Am J Crim Just (2008) 33:130–147 DOI 10.1007/s12103-008-9033-3
Social Disadvantage and Family Violence: Neighborhood Effects on Attitudes about Intimate Partner Violence and Corporal Punishment Deeanna M. Button
Received: 1 September 2007 / Accepted: 15 December 2007 / Published online: 10 April 2008 # Southern Criminal Justice Association 2008
Abstract Social disorganization theory asserts that neighborhood composition affects levels of violence within the community. The purpose of this article is to analyze the bivariate effects of social disorganization, crime, and collective efficacy, in addition to the individual factors of gender, race, and a history of child maltreatment, on the acceptance of using violence within the family. Data from the Norfolk Police Department (2000–2004), 2000 Census, and 2006 Norfolk Residents’ Attitudes about Crime Survey were used to determine differences in approval of family violence. Results indicated that approval for family violence is an individuallevel phenomenon as well as a community-level occurrence. Various aspects of family violence elicit different levels of tolerance by both micro- and macro-level characteristics. Implications are discussed. Keywords Social disorganization . Neighborhoods . Attitudes . Family violence . Intimate partner violence . Corporal punishment
Introduction Family violence is widespread and occurs everyday in the United States. Indeed, “1,551,143 incidents of family violence were reported to the National IncidentBased Reporting System between 1996 and 2001” (National Criminal Justice Reference Service 2006). Family violence may include the physical abuse of a child by an adult, intimate partner violence, violence between siblings, and elder abuse (Gelles 1990; Payne and Gainey 2005). Likewise, over 90% of the United States population has been corporally punished (Straus 2001a). The consequences of the various forms of family violence and physical discipline are both immediate and long lasting. Although there is a great deal of variation in the effects on the
D. M. Button (*) Department of Sociology and Criminal Justice, University of Delaware, Newark, DE 19711, USA e-mail:
[email protected]
Am J Crim Just (2008) 33:130–147
131
individual, the costs to society are steadfast and detrimental (Payne and Gainey 2005; Polonko 2005; Straus 2001a). Researchers emphasize the importance of individual level characteristics such as age, race, income, education etc., and the interplay that these factors have on the prevalence of family violence and the pervasiveness of corporal punishment (Hussey et al. 2006; Kolko 2002; Stiffman et al. 2002; Straus 2001a). Additionally, scholars have determined that the presence of mental illness, substance use, and a history of abuse impact the incidence of family violence (Bolton 1987; Kolko 2002; Straus 2001a). Previous research has predominately centered on micro-level characteristics, and, as a result, policy implications have been geared toward individuals. Little research has been done at the aggregate level. To determine if family violence is “as much a spatial phenomenon as it is a matter of individual characteristics” (Almgren 2005, p. 218), the focus needs to shift from individual psychological factors to structural variables. As nearly every family is victimized by some type of family violence (Payne and Gainey 2005), it is important that alternative dynamics of family life be explored. Determining the influence that community-level processes have on family violence allows for the development of very different social policy interventions.
Literature Review Social Disorganization Theory and Family Violence Social disorganization theory, defined as “the inability of a community structure to realize the common values of its residents and maintain effective social controls” (Sampson and Wilson 1995, p. 45) has been used to clarify the occurrence of violence within local communities (Sampson and Groves 1989). Originating in the Chicago school in the 1940s, researchers Shaw and McKay (1942) argued that the presence of concentrated economic disadvantage, ethnic heterogeneity, and residential instability leads to a disturbance in a community’s organizational structure and social control (Sampson and Groves 1989). Historically, the presence of these variables has been linked to weak social control that, in turn, leads to increased levels of crime and interpersonal violence (Bursik and Grasmick 1993; Miles-Doan 1998; Patillo 1998). Bursik and Grasmick (1993a) contend that “the differential rates of criminal behavior and victimization among neighborhoods... represent variations in the ability of neighborhoods to regulate themselves” (p. 4). These authors describe social control as the community’s effort to regulate the behavior of residents and visitors (Bursik and Grasmick 1993a). Organized neighborhoods share common values that create a degree of social cohesion. This community-wide unity provides an essential framework for the actions and behaviors of its residents. Deviant behaviors, such as delinquency, crime, and violence, are hampered by cohesive neighborhood standards. Neighborhoods with increased levels of social disorganization have weakened social controls that may be necessary to prevent violence and crime (Patillo 1998).
132
Am J Crim Just (2008) 33:130–147
Areas of concentrated disadvantage—the neighborhoods with the lowest incomes, increased rates of unemployment, and institutional disinvestment—lack adequate resources and financial support. Neighborhood level poverty results in a weakened ability to maintain basic social control agents. Institutions such as churches, schools, and community organizations, struggle to prosper, and lose the ability to exercise control over the community (Benson et al. 2004; Sampson and Groves 1989; Sampson and Raudenbush 1999). The indirect effects of economic deprivation are of similar theoretical importance (Bursik and Grasmick 1993). A neighborhood’s socioeconomic status has been negatively linked to the residential stability and ethnic heterogeneity of the community. Warner and Rountree (1997) state, “[n]eighborhood instability makes the development of deep and lasting affective relational networks very difficult, while heterogeneity in the area limits the breadth of such networks” (p. 521). Residents living in indigent communities relocate to neighborhoods that are more desirable. This instability compromises the organization of the neighborhood. Mobility stands as a barrier to the development of relationships that create informal neighborhood control (Bursik and Grasmick 1993b; Tittle 1983). Heterogeneity among residents decreases communication and familiarity among inhabitants. As residents fear and distrust their diverse neighbors, common problems (i.e., crime and violence) and possible solutions go unrecognized (Van Wyk et al. 2003). The community lacks the ability to organize itself, and consequently, control unwanted behavior (Sampson and Groves 1989; Patillo 1998; Tittle 1983). Sampson et al. (1997) argue that the presence of social control mediates the relationship between social disorder and violence. Informal social control is described as a community’s outward acknowledgement of shared goals within their neighborhood. Collective efficacy, defined as “social cohesion among neighbors combined with their willingness to intervene on behalf of the common good” (Sampson et al. 1997, p. 918) has been inversely linked to neighborhood crime and violence (see also Sampson and Raudenbush 1999). Essentially, collective efficacy is conceptualized as the social networks within a community—kinship, friendship, and acquaintanceship—and the degree of resident involvement in community-based organizations (Browning 2002, p. 834). Sampson et al. (1997) and Sampson and Raudenbush (1999) assert that the motivation to intercede deviance and violence is based upon the mutual trust that the individuals of the neighborhood share with one another. There must be solidarity in recognizing and addressing delinquent behavior, criminal activity, and violent acts. Research shows that lower rates of violence may be predicted by higher levels of collective efficacy (Sampson et al. 1997). While little research has been conducted regarding the effects of collective efficacy on family violence, Van Wyk et al. (2003) found an indirect relationship between neighborhood interaction and partner violence. The more contact residents had with one another, the lower the likelihood of intimate partner violence. Likewise, Browning’s (2002) research shows that while intimate partner homicide rates were connected to concentrated disadvantage, collective efficacy mediated the relationship. These studies suggest violence that occurs behind closed doors may be reduced through the willingness of neighbors to uphold the common values of the community.
