Psychometric properties of the Spanish-language child depression inventory with Hispanic children who are secondary victims of domestic violence.

Molina, Carmen Soto ; Gomez, Jose Rodriguez ; Pastrana, Maria C. Velez 等

It is important to be aware of the problems of domestic violence and depression in the Puerto Rican population. Depression in teenagers has been associated with conflictive and dysfunctional families (Arzola-Colon, Gonzalez-Villanova, and Rosello, 2000). Fendrich, Warner, and Weissman (1990) pointed out that children from dysfunctional families are at greater risk of developing psychopathological disorders than children from stable families.

Likewise, it has been found that when stressors in the family increase, the functioning of the children and teenagers deteriorates (Forehand, Wierson, McCombs, Armistead, Kempton, & Neighbors, as cited in Saez and Rosello, 2001).

Affective disorders in children and teenagers represent a serious mental health problem. Many researchers believe that mood disorders in children and teenagers present a low prevalence within the group of psychiatric illnesses (Rosello, 1993). Some of the reasons for this might be the following: children do not always express their feelings, the symptoms of mood disorders are different in children and in adults, mood disorders may be accompanied with other psychiatric disorders which might mask symptoms of depression and, lastly, many psychiatrists tend to think that depression and other mood disorders are adult illnesses (Rosello, 1993).

Depression is a disorder that affects people of all ages, including infants, boys, girls, teenagers and mainly adults (Rosello & Martinez, 1997). Furthermore, it is also one of the most frequent reasons for which psychological services are sought in Puerto Rican society (Bernal, Rosello, & Martinez, 1992; Rodriguez & Alsina, 1994).

There are important reasons to study and treat juvenile depression. First, there has been an increase in the incidence of depression and suicide in this population. Second, depression interferes with developmental tasks, thereby causing additional problems. Third, depression, if left untreated, tends to be recurrent. Finally, depression is a condition that causes great suffering to those who go through it and to their families (Rosello & Martinez, 1997).

The family is the institution which most influences human socialization. It is the first school of emotional and cognitive learning for a child and where the child experiences the first models of behavior (Nevarez, as cited in Ortiz, 2001). However, families today may well be either the network of economic and emotional sustenance or the most violent context for its members (Silvia, as cited in Ortiz, 2001).

The Puerto Rican Domestic Violence Prevention and Intervention Act of 1989, also known as Public Law #54, defines domestic violence as "a constant pattern of using physical force or psychological violence, intimidation or persecution against a person by that person's partner, ex-partner, the person with whom he or she lives or has lived, with whom he or she has or has had a consensual relationship or a person with whom he or she has had a child, to cause him or her bodily harm, or to harm his or her property or someone else in order to cause him or her severe emotional damage." This act also states, in its Statement of Purpose, that domestic violence is considered an antisocial behavior that affects the entire family, especially the children. Public Law #54 (1989) maintains that children who come from homes where domestic violence takes place, carry with them the traces of violent patterns for their entire lives.

Because depression is such a severe mental health condition associated with dysfunctional families, it is important to evaluate the relationship between the development of symptoms associated with depression and living in homes where domestic violence is observed.

Studies and Statistics on Domestic Violence and Depression in Puerto Rico

The problem of domestic violence has been a serious one in Puerto Rico. According to statistics of the Puerto Rico Police Department (2001), 17,770 complaints were registered in Puerto Rico in 2001. Arpong the reported cases, there was a total of 55% (9,713) corresponding to abuse, of which 37% (6,542). involved physical force, followed by 31 reported cases in which a sharp weapon was used. More than half of domestic violence incidents reported in 2001 occurred in the victim's residence with a total of 51% (9,066). Recent statistics do not demonstrate great changes in the prevalence of the phenomenon.

An epidemiological study conducted in Puerto Rico estimates that 10% of Puerto Rican children suffer from a psychiatric disorder (Bird, Canino, Stipec, Gold, Ribera, Woodberry, Huertas, Goldman, Pagan, Sanchz-Lacay, & Moscoso, 1988). Among that group, 4% suffer from depression. A study was performed in Puerto Rico for UNICEF on depression in Puerto Rican children and its possible causes. The results showed that one-third of the children and teenagers had symptoms of depression. It as also found that 28% of the participants, (140,000 children and teenagers) acknowledged that they observed scenes of violence in their homes such as yelling and hitting (Figueroa, 2001).

