Risk of Death or Life-Threatening Injury for Women with Children Not Sired by the Abuser
- First Online:
- Cite this article as:
- Miner, E.J., Shackelford, T.K., Block, C.R. et al. Hum Nat (2012) 23: 89. doi:10.1007/s12110-012-9129-9
- 189 Views
Women who are abused by their male intimate partners incur many costs, ranging in severity from fleeting physical pain to death. Previous research has linked the presence of children sired by a woman’s previous partner to increased risk of woman abuse and to increased risk of femicide. The current research extends this work by securing data from samples of 111 unabused women, 111 less severely abused women, 128 more severely abused women, and 26 victims of intimate partner femicide from the Chicago Women’s Health Risk Study to document an ordinal trend in the risk of experiencing more severe forms of violence for women who have children in the household sired by a previous partner. The discussion addresses two potential explanations for this trend and highlights directions for future research.
KeywordsDomestic violenceIntimate partner abuseIntimate partner homicideStepchildren
Intimate partner abuse is widespread in both industrialized and unindustrialized settings. A World Health Organization multi-country study found that the lifetime prevalence of physical partner violence among ever-partnered women ranges from 13% to 61% (Garcia-Moreno et al. 2006). Female victims of their male intimate partner’s abuse incur many costs, ranging from minor physical wounds to death (for review, see Campbell 2002; Heise et al. 1994). Research guided by various theoretical perspectives, from developmental to feminist, has identified numerous correlates of men’s abuse of female intimate partners, including alcohol consumption, violence in the woman’s childhood home, race, employment status, marital status, and socioeconomic status (e.g., Coker et al. 2000; Vest et al. 2002). Guided by an evolutionary perspective, other researchers hypothesized that men abuse their partners in order to control their partner’s sexual behavior (Daly and Wilson 1988; Wilson and Daly 1996). From this perspective, men’s abuse of their partners is one manifestation of male sexual jealousy.
In fact, research indicates that the composition of a woman’s household can trigger or exacerbate her partner’s sexual jealousy. Daly et al. (1993:209) suggested that men whose partners are raising other men’s children “may resent their predecessors’ children as living violations of their monopoly” over their partner’s reproductive opportunities. In addition, men who are genetically unrelated to their partner’s children may be less willing to invest time and resources in those children than are the mothers, whose genetic relationship to the children is certain (Daly and Wilson 1996; Emlen 1997). This difference in willingness to invest in the woman’s children has been hypothesized to lead to conflicts over parental investment and to increased abuse of their intimate partner by men (Brewer and Paulsen 1999; Daly et al. 1997).
Previous research documents that women residing with children sired by previous partners are overrepresented among the victims of nonlethal intimate partner abuse (Brownridge 2004; Daly et al. 1993; Figueredo and McCloskey 1993). Brownridge (2004) analyzed women’s experiences of nonlethal intimate partner abuse by the method of abuse (slapping, throwing object, etc.) and by the relatedness of the children in the household to the woman and her current partner. Brownridge found that women in stepfamilies are at greater risk of nonlethal intimate partner abuse than women in biological families, and also at greater risk of the most severe, nonlethal forms of physical abuse. For example, Brownridge found that women in stepfamilies are twice as likely as women in biological families to be pushed and three to four times as likely to be choked or beaten. However, the method of family classification makes interpretation difficult. Brownridge classified stepfamilies in terms of legal marriages and common-law marriages in which at least one of the children in the household was from a previous relationship of one of the parents. Women who were unrelated to at least one child in the household and men who were unrelated to at least one child in the household were grouped together. Biological families included legal marriages and common-law marriages in which all of the children in the household were from the current relationship. Although most stepfamilies in Brownridge’s study likely involved stepfathers (most custody awards are to women), because of the inability to distinguish between stepmothers and stepfathers, it not possible to determine with certainty how the presence of children sired by a woman’s previous partner affects her risk of nonlethal intimate partner abuse.