Am J Crim Just (2008) 33:130–147
133
Attitudes Toward Family Violence Researchers “look to the values, beliefs, and norms of a society to better understand variation in family violence” (Payne and Gainey 2005, p. 47). There is evidence to suggest that the approval of aggressive behavior is correlated with the use of violent acts. Attitude alone does not determine whether a one will hit, but it is a significant predictor (Ateah et al. 2003). For example, Kantor et al. (1994) determined that the odds of adult male physical aggression is increased by 2.17 times with one’s approval of aggression. Straus (2001a) asserts that the more approval a parent has for physical punishment (measured by the Corporal Punishment Approval index), the more likely a child will be punished with physical violence. Regarding discipline strategies and attitudes about physical punishment, Vittrup et al. (2006) surveyed 132 mothers, and found that, over time, spanking was significantly correlated with mothers’ attitudes toward physical punishment (p. 2055). Ashton (2001) found that individuals with higher scores for approval for corporal discipline were less likely to perceive child maltreatment, and subsequently, less likely to report abusive behaviors. Gracia and Herrero (2006) determined that community characteristics influence attitudes toward reporting child physical abuse. Individuals living in neighborhoods with low or medium levels of disorder were significantly more willing to report maltreatment than those who characterized their community as highly disorganized. Because attitudes towards violence explain a significant proportion of aggressive behaviors, it is necessary to examine the impact of neighborhood characteristics on the approval of family violence. To create a framework in which structural features are used to explain family violence, micro- and macro-level factors need to be integrated. Previous research shows that age, race, income, education, and a history of child maltreatment are significant predictors of intimate partner violence and the use of corporal violence. Intimate Partner Violence According to Schneider (2005), partner aggression consists of the “physical, sexual, or psychological harm to another by a current or former partner or spouse.” Intimate partner violence is conceptualized as intentional violent acts by intimate partners that result in physical pain or injury of the other intimate partners (Straus 2001b). Age is a consistent risk factor for intimate partner violence. Indeed, “for every 10 years increase in age, risk of mild male partner aggression decrease[d] by 29% and risk of severe aggression decrease[d] by 19%” (Schumacher et al. 2001, p. 286). Research from Leonard and Blane (1992) shows a significant predictor of adult male aggression is being non-White. However, Kantor et al. (1994) found that race did not significantly predict intimate partner violence when other factors were controlled. Regarding income and education, Coleman et al. (1980) and McKenry et al. (1995) found a significant negative correlation between family income and intimate partner violence. Chase et al. (2003) recruited 103 couples from four addiction treatment facilities and found that individuals who engaged in partner violence, compared to nonviolent counterparts, reported less education. Furthermore, Sugarman and Hotaling’s (1989) nationally representative sample of 608 men validated previous research: “experiencing child abuse in the family of origin was a significant predictor
134
Am J Crim Just (2008) 33:130–147
of adult male partner aggression” (Macewen and Barling 1988; O’Leary and Curley 1986; Schumacher et al. 2001, p. 290). Corporal Violence of Children Similar to intimate partner violence, a considerable amount of research has been conducted on predictors of child corporal violence. According to Straus (2001b), corporal punishment is “an act carried out with the intention of causing a child to experience physical pain, but not injury, for purposes of correction or control of child’s behavior” (p. 187). Research shows that younger parents use physical discipline more frequently than older parents (Straus 2001a). In fact, Giles-Sims et al. (1995) assert that mothers under 25 are most likely to use corporal violence. Regarding race, Cazenave and Straus (1990) assert that White parents report more spanking than non-White parents. However, in a review of the literature, Straus (2001a) reports findings that suggest that the same proportion of Black parents as White parents admit to the use of physical discipline. Pinderhughes et al. (2000) found that harsh parenting strategies (i.e., corporal punishment) are predicted by lower socioeconomic status, but when the age of parent, ethnicity of the family, and violence between caregivers is accounted for, there is no relationship between socioeconomic status and corporal punishment (Straus 2001a). Research also shows that a history of child maltreatment is associated with the use of corporal punishment (Ateah et al. 2003). Kaufman and Zigler (1986) assert that a history of abuse increases the risk for the mistreatment of one’s own offspring. Parents who were hit as adolescents have a greater likelihood of corporally punishing their own children (Straus 2001a). To extend social disorganization theory and the notion of collective efficacy to family violence implies that the control of family violence is a common value at the community level. To explore this notion, it is necessary to examine the variation in attitudes toward family violence at the macro-level. The purpose of this study is to analyze the bivariate and multivariate effects of social disorganization, crime, and collective efficacy, in addition to the individual factors of gender, race, and a history of child maltreatment, on the acceptance of using violence within the family.
Methods Research Design Three data sources were used in the analyses: 2000 decennial census data, 2000– 2004 Norfolk police data, and the 2006 Norfolk Residents’ Attitudes about Crime Survey. Address based data on aggravated assaults was provided by the Norfolk police department. This study specifically looked at aggravated assault rates as an indication of criminal violence within a community. The Norfolk police department utilizes the National Incident Based Police Reporting System (NIBRS) which “aggregates the number of incidents by offense type, [and] provides...information about offenses, victims, and offenders” (Bureau of Justice Statistics 2006). Several years of data were used to obtain a more stable estimate of crime in each neighborhood.