Effects of Domestic Violence on Children

It has been determined that both domestic violence and juvenile depression required special attention, not only because of the individual impact, but their social impact. International studies have found a relation between exposure to domestic violence and the development of depression-related symptoms in children (Afifi, Brownridge, Cox, Sareen, 2006; Hussey, Chang, & Kotch, 2006).

Authors such as Edlesson (1999), point out that children can witness domestic violence in several ways. They can be a direct witness by being present when the aggression occurs or they can be an indirect witness by hearing arguments between their parents or by observing the consequences of physical assault suffered by the victim.

Hughes (1983) conducted a study in which he found that in 90% of domestic violence incidents, the child was in the same or next room. Jaffe et al., (1990) conducted a series of interviews with children in Canada. These revealed that most of the children could give a detailed description of domestic violence incidents, even though their parents thought that the children were not aware of what had happened.

The impact on the children and their response to domestic violence varies according to the child's age, gender, developmental stage, and according to the frequency and intensity of the violent incidents against the child's mother (Jaffe et al., 1990). According to Jaffe, children who grow up in homes where domestic violence is observed learn that it is the correct way to solve problems. Girls also learn that being victims is inevitable and that nobody can change those patterns. Boys, in turn, tend to practice what they learned by behaving aggressively in the home and at school.

Exposure of children to abuse toward women by their partners fosters the development of emotional problems in the children (secondary victims). These children grow up ashamed of the situation at home, thinking it is a family secret, which in turn keeps them from seeking help (Jaffe et al., 1990).

According to Jaffe, thoughts such as "If I behaved better, Dad wouldn't hit Mom," prevail among these children. The experiences of these children and the way they construe them slowly destroy their self-esteem and their confidence in themselves and in the future. Feelings of guilt are characteristic of these children because they think they can prevent and avoid the violent incidents. They also manifest concentration problems which affect their academic performance.

Cummings and Davies (1994) found that children of abused women are at higher risk of exhibiting behavior problems than children who do not report violence in their homes. Those children who have been exposed to domestic violence incidents manifest more psychological adjustment problems such as anxiety (Christopolous, Cohn, Shaw, Joyce, Sullivan, Kraft, & Emergy, 1987), depression (Sternberg, Lamb, Ciccetti, Manela, Crispin, & Loret, 1993) and low self-esteem (Hughes & Barad, 1983).

Edleson (1999) proposes that children who have been exposed to domestic violence will show greater behavioral and emotional problems than children who have not. Children who grow up witnessing domestic violence exhibit aggressive and antisocial behavior (so-called externalizing behavior), as well as inhibitions and fearful behavior (internalizing behavior). These children also show higher levels of anxiety, depression, symptoms of trauma and temperament problems than children who do not observe violence in their homes (Fantuzzo, et al., 1991; Hughes & Barad, 1983) and demonstrate low social competence (Adamson & Thompson, 1998; Fantuzzo et al., 1997).

A relation has been observed among the nature of the conflict to which the child is exposed, the type of residence, and the child's adjustment problems. Children who have been exposed to verbal and physical conflicts and who lived in shelters for abused women had clinically significant behavioral problems, high levels of emotional problems, and low scores in social functioning (Fantuzzo et al., 1997).

Silvern (1995) in a study with adults who had witnessed domestic violence in their homes during childhood found it to be related to their current symptoms of depression, low self-esteem (among women), and symptoms of trauma (in males).

Sternberg et al., (1993) explored how different types of domestic violence (being a victim of abuse, observing abuse toward women and being a victim of and observing abuse) affect behavior and development of symptoms of depression. In the study, participating children were divided into three groups according to type of violence they had been exposed to and a comparison group. According to Sternberg, depression in children who grow up in homes where there is domestic violence depends on the type of violence. The study found that depression in children is linked with witnessing domestic violence against their mothers. When the children who had been victims of domestic violence were compared with the other group, the differences in symptoms of depression were not significant, although the group that had witnessed domestic violence had a higher CDI score. Sternberg et al. (1993) points out that when evaluating the relationship between domestic violence and depression, it is important to consider the type of violence as well as who reports the child's symptoms--the child or someone else.

