The study showed that 17 women (14.7%) had been diagnosed as PTSD.
of PTSD correlated with minor psychological abuse, severe psychological abuse, minor physical abuse, severe physical abuse, and physical injury.

Also, 18% of women showed moderate to sever anxiety symptoms. The results showed that total scores of anxiety (TMAS) were correlated with minor physical abuse, severe physical abuse, and physical injury.

According to the study, 17 women showed moderate to severe depression (14.7%) (0.9%). The results showed that total scores of depression (BDI-II) correlated with severe psychological abuse, minor physical abuse, severe physical abuse, and physical injury.

Sugihara & Warner (2002) in study of gender differences in dominance and aggressive behavior in intimate relation-ships among Mexican Americans found that 99% of women used negotiation in their relationships .Eighty five percent of women reported the use of psychological aggression, whereas 48% reported that they had physically assaulted their partners at least once in the last year. It was also found that 17% of women reported that they had inflicted injury upon their partners during the past year. McCloskey et al (2002) in a comparative study of 82 battered women with children who sought counseling in Italy and the United States. The result showed that Italian women (30%)reported sexual abuse from their partners than either Hispanic or Anglo American women (more than 70%).

Recent data from the Bureau of Statistics found that 691,710 nonfatal and

1,247 fatal violent victimizations were committed by intimate partners in the

United States in 2001.This number only makes up 20%of the violent crimes

against women in the United States (Rennison, 2003).

A number of domestic violence researchers have noted that most research on determinants and correlates of domestic violence has overemphasized both distal factors (e.g., early exposure to violence) and individual traits (e.g., alcohol abuse, self-esteem), and has substantially underemphasized relational patterns and proximal interaction processes (Feldman & Ridley,1995; Infante & Wigley, 1986; Lloyd,1990; Margolin et al.,1988).

Violence has negative consequences on physical and mental health. Women who have been victims of recent or past violence, besides experiencing the short and long term consequences of physical injuries, are more likely to suffer from a variety of health problems, including post traumatic stress, depression, drug abuse, eating, sexual, gynecological, and gastrointestinal disturbances (Fleming, Mullen, Sibthorpe, & Bammer, 1999; Kilpatrick, Acierno, Resnick, Saunders, & Best, 1997; Plichta & Abraham, 1996).

Recent studies revealed that there are variety of mental health problems is women exposed to violence including depression, PTSD and anxiety (Leisring et al, 2003; Sharhabani-Arzy et al, 2004; Feerick & Snow, 2005; Pillay and Kriel, 2006; Logan et al, 2006; )

The aim of the study was to investigate types and level of domestic violence against Palestinian women and mental health problems such as depression, anxiety, and PTSD.

Subjects and method

Subjects

The study population sample included 125 Palestinian women in the Gaza Strip aged from 18 to 50 years (mean age = 29.55 years, SD = 7.75). They were randomly selected as a pilot sample of survey of domestic violence among women in the Gaza Strip. A group of 10 community workers who had been working in women empowerment project visited the women in their homes for the first time in July 2006. Each woman was asked to read the formal written consent form before signing for agreement to participate in this study. From total 125 women selected 116 of them agreed to participate in this study with response rate was 92.8%. This study was conducted on July 2006. A total of 116 questionnaires were returned, giving a response rate of 92.8%.

Instruments

The instruments used in this study were Arabic translations Conflict Tactics Scale (CTS-2) without sexual coercion subscale, BDI-II, Manifest Anxiety Scale, Posttraumatic Stress Disorder Checklist (PCL), and a Demographic Questionnaire developed by the authors of the present study.

Demographic and social characteristics

Data were collected on the women’s socio-demographic characteristics, including age, number of children, place of residence, education, marital status, and occupational status, and husband job.

