are not unique to traumatic stress studies and have been reported in other research areas as well. Skog (1992) found large inconsisten- cies between self-reported drug use over time; of those reporting lifetime drug use, approximately 50% reported lifetime drug use when they were asked again a year later. Similarly, major inconsistencies in self-reported age of first alcohol use and first smoking were reported by engels, Knibbe, and Drop (1997). Hence, the incon- sistencies in traumatic stress studies may be no larger than those in other research areas of the human and social sciences.

Inconsistencies in reporting PTes may have serious implications for estimated prevalence rates and for evaluating the consequences of violence, sexual assault, and other PTes. How can such inconsistencies be explained? Some studies have tried to investigate predictors of inconsistencies in self-reports of traumatic events. These include

702 Violence against Women

gender, level of posttraumatic stress symptoms and dissociation, substance abuse, memory functions, compensation seeking, initial underreporting, and embarrassment owing to stigma and shame (e.g., Della Femina, Yeager, & Lewis, 1990; Frueh, Hamner, Cahill, & gold, 2000; goodman, Corcoran, Turner, Yuan, & green, 1998; Hepp et al., 2006; Krinsley, gallagher, Weathers, Kutter, & Kaloupek, 2003; Ouimette, Read, & Brown, 2005). However, little is known about how informants respond to PTe questions, if they find PTe questions difficult to answer, and what active choices informants make in the decision-making process when they choose between a “yes” and a “no” answer.

This study was undertaken to investigate the acceptability and comprehensibility of a brief screening instrument for PTes, including violence and sexual abuse in women. The research questions in this study were as follows:

1. Do women find it acceptable to be asked questions regarding violence, sexual abuse, and other PTes?

2. Do women find it difficult to answer yes or no to such questions? 3. For those women who find it difficult to answer yes or no, why is it difficult?

To answer these questions, a combination of quantitative and qualitative research methods was used.

Method

Procedure and Participants

The research was designed as a pilot study for a large, prospective epidemiologi- cal study of women and children in Norway. The pilot questions were intended to be included at the time point when the children would be 6 to 8 years of age. as the mean age of giving birth in Norway was 27 years in 2000, we can estimate that the mean age of the women in our study population was approximately 34 years. Twenty-five percent of 34-year-old women in Oslo are immigrants (those who either were not born in Norway or whose parents were not born in Norway; M. T. Dzamarija, Statistics Norway, personal communication, March 17, 2008). Seventy-nine percent of them work either part-time or full-time (B. Olsen, Statistics Norway, personal communication, March 18, 2008). Hence, the study population in the current study consisted mostly of working women living in Oslo who were mothers of 6- to 8-year-old children, with a predominantly Norwegian background. The study was anonymous and did not record any personal data.

The Board of education in the Municipality of Oslo approved the study and for- warded a letter to all schools in Oslo, asking them to distribute the letter to mothers of children in the first, second, and third grades. The letter described the purpose of the study, asking the women to assist the researchers in constructing clear questions regarding “negative life events, such as violence, sexual assault, and other traumatic

Thoresen, Øverlien / Trauma Victim: Yes or No? 703

events.” To do this, the women had to enter a Web site and answer a Web questionnaire. This procedure was considered sufficient for the purpose of the pilot study, although it was obvious that a rather large proportion of the population would not be able to participate, such as women with no easy access to the Internet and women not suf- ficiently fluent in the Norwegian language (approximately 10% of the Oslo popula- tion have a language other than Norwegian as their first language). The response rate was low in this study but could not be calculated exactly. altogether, 99 schools were asked by the Board of education to distribute the letter. We contacted all schools and found that 55 schools, attended by approximately 8,831 children, had actually asked their teachers to distribute the letter. However, school administrators had not recorded how many of their teachers actually distributed the letter to their pupils. In addition, in the case of siblings, efforts were made not to distribute to more than one child of the same mother. However, the proportion of siblings was not known. altogether, 628 women responded to the Web survey, forming a nonrepre- sentative and probably highly self-selected sample of the population in question. Results of this study should not be taken as representative of any larger group, but rather be considered as examples from a potential range of considerations individu- als may have when answering traumatic event questionnaires.

Measures

PTEs. The purpose of the pilot study was to evaluate the appropriateness of the PTe questions to be included in a large questionnaire study of women (and their children). From several available instruments, we chose the Stressful Life events Screening Questionnaire (SLeSQ; goodman et al., 1998), a 13-item questionnaire that includes behaviorally formulated items, focuses more than other instruments on interpersonal trauma, and has been carefully researched for psychometric properties (Norris & Hamblen, 2004) and cross-cultural validity (green, Chung, Daroowalla, Kaltman, & DeBenedictis, 2006). The SLeSQ was adapted for our purposes in several steps, including translation to Norwegian, restriction of certain items, and simplification of response alternatives, including two items that could be especially relevant for women, and some changes of wording as a result of pilot interviews with 29 women exposed to violence and/or sexual abuse (see appendix). Changes made in the SLeSQ-adapted were carefully recorded and are available from the authors on request. The SLeSQ-adapted consists of 13 questions regarding serious accident; stillbirth or spontaneous abortion; loss of a child; loss of a family mem- ber in an accident, murder, or suicide; rape; sexual assault; other unwanted sexual experiences; violence in childhood; violence in adulthood; emotional violence; witnessing assault; being a victim of criminal acts; and other life-threatening events. Back-translation of the SLeSQ-adapted was performed by an independent agency.