Depression in Puerto Rican Children

Several studies about depression in children and teenagers have been conducted in Puerto Rico with the purpose of identifying associated factors. Rosello et al. (1992) developed a translation and adaptation of the Child Depression Inventory (CDI) for Puerto Rico. It had an internal consistency index of .95 and concurrent validity was documented with the CESD-C.

Several studies have been conducted in Puerto Rico using the Spanish language CDI with the aim of identifying factors associated with juvenile depression. The following were found to be related: Martinez and Rosello (1995) report poor family functioning. The perception that the children and teenagers have of the marital conflicts between their parents and the perceived criticism was reported by Saez and Rosello (1997). The marital status of the parents was reported by Arzola Colon et al. (2000) as well as their parents' marital conflicts (Saez & Rosello, 2001). All of these factors have been linked to the development of depressive symptomatology in Puerto Rican youth and children.

THEORETICAL BACKGROUND

Beck's Cognitive-Behavioral Theory

Beck et al., 2001 defines depression in cognitive terms. It is based on the underlying theoretical assumption that the affection and behavior of an individual are determined in great measure by the way the individual structures the world. His or her cognitions are based on attitudes or assumptions developed from previous experiences. The cognitive model states three specific concepts to explain the psychological substrata of depression: (1) cognitive triad, (2) schemes, (3) cognitive errors.

The cognitive triad consists of three main cognitive patterns: (a) patients view themselves negatively, (b) they interpret their experiences negatively, (c) they have a negative view of the future. The second component of the cognitive model is the structural organization of thought which Beck called schemes. Schemes are relatively stable cognitive patterns that constitute the basis for the interpretations regarding a determined set of situations. This is a fundamental principle of the cognitive model. Beck et al. (2001) states that a schematic interpretation always mediates between experience and the emotional responses to it. A person's negative and distorted cognitions in a concrete situation are considered errors in the processing of the information, also called "automatic thoughts."

METHOD

Participants

A sample of 100 children was selected. Of these, fifty were girls and fifty were boys. The ages of the participants were between 8 and 12 years, with an average age of 9.92 years (SD = 1.50) in both groups. Participants included 2 groups of 50, one consisting of children who were secondary victims of domestic violence (SVDV). These children participated in a program that offered services to victims in the San Juan Metropolitan area as well as in several towns on the island during the months of July to November 2003.The remaining fifty children were non-victims (NV) recruited among those attending a private school in the Metropolitan area.

Participants ranged from grades third to seventh; 34% attended public school and 66% private school. The group of secondary victims of domestic violence were receiving psychological treatment at the time.

Regarding the parents of the participants, 38% were married, 20% were separated, 32% were divorced, and 10% noted that they cohabited; 78% of the mothers indicated that they were Catholic, 16% Protestant, and 3% indicated other; 29% of mothers indicated that they had a Bachelor's Degree.

The instruments used assessed variables including socio-demographic information and depressive symptomatology. Additionally, a scale to measure conflict management tactics was administered. The instruments used in the study are described below.

The Demographic Information Sheet includes personal and social backgrounds of the participants, including age of the mother, father, and child, parents' marital status, religious affiliation, family income, whether they received financial aid or child support, treatment for mental health problems, grade, number of siblings, and whether it is the first time they received services in the program and the number of sessions for those receiving services. Additionally, questions were included to identify mothers and their children who had not been victims of domestic violence to be part of the comparison group. Some of the questions were: Does your partner insult you in front of your child? Does your partner use curse words with your child? Has your partner hit you in front of your child? Does your partner not support your looking for a job? Does your partner constantly tell you what to do? Have you had to file or withdraw a police report in the last five years?

