This study used meta-analysis to examine the relationship between childhood exposure to domestic violence and children's internalizing, externalizing, and trauma symptoms. Results from 60 reviewed studies revealed mean weighted effect size d-values of .48 and .47 for the relationship between exposure to domestic violence and childhood internalizing and externalizing symptoms, respectively, indicating moderate effects. A larger mean weighted effect size d-value of 1.54 was obtained for the relationship between exposure to domestic violence and childhood trauma symptoms, though this figure was based on only six studies. Moderator analyses for gender showed that the relationship between exposure to domestic violence and externalizing symptoms was significantly stronger for boys than for girls. Further analyses examining age, age by gender, and recruitment setting variables revealed no significant effects. Descriptive information obtained from this meta-analytic review suggests that more recent research within this area is beginning to address some of the significant methodological limitations of past research. Recommendations for future research in the area are discussed. Keywords: domestic violence, child witnesses, meta-analysis, child behavior problems, family violence, child abuse Each year, approximately 4.8 million acts of physical or sexual aggression are perpetrated against women while 2.9 million physically aggressive acts are perpetrated against men in the United States. What makes these figures even more disturbing is the realization that many of these incidents take place in the presence of children. In fact, researchers estimate that between three and 17.8 million children are exposed to at least one incident of domestic violence each year. Moreover, studies using adults' retrospective reports indicate that 20% to 40% reported exposure to domestic violence during childhood or adolescence. Despite high prevalence of children exposed to domestic violence, researchers have only recently begun to investigate the effects of this exposure. While the first case study examining the negative impact of childhood exposure to domestic violence was published over thirty years ago, the first empirical studies did not appear until the early 1980's. This “first generation” of research, published from 1980’s to the early 1990’s, was primarily concerned with documenting the association between male-perpetrated violence towards females and various types of childhood symptomatology. In 1989, Fantuzzo and Lindquist published a qualitative review of this first generation of empirical literature. Their summary highlighted inconsistencies and methodological limitations, including little precision in describing the types of violence to which children were exposed, the use of unstandardized measures of exposure to domestic violence, and a failure to assess moderating variables such as age and gender. This review, along with the equivocal results of the previous literature, spawned a second generation of research, primarily published since 1990. These studies employed more sophisticated research designs and tested models that included mediating and moderating variables. The most recent empirical studies have continued these trends by extending investigations to young children and adolescents while continuing to addressThis study used meta-analysis to examine the relationship between childhood exposure to domestic violence and children's internalizing, externalizing, and trauma symptoms. Results from 60 reviewed studies revealed mean weighted effect size d-values of .48 and .47 for the relationship between exposure to domestic violence and childhood internalizing and externalizing symptoms, respectively, indicating moderate effects. A larger mean weighted effect size d-value of 1.54 was obtained for the relationship between exposure to domestic violence and childhood trauma symptoms, though this figure was based on only six studies. Moderator analyses for gender showed that the relationship between exposure to domestic violence and externalizing symptoms was significantly stronger for boys than for girls. Further analyses examining age, age by gender, and recruitment setting variables revealed no significant effects. Descriptive information obtained from this meta-analytic review suggests that more recent research within this area is beginning to address some of the significant methodological limitations of past research. Recommendations for future research in the area are discussed. Keywords: domestic violence, child witnesses, meta-analysis, child behavior problems, family violence, child abuse Each year, approximately 4.8 million acts of physical or sexual aggression are perpetrated against women while 2.9 million physically aggressive acts are perpetrated against men in the United States. What makes these figures even more disturbing is the realization that many of these incidents take place in the presence of children. In fact, researchers estimate that between three and 17.8 million children are exposed to at least one incident of domestic violence each year. Moreover, studies using adults' retrospective reports indicate that 20% to 40% reported exposure to domestic violence during childhood or adolescence. Despite high prevalence of children exposed to domestic violence, researchers have only recently begun to investigate the effects of this exposure. While the first case study examining the negative impact of childhood exposure to domestic violence was published over thirty years ago, the first empirical studies did not appear until the early 1980's. This “first generation” of research, published from 1980’s to the early 1990’s, was primarily concerned with documenting the association between male-perpetrated violence towards females and various types of childhood symptomatology. In 1989, Fantuzzo and Lindquist published a qualitative review of this first generation of empirical literature. Their summary highlighted inconsistencies and methodological limitations, including little precision in describing the types of violence to which children were exposed, the use of unstandardized measures of exposure to domestic violence, and a failure to assess moderating variables such as age and gender. This review, along with the equivocal results of the previous literature, spawned a second generation of research, primarily published since 1990. These studies employed more sophisticated research designs and tested models that included mediating and moderating variables. The most recent empirical studies have continued these trends by extending investigations to young children and adolescents while continuing to address