Am J Crim Just (2008) 33:130–147
135
The 2000 census data provided demographic information at the block group level. A census block group generally consists of 500 households and contains between 600 and 3,000 individuals. Local residents, providing a more accurate breakdown of neighborhoods, establish census block groups (U.S. Census Bureau n.d.). Information obtained from this data provided racial/ethnic, socioeconomic, and family structure characteristics of the neighborhoods surveyed. The 2006 Norfolk Residents’ Attitudes about Crime Survey consists of a sample of Norfolk, Virginia residents and focuses on respondents’ various attitudes regarding crime and criminal justice issues. Data were gathered from 20 Norfolk neighborhoods. Neighborhoods were defined by census block groups. Block groups were selected based on population size (block groups with over 1000 persons) and aggravated assault rates obtained from the 2000–2004 Norfolk Police Data. A neighborhood was defined as high crime if the aggravated assault rate exceeded 300 per 100,000 persons, and was defined as low crime if the aggravated assault rate fell below 100 per 100,000 persons. The sampling technique consisted of randomly selecting streets within block groups, households on streets, and adult residents (one 18-year-old or older per household) within households. This method was previously utilized by Browning (2002). Interviewers hand delivered the questionnaire, enclosed with a self-addressed stamped envelop to selected households. The 2006 Norfolk Residents’ Attitudes toward Crime Survey achieved a final response rate of 20.9%. Due to the purposive nature of this sample, it was somewhat expected to receive lower than usual response rates. First, mail surveys typically elicit extremely low response rates. Second, research recruitment of low-income minorities is difficult (Jackson and Ivanoff 1999). Neighborhoods were selected according to crime rates. Half of the communities sampled had crime rates two standard deviations above the mean; while the other half had crime rates two standard deviations below the mean. High crime neighborhoods are often characterized by high proportions of low-income minority residents (Sampson and Wilson 1995). Researchers have addressed the issue of low participation by minority residents (Hatchett et al. 2000; Jackson and Ivanoff 1999; Koch and Cebula 2004). The literature suggests that African Americans are hesitant to participate in empirical studies due to a common mistrust toward science (Hatchett et al. 2000; Jackson and Ivanoff 1999). Outside of this widespread suspicion, lack of awareness and overlooked opportunity are frequently cited reasons for forgoing research participation (Hatchett et al. 2000). Many of these neighborhoods receive limited public service announcements regarding research initiatives. When residents are approached, they are likely to decline because they are unaware of the empirical importance and distrustful of the research credibility. Measurement of Variables Dependent Variables Family violence may be conceptually defined as “an act carried out with the intention or perceived intention of causing physical pain or injury to another person” (Straus 2000) within the familial structure. As the most common form of family violence is inter-spousal (Gelles 1990), this research focuses on the violence
136
Am J Crim Just (2008) 33:130–147
between intimate partners. Regarding corporal punishment, Straus (2001a) states that corporal punishment may include “the use of physical forces with the intention of causing a child to experience pain, but not injury, for the purpose of correction or control of the child’s behavior” (p. 4). The 2006 Norfolk Residents’ Attitudes about Crime Survey included items from the Personal and Relationships Profile. The Personal and Relationships Profile (PRP) can be used to research family violence. It measures both individual-level and relationship-level characteristics (Straus and Hamby 1999). The present investigation used the violence approval scale of the PRP. The violence approval scale measures “the extent to which use of physical force is acceptable in a variety of interpersonal situations” (Straus and Hamby 1999, p. 5). According to Straus and Mouradian (1999), the violence approval scale has good reliability (AlphaQ.70). This study utilized the family violence subset of the violence approval scale. Approval of intimate partner violence was indicated if a respondent agreed or strongly agreed with the statement “I can think of a situation when I would approve of a wife slapping a husband’s face” or “I can think of a situation when I would approve of a husband slapping a wife’s face” (Straus and Hamby 1999, p. 14). Likewise, approval of corporal punishment was indicated if a respondent agreed or strongly agreed with the statement “It is sometimes necessary to discipline a child with a good, hard spanking” or “It is sometimes necessary for parents to slap a teen that talks back or is getting into trouble” (Straus and Hamby 1999, p. 14).
Independent Variables The presence of social disorder has been linked to tangible neighborhood features (Patillo 1998; Sampson and Raudenbush 1999; Veysey and Messner 1999). Neighborhood social disorganization was operationalized by responses (four-point scale) indicating the presence of: (a) litter, (b) signs of vandalism, (c) broken/ boarded windows, (d) burglar bars, (e) unsupervised youth, (f) sloppy repairs, (g) noisy neighbors, (h) public lighting, (i) overcrowding, (j) public drinking, (k) run down buildings, (l) gang activity, and (m) lack of police. Collective efficacy was scaled through an index of several items. Respondents were asked to specify the likelihood (five-point scale) of a neighbor intervening if (a) children were skipping school, (b) children were showing disrespect to an adult, (c) a fight broke out in front of their house, and (d) children were spray painting a graffiti on a local building. To ascertain if there is a relationship between neighborhood context and approval of intimate partner violence and/or corporal punishment, the influence of individual factors needs to be controlled. History of family violence was measured by the violent socialization subset of the PRP, and was indicated if a respondent agreed or strongly agreed with the statement “When I was less than 12 years old, I was spanked or hit a lot by my mother or father” or “When I was a teenager, I was hit a lot by my mother or father” (Straus and Hamby 1999, p. 14). Parents with a history of abuse are more than twice as likely to maltreat their own children (Pears and Capaldi 2001). Ringwalt et al. (1989) found a significant relationship between a parent’s history of physical discipline their approval of corporal punishment.
Am J Crim Just (2008) 33:130–147
137
Demographic variables consisted of respondent’s gender, a dichotomous variable (0=male, 1=female) and race (0=other, 1=White).