The Conflict Tactic Scale (CTS) is designed to measure coping techniques used when faced with a conflict with one's partner. The purpose of the scale is to evaluate the type and grade of domestic violence between married couples and/or partners who live together (Strauss, Homby, & Boney-McCoy, 1996). The scale is composed of 19 items that range from behaviors of lesser cohesion to those of greater cohesion. The scale in turn has three subscales: reasoning, verbal aggression, and physical aggression. The physical subscale, in turn, is subdivided into severe physical aggression (i.e., he kicked you, he bit you or he punched you). Scores range from 0 = has never happened, 1 = twice last year, 2 = twice last year, 3 = 3-5 times last year, 4 = 6-10 times last year, 5 = 11-20 times last year, and 6 = more than 20 times last year.

Upon evaluating the psychometric properties of the CTS, the following was found: the CTS obtained an internal consistency index in a sample of North Americans of an alpha between .42 and .61 on the reasoning scale, .62 and .83 on the verbal aggression scale, and .69 and .88 on the physical aggression scale. In Puerto Rico, the Sociodemographic Research and Evaluation Center translated the CTS. They reported an alpha of .87 on the moderate aggression subscale and an alpha of .84 on the severe aggression subscale, indicating that the instrument has a strong internal consistency.

In the present study the CTS was used to screen for the presence of domestic violence in the process of selecting the group of non-domestic violence participants (comparison group). Those children whose mothers marked zero on all items were selected because they met the previously established inclusion criteria of no presence of domestic violence.

The Child Depression Inventory (CDI) was developed by Kovacs (1992) to measure signs and symptoms of depression in children and teenagers between the ages of 6 and 17 years. The CDI is an instrument with a self-report format containing 27 multiple choice items about a specific category of signs and symptoms associated with depression. Each item consists of three self-evaluation sentences with a score of 0, 1 or 2 in the direction of severity of the symptom. The margin of final scores ranges from 0 to 54 points, in which the greater the score, the greater the level of depression. Scores between 0 and 11 suggest an absence of depressive symptomatology; scores between 12 and 18 suggest moderate depressive symptomatology; and scores higher than 19 are considered severe (Kovacs, 1992). The CDI has an internal consistency between .71 and .89, based on studies carried out with samples from the United States (Kovacs, 1992).

Research using the CDI suggests that depression is a complex and multidimensional phenomenon since this instrument has a multifactor structure. Kovacs reports five factors that are consistently suggested by the research: negative mood, interpersonal problems, uselessness, anhedonia, and negative self-esteem (Bernal et al., 1992).

Kovac's study found no statistically significant differences in the total scores of the CDI by gender. Internal reliability had favorable support. The internal consistency indexes of the CDI were excellent both for the community sample and for the clinical sample (Bernal et al., 1992). Regarding the internal reliability of the CDI in community samples, the following were the highest items: somaticism, fatigue, irritability, and difficulty with schoolwork. The average for all items was .51, with a minimum of .16 and a maximum of .74. The items that highly correlated with the total score of the CDI were those of loneliness, comparison to others, disobedience, feelings of lack of affection, body image, and sadness. The average of the correlations between items was .14 with an interval that fluctuated between -.10 and .52. Internal reliability, according to Cronbach's Alpha Internal Consistency Index, obtained for this sample was .829 (Bernal, 1992).

Kovacs (1992) developed a Spanish translation of the CDI, a study which was used to evaluate its psychometric properties for the population of Puerto Rican children (Beall, 2001). Teachers also completed the Inventario de Comportamiento-Version Escolar (IDC-E, Bauermeister). The IDC-E was developed and validated in Puerto Rico by Jose Bauermeister to determine and measure levels of attention deficit and hyperactivity, as well as symptoms of other disorders that may be present in childhood. The IDC-E includes a sub-scale for depressive symptoms. Participants were evaluated using the CDI, and the results were compared with the scores given by the teachers through the IDC-E. The results showed a positive and significant correlation between the scores obtained on the CDI and those on the IDC-E (Beall, 2001). According to Beall (2001), the positive correlation found at .01 level of significance between the CDI and the IDC-E suggests a concurrent validity between both instruments.