In addition to being pushed and beaten by their intimate partners, women are sometimes killed by them, thus becoming victims of lethal intimate partner abuse. Among the various risk factors for both lethal and nonlethal intimate partner abuse is that of living with a child sired by a previous partner. Women living with children sired by previous partners are overrepresented as victims of lethal intimate partner abuse relative to women living with children sired only by their current partner (Brewer and Paulsen 1999; Campbell et al. 2003; Daly et al. 1997). This result holds whether the comparison sample is unabused (Brewer and Paulsen 1999; Daly et al. 1997) or abused (Campbell et al. 2003) women. Because of this common risk factor, Goetz and colleagues (2008) suggested that future research might investigate how such risk factors as the presence of children in the household who are sired by previous partners differentiate unabused women from victims of nonlethal and of lethal intimate partner abuse. If we consider homicide to be the most severe form of intimate partner abuse, we might hypothesize a relationship between the presence of children sired by previous partners and the risk of violence severity. Although each of the comparisons made in the current study (e.g., unabused versus abused, less severely abused versus more severely abused) have been documented previously, no study that we are aware of has tested these hypotheses simultaneously with samples drawn from the same population, which is important for considering the size of the stepchild effect across groups. The current research tests the hypothesized relationship between presence of children sired by a previous partner and violence severity using two samples of data obtained from the Chicago Women’s Health Risk Study (Block 2000a). The first sample contains data pertaining to women who were victims of nonlethal intimate partner violence (abused women) and women who were not (unabused women). The second sample contains data pertaining to victims of lethal intimate partner abuse. Both samples were combined for analysis to examine the hypothesized ordinal relationship between the presence of children sired by previous partners in the household and the severity of intimate partner abuse, from nonlethal to lethal forms of abuse.
The two samples of data analyzed in the current research were obtained from the Chicago Women’s Health Risk Study, 1995–1998 (Block 2000a), which was designed to include “hidden” women—that is, women who were not part of agency or shelter populations. The first sample consists of data secured from interviews of unabused women and abused women who were questioned at specific sites in Chicago chosen because they are in areas where the population-wide levels of intimate partner homicide are high (Block 2000b). A total of 2,740 women entering hospitals or health clinics was screened using three questions about physical violence, sexual violence, and fear of returning home. All women who were 18 years of age or older and who answered “yes” to any of these questions were interviewed. In addition, a random sample of women who answered “no” to all questions and who had been in an intimate relationship in the past year was interviewed. Following the screening, women were asked 11 questions using a modified version of the Conflict Tactic Scale (CTS; Straus 1979) to classify them as having been either abused or not abused in the past year.
The second sample consists of data secured from interviews with knowledgeable relatives or friends about all cases of intimate partner homicide that involved a woman (whether victim or offender, in a heterosexual or a homosexual relationship, whether abused or not abused) during 1995 and 1996 in Chicago (Block 2000b). For further information about the Chicago Women’s Health Risk Study, see Block (2000b).
Categories of Severity of Abuse
Unabused women are those women who did not report any incidents of physical abuse or threats of physical abuse by an intimate partner in the previous year (Block 2000b). Abused women are those women who reported at least one incident of physical abuse or threat of physical abuse by an intimate partner in the previous year (Block 2000b).
Abused women completed a 12-month retrospective calendar that requested details about all violent incidents within that period (Block 2000b). Interviewers coded the severity of each incident based on either the act against the woman or the injuries sustained as a consequence of a partner’s physical abuse against them using the highest number category (Block 2000b; see below for details). Interviewers separately noted incidents of forced sex with or without accompanying physical abuse (Block 2000b). The most severe incidents recorded for the previous year that were not life-threatening involved (1) a threat to hit with a fist or anything else that could hurt them; (2) slapping, pushing, or throwing something that could hurt them, or no injury or lasting pain; and (3) punching, kicking, or bruises, cuts, or continuing pain. For women who experienced at least one potentially life-threatening incident, the most serious incident was (1) “beaten up,” choked, or burns, broken bones, or severe contusions; (2) a threat to use a weapon, or an incident resulting in a head injury, loss of consciousness, permanent injury; or internal injury, or (3) an incident where a weapon was used or which resulted in wounds from a weapon.