Results Univariate Analysis Table 1 shows the descriptive statistics. The majority of participants live in neighborhoods with low aggravated assault rates (60.8%). Most respondents were between the ages of 25–55 (59.8%). Nearly three-fourths of participants identified themselves as White, and almost two-thirds of the sample is female (62.5%). While there is consistent variation across the 11 categories of income (8–10% per group), the greater part of respondents earned $50,000 or more annually. Whereas only 10.1% of the sample indicated high approval for intimate partner violence, 26.5% of individuals Table 1 Descriptive characteristics
Variable Crime rate High Low Social disorder Very low Low High Very high Gender Male Female Age 18–24 25–39 40–54 55–69 70 years or more Race White Other Intimate partner violence Low approval Moderate approval High approval Corporal punishment Low support Moderate support High support Collective efficacy Low Medium High History of child maltreatment No Yes
n
%
82 127
39.2 60.8
125 43 21 11
62.5 21.5 10.5 5.5
78 130
37.5 62.5
11 52 73 43 30
5.3 24.9 34.9 20.6 14.4
152 57
72.7 27.3
158 28 21
76.3 13.5 10.1
104 46 54
51.0 22.5 26.5
57 34 117
27.4 16.3 56.3
156 21
88.1 11.9
138
Am J Crim Just (2008) 33:130–147
suggested strong support for corporal punishment. Most residents feel positively about the degree of social cohesion within their communities; 56.3% indicated high levels of collective efficacy. Ten percent of individuals indicated a prior history of child maltreatment. This is consistent with previous research (Polonko 2005). Bivariate Analysis Pearson’s r was used to determine the relationship between the different forms of family violence. The data shows a positive zero-order relationship between one’s approval of intimate partner violence and one’s support for corporal punishment (r=.269; p=.000). To determine if there are differences between individuals and neighborhoods in the approval of family violence, independent samples t-tests were conducted. Differences in attitudes were examined by levels of perceived social disorder, levels of crime, levels of perceived collective efficacy, history of child maltreatment, gender, and race. Additionally, ANOVA scores were analyzed to determine differences in attitudes by level of social disorganization and by level of collective efficacy. See Tables 2 and 3 for a summary of results. Intimate Partner Violence Results indicate that those residing in neighborhoods with high levels of crime approve of intimate partner violence significantly more than those living in communities with low levels of crime. There is a significant difference in attitudes toward intimate partner violence between individuals with and without a history of child maltreatment. As expected, individuals who were hit as children had higher
Table 2 Mean scores for IPV and CP by dichotomized independent variables Approval of IPV
Disorder Low High Level of crime Low High Child maltreatment History No history Gender Male Female Race White Other *p<.05, **p<.01
x
SD
3.48 3.38
1.34 1.45
3.46 3.55
1.30 1.53
4.29 3.37
1.79 1.30
3.66 3.40
1.34 1.42
3.61 3.20
1.43 1.25
Support for CP t
x
SD
4.48 5.00
1.43 1.65
4.45 4.73
1.47 1.47
4.85 4.41
1.63 1.46
5.04 4.26
1.40 1.44
4.41 4.98
1.43 1.52
t −1.809*
0.408 −2.413**
0.429
−2.899**
−1.241
1.292
3.767**
−1.860*
2.491**
Am J Crim Just (2008) 33:130–147
139
levels of support for aggression between intimates than those who did not experience violence. Residents who identified themselves as White have significantly more support for partner aggression than non-White residents. Corporal Punishment Independent samples t-test showed variation in means of support for corporal punishment between neighbors with low and high levels of perceived disorder. Similarly, as shown in Table 3, ANOVA scores indicate significant differences in support for corporal discipline among the four levels of perceived social disorder. Residents with very low levels of perceived social disorder have the least amount of support for physical punishment. Mean scores increase with the degree of perceived disorder in a community. Note that although residents who perceive their neighborhoods as highly disordered support the use of corporal violence more than residents in communities with very low levels of perceived disorder, it appears as if neighborhoods with moderate levels of perceived disorder support the use of physical discipline more than communities that are at the extremes of perceived (dis)organization. While gender does not play a significant role in predicting attitudes about intimate partner violence, one’s sex does significantly influence support for corporal punishment. Men show more support for spanking than women. Non-White individuals were significantly more likely to approve of spanking than White residents. Multivariate Analysis Multiple regression was used to determine the amount of variation in the approval of family violence that is explained by the following micro- and macro-level variables: (a) support for corporal punishment/approval for intimate partner violence, (b) perceived social disorder, (c) perceived collective efficacy, (d) history of child maltreatment, (e) gender, and (f) race. To determine what individual-level and community-level characteristics affect attitudes about family violence, the seven independent variables where put into two linear regression models.
Table 3 ANOVAS for IPV and CP by social disorder and collective efficacy levels Approval of IPV
Social disorder Very low Low High Very high Collective efficacy Low Medium High *p<.05, **p<.01
x
SD
3.43 3.63 3.48 3.18
1.33 1.36 1.17 1.94
3.40 3.94 3.41
1.36 1.50 1.36
Support for CP F
x
SD
4.35 4.88 5.24 4.50
1.38 1.50 1.64 1.65
4.48 4.69 4.56
1.40 1.26 1.57
0.395
F 3.064*
2.065
0.196
140
Am J Crim Just (2008) 33:130–147
Table 4 Model 1. Regression of predictor variables on approval of intimate partner violence Variables
B
Std. error
Beta
t
Sig.