Through a preliminary factor analysis, Beall (2001) identified five factors of the CDI: Component one and its items (5, 8, 12, 15, 18, 21, and 25), Component two (1, 2, 14, and 23), Component three (4, 7, 9, 13, 17, and 22), Component four (6, 11, and 24) and Component five (10 and 20).

Kovacs' Spanish translation of the CDI was used in the present study.

Procedure

The study was conducted in accordance with ethical standards for research, and the methodology was previously approved by an institutional review board. To select the children participants for the group of Secondary Victims of Domestic Violence (SVDV), written authorization was requested from the director of the program in the metropolitan area and different towns across the island. The program offers psychological services to women who have been victims of domestic violence and their children. To select the group of children for the no domestic violence comparison group (NV), the researchers obtained authorization of the administration of a private school to recruit a sample among students attending that school.

Once the participants of both groups were identified and informed consent obtained, the mothers of all participating children were interviewed. As part of the interview, the purpose of the study was explained to the children and mothers who signed the participation consent form and filled out the sociodemographic data sheet. To screen and identify the mothers who were not victims of domestic violence, they completed the CTS scale used to determine the use of conflict solution strategies. Afterward, the CDI was administered only with those children whose parents had signed the consent form.

Statistical Analyses

Factor analysis. A confirmatory factor analysis was performed to examine the construct validity of the Child Depression Inventory. Principal component analysis (PCA) was used as the extraction method, which consists of forming linear combinations with the variables (items). The Varimax rotation method was used, which helps to simplify interpretation of the factors. The Varimax method of orthogonal rotation helps in the interpretation of the factors by minimizing the number of items with loadings on a given factor. Five factors were extracted, according to the findings of previous studies (Kovacs, 1992; Beall, 2001).

Reliability. To evaluate the reliability of the instrument, the internal consistency coefficient for the total score of CDI was obtained using Cronbach's Alpha.

RESULTS

The participants (N = 100) obtained scores on the Child Depression Inventory which ranged between 0 and 39, with an average score of 9.48 for the SVDV group and 8.52 for the NV group.

Factor analysis. The analysis generated five factors that explained the 66.379% of the variance in the scores of the CDI. These factors had Eigenvalues greater than 1.000. The factors include a first factor that explains 29% of the variation in the scores of the CDI. Factors 2, 3, 4, 5 explain 8.0%, 6.2%, 5.6%, and 5.0% of the variance, respectively. Table 1 shows the five factors with their Eigenvalues ant the percentages of variance explained by each.

Table 2 shows the final matrix of factors rotated through the Varimax method. Five factors were extracted, in keeping with the findings of previous studies (Kovacs, 1992; Beall, 2001).

Reliability. Cronbach's Alpha internal reliability index was used to assess the consistency of the answers throughout the different items of the test. Reliability obtained in the CDI was .88. This suggests that the instrument has a strong and positive consistency between each of the items with the total of the scale. This further suggests that the CDI can be considered as an instrument with the necessary properties to reliably identify children and teenagers who show symptoms associated with depression and those who do not.

DISCUSSION

This section discusses the findings and compares them with those of other research on the factor structure of the CDI. Published research using samples of U.S. and Hispanic children (Kovacs, 1992), as well as with Puerto Rican children (Beall, 2001) were also included.

In the present study, five components with Eigenvalues greater than 1.00 were extracted which explained 66% of the variance in Child Depression Inventory (CDI) scores. The factors include a first factor which explains the 28% variation, identified as Negative Mood, and which includes 7 items. Factors 2, 3, 4, and 5 explained 8.0%, 6.2%, 5.6%, and 5.0% of the variation, respectively. The first was identified as Negative Mood (including items 16, 7, 10, 1, 18, 11, and 26). The second was Behavior Problems (items 25, 27, 3, 5, 15, 14, and 24). The third was identified as School Problems (items 21, 22, 24, and 23). The fourth was Insecurity (2, 13, 8, and 12). The fifth was Worries (6, 19, 17, and 20). As mentioned, Kovacs (1992) translated the CDI into Spanish. A total of 1,266 children were used with that translation. When performing the factor analysis, she obtained five factors. The first obtained an Eigenvalue of 6.3, which explained the 23% variation obtained (the variation of the other factors was not specified).