Acceptability. Respondents were asked to report the degree to which they would respond with the following reactions if they were asked such PTe questions in a

704 Violence against Women

comprehensive survey: (a) I would be upset or uneasy, (b) I would react negatively to being asked such questions, (c) I would react negatively to not being asked about something I have experienced, (d) I would react positively to being asked such ques- tions, (e) It would be dangerous for me to answer such questions, and (f) If I was asked such questions, I would have a need to talk to somebody professional, or get other forms of support. all items were rated on a 5-point Likert-type scale ranging from not at all to to a very high degree.

Difficulties in deciding how to answer PTE questions. The respondents were asked if they had difficulties deciding whether to answer yes or no to any of the PTe questions and, if yes, to specify to which of the questions these difficulties applied. The respond- ents were then asked to describe why it was difficult to decide how to answer.

Analyses and Statistical Procedures

Statistical analyses. Differences between groups were analyzed using chi-square tests (for categorical data) and Pearson’s r (for continuous data). all analyses were performed in SPSS-14.

Qualitative analyses. The qualitative data consisted of an elicited text, that is, a text written in response to the researchers’ request. Using a grounded theory approach to analyzing our data (glaser & Strauss, 1967; Strauss & Corbin, 1990), we first asked the pivotal question, “What is happening here?” (glaser, 1978). To answer this question, the two authors separately performed a line-by-line coding followed by a focused coding in which the most frequent earlier codes were grouped together in categories (Charmaz, 2006). When comparing the categories from the two different analyses, nine categories emerged, which were later organ- ized into three metacategories.

Results

Acceptability of PTE Questions

as shown in Table 1, informants in this sample reported a high overall acceptance of PTe questions. Only 3% reported that they would react negatively to being asked such questions. acceptance was equally high for those who had been and those who had not been exposed to violence and/or sexual abuse.

Difficulties in Deciding How to Answer PTE Questions

Table 2 shows the proportion of respondents who confirmed having experienced the different PTes, and the proportion of respondents who reported that they found it

Thoresen, Øverlien / Trauma Victim: Yes or No? 705

difficult to decide how to answer the questions. For 10 of the 13 items, the “exposed” group reported difficulties in deciding how to answer significantly more often than the “not exposed” group.

Of the 628 informants, 65% (n = 410) reported no difficulties in answering any of the PTe items, 16% (n = 98) reported difficulty on only one item, 10% (n = 65) reported difficulties on two items, and 9% (n = 55) reported difficulties on three to six of the items.

Difficulties in answering PTe items occurred more frequently on the sexual assault and violence items (29% reported difficulties on at least one of Items 5 to 10) compared with the other items (15% reported difficulties on at least one of Items 1 to 4 or 6 to 13) (p < .001).

Having difficulties answering the questions was reported much more frequently by those who reported any trauma (43%) compared with those who reported no trauma (14%) (p < .001). Of the women who reported being exposed to sexual assault or violence, 50% had difficulties answering at least one item. In addition, the number of traumas reported was significantly associated with the number of items that respondents had difficulties answering (Pearson’s r = .34, p < .001).

Why It Was Difficult to Answer PTE Questions: Qualitative Analysis

Of the 218 women who reported that they had difficulties choosing between a “yes” and a “no” answer to at least one of the questions, 174 (80%) described why

Table 1 Reported Reactions to PTE Questions in Those Exposed

and Not Exposed to Violence and/or Sexual Abuse

Reactions (to a High or Very High Degree)a exposed, % (n) Not exposed, % (n) χ2,a

I would be upset or uneasy 5 (16) 3 (8) 1.54 I would react negatively 3 (9) 3 (8) 0.01 to being asked such questions I would react negatively to not 8 (28) 5 (15) 2.18 being asked about something I have experienced I would react positively to being 73 (250) 68 (194) 2.46 asked such questions It would be dangerous for me to 0.9 (3) 0.3 (1) 0.70 answer such questions I would have a need to talk to 8 (26) 7 (20) 0.10 somebody professional, or get other forms of support

Note: PTe = potentially traumatic event. a. None of the differences was significant at the .05 level.

706 Violence against Women

this was difficult for them. In this section, we present the results of an analysis of these descriptions. We focused on the six violence and sexual assault questions, as difficulties occurred more often on these items (Items 5-10). The nine categories that emerged from our qualitative analysis of “reasons for difficulties” were sever- ity, frequency, force, responsibility, reactions, intent of abuser, memory, protecting oneself, and protecting others. Categories representing the most frequent descrip- tions are presented first.

Severity: When does punishment become violence? Informants who found it difficult to answer yes or no to the question regarding violence in childhood reported difficulties in deciding if their experience would be considered serious enough to fit the intention of the question. Four examples are provided below, but a series of similar descriptions were given.

