Exploring the Role of Victim Sex,Victim Conduct,and Victim–Defendant Relationship in Capital Punishment Sentencing

Abstract

Disparities in the administration of capital punishment are a prominent social and political issue. Recent studies indicate that victim characteristics of sex and race produce interactive effects on capital-sentencing outcomes. Extending this line of research, the current analysis explores the intersection of victim sex with victim conduct and victim–defendant relationship, utilizing a population of North Carolina capital cases spanning the years 1977 to 2009 (N = 1,285). Findings indicate that cases with a female victim who was not involved in illegal activity at the time of the murder and acquaintance female victim cases are most likely to result in a death recommendation. Potential reasons for these findings are discussed.

Keywords

death penalty gender sentencing victim/offender relationship victim conduct

The influence of extra-legal factors in capital-sentencing decisions is of particular concern for social scientists, policy makers, and those involved in the criminal justice process. In recent years, examinations of sentencing and jury decision-making have increasingly focused on the role of the crime victim. Perhaps influenced by the victims’ rights movement, which campaigned primarily for the inclusion of victims in criminal justice processes (for a review see Karmen, 2010; Sebba, 1996), the importance of examining the effects of victim characteristics in sentencing outcomes has been emphasized (Paternoster & Deise, 2011). The potential for aspects of a victim’s character, appearance, or