The first factor was identified as Negative Mood including the following items: 1, 6, 8, 10, 11, and 13. The second was identified as Interpersonal Problems (5, 12, 26, and 27). The third was Ineffectiveness (3, 15, 23, and 24). The fourth was Anhedonia (4, 16, 17, 18, 19, 20, 21, and 22). The fifth was Negative Self-esteem (2, 7, 9, 14, and 25).

A study was performed in Puerto Rico to explore the psychometric properties of Kovacs' Spanish version of the CDI (Beall, 2001). Their interest was to explore whether the Spanish CDI presented the same factor structure as the original English version. Beall also was interested in evaluating the concurrent validity of the instrument with one that was standardized for Puerto Rican children. The sample consisted of 100 children between the ages of 6 and 13 years attending private schools.

Beall (2001) found that the factor structure and the importance of the factors did not correspond to those reported by Kovacs. Beall (2001) identified the factors (components) as follows: Component one included items 5, 8, 12, 15, 18, 21, and 25; Component two included items 1, 2, 14, and 23; Component three included items 4, 7, 9, 13, 1, 7, and 22; Component four included items 6, 11, and 24; and Component five items 10 and 20. The analysis was preliminary and Beall pointed out the need for a larger and more representative sample to replicate findings by Kovacs.

Upon evaluating the results obtained in the present investigation in contrast with the original evidence, it was apparent that the items were not grouped in the same way and the factor structure that emerged in the three studies differed. However, although the factor structure was not identical, some similarities were found. For instance, it should be noted that items such as 1, 8, 10, and 11 were grouped together in factor one (negative mood) both in Kovacs' and in the present research. Their contents reflect the emotional component of symptoms associated with depression: sadness, guilt, crying, and irritability. This was not consistent with Beall's (2001) results. Likewise, when the results were compared to those obtained by Beall (2001) with a sample of Puerto Rican children, no correspondence was found between his results and those of Kovacs' (1992). When evaluating and interpreting these results, it should be noted that the language as well as ethnic origin of the participants in the three studies was different, the number of participants was not the same in the three research projects, and their ages also differed. The results obtained suggest that when using the CDI, precautions should be taken since the standards for the instrument were not developed for use with Puerto Rican children.

The present research has limitations in terms of the sampling method since a convenience sample was used. This, as well as the size of the sample (N = 100), which is relatively small for this purpose, limits the generalizability of findings. Replicating the research with a broader community-based sample selected at random is recommended. At the same time, variables such as intensity of the domestic violence incidents to which the children are exposed should be taken into consideration. To this end, using instruments that evaluate the amount and severity of the incidents to which the children are exposed and a clinical interview of all participants are also recommended.

REFERENCES

Afifi, T. O., Brownridge, D. A., Cox, B. J., & Sareen, J. (2006). Physical punishment, childhood abuse and psychiatric disorders. Child Abuse Neglect, 30, 1093-1103.

American Psychiatric Association. (1994). Diagnostic manual of mental disorders (4th ed.) Washington, DC, EE.UU: APA.

Adamson, J. L., & Thompson, R. A. (1998). Coping with interparental conflict by children exposed to spouse abuse and children from nonviolent homes. Journal of Family Violence, 13, 213-232.

Arzola-Colon, G. M., Gonzalez Villanova, S., & Rosello, J. (2000). Depresion juvenil y su relacion al estado marital de los padres, funcionamiento familie y adaptacion social [Juvenile depression and the relationship with parents' marital status, familial functioning and social adaptation]. Ciencias de la Conducta, 15, 53-67.

Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (2001). Terapia cognitiva de la depression (13th ed.) [Cognitive therapy for depression]. New York: Guilford.

Beall, S. (2001). Assessing children and adolescents with the children's depression inventory in San Juan, P.R. Unpublished dissertation. Universidad Carlos Albizu, San Juan.