Constant Support for CP Social disorder Crime Collective efficacy History of child maltreatment Gender Race F R2
1.924 .236 −.004 −.171 −.016 .119 −.069 .655 3.745 .124
.649 .069 .011 .218 .023 .068 .205 .242
.258 −.026 −.062 .051 .126 −.025 .213
2.966 3.414 −.319 −.788 −.705 1.751 −.339 2.704
.003 .001 .750 .432 .482 .082 .735 .007 .001
Model 1: Intimate Partner Violence Table 4 shows the results for Model 1. This model significantly explains 12.4% of the variation in one’s approval of intimate partner violence (F=3.745; p=.001; R2 =.124). Support for corporal punishment and race are the only two variables that significantly contribute to variation in approval of intimate partner violence. It was expected that one’s race will have a significant effect on the approval of intimate partner violence. Race does predict approval of intimate partner violence. However, previous literature suggests that individuals who self-identify as nonWhite are more likely to engage in partner abuse. Consistent with bivariate results, this model suggests that White residents have greater approval of partner abuse. Note, too, that one’s attitude toward corporal punishment and one’s race have the greatest impact on approval of intimate partner violence. Perceived social disorder, crime, perceived collective efficacy, history of child maltreatment, and gender do not significantly predict approval of intimate aggression. Model 2: Support for Corporal Punishment Table 5 shows the results for Model 2. Model 2 significantly explains 21.9% of the variation in the support for corporal punishment (F=7.414; p=.000; R2 =.219).
Table 5 Model 2. Regression of predictor variables on support for corporal punishment Variables
B
Std. error
Beta
t
Sig.
Constant Approval of IPV Social disorder Crime Collective efficacy History of child maltreatment Gender Race F R2
3.477 .251 .021 .105 .022 .130 −.848 −.740 7.414 .219
.635 .074 .012 .225 .024 .070 .202 .249
.230 .138 .034 .063 .127 −.279 −.220
5.472 3.414 1.803 .466 .921 .1868 −4.200 −2.973
.000 .001 .073 .642 .358 .063 .000 .003 .000
Am J Crim Just (2008) 33:130–147
141
Resembling Model 1, support for corporal punishment is predicted by one’s approval of intimate partner violence. This is consistent with bivariate findings. As with bivariate findings, one’s gender and race significantly affect the support for physical discipline. Men have significantly more support for spanking than women, when controlling for other factors. Individuals who self-identified as nonWhite rate a greater amount of support of corporal violence. While previous literature suggests that race is an important variable, earlier results indicate that White participants have greater support for spanking than non-White participants (Cazenave and Straus 1990; Straus 2001a) Notice, also, that one’s attitude toward intimate partner violence, one’s gender, and race have the greatest impact on support for corporal punishment. Perceived social disorder, crime, perceived collective efficacy, and a history of child maltreatment do not significantly predict any proportion of support for corporal punishment.
Discussion Tolerance for one type of violence is generally coupled with acceptance of another type of violence (Payne and Gainey 2005). This paper found that individuals who approve of aggression between intimate partners were likely to support aggressive discipline strategies. Collectively, these findings have important implications for policy, research, and theory. Approval of intimate partner violence varied by level of crime within a neighborhood, but did not vary by level of perceived social disorder. This suggests that attitudes are not necessarily shaped by the lack of perceived social control within an area, but rather that attitudes may be influenced by the presence of actual crime. Individuals residing in neighborhoods with increased violent activity may assimilate to the nature of violence. As approval of violence is associated with aggressive behavior (Black et al. 1999; Schumacher et al. 2001), social control agents should be aware that neighborhood crime is a risk factor for the violence that individuals endure behind closed doors. The opposite is true for the support for corporal violence. Support for physical punishment did not vary by neighborhood crime levels, but did differ by level of perceived social disorder. Corporal punishment is used as a method of social control. Parents attempt to facilitate children’s behavior with the use of physical violence. Those residing in high crime neighborhoods are no more likely to approve of spanking than those who live in low crime neighborhoods. However, in communities perceived as socially disordered, the use of corporal violence to manage children is seen more favorably. In disorganized neighborhoods, residents feel as if they have renounced their ability to organize the community (Sampson et al. 1997). Perhaps, parents fear the loss of control over their own children, and employ more extreme methods of management. Notice that there may be a ceiling on the effects of perceived social disorder. Approval for corporal violence increases from disorder perceived as very low to high, but tapers off for neighborhoods perceived as very highly disordered. This implies that although less disorder influences attitudes
142
Am J Crim Just (2008) 33:130–147
toward violence favorably, there is a point where the negative effects of social disorganization maximize. When implementing policy, it should not just be the very worst communities that receive services. Programs should target all communities that struggle with macro-level disorder. Attitudes toward family violence (i.e., approval of intimate partner violence and support for corporal punishment) did not differ by perceived level of collective efficacy. Collective efficacy is cultivated on the presence of comparable values and like ideals (Sampson et al. 1997). Whether residents feel positively or negatively towards violence carries little significance, as long as individuals have similar views and beliefs. As the data suggest, attitudes towards family violence, although different measures, vary by the presence of crime and disorder. Neighborhoods are cohesive in their tolerance for family violence. The data affirm that it is not the particular value, but, rather, the interconnectedness of the belief that influences social cohesion. To further the point, perceived collective efficacy did not significantly vary by crime (Low crime x=13.86 and High crime x=12.44; t=1.355; p=.089). Although, neighborhoods characterized with high crime rates tend to be more disorganized, not all violent neighborhoods lack social control. There are communities that employ the use of crime advantageously (see for example, Patillo 1998). Yet again, it is not the belief itself that ties a community; it is the consistency in which residents support the belief. Although no differences were found in support for corporal punishment, those with a history of child abuse were more likely to approve of intimate partner violence. This is consistent with previous research (Kaufman and Zigler 1986; Pears and Capaldi 2001; Ringwalt et al. 1989). The effects of child maltreatment extend well into adulthood (Kolko 2002; Polonko 2005). Adult children are socialized into a world of violence where reactions repeatedly consist of aggression. Mental health practitioners and community workers should focus on re-socializing survivors’ expectations and tolerance of violent reactions. Therapy provides an opportunity for individuals to learn healthy, realistic responses to stressful situations that could potentially escalate into violence (Boschert 2004). It is reassuring to note that both men and women have low tolerance for intimate partner violence. Gender is one of the most salient predictors of partner aggression (Payne and Gainey 2005). Likewise, the literature suggests that men maintain more accepting attitudes towards violence within relationships (Black et al. 1999; Schumacher et al. 2001). The data presented here suggest that the gender gap, regarding partner violence, may be closing. However, when examining the effects of gender on approval for corporal violence, we see that men have significantly more tolerance for the use of physical punishment than do women. Men, in general, are more likely to accept violence as an acceptable outcome than women are (Johnson and Sigler 1996). The fact that men have more support for corporal violence is not necessarily surprising (Straus 2001a). The reason that there is a significant gender difference for support for corporal punishment and not approval of intimate partner violence may be due to the covert patriarchal beliefs of our society. The role of fatherhood traditionally includes maintaining strict control of one’s offspring. Corporal violence is seen as a means to