Table 2 Proportion of Respondents Exposed to Potentially Traumatic Events

and Proportion of Respondents Reporting Difficulties Answering Proportion Reporting Difficulties answering

Potentially Traumatic Proportion Reporting Total Sample, exposed, Not exposed, event exposure, % (n) % (n) % (n) % (n) χ2,a

1. Serious accident 14 (89) 4 (27) 14 (12) 3 (15) 21.14* 2. Stillbirth or spontaneous 12 (76) 2 (10) 1 (1) 2 (9) 0.04 abortion 3. Loss of a child 2 (12) 1 (7) 8 (1) 1 (6) 5.77 4. Loss of a family member 8 (51) 3 (20) 8 (4) 3 (16) 3.89 in accident, murder, or suicide 5. Rape 16 (100) 7 (44) 26 (26) 3 (18) 65.55* 6. Sexual assault 23 (142) 11 (70) 28 (40) 6 (30) 53.52* 7. Other unwanted sexual 15 (95) 9 (55) 24 (23) 6 (32) 33.35* experiences 8. Violence in childhood 21 (129) 9 (56) 23 (29) 5 (27) 36.66* 9. Violence in adulthood 20 (124) 3 (18) 10 (12) 1 (6) 25.68* 10. emotional violence 21 (130) 10 (62) 27 (35) 5 (27) 53.26* 11. Witnessing violence 12 (73) 3 (20) 12 (9) 2 (11) 22.35* or sexual assault 12. Victim of criminal 11 (71) 3 (19) 10 (7) 2 (12) 12.71* offense 13. Other life-threatening 11 (71) 4 (23) 18 (13) 2 (10) 48.57* events

a. χ2 for the difference between exposed and not exposed. *p < .01.

Thoresen, Øverlien / Trauma Victim: Yes or No? 707

Because it was a part of child raising, and it wasn’t perceived or intended as maltreatment.

My father punished me, slapped my back when I did something wrong. How much is “injured”? Is a burning chin enough? I experienced “controlled” spanking as a child, but cannot remember that I perceived

this as abuse.

The informants seemed to struggle with the borderline between child maltreatment and physical punishment, which they perceived to be part of child raising.2

Contrary to the questions regarding violence in childhood, only a few informants reported difficulties in answering the question regarding violence in adulthood. The few comments from informants related to the degree of severity of the violent act and the degree of injuries.

In relation to sexual abuse, some informants found it difficult to decide if their expe- riences were severe enough to qualify for a “yes” answer. In addition, informants strug- gled to determine whether the act involved could be termed “sex”; whether the parts of the body touched, or the parts of the body they had to touch, qualified as “private parts”; and if “touching” was a correct term if their clothes were on. For example,

I don’t think it was serious enough. Because the rape was not completed. I’m not sure if that . . . could be called intercourse. Because it was outdoors, with my clothes on. I was forced to touch parts other than the private parts.

Frequency: What if it happened only once? Informants reported problems in deciding if what they experienced occurred sufficiently often, or for a sufficiently long period, to qualify for a “yes” answer. Such considerations related to violence in childhood, psychological violence, and sexual abuse. For instance,

It happened only once; I got away from it. Violence is a strong word. Is one single blow violence?3

Pressured to intercourse, but only one single episode. . . . it wasn’t systematic. Only once in a while . . .

Force: Hard to define. Similarly, informants reported it difficult to decide the degree of force or pressure that had been used against them. However, this related only to sexual abuse.

What if you are afraid of what will happen if you don’t give in? Force, but not power. How much threat . . . it’s hard to say. It’s hard to evaluate how much force . . . in hindsight—yes, but back then I didn’t think

of it like that.

708 Violence against Women

To what extent did the abuser induce fear, use physical or psychological threat, or use physical force? as can be seen from the examples, these terms were not perceived as categorical phenomena, which made it difficult to answer yes or no to the questions. as indicated below, the force or pressure used by the abuser was perceived to occur in an interaction with the participant, and many informants consequently questioned their (in)ability to stop the event from occurring.

Responsibility: Perhaps it was my own fault? In this study, considerations regard- ing responsibility occurred frequently in association with sexual assault questions, but were otherwise almost nonexistent. Considerations regarding responsibility ranged from giving in because of fear of the consequences of saying no, changing states of desire and fear, a reported inability to protest (e.g., because of being under- age), an inability to fight back, unspoken threats, and taking responsibility for “building up to something.” For example,

Because I was partly willing. . . . was it something I chose to put myself through . . . or was it force . . .? I sort of built up to it myself . . . flirting, kissing . . . hadn’t thought it would go so far

as to having sex. Locked the car doors, touched me, and I protested. Now I know I just wasn’t tough enough to say no . . . Because I didn’t fight back. Because my reasons to refuse were moral, not due to lack of desire . . . but no doubt

there was pressure . . . was physically held down in spite of loud protests.

Parallel to the perception of force not being a categorical phenomenon, desire, doing something against one’s will, saying no, and resistance were perceived as states and actions that could be present to varying degrees. Informants expressed difficulties in deciding if it was the abuser who used force or pressure, or if it was the informant who allowed it to happen. For the informant, her own actions in the situation leading up to the event seemed to be of pivotal importance for how she understood and conceptualized the event that followed, and hence, how she responded to questions about sexual abuse.

Reactions: When it doesn’t bother you, does it count? another aspect of the women’s decision-making process was connected to their reactions as victims.

I . . . have not carried it with me to any great extent. Have not seen the events as being categorized as traumatizing. I don’t have any problems attached to this, and it happened a long time ago. When you are not particularly bothered by it afterwards, then does it count? I remember being hit by my mother, but this is not something I have a bad feeling about

today.

Thoresen, Øverlien / Trauma Victim: Yes or No? 709

This category of answers has nothing to do with the abuser or the situation itself, but relates to the victim. absence of negative emotional reactions seemed to influence the informant toward a “no” answer. For a situation of coerced sex to be defined as abuse, for example, there seemed to be a preconceived notion that the victim would not only have strong negative reactions but that these reactions should be long lasting.

The intention of the abuser: What if he didn’t do it on purpose? Our informants seemed to take into account the purpose or intentions of the abuser.

What about when it is done with the best intentions? I don’t think my mother did it because she is evil. Not systematic, and against his intelligence, thereby not planned. The behavior was a part of him and not directed only at me.