Bernal, G., Rosello, J., & Martinez, A. (1992). E1 inventario de depresion para ninos y ninas: Propiedades psicometrias en dos muestras puertorriquenas [Inventory of depression for boys and girls: Psychometric properties of two Puerto Rican samples]. Revista de Psicologia Contemporanea, 4, 12-23.

Bird, H., Gould, M. S., Yager, T., Staghezza, B., & Canino, G. (1989). Risk factors of maladjustment in Puerto Rican children. American Academy of Child and Adolescent Psychiatry, 28, 847-850.

Christopoulos, C., Cohn, D. A., Shaw, D. S., Joyce, S., Sullivan Hanson, J., Kraft, S. P., & Emery, R. E. (1987). Children of abused women: Adjustment at time of shelter residence. Journal of Marriage and the Family, 49, 611-619.

Cummings, E. M., & Davies, P. (1994). Marital conflict and child development. Children and marital conflict. The impact of family dispute and resolution (pp. 1-12). New York: Guilford.

Departamento de Policia de Puerto Rico (2001). Division de Estadisticas. Informe Global. Cuartel General. Hato Rey, P.R.

Edleson, J. L. (1999). Children of adult domestic violence. Journal of Interpersonal Violence, 14, 839-870.

Fantuzzo, J. W., De Paola, L. M., Lamber, L., Martihno, T., Boruch, R., Beriama, A., Atkins, M., & Marcus, S. (1997). Domestic violence and children: Prevalence and risk in five major U.S. cities. Journal of American Academy of Child and Adolescent Psychiatry, 36, 116-122.

Fantuzzo, J. W., & Mohr, W. K. (1999). Prevalence and effects of child exposure to domestic violence. The Future of the Child, 9, 21-29.

Fendrich, R. E., Warnes, V., & Weissman, M. M. (1990). Family risk factors, parental depression and psychopathology in offspring. Developmental Psychology, 26, 40-50.

Figueroa, J. (2001, October 24). Deprimidos mas de un tercio de los jovenes Puertorriquenos [More than a third of Puerto Rican adolescents are depressed]. Puerto Rico: El Nuevo Dia.

Hernandez, S. R., Fernandez, C. C., & Baptista, L. P. (1998). Metodologia de la Investigacion (2nd ed.) [Research methoodology] Mexico: McGraw Hill.

Hughes, H., & Barad, S. (1983). Psychological functioning of children in a battered women's shelter: A preliminary investigation. American Journal of Orthopsychiatry, 53, 525-531.

Hussey, J. M., Chang, J. J., & Kotch, J. B. (2006). Child maltreatment in the United States: Prevalence, risk factors, and adolescent health consequences. Pediatrics, 118, 933-942.

Jaffe, P. G., Wolfe, D. A., & Wilson, S. K. (1990). Children of battered women. London: Sage Publications.

Junta Examinadora de Psicologos de Puerto Rico. (1992). Codigo de Etica de Psicologos de Puerto Rico. Officina de Reglamentacion de Profesionales de la Salud, Departamento de Salud, Estado Libre Associado de Puerto Rico.

Kline, P. (1993). The Handbook of Psychological Testing. New York: Routledge.

Kovacs, M., & Beck, A. T. (1977). An empirical clinical approach to childhood depression. In J. C. Schulter & A. Rasking (Eds.). Depression in childhood: Diagnostic, treatment and conceptual models (pp. 1-25). New York: Raven Press.

Kovacs, M. (1992). Children's Depression Inventory (CDI) manual. New York: Multi-Health Systems, Inc.

Ley para la Prevencion e Intervencion con la Violencia Domestica. (1989). Leyes de Puerto Rico Anotadas, Titulo 89, 601.

Martinez, A., & Rosello, J. (1995). Depresion y functionamiento familiar en nifios puertorriquenos [Depression and family functioning in Puerto Rican children]. Revista Puertorriquena de Psicologia, 10, 215-245.

Ortiz, R. M. (April 10, 2001). La realidad de la violencia en la relacion de pareja [The reality of violence in couples relationships]. Gente Especial [on the web] Available at: www.saludNutricion.com

Rodriguez, J. R., & Alsina, S. (1994). The suicide phenomenon in the Puerto Rico elderly population: A ten-year retrospective study (1980-1990). Revista de Psiguiatria, 1, 9-11.