Revised Conflict Tactics Scale (CTS2; Straus, Hamby, Boney-McCoy & Sugarman, 1996)

We used CTS to assess the prevalence of domestic violence among sampled participants. The previous version of CTS was used in previously in Arabic culture (Haj Yehia, 2000). The Arabic translated version of the instrument was used which was translated and back translated and only minor differences between the two translations was found. Later on, this Arabic version was given to a panel of four bilingual and bicultural Arabic researchers to ensure congruence across both the Arabic and English versions of the instrument. We excluded the sexual violence items and the final Arabic version consisted of 33 items after exclusion of the six items of the sexual coercion. This self-report instrument measures the frequency of physical abuse, verbal violence, sexual violence, reasoning tactics, and domestic violence injury within an intimate relationship. Participants were asked to report the frequency with which they performed the various behavioral conflict resolution tactics in the previous year. Frequencies are on a 7-point scale ranging from never (0) to more than 20 times a year (6). On the scale, 7 represents that abusive tactics did not happen in the previous year but they had occurred in the past. For the present study, 7 on the scale was only used to determine the prevalence of lifetime physical abuse among the participants. This study only examined prevalence of partner abuse among male participants; therefore, scale items about partner behavioral tactics during a conflict were omitted. The Cronbach’s alpha coefficient for the CTS-2 scales ranges from .79 to .95 for the U.S. population (Straus et al., 1996). For the present study, the scale had acceptable reliability coefficients, ranging from .92 to .74.

PTSD (Posttraumatic Stress Disorder Checklist)

To provide background information, the PCL contains 17 items adapted from Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV; American Psychiatric Association, 1994) PTSD symptom criteria, taking approximately 5 minutes to administer. Respondents are asked to rate on a 5-point Likert scale (1 = not at all to 5 = extremely) the extent to which symptoms bothered them in the previous month. The PCL demonstrates adequate internal consistency (alpha = .94) (Blanchard, Jones-Alexander, Buckley, & Forneris, 1996). Diagnostic efficiency in PTSD detection has using cutoff scores of 50, with samples of victims of motor vehicle accidents and sexual assault (Blanchard et al., 1996). In this study internal consistency of the scale, calculated using Cronbach’s alpha was PTSD .89, and split-half was .83.

Manifest Anxiety Scale (MAS) (Reynolds & Richmond, 1978, 1997)

Taylor (1953) developed one of the first measures of chronic, manifest

anxiety, Taylor’s Manifest Anxiety Scale (MAS). Taylor’s scale consisted of

items selected from the Minnesota Multiphase Personality Inventory

Taylor’s MAS was reported to be useful in identifying adults with chronic anxiety (Reynolds & Richmond, 1978, 1997). We used the Arabic version with 50 items and answer is "Yes" or "No". The score ranged from 0-26 (no anxiety), 27-32 (Mild anxiety), 33-40 (severe anxiety), and 41 and above (very severe anxiety). In this study internal consistency of the scale, calculated using Cronbach alpha was = .83 and split-half was .78.

Beck Depression Inventory (Beck et al, 1988) Arabic version

The original form of (BDI-II) contains 21 items and aims to assess quantitatively the severity of depression are used in this study, it also has a great benefit in clarification of the cognitive aspects of depression. The severity of depression is classified on the basis of the total score as the following: In normal community sample, a BDI score <20 suggests no or minimal depression, 21 to 31 represents mild to moderate depressive affects, 32 to 41 is moderate to severe, and <= 43 indicates severe levels of depression (Gareeb, 2000). In this study, the split half reliability of the scale was high (r = 0.62) internal consistency of the scale calculated as Cronbach alpha, was also high (alpha = 0.82).

The SPSS Version 11.0 computer program was used in the analysis of the data.

The frequencies of categorical data are presented. Established cut-off scores were used to provide rates of PTSD, depression, and anxiety. The relationship between domestic violence, demographic variables, PTSD, depression, and anxiety was investigated by Spearman correlation since the continuous scores were not normally distributed. Differences in means of violence and mental health problems and sociodemographic variables were tested by One way ANOVA. The associations between violence, mental health problems and sociodemographic data were explored by linear regression analyses, with all variables entered together in each model.

The study sample included 116 Palestinian women in the Gaza Strip aged from 18 to 50 years (mean age = 29.55 years, SD = 7.75). According to marital status, 22.6% of women were single, 74% were married, 7% were divorced, and 7% were worried. According to education 66% of them finished secondary education, 7% finished high diploma, and 6.1% finished university education. The sample distributed between cities and camps, 48.3% live in cities, 38.8% live in camps, and 12.9% live in village. In looking to husband of married women, 68% finished secondary education, 11.8% finished high diploma, and 19.4% finished university level. Eighty percent women were housewives, 4.3% were simple worker, and 12.9% were employee.

Age was recoded in to three categories (18-29 years, 30-39 years, and 40-50 years). The results showed that there were no statistically significant differences between different groups of age of women and violence.