If the respondent understood the intention of the abuser to be good, spontaneous, or as against his or her better knowledge, the respondent may have been inclined to answer no to the questions of abuse and violence.

Memory: What if I can’t remember it properly? Problems with recollection of specific events were mentioned by a few, although this topic did not dominate the women’s descriptions. “Memory considerations” ranged from very vague and gen- eral to a more confident perception that “something” happened but that recollections of this event were not adequately precise. Reasons for memory problems were attrib- uted to the long period of time that had passed since the event, repression, and problems in separating recollections of several different events. For instance,

Think you forget things that you don’t want to remember. Sometimes I get the feeling that something happened when I was a child—an uncle and

a basement. I’m so unsure. . . . Did anything happen, did I repress it, or do I just imagine?

I experienced something when I was a child. Things get repressed, especially when there are several things . . . not sure. It’s a long time ago. I buried these things far back in my brain.

These examples relate to both violence and sexual assault items. Several of the women with “memory considerations” reported having difficulties choosing between “yes” and “no” on several items.

Protecting oneself: Me, a victim? a few of our respondents did not have the energy or strength to answer our questions, or found it hard to be reminded of an incident that belonged in their past. Others pondered whether a positive answer defined them in a way they did not want to be defined.

710 Violence against Women

Don’t have the strength. I don’t like to dwell on things that have been difficult. Feels a bit like I’m exposing myself. Because it’s embarrassing to tell that both as a child and as an adult you have been a

victim of . . . humiliations. Hard to admit to oneself. I don’t see myself as a victim.

answering no to the survey question can be understood as a way of protecting oneself from painful memories and from putting a label on oneself that may feel negative and not in accordance with the respondent’s self-image. What does it do for a person to answer yes to a question that defines her or him as a victim? What con- sequences does the check mark in the “yes” box have for the person’s identity?

Protect others: If I tell, he will look bad. The last category found among our respondents has to do with wanting to protect others. Defining someone as an abuser also labels the person in a way the respondent may not wish to do.

gives a negative impression of him. By marking “yes” . . . [it] feels like involving the family. I don’t define it as violence but as my mother’s helplessness. I told my mother, and she said I should be careful not to say such things.

Comments such as these give the impression that the respondent wants, or is told by others, to protect the abuser from being put in a bad light. By answering yes to the survey question, the respondent not only defines the person as an abuser, an offender, or a person committing a crime (i.e., as a “bad” person) but also as a person with “bad” intentions. By redefining violence into helplessness, the mother cannot be defined as an abuser, but rather as a victim who needs protection herself. What consequences does a “yes” have for not only me but also the person who did this to me and the image I have of him or her?

Three Metacategories

We grouped the nine categories described above into three metacategories, relating to “event fit,” “me (victim?),” and “you (abuser?).” “event fit” relates to the different aspects of deciding whether the individual experience belongs to the population of events intended in questions, such as “severity” (e.g., how hard was the blow, how much injury was there), “force” (e.g., how much physical force was used, did what he said qualify as a threat), and “frequency” (e.g., it happened only once, it was not typi- cal). “event fit” was by far the most frequent reason given for difficulties in deciding how to answer violence and sexual abuse questions. “Me (victim?)” relates to aspects

Thoresen, Øverlien / Trauma Victim: Yes or No? 711

of the informant, such as “responsibility” (e.g., communication of unwillingness to participate), “reactions” (e.g., using the current absence of stress reactions to define the event as not qualifying for a “yes” answer), “protection” (e.g., protecting against the self-perception of being a victim and against stressful reminders), and “memory” (e.g., lack of precise recollections). The last metacategory, “you (abuser?),” concerns the abuser and the informant’s relationship to him or her, such as the “intention” (e.g., informants seemed reluctant to give a “yes” answer for behaviors that were perceived as unintentional or explained by other factors, such as helplessness), and “protection” (e.g., wanting to protect the abuser from looking bad). Certain aspects of these meta- categories overlapped to some extent. For example, for some, the issue of frequency was related to the evaluation of severity, and responsibility for the event was related both to the use of force and to the victim’s inability to stop the event from happening.

although informants reported more difficulties with violence and sexual abuse questions, there were also several comments on the other PTe items. Most related to different aspects of “event fit,” such as, what is to be considered a “serious” acci- dent; what is to be considered “an injury”; what happens when you witness sexual acts, but you don’t know if they were abusive or not; what if you knew your girl- friend was abused by her dad, but you never actually saw it; how are you to evaluate if a threat was serious or not; and in a natural disaster, such as an earthquake, how do you know if your life is threatened or not.

Discussion

The aim of this study was to investigate whether questions regarding violence, sexual abuse, and other PTes were acceptable, if such questions were difficult to answer, and if so, why they were difficult.

Was it acceptable to be asked such questions? The vast majority of informants in the current study reported positive attitudes to being asked questions regarding vio- lence, sexual abuse, and other PTes. This is in line with previous research that shows that questions regarding disaster experiences (galea et al., 2005) and childhood victimization (Newman et al., 1999) are well tolerated, and that most informants judge their participation favorably (Newman & Kaloupek, 2004). acceptance was equally high in those exposed to violence and/or sexual abuse and those not exposed, which has also been previously reported (Black et al., 2006). It does not seem likely, then, that the reported difficulties in answering the PTe questions in this study could be explained by lack of acceptance.