Rosello, J. (1993). Acercamientos terapeuticos para la depresion en adolescentes puertorriquenas: Dos estudios de casos [Therapeutic approaches to depression in Puerto Rican adolescents: Two case studies]. Revista Interamericana de Psicologia, 72, 163-180.

Rosello, J., Guisasola, E., Ralat, S., Martinez, S., & Meves, A. (1992). La evaluacion de la depresion en un grupo de jovenes puertorriquenos [The evaluation of depression in a Puerto Rican male adolescent group]. Revista Puertorriquena de Psicologia, 8, 55-62.

Rosello, J., & Martinez, A. (1997). Depresion en la ninez y adolescencia: Aportes de la investigacion [Children and adolescent depression: A research contribution]. Revista Psicologia Contemporanea, 4, 64-74.

Saez, S. E., & Rosello, J. (1997). Percepcion sobre los conflictos maritales de los padres, ajuste familar y sintomatologia depresiva en adolescentes puertorriquenos [Perception of marital conflicts, family adjustment and depressive symptoms in Puerto Rican adolescents]. Revista Interamericana de Psicologia, 31, 279-291.

Saaz, S. E., & Rosello, J. (2001). Relacion entre el ambiente familiar, los sintomas depresivos y los problemas de conducta en adolescentes puertorriquenos/as [Relationship between family environment, depressive symptoms and behavioral problems in Puerto Rican adolescents]. Revista Interamericana de Psicologia, 35, 113-125.

Sanchez, V. (1994). Fundamentos del razonamiento estadistico (4th ed.) [Fundamentals of Statistical Reasoning]. Republica Dominicana: D'Vinni Editorial.

Silvern, L., Karyl, J., Waelde, L., Hodges, W. F. et al. (1995). Retrospective reports of parental partner abuse: Relationships to depression, trauma symptoms and self-esteem among college students. Journal of Family Violence, 10(2), 177.

Sternberg, K. J., Lamb, M. E., Greenbaum, C., Ciccetti, D., Dawud, S., Manela Cortes, R., Krispin, O., & Lorey, F. J. (1993). Effects of domestic violence on children's behavior problems and depression. Developmental Psychology, 29, 44-52.

Strauss, M. A., Hamby, S., & Boney-McCoy, S. (1996). Revised Conflict Tactic Scale. Journal of Family Issues, 17, 283-316.

Carmen Soto Molina, Ph.D. and Maria C. Velez, Pastrana Ph.D., Carlos Albizu University, P.O. Box 9023711, Old San Juan Station, San Juan PR 00902.

Reprint requests should be sent to Jose Rodriguez Gomez, M.D., University of Puerto Rico, General Social Sciences Dept., Rio Piedras Campus, San Juan PR 00936.

Table 1
Five Factors Extracted from the Spanish language Child Depression
Inventory (CDI) using Principal Components Analysis and Varimax
rotations (n = 100 children).

 Factor Eigenvalues % variance cumulative
 explained % variance
 explained

I. Negative mood 7.811 28.93 28.93
II. Behavior problems 1.163 8.01 36.94
III. School problems 1.679 6.22 43.16
IV. Insecurity 1.524 5.65 48.81
V. Worries 1.358 5.03 53.84

Table 2
Final matrix of five factors rotated using the Varimax method, for the
factor analysis of 26 items composing the Child Depression Inventory
(CDI; N=100 children).

 Factor loadings
CD
items Negative Behavior School Insecurity Worry
 Mood Problems Problems

Item 16 .70
Item 7 .68
Item 10 .67
Item 1 .66
Item 18 .56
Item 11 .54
Item 26 .44
Item 25 .77
Item 3 .64
Item 27 .63
Item 5 .59
Item 24 .54
Item 15 .50
Item 14 .46
Item 21 .77
Item 22 .72
Item 4 .62
Item 23 .58
Item 2 .66
Item 13 .58
Item 8 .51
Item 12 .44
Item 6 .56
Item 17 .52
Item 20 .48
Item 19 .47