According to type of residence, post hoc showed differences toward women living in her own home than living with extended family in total negotiations scores (F= 3.50, p =.03).

According to marital status of women, post hoc test showed no statistically significant differences in scores of violence according to marital status.

According to place of residence (village, city, and camp), post hoc test showed no significant differences in scores of violence according to place of residence.

According to monthly family income, post hoc test showed no significant differences in scores of violence according to monthly family income.

According to women education, post hoc test showed no significant differences in scores of violence according to women education.

According to type of work, post hoc test showed no significant differences in scores of violence according to women type of work.

Post traumatic stress symptoms and level in women exposed to domestic violence

Our results showed that the most common PTSD reactions were avoidance ideas and feelings concerning the abuse (40.5%), avoidance of places and people reminding of the abuse (39.7%), upset when reminded of the violence (32.1%), while the least common symptoms were reliving the abuse as it is happening again (7.7%) and numbness feelings (10.3%). Our result shoed that mean PCL was 36.8 (SD= 12.09), intrusion mean = 10.6 (SD = 4.2), avoidance mean = 15.32 (SD = 5.3), hyperarousal mean = 10.9 (SD = 4.14). Considering the cutoff point of 50 for PCL, 17 women (14.7%) had been diagnosed as PTSD, and 99 had no PTSD (85.3%).

In order to investigate the differences in PTSD scores and socioeconomic variables such as age, type of residence, education, place of residence, marital status, number of children, and monthly family income, a One way ANOVA test was performed in which each of the demographic variable was entered as independent variable and means of PTSD as dependent variables. The results showed no differences in type of residence, marital status of women, being in extended family, type of education, and monthly income. and the results showed in table.

In order to investigate the type of violence which may predict PTSD a linear regression analysis was performed in which total scores of PTSD was entered as dependent variable and each of the violence items without negotiations as dependent variable. The result showed that the following items predicated PSTD later on: destroying something of mine (B = 0.22, p = 0.03) and burning and ironing me (B= 0.28 , p = 0.02).

A One way ANOVA test was performed in which each of the demographic variable was entered as independent variable and means of anxiety as dependent variables. The results showed no differences in type of residence, marital status of women, being in extended family, type of education, and monthly income.

A set of linear regression analysis was performed to find the prediction of total anxiety by types of violence against women. The results showed that the following items predicted the anxiety: shouted or yelled at me (B = 0.41, p = 0.02), twisted my arm or hair (B = 0.35, p = 0.02), I need to go to physician but he did not take me (B = 0.42, p = 0.003), and I had breaking bone due to the quarrel was negatively predicted to anxiety (B = -0.38, p = 0.02).

Considering the cutoff point of Depression (Gareeb, 2000), 61 of women showed no depression (52.6%), 38 showed mild depression (32.8%), 16 showed moderate depression (13.8%), and only one showed severe depression (0.9%).

Depression symptoms ranged from minimum 2 symptoms to 44 symptoms with mean depression = 19.68, SD = 10.6.

In order to investigate the type of violence which may predict depression scores a linear regression analysis was performed in which total scores of depression (BDI-II) scores was entered as dependent variable and each of the violence items without negotiations as dependent variable. The result showed that the following domestic violence items predicted depression in women: I did something to spite my husband. (B= 0.26, p = 0.04), threatened to hit or throw (B= 0.35, p = 0.01), he beat up me. (B= 0.32, p = 0.02) predicted depressive symptoms.

Relationship between domestic violence and PTSD, depression, anxiety

In order to find the relationship between type of violence and PTSD, anxiety, and depression, a correlation coefficient Spearman test was done. The results showed that total scores of PTSD were correlated with total domestic violence (r = 0.55, p = 0.001), negotiation (r = 0.28, p = 0.001), psychological abuse (r = .46, p = 0.001), physical assault (r = 0.53, p = 0.001), and physical injury (r = 0.55, p = 0.001). The results showed that total scores of anxiety (MAS) were correlated with total violence (r = 0.26, p = 0.001), psychological aggression (r = 0.23, p = 0.001), physical assault (r = 0.30, p = 0.001), and physical injury (r = 0.43, p = 0.001). The results showed that total scores of depression (BDI-II) were correlated with total domestic violence (r = 0.40, p = 0.001), psychological aggression (r = .23, p = 0.001), physical assault (r = 0.30, p = 0.001), and physical injury (r = 0.31, p = 0.001).

their parents and they will not return unless the husband came and she can return to husband home under many conditions including not to be abused again and to treat her well. Also the pressure from the extended families the Palestinian couple live will buffer the negative effect of domestic violence against women.