Am J Crim Just (2008) 33:130–147
143
end in gaining youth compliance. Men may be more accepting of this form of punishment because it provides an efficient method of fulfilling their societal role (Barker and Loewenstein 1997). Race plays a significant role in forming attitudes toward intimate partner violence and corporal punishment. It is imperative that researchers accurately understand the role that race plays in promoting attitudes about family violence. Failing to understand this relationship, and overlooking the real causes of aggression within the family, could result in higher levels of victimization among minorities (Penn 2006). Previous research shows that when other factors are controlled, the relationship between race and aggression approval is spurious (Kantor et al. 1994; Straus 2001a). Future investigations may focus on what other individual-level and community-level variables mediate this relationship (i.e., income, concentrated disadvantage). Intimate Partner Violence The bivariate results of this study suggest that approval of intimate partner violence is rooted in both individual- and neighborhood-level factors. Support for corporal punishment, neighborhood crime levels, a history of child maltreatment, and race significantly contribute to the variation in approval of intimate partner violence at the bivariate level. On the other hand, there were no significant differences in approval of partner aggression for neighborhood social disorder, neighborhood collective efficacy, and gender. However, the multivariate results show that when controlling for all other factors, neighborhood crime level and a history of child maltreatment lose salience in predicting attitudes toward intimate partner violence. Only support for corporal punishment and race remain significant. This suggests that neighborhood factors, or, at least, the perceptions of neighborhoods may not be as important in determining one’s acceptance of partner abuse. Individual level factors (i.e., attitudes toward spanking and race) are more relevant. Corporal Punishment When examining the bivariate results for support for corporal punishment, we see that there are as many similarities as there are differences with acceptance of interpersonal violence. Again, attitudes toward spanking are a function of individual characteristics and perceived neighborhood factors. Support for corporal punishment, perceived neighborhood social disorder, gender, and race significantly contribute to the variation in support for corporal punishment at the bivariate level. On the other hand, there were no significant differences in tolerance of violent discipline for neighborhood crime level, collective efficacy, and a history of child maltreatment. When controlling for all other factors, the multivariate results indicate that perceived neighborhood social disorder becomes unimportant in determining attitudes toward spanking. Approval of intimate partner violence, gender, and race remain significant. Again, this implies that neighborhood factors, or at least perceptions of neighborhoods may not be as important in determining one’s support
144
Am J Crim Just (2008) 33:130–147
for corporal discipline. As with approval of intimate partner violence, individual-level factors (i.e., attitudes toward partner aggression, gender, and race) are more relevant. We see that both models affirm the importance of individual-level characteristics. What must be considered is how these factors are influenced by the macro-level. Although tolerance for family violence is an individual-level attribute, the fact that our society as a whole condones the use of violence for certain situations affects how individuals view the role of violence in a variety of circumstances. For example, because the United States mandates the use of corporal violence, individuals believe that it is an appropriate discipline strategy. Consider sibling abuse. There is no legislation to control aggression between siblings. A frequently used layperson explanation for the violence that occurs between siblings consists of “that’s just what brothers and sisters do...its normal.” If the United States government, a macro-level institution, spotlighted the deviance, despite the commonness, of violence between siblings, individual attitudes and efforts to curb this type of abuse would be different. Race remained significant for both approval of intimate partner violence and support for corporal punishment while all other factors were controlled. As stated, previous research shows that when other factors are controlled, the relationship between race and aggression approval is spurious (Kantor et al. 1994; Straus 2001a). The models did not include measures for economic disadvantage, nor did they include a measure for educational attainment. These variables, both of which are confounded with race and linked to intimate partner violence and corporal violence, may mediate the relationship between race and tolerance of family violence. It must be reiterated that researchers must accurately understand the role that race plays in promoting attitudes about family violence. Failing to understand this relationship, and overlooking the real causes of aggression within the family, could result in higher levels of victimization among minorities (Penn 2006). Gender remained significant in Model 2 with men having more support for corporal punishment. This emphasizes the importance of that gender roles play in perpetuating violence. Initiatives should emphasize that violence, regardless of sex or situation, is not tolerable. Community programs should focus on educating men in particular. For example, parenting classes that target men could afford fathers with different parenting strategies and techniques that could reduce the use of physical discipline. Although the results did not indicate the significance of community-level experiences or perceptions, they did not deny the importance of neighborhoods in our society. To end family violence, there needs to be an end in the societal toleration for the aggression that takes place between family members. A shift in attitudes needs to occur, and this shift must begin at the very top. The enactment of legal statutes prohibiting the use of spanking and other forms of family violence should be coupled with public education. Uniting micro-level and macro-level initiatives allows neighborhoods to be effective institutions in preventing family violence. Encouraging the growth of social networks and fostering violence-free value systems can lead to community-wide deviance control. Collective efficacy, initiated by a value system intolerant of aggression, could be a very powerful tool in preventing violence that occurs behind closed doors.