Cases of same-sex intimate partner homicide and cases in which women were the offenders rather than the victims of intimate partner homicide were excluded. The intimate partner homicide cases in our sample consisted of women who had been killed by male intimate partners. We constructed an ordinal variable indicating the severity of abuse such that 1 = unabused women, 2 = abused, but no life-threatening incidents, 3 = abused with at least one life-threatening incident, and 4 = abused homicide victims.
The first sample includes 205 unabused women, 197 less severely abused women, and 300 more severely abused women. Three abused women were excluded because missing calendar data made the severity of abuse impossible to determine. Seven women from the first sample were excluded because their named partner was a woman (see below). Of the 87 cases of intimate partner homicide in the second sample, 28 cases were excluded because the woman was the offender rather than the victim, leaving 59 cases of opposite-sex perpetrated intimate partner homicide.
Categories of Children
During the interviews, each woman named a recent intimate partner. For the unabused women, the named individual was a man who had been their intimate partner in the past year. Women who had multiple intimate partners in the past year were asked to name the partner that they spend the most time with currently and that they feel closest to. For the women who had experienced abuse in the past year from one intimate partner, the named individual was the abusive intimate partner, whether a current or a former partner. Women who had experienced abuse in the past year from more than one intimate partner were asked to select as the named partner “the person responsible for the most serious of the incidents we have been talking about, or the incident(s) that bothered you the most” (Block 2000c:35, Clinic/Hospital Interview, emphases in original). For the cases in the intimate homicide sample, the named partner is the purported offender (using proxy data, police summary files, medical examiner files, court case files, and newspaper articles; Block 2000b). For both samples of women, the named partner could have been a current or a former partner.
Interviewers collected information about the relationship of every child in each woman’s household to both the woman as well as the intimate partner that she named. For women who were victims of intimate partner homicide, interviewers collected the same information from proxies. We constructed a dichotomous variable for both samples such that 0 = some of the woman’s genetic children in the household were sired by another partner and 1 = all of the woman’s genetic children in the household were sired by the partner.
Women from both samples were excluded if they did not have genetic children living in the household. The many reasons for not living with genetic children included the following: they did not have children, their children were adults who lived elsewhere, their children had died, and their children had been removed from the home by the Department of Children and Family Services. For these reasons, 92 unabused women, 83 less severely abused women, and 170 more severely abused women were excluded from analyses. For the homicide victim sample, 31 of the 59 remaining women were excluded because they did not have genetic children living in the household. Women were also excluded if they were in a same-sex relationship with the named partner. This exclusion criterion eliminated seven individuals from the first sample and two individuals from the second sample. The final sample therefore includes 111 unabused women, 111 less severely abused women, 128 more severely abused women, and 26 homicide victims.
Both of the variables in the current study—the severity of abuse and the presence of genetic children in the household sired by someone other than the partner—are ordinal variables. For this reason, we performed a Mantel-Haenszel chi-square (Mantel and Haenszel 1959; Mantel 1963). This one-degree-of-freedom summary statistic tests the linear relationship between the row and column variable—that is, whether the increase in the row variable associated with the increase in the column variable is greater than expected by chance.
The results of the current research indicate that women who have genetic children in the household sired by a previous partner experience increased risk of severe forms of abuse when compared with women whose children are all sired by their current partner. Women with some children sired by a previous partner are increasingly overrepresented among victims of abuse as the severity of abuse increases. The current study connects research on nonlethal forms of woman abuse (e.g., Brownridge 2004; Daly et al. 1993; Figueredo and McCloskey 1993) with lethal forms of woman abuse (e.g., Brewer and Paulsen 1999; Campbell et al. 2003; Daly et al. 1997) using the presence of children sired by a previous partner as a common risk factor uniting the four groups (no abuse, nonlethal abuse, life-threatening nonlethal abuse, and lethal abuse).