Was it difficult to answer these questions? The majority of women in this study (65%) did not find it difficult to answer any of the PTe questions. However, sig- nificantly more informants reported difficulties in choosing between “yes” and “no” when answering violence and sexual abuse questions compared with other PTe items. In addition, having difficulties answering was related to trauma; difficulties

712 Violence against Women

were reported significantly more often in informants who had experienced violence or sexual assault, and the number of traumatic events was associated with the number of items that were difficult to answer.

Why was it difficult to answer? When analyzing the comments made by the women about why the questions were difficult to answer, nine different—but also clearly linked—categories were found, out of which three metacategories emerged: (a) “event (fit?)” (severity, frequency, force); (b) “Me (victim?)” (responsibility, protecting oneself, memory, reactions), and (c) “You (abuser?)” (protecting the abuser, intention of the abuser). The first was by far the most dominant category.

The majority of reasons given for why it was difficult to answer concerned the question of whether the personal experience would qualify for a “yes” answer: Was the experience “serious” enough; was the physical contact sufficiently hard; did it last long enough or occur with sufficient frequency; was enough force or threat used? Informants in this study seemed to possess a more multidimensional view of potentially traumatic experiences than a brief screening instrument could encom- pass; several dimensional phenomena (e.g., “serious,” “life threat,” “force”) were summarized into a qualitative (dichotomous) question demanding a “yes” or “no” answer. Confronted with this rather difficult task, it is possible that only prototype events will elicit a straightforward “yes” answer. Other, less prototypical events may leave the informant with a need to interpret and evaluate several aspects of the event. These results are in line with a previous study that showed that reporting of abuse depends, among other factors, on the severity and frequency of the abusive acts (Katerndahl, Burge, & Kellogg, 2006).

Responsibility issues were uniquely related to sexual abuse questions. Informants gave descriptions of a variety of situations in which their own behavior may have been related to the event occurring. These comments seemed to range from a bal- anced retrospective analysis of responsibility to self-blame. Victims who take responsibility for the violence and blame themselves have been discussed by Loseke (2001) and Leisenring (2006), among others. These issues are similar to the “issues of consent” described by Harned (2005) as one important aspect of women labeling unwanted sex as abuse and may well be one reason for getting a “no” response to events in which the researcher would have expected a “yes” answer. However, less attention has been given to the dimensional aspects of consent that were discussed by informants in this study. Lack of negative emotional long-term reactions was used by some informants to retrospectively conceptualize the event as “not qualifying” for a “yes” answer. If this reasoning is widespread in research participants, it may be a serious threat to the validity of the association between traumatic exposure and subsequent mental health or social problems. Problems with recollection of specific events were mentioned by some informants in this study, although much less fre- quently than we expected considering the extensive literature on memory and trauma.

Only a few informants commented that they did not have the strength to answer, or that answering would activate reminders. More frequently, questions regarding

Thoresen, Øverlien / Trauma Victim: Yes or No? 713

violence and sexual abuse seemed to touch on the informants’ self-perception and identity: Who am I to be abused? Some informants seemed to perceive answering the Web-based questionnaire as interactive, and felt that selecting “yes” would affect the way they perceived themselves and were perceived by others (in this case, the researchers). The informants were not only hesitant to answer yes because they did not wish to expose themselves to a possible negative emotional memory, but also because they did not wish to identify themselves as victims. We understand this as protection against the negative attitudes and values of being a victim of violence and sexual abuse in society as a whole. as Leisenring (2006) argues, there are strong cultural discourses that depict abused women as blameworthy and powerless. Berns and Schweingruber (2007) found that informants who have not been victimized describe victims of abuse as weak and lacking personal qualities that could have prevented the abuse. Consequently, victims are stigmatized (goffman, 1968). Claiming a victim identity is thereby risky business.

The relationship between the abuser and the abused is of great importance in understanding the decision-making process of the women (e.g., Theran, Sullivan, Bogat, & Sutherland, 2006). Several of our informants gave explicit comments that they felt reluctant to answer yes because this might give a negative impression of the person who did this to them. In addition, the presumed intention of the “abuser” was taken into account when answering the questions. Can you be an abuser without having the intention to abuse? Similar findings were reported by Sedlak (1988). In her examination of the labeling of violence, she found that the decision-making process was influenced by the victim’s attributions of the abuser’s intelligence and motivation. as researchers, we would prefer our informants report their experiences with violence and abuse without considering how the abuser may be perceived by the researchers and without speculations as to the intention of the abuser. However, this and previous studies indicate that some informants do take these factors into consideration.

Implications

Several previous studies have focused on issues regarding the informant to explore reasons for inconsistent reporting of PTes, including memory and mental health symptoms (e.g., Della Femina et al., 1990; Ouimette et al., 2005). In our opinion, the results of this study may contribute to a move away from the informant and toward the way PTe screening instruments are designed. The “yes” or “no” response categories consistently used in screening instruments when referring to dimensional phenomena may leave the informant with a less-than-optimal under- standing of how to answer the questions. This may be one explanatory factor for the observed inconsistency in PTe reporting.