Am J Crim Just (2008) 33:130–147
145
Since the 1960s, policy action has increasingly portrayed the lack of tolerance for domestic violence. In contrast, corporal punishment is a widely accepted discipline strategy (note that there was significantly more support for the use of physical discipline than there was approval of intimate partner violence). While research illustrates the negative effects of corporal violence, there has been little to no statutory action to curb the use physical discipline (Straus 2001a). While individual attitudes toward violence against intimates may be changing in our society, the manner in which we view children remains the same. Our society does not afford the same rights of safety to children, and thus the value we place on children is diminished. By denying children the protection they need and deserve, they are stripped of dignity and put in a position to be victimized (Polonko 2005). Moreover, our society provides parents, those responsible for the wellbeing and security of children, the legal permission to use physical violence against children. Perhaps, the same development of programs and implementation of legislation that has effectively influenced individual attitudes, and consequently, deterred aggressive actions towards intimates should be employed for corporal punishment. This study indicates that approval for family violence is an individual-level phenomenon as well as a community-level occurrence. The results are limited in generalizability, and, as such, future research should attempt to utilize a more representative sample that includes more neighborhood-level variation. The purpose of this paper was to identify the differences in the approval of family violence by micro-level characteristics while considering and a macro-level measurement of crime. This study has brought up a number of additional research questions. Do neighborhood characteristics influence attitudes toward other forms of family violence (i.e., elder abuse or sibling violence)? Do social institutions within neighborhoods (churches, schools, businesses) mediate the relationships between attitudes toward family violence and neighborhood characteristics? Additionally, do the individual tenets of social disorganization affect attitudes toward family violence differently? Future research should too consider utilizing empowerment approaches to engage minority communities in empirical efforts. An empowerment approach would link several agencies and community officials for the specific purpose of increasing research participation.
References Almgren, G. (2005). The ecological context of interpersonal violence: From culture to collective efficacy. Journal of Interpersonal Violence, 20(2), 218–224. Ashton, V. (2001). The relationship between attitudes toward corporal punishment and the perception and reporting of child maltreatment. Child Abuse & Neglect, 25, 389–399. Ateah, C. A., Secco, M. L., & Woodgate, R. L. (2003). The risks and alternatives to physical punishment use with children. Journal of Pediatric Health Care, 17, 126–132. Barker, G., & Loewenstein, I. (1997). Where the boys are: Attitudes related to masculinity, fatherhood, and violence toward women among low-income adolescent and young adult males in Rio de Janeiro. Youth & Society, 29(2), 166–197. Benson, M. L., John, W., Thistlethwaite, A. B., & Fox, G. L. (2004). The correlation between race and domestic violence is confounded with community context. Social Problems, 51(3), 326–342. Black, D. A., Schumacher, J. A., Smith Slep, A. M.,, & Heyman, R. E. (1999). Review of partner physical aggression risk factors. http://www.nnh.org/risk/chap2_RiskFactorsforMale.html. Retrieved 12 October 2006.
146
Am J Crim Just (2008) 33:130–147
Bolton Jr., F. G. (1987). The father in the adolescent pregnancy at risk for child maltreatment. I. helpmate or hinderance? Journal of Family Violence, 2(1), 67–80. Boschert, S. (2004). Early life trauma affects chronic depression TX: Combining particular form of CBT and Nefazodone led to highest remission rates. Clinical Psychiatry News, 32(5), 1–3. Browning, C. R. (2002). The span of collective efficacy: Extending social disorganization theory to partner violence. Journal of Marriage and Family, 64(4), 833–850. Bureau of Justice Statistics. (2006). National incident-based reporting system (NIBRS) implementation program. http://www.ojp.usdoj.gov/bjs/nibrs.htm. Retrieved 18 September 2006. Bursik, R. J., & Grasmick, H. G. (1993a). Economic deprivation and neighborhood crime rates: 1960– 1980. Law & Society Review, 27(2), 263–284. Bursik, R. J., & Grasmick, H. G. (1993b). Neighborhoods and crime. New York: Lexington Books. Cazenave, N. A., & Straus, M. A. (1990). Race, class, network embeddedness, and family violence: A search for potent support systems. In M. A. Straus & R. J. Gelles (Eds.), Physical violence in American families (pp. 321–339). New Brunswick, NJ: Transaction Publishers. Chase, K. A., O’Farrell, T. J., Murphy, C. M., Fals-Stewart, W., & Murphy, M. (2003). Factors associated with partner violence among female alcoholic patients and their male partners. Journal of Studies on Alcohol, 64(1), 137–150. Coleman, K. H., Weinman, M. L., & Bartholomew, H. P. (1980). Factors affecting conjugal violence. Journal of Psychology, 105, 197–202. Gelles, R. J. (1990). Domestic criminal violence. In N. A. Weiner, M. A. Zahn, & R. J. Sagi (Eds.), Violence: Patterns, causes, and public policy (pp. 106–122). New York: Harcourt, Brace, Jovanovich. Giles-Sims, J., Straus, M. A., & Sugarman, D. A. (1995). Child, maternal, and family characteristics associated with spanking. Family Relations, 44(2), 170–176. Gracia, E., & Herrero, J. (2006). Perceived neighborhood social disorder and residents’ attitudes toward reporting child physical abuse. Child Abuse & Neglect, 30, 357–365. Hatchett, B. F., Holmes, K., Duran, D. A., & Davis, C. (2000). African Americans and research participation: The recruitment process. Journal of Black Studies, 30(5), 664–675. 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(3), 933–944. Jackson, A. P., & Ivanoff, A. (1999). Reduction of low response rates in interview surveys of poor African–American families. Journal of Social Service Research, 25(1/2), 41–60. Johnson, I., & Sigler, R. T. (1996). Public perceptions of interpersonal violence. Journal of Criminal Justice, 24(5), 419–430. Kantor, G. K., Jasinski, J. L., & Aldarondo, E. (1994). Sociocultural status and incidence of marital violence in Hispanic families. Special issue: Violence against women of color. Violence Victims, 9, 207–222. Kaufman, J., & Zigler, E. (1986). Do abused children become abusive parents? American Journal of Orthopsychiatry, 57(2), 186–193. Koch, J. V., & Cebula, R. J. (2004). The final 2000 census state response rates: Myths and realities. The Social Science Journal, 41, 575–585. Kolko, D. J. (2002). Child physical abuse. In J. E. B. Myers, L. Berliner, J. Briere, C. T. Hendrix, C. Jenny, & T. Reid (Eds.), The APSAC handbook on child maltreatment (2nd ed., pp. 3–20). Thousand Oaks, CA: Sage Publications. Leonard, K. E., & Blane, H. T. (1992). Alcohol and marital aggression in a national sample of young men. Journal of Interpersonal Violence, 7, 19–30. Macewen, K. E., & Barling, J. (1988). Multiple stressors, violence in the family of origin, and marital aggression: A longitudinal investigation. Journal of Family Violence, 3, 73–87. McKenry, P. C., Julian, T. W., & Gavazzi, S. M. (1995). Toward a biopsychosocial model of domestic violence. Journal of Marriage and Family, 57(2), 307–320. Miles-Doan, R. (1998). Violence between spouses and intimates: Does neighborhood context matter? Social Forces, 77(2), 623–645. National Criminal Justice Reference Service (2006). Family violence—facts and figures. http://www.ncjrs. gov/spotlight/family_violence/facts.html. Retrieved 30 September 2006. O’Leary, K. D., & Curley, A. D. (1986). Assertion and family violence: Correlates of spouse abuse. Journal of Marital and Family Therapy, 12, 281–289. Patillo, M. E. (1998). Sweet mothers and gangbangers: Managing crime in a Black middle-class neighborhood. Social Forces, 76(3), 747–774. Payne, B. K., & Gainey, R. R. (2005). Family violence & criminal justice: A life-course approach (2nd ed.). Cincinnati, OH: Anderson.