The results of the current research identify an ordinal trend of increasing representation of women with children sired by a previous partner as victims of abuse as the severity of abuse increases. It remains unclear, however, whether jealousy associated with the presence of the stepchildren themselves is the cause of this trend. An alternative explanation is that individuals who become stepparents have certain undesirable characteristics that make them less desirable mates and, therefore, more likely to pair with other lower-mate-value individuals, such as women with children from a previous relationship. Figueredo and McCloskey (1993) suggest that such low-mate-value men have a disadvantage in the mating market and may pursue mating and parental strategies that include the use of intimate partner violence. There is some evidence for this hypothesis, because men who become stepfathers differ in systematic ways from men who do not become stepfathers. Stepfathers are less educated, have lower income, and are more likely to have been divorced and to already have children themselves (Anderson 2000). Thus, it seems possible that men who have been excluded from the mating market may be pursuing a coercive mating strategy, sometimes to their own potential detriment, as in the case of femicide. Future research should attempt to differentiate the two major evolutionary explanations for the relationship between the presence of children sired by a previous partner and abuse severity.
One way to differentiate and test these two hypotheses might be to examine the influence of stepchild residence on the severity of intimate partner violence. If children sired by a woman’s previous partners provide a daily reminder of the woman’s partner’s inability to dominate his partner, as suggested by Daly et al. (1993), then women whose children from previous relationships live outside their household should experience less severe abuse than women whose children from previous relationships live in their household. Alternatively, the competitively disadvantaged male hypothesis does not clearly predict a difference in abuse severity between women in these two groups.
Future research might also address whether the number of children sired by previous partners or the number of sires of children in the household affects the relationship between the presence of children sired by a previous partner and the severity of partner-directed abuse. If Daly and colleagues (1993) are correct regarding the male partner’s inability to dominate, then men may inflict more severe abuse against their partner when there are more children sired by previous partners in the household or when the number of sires of children in the household is greater, because each of these scenarios may activate a stronger response of male sexual jealousy. Alternatively, the number of children in the household that are sired by a previous partner may be an indication of the woman’s mate value, with more children corresponding to a lower mate value. If this is the case, then higher numbers of children in the household sired by a previous partner may be indicative of the extent to which the man is competitively disadvantaged and, therefore, the extent to which he may opt for coercive sexual or parental strategies. Thus, both evolutionary explanations predict the same pattern between number of children in the household sired by a previous partner and abuse severity.
There are several limitations of the current research, some of which can be addressed in future research. Because of the way the questions were posed in the Chicago Women’s Health Risk Study, the severity of a particular incident in the nonlethally abused group of women is defined by the intended action of the abuser against the woman as well as the resulting injury (e.g., “slapping, pushing, throwing something that can hurt you, or no injury; no lasting pain”), whereas the severity of abuse for the lethal group is defined only by the consequences of the action. For example, if a man kicked his partner and she was bruised, the incident would have been classified as an indication of abuse, but not a life-threatening incident. If, however, a man kicked his partner and she happened to fall and fatally hit her head, the incident would have been classified as lethal abuse. Future research might separate the two definitions of abuse severity used for the nonlethally abused group of women into two measures to determine if the intent to severely harm is more strongly related to the presence of children sired by a previous partner than is the reported severity of harm, after the cases of unintended severe harm are coded by intent alone.
In addition to raising several questions for future research, the current research identifies a predicted significant, positive ordinal trend in the overrepresentation of women with children sired by a previous partner as victims of severe forms of partner abuse, linking research on nonlethal domestic abuse with research on intimate partner femicide.