The main implication of this study is an increased awareness of the complexity of such phenomena and a call for caution in how epidemiological results in this field may be interpreted. Previous research has shown that behaviorally specific questions,

714 Violence against Women

compared to labeling questions, are more likely to result in a “yes” response (Harned, 2004; Thombs et al., 2006). In the present study, labeling terms such as “rape” were avoided, and efforts were made to include behaviorally specific questions. However, a possible way to further reduce ambiguity in the questions may be to operationally define important concepts such as “force” and “pressure” (O’Sullivan, Byers, & Finkelman, 1998). One important issue for this research field may be the balance between brevity and precision. On one hand, we strive to keep screening instruments brief and the questions short and easy to read. On the other hand, we strive to achieve precise measurement, which implies that at the least, the informants clearly understand if their experiences fall into or outside of the category covered by the question. The “ease” of answering such questions and the balance between brevity and precision could be considered an empirical question that has not been suffi- ciently investigated and one that could be further researched using both qualitative (e.g., green et al., 2006) and quantitative (e.g., Thombs et al., 2006) methods. The difficulty of informants’ using their own emotional reactions to the event and their interpretation of abuser intent as guides for whether to answer yes to a PTe question may possibly be overcome by specific instructions to the informant not to consider such factors. However, such a proposal needs to be investigated empirically.

Limitations

The current study was not designed to be representative of any specific group. The reported acceptance, proportions exposed to different PTes, and proportion who found it difficult to answer the questions are all results that may be highly affected by the selected population (women with children of a certain age), response bias, research method, time period, and culture.

The aim of this study was to shed light on the decision-making process of women informants and the reasons why they found it difficult to answer PTe ques- tions. Results should be considered examples from a potential range of considera- tions that individuals may have when answering traumatic event questionnaires. The considerations described in this study may well have relevance for informants of another gender or age, in another setting, or responding to somewhat differently formulated questions. However, improved clarity of the questions might have been achieved through defining concepts such as “pressure” and “force.” although violence and sexual abuse of women is a worldwide social problem, it is also a temporally and culturally dependent phenomenon. Issues regarding violence and sexual abuse have been understood differently in different time periods, and are understood differently in different cultures. The population in the current study comprised mostly working women in a certain age group, with minor children, and a predominantly Norwegian background. The willingness to respond to these sen- sitive issues might be regarded as highly culture or sample dependent. However, we consider it less likely that the difficulties answering survey questions regarding these sensitive taboo issues would be restricted to Norwegian women and would

Thoresen, Øverlien / Trauma Victim: Yes or No? 715

expect similar difficulties to be found among respondents living in other areas of europe or North america. This expectation requires empirical investigations with other informant groups.

The qualitative data were restricted to the rather short descriptions given by the informants, without the possibility of further exploration or reflection, which could have been given in an interview setting. However, anonymous written comments, such as those made by the informants in this study, may foster frank disclosures that would not emerge during an interview, especially if the topic is sensitive and could result in the informant feeling shame or disgrace (Charmaz, 2006).

Other strengths of this study include the rather large sample size, the large propor- tion of women who reportedly found it difficult to answer the questions and actually added a description of why this was difficult, the large spectrum of descriptions given by the informants, and the fact that to our knowledge this is the first study of its kind.

Concluding Thoughts

Our intention with this article is by no means to suggest that informants have misunderstood the survey questions or have a confused understanding of the phe- nomena. Finding it difficult to choose between “yes” and “no” is not the same as giving “incorrect” answers. On the contrary, we consider the informants’ comments to be important contributions to the ongoing discussion of defining and measuring violence and sexual abuse.

Appendix SLESQ-Adapted

1. Were you ever in a serious car accident or other accident in which you were seri- ously injured or your life was in danger?

2. Did you ever experience stillbirth or missed abortion after week 12 of pregnancy? 3. Have you ever lost a child (from death)? (biological child, foster child, adopted

child, step child or partner’s child that you have lived together with)? 4. Has an immediate family member or romantic partner died because of accident,

homicide, or suicide? 5. Has anyone (parent, other family member, romantic partner, stranger, or someone

else) ever forced you to have intercourse, or to have oral or anal sex against your wishes? (either by physically forcing you, by threatening you, or by taking advan- tage of a situation in which you could not defend yourself, e.g. because you were asleep or unconscious/intoxicated, or that you were underage.)

6. Other than experiences mentioned in earlier questions: Has anyone ever touched private parts of your body or made you touch private parts of their body against your wishes?

(continued)

716 Violence against Women

7 Other than experiences mentioned in earlier questions: Have you ever been pres- sured to engage in other sexual situations or acts against your wishes?

8. Before age 16: Did a parent, caregiver, or other adult person ever kick you, hit you, or otherwise attack or harm you?

9. after age 16: Have you ever been kicked, beaten, slapped around, or otherwise physically harmed by a romantic partner, date, family member, stranger, or some- one else?

10. Has a parent or a romantic partner systematically ridiculed you, put you down, or told you you were no good?

11. Have you ever witnessed another person being killed, seriously injured, or sexually or physically assaulted?

12. Other than the experiences mentioned in earlier questions: Were you ever exposed to a criminal act in which you could have been seriously injured or your life was in danger (robbed, taken hostage, kidnapped, or attacked or threatened with a weapon)?

13. Other than the experiences mentioned in earlier questions: Have you ever been in any other situation where you could have been seriously injured or your life was in danger (e.g. experienced war, natural disaster, fire)?

all questions had the response categories “yes” or “no, never.” If a “yes” answer was given, the respondent was asked to specify in which time period the event occurred: Before age 16, after age 16, and past 12 months (the specification of time period was omitted for Item 8 (violence in childhood), and the time period “before age 16” was omitted for Item 9 (violence in adulthood). Item 5 (rape) and Item 9 (violence in adult- hood) asked in addition for frequency (“Please specify how many times this happened to you”: 1, 2-4, 5-10, more than 10 times). Item 9 (violence in adulthood) asked in addition for the identity of the perpetrator (“stranger, family or relative, other known person”).