Am J Crim Just (2008) 33:130–147
147
Pears, K. C., & Capaldi, D. M. (2001). Intergenerational transmission of abuse: A two-generational prospective study of an at-risk sample. Child Abuse & Neglect, 25, 1439–1461. Penn, E. B. (2006). Black youth: Disproportionality and delinquency. In E. B. Penn, H. T. Green, & S. L. Gabbidon (Eds.), Race and juvenile justice (pp. 47–64). Durham, NC: Carolina Academic Press. Pinderhughes, E. E., Dodge, K. A., Bates, J. E., Pettit, G. S., & Zelli, A. (2000). Discipline responses: Influences of parents’ socioeconomic status, ethnicity, beliefs about parenting, stress, and cognitiveemotional processes. Journal of Family Psychology, 14(3), 380–400. Polonko, K. A. (2005). The causes and consequences of child abuse: Shedding more light on the cycle of violence and neglect. Unpublished paper, Old Dominion University. Ringwalt, C. L., Browne, D. C., Rosenbloom, L. E., Evans, G. A., & Kotch, J. B. (1989). Predicting adult approval of corporal punishment from childhood parenting experiences. Journal of Family Violence, 4(4), 339–343. Sampson, R. J., & Groves, W. B. (1989). Community structure and crime: Testing social disorganization theory. The American Journal of Sociology, 94(4), 774–802. Sampson, R. J., & Raudenbush, S. W. (1999). Systematic social observation of public spaces: A new look at disorder in urban neighborhoods. American Journal of Sociology, 105, 603–651. Sampson, R. J., Raudenbush, S. W., & Earls, F. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 227(5328), 918–924. Sampson, R. J., & Wilson, W. J. (1995). Toward a theory of race, crime, and urban inequality. In J. Hagan & R. D. Peterson (Eds.), Crime and inequality. Stanford, CA: Stanford University Press. Schneider, M. E. (2005). Costs of intimate partner violence (policy & practice). OB GYN, 40(23), 31. Schumacher, J. A., Feldbau-Kohn, S., Smith Slep, A. M., & Heyman, R. E. (2001). Risk factors for male-to-female partner physical abuse. Aggression and Violent Behavior, 6, 281–352. Shaw, C. R., & McKay, H. D. (1942). Juvenile delinquency and urban areas. Chicago: University of Chicago Press. Stiffman, M. N., Schnitzer, P. G., Adam, P., Kruse, R. L., & Ewigman, B. G. (2002). Household composition and risk of fatal child maltreatment. Pediatrics, 109(4), 615–622. Straus, M. A. (2000). Family violence. In E. F. Borgatta & M. L. Borgatta (Eds.), Encyclopedia of sociology (2nd ed., Vol. 2, pp. 981–987). New York: Macmillan Publishing Co. Straus, M. A. (2001a). Beating the devil out of them: Corporal punishment in American families and its effects on children (2nd ed.). New Brunswick, NJ: Transaction Publishers. Straus, M. A. (2001b). Physical aggression in the family: Prevalence rates, links to non-family violence, and implications for primary prevention of societal violence. In M. Martinez (Ed.), Prevention and control of aggression and the impact on its victims (pp. 181–200). New York: Kluwer Academic/ Plenum Publishers.Gotz. Straus, M. A., & Hamby, S. L. (1999). The personal relationships profile (PRP). Family Research Laboratory, University of New Hampshire. http://pubpages.unh.edu/~mas2/PR10M.pdf#search='prp% 20straus'. Retrieved 24 September 2006 . Straus, M. A., & Mouradian, V. E. (1999). Preliminary psychometric data for the personal and relationships profile (PRP): A multi-scale tool for clinical screening and research on partner violence. American Society of Criminology. http://pubpages.unh.edu/~mas2/PR15.pdf. Retrieved 30 September 2006. Sugarman, D. B., & Hotaling, G. T. (1989). Violent men in intimate relationships: An analysis of risk markers. Journal of Applied Social Psychology, 19, 1034–1048. Tittle, C. R. (1983). Social class and criminal behavior: A critique of the theoretical foundation. Social Forces, 62(2), 334–358. U. S. Census Bureau. n.d. Census block groups: Cartographic boundary files descriptions and metadata. http://www.census.gov/geo/www/cob/bg_metadata.html. Retrieved 18 September 2000. Van Wyk, J. A., Benson, M. L., Fox, G. L., & DeMaris, A. (2003). Detangling individual-, partner- and community-level correlates of partner violence. Crime and Delinquency, 49(3), 412–438. Veysey, B. M., & Messner, S. (1999). Further testing of social disorganization theory: An elaboration of Sampson and Groves’ ‘community structure and crime.’ Journal of Research in Crime and Delinquency, 36, 156–174. Vittrup, B., Holding, G. W., & Buck, J. (2006). Attitudes predict the use of physical punishment: A prospective study of the emergence of disciplinary practices. Pediatrics, 117(6), 2055–2064. Warner, B. D., & Rountree, W. P. (1997). Local social ties in a community and crime model: Questioning the systemic nature of informal social control. Social Problems, 44(4), 520–536.