Victim-related characteristics stemming from socioeconomic status (SES) have been identified as potentially affecting the likelihood of violence. Low socioeconomic status generally been associated with violence against women (Barnett et al, 2000). The results showed that women aged 18-29 years exposed to severe psychological abuse than the older age group above 40 years, also women living in her own home than living with extended family in total negotiations scores. This results inconsistent with study of national sample of women in which living in poverty, being unemployed, being student, being young, being single were associated with higher rates of subsequent assault (Kilpatrick et al, 1998). This also was found in study of sample of USA women, results indicated that women experienced increase risk victimization when income is below poverty level and when newly divorced (Byrne et al, 1999). In study of Logan et al (2006) of 757 women who were enrolled in a study of w omen with protective orders against a male intimate partner, more women in the severe violence and the severe violence plus stalking groups were from the rural area compared to the moderate violence group. Our study was inconsistent with other studies could be to the continuous conflict and war in the area and not only women exposed to domestic violence, but other types of political and community violence which masked the real situation of Palestinian women. Our study showed that 16.5% of women said that she agreed to try a solution for the problems, 14.7% showed respect for his feelings, and 13.8% would agree on the solution he proposed. Our findings that Palestinian women had mental health problems due to domestic violence were consistent with studies of battered women. Our study found that 14.7% of women developed PTSD, and 14.7% met the criteria for depression this rate is less than the findings of others, in a meta-analysis of 11 studies, Golding (1999) reported that 31% to 84.4% of women who experienced IPV met PTSD criteria

(weighted mean prevalence = 63.8%). Cascardi et al (1999) in study of depression and self-esteem of battered women found that 30% of their sample met the criteria for PTSD, 32% met the criteria for depression, and 17% met the criteria for both.

A similar study in Asian country was conducted, Fikreeu and Ghatti (1999) in study to assess of the prevalence and health consequences of domestic violence among women in Karachi, Pakistan found that pregnant and 72% of physically abused women were anxious depressed. Physical abuse was a major predictor of anxiety and depression. In another study Leisring et al (2003) reported that approximately 45% of women entering their anger management program experienced clinically significant levels of PTSD symptoms.

Our study is consistent with Sharhabani-Arzy et al (2004) in study of self-criticism, dependency and posttraumatic stress disorder among a female group of help-seeking victims of domestic violence in Israel found that that 74.7% of the participants in the sample had intrusion symptoms, 62.6% had avoidance symptoms and 78% had hyperarousal symptoms. 51.6% of the participants had full PTSD.

Our study rate of PTSD and depression were inconsistent with study of Logan et al (2006) of 757 women who were enrolled in a study of women with protective orders against a male intimate partner, found that 23.5%, 35.7% of moderate and severe violence met criteria for post traumatic stress disorder, while 38.2%, 46.5% of moderate and severe violence met criteria for current depression. In study of women exposed to childhood sexual abuse, women with history of sexual abuse reported more symptoms of anxiety and post traumatic stress disorder than other women (Feerick & Snow, 2005). Our results consistent with Pillay and Kriel (2006) in study of various sociodemogrophic and clinical variables pertaining to 422 women using district-level clinical psychology services in Pietermaritzburg, South Africa. The study showed that over one-third had relationship problems, 21% depression, and 14% suicidal behaviour. Also, our study was inconsistent with Stuart et al (2006) study which examined the prevalence of women arrested due to domestic violence showed that women reported high levels of Axis I psychopathology. Of the women, 44% met or exceeded the PTSD cut score, 35% met or exceeded the depression cut score, 34% met or exceeded the GAD cut score, and 28% met or exceeded the Panic disorder cut score. Our findings of the strong correlation between PTSD, anxiety, and depression and all domestic violence domains which reflect the impact of domestic violence on Palestinian women mental health. This is consistent with most of the studies concerning domestic violence and women mental health (Sharhabani-Arzy et al , 2004; Feerick & Snow, 2005; Stuart et al, 2006; Logan et al, 2006).