Notes

1. The use of the word victim in this article may seem counterproductive, as our aim is by no means to further emphasize the status of women as victims. Furthermore, the Web survey did not use the word victim. However, our analysis shows that the women relate to the question of being a victim when decid- ing if they should answer yes or no to the survey questions. The choice of the word victim in the title of the article, for example, is thereby part of our analytical work.

2. Spanking, and other physical punishment of children, is forbidden by Norwegian law. 3. The word violence was not used in the question.

References

Becker-Blease, K. a., & Freyd, J. J. (2006). Research participants telling the truth about their lives. The ethics of asking and not asking about abuse. American Psychologist, 61, 218-226.

Berns, N., & Schweingruber, D. (2007). “When you’re involved, it’s just different”: Making sense of domestic violence. Violence Against Women, 13, 240-261.

Appendix (continued)

Thoresen, Øverlien / Trauma Victim: Yes or No? 717

Black, M. C., Kresnow, M. J., Simon, T. R., arias, I., & Shelley, g. (2006). Telephone survey respond- ents’ reactions to questions regarding interpersonal violence. Violence and Victims, 21, 445-459.

Bramsen, I., Dirkzwager, a. J., van esch, S. C., & van der Ploeg, H. (2001). Consistency of self-reports of traumatic events in a population of Dutch peacekeepers: Reason for optimism? Journal of Traumatic Stress, 14, 733-740.

Charmaz, K. (2006). Constructing grounded theory. A practical guide through qualitative analysis. London: Sage.

Creamer, M., Burgess, P., & McFarlane, a. C. (2001). Post-traumatic stress disorder: Findings from the australian National Survey of Mental Health and Well-Being. Psychological Medicine, 31, 1237-1247.

Della Femina, D., Yeager, C. a., & Lewis, D. O. (1990). Child abuse: adolescent records vs adult recall. Child Abuse & Neglect, 14, 227-231.

Dube, S. R., anda, R. F., Whitfield, C. L., Brown, D. W., Felitti, V. J., Dong, M., et al. (2005). Long-term consequences of childhood sexual abuse by gender of victim. American Journal of Preventive Medicine, 28, 430-438.

engels, R. C., Knibbe, R. a., & Drop, M. J. (1997). Inconsistencies in adolescents’ self-reports of initia- tion of alcohol and tobacco use. Addictive Behaviors, 22, 613-623.

Fergusson, D. M., Horwood, L. J., & Woodward, L. J. (2000). The stability of child abuse reports: a longitudinal study of the reporting behaviour of young adults. Psychological Medicine, 30, 529-544.

Follette, V. M., Polusny, M. a., Bechtle, a. e., & Naugle, a. e. (1996). Cumulative trauma: The impact of child sexual abuse, adult sexual assault, and spouse abuse. Journal of Traumatic Stress, 9, 25-35.

Frans, O., Rimmo, P. a., aberg, L., & Fredrikson, M. (2005). Trauma exposure and post-traumatic stress disorder in the general population. Acta Psychiatrica Scandinavica, 111, 291-299.

Frueh, B. C., Hamner, M. B., Cahill, S. P., & gold, P. B. (2000). apparent symptom overreporting in combat veterans evaluated for PTSD. Clinical Psychology Review, 20, 853-885.

galea, S., Nandi, a., Stuber, J., gold, J., acierno, R., Best, C. L., et al. (2005). Participant reactions to survey research in the general population after terrorist attacks. Journal of Traumatic Stress, 18, 461-465.

glaser, B., & Strauss, a. L. (1967). Discovery of grounded theory. Chicago: aldine. glaser, B. (1978). Theoretical sensitivity. Mill Valley, Ca: Sociology Press. goffman, e. (1968). Stigma: Notes on the management of spoiled identity. Harmondsworth, UK: Penguin. goodman, L., Corcoran, C., Turner, K., Yuan, N., & green, B. (1998). assessing traumatic event expo-

sure: general issues and preliminary findings for the stressful life events screening questionnaire. Journal of Traumatic Stress, 11, 521-542.

graham-Bermann, S. a., & Seng, J. (2005). Violence exposure and traumatic stress symptoms as addi- tional predictors of health problems in high-risk children. Journal of Pediatrics, 146, 349-354.

green, B. L., Chung, J. Y., Daroowalla, a., Kaltman, S., & DeBenedictis, C. (2006). evaluating the cultural validity of the Stressful Life events Screening Questionnaire. Violence Against Women, 12, 1191-1213.

green, B. L., goodman, L. a., Krupnick, J. L., Corcoran, C. B., Petty, R. M., Stockton, P., et al. (2000). Outcomes of single versus multiple trauma exposure in a screening sample. Journal of Traumatic Stress, 13, 271-286.

Hardt, J., & Rutter, M. (2004). Validity of adult retrospective reports of adverse childhood experiences: Review of the evidence. Journal of Child Psychology and Psychiatry, 45, 260-273.

Harned, M. S. (2004). Does it matter what you call it? The relationship between labeling unwanted sexual experiences and distress. Journal of Consulting and Clinical Psychology, 72, 1090-1099.

Harned, M. S. (2005). Understanding women’s labeling of unwanted sexual experiences with dating partners. a qualitative analysis. Violence Against Women, 11, 374-413.

Hepp, U., gamma, a., Milos, g., eich, D., aidacic-gross, V., Rossler, W., et al. (2006). Inconsistency in reporting potentially traumatic events. British Journal of Psychiatry, 188, 278-283.

Jarvis, K. L., gordon, e. e., & Novaco, R. W. (2005). Psychological distress of children and mothers in domestic violence emergency shelters. Journal of Family Violence, 20, 389-402.

Katerndahl, D., Burge, S. K., & Kellogg, N. (2006). Factors that predict how women label their own childhood sexual abuse. Journal of Child Sexual Abuse, 15, 43-54.

718 Violence against Women

Kessler, R. C., Sonnega, a., Bromet, e., & Nelson, C. B. (1995). Posttraumatic stress disorder in the National Comorbidity Survey. Archives of General Psychiatry, 52, 1048-1060.

Krinsley, K. e., gallagher, J. g., Weathers, F. W., Kutter, C. J., & Kaloupek, D. g. (2003). Consistency of retrospective reporting about exposure to traumatic events. Journal of Traumatic Stress, 16, 399-409.

Leisenring, a. (2006). Confronting “victim” discourses: The identity work of battered women. Symbolic Interaction, 29, 307-330.

Loseke, D. (2001). Lived realities and formula stories of “battered women.” In J. F. gubrium & J. a. Holstein (eds.), Institutional selves: Troubled identities in a postmodern world (pp. 106-126). New York: Oxford University Press.

Nemeroff, C. B., Bremner, J. D., Foa, e. B., Mayberg, H. S., North, C. S., & Stein, M. B. (2006). Posttraumatic stress disorder: a state-of-the-science review. Journal of Psychiatric Research, 40, 1-21.

Newman, e., & Kaloupek, D. g. (2004). The risks and benefits of participating in trauma-focused research studies. Journal of Traumatic Stress, 17, 383-394.

Newman, e., Walker, e. a., & gefland, a. (1999). assessing the ethical costs and benefits of trauma- focused research. General Hospital Psychiatry, 21, 187-196.

Norris, F. H. (1992). epidemiology of trauma: Frequency and impact of different potentially trauma- tic events on different demographic groups. Journal of Consulting and Clinical Psychology, 60, 409-418.

Norris, F. H., & Hamblen, J. L. (2004). Standardized self-report measures of civilian trauma and PTSD. In J. P. Wilson & T. M. Keane (eds.), Assessing psychological trauma and PTSD (2nd ed., pp. 63-121). New York: guilford.

O’Sullivan, L. F., Byers, e. S., & Finkelman, L. (1998). a comparison of male and female college stu- dents’ experiences of sexual coercion. Psychology of Women Quarterly, 22, 177-195.

Ouimette, P., Read, J., & Brown, P. J. (2005). Consistency of retrospective reports of DSM IV criterion a traumatic stressors among substance use disorder patients. Journal of Traumatic Stress, 18, 43-51.

Resnick, H. S., Kilpatrick, D. g., Dansky, B. S., Saunders, B. e., & Best, C. L. (1993). Prevalence of civilian trauma and posttraumatic stress disorder in a representative national sample of women. Journal of Consulting and Clinical Psychology, 61, 984-991.

Roemer, L., Litz, B. T., Orsillo, S. M., ehlich, P. J., & Friedman, M. J. (1998). Increases in retrospective accounts of war-zone exposure over time: The role of PTSD symptom severity. Journal of Traumatic Stress, 11, 597-605.

Sedlak, a. (1988). The effects of personal experience with couple violence on calling it “battering” and allocating blame. In T. gerald, D. Finkelhor, J. T. Kirkpatrick, & M. a. Strauss (eds.), Coping with family violence: Research and policy perspectives (pp. 31-59). Thousand Oaks, Ca: Sage.

Skog, O. J. (1992). The validity of self-reported drug use. British Journal of Addiction, 87, 539-548. Southwick, S. M., Morgan, C. a., 3rd, Nicolaou, a. L., & Charney, D. S. (1997). Consistency of memory

for combat-related traumatic events in veterans of Operation Desert Storm. American Journal of Psychiatry, 154, 173-177.

Spinhoven, P., Bean, T., & eurelings-Bontekoe, L. (2006). Inconsistencies in the self-report of traumatic experiences by unaccompanied refugee minors. Journal of Traumatic Stress, 19, 663-673.

Strauss, a., & Corbin, J. a. (1990). Basics of qualitative research: Grounded theory procedures and techniques. Newbury Park, Ca: Sage.

Theran, S. a., Sullivan, C. M., Bogat, g. a., & Sutherland, C. S. (2006). abusive partners and ex-partners. Understanding the effects of relationship to the abuser on women’s well-being. Violence Against Women, 12, 950-969.

Thombs, B. D., Bernstein, D. P., Ziegelstein, R. C., Scher, C. D. Forde, D. R., Walker, e. a., et al. (2006). an evaluation of screening questions for childhood abuse in 2 community samples. Implications for clinical practice. Archives of Internal Medicine, 166, 2020-2026.

Thoresen, Øverlien / Trauma Victim: Yes or No? 719

Siri Thoresen is a researcher at the Norwegian Centre for Violence and Traumatic Stress Studies. She was trained as a clinical psychologist and received her PhD from the University of Oslo, Norway. Her research interests include traumatic stress, gender, suicide, social psychology, and public health.

Carolina Øverlien is a researcher at the Norwegian Centre for Violence and Traumatic Stress Studies. She has a PhD in Child Studies from Linköping University, Sweden. Her research interests include chil- dren and adolescents exposed to domestic violence, physical violence and sexual abuse, qualitative methods, and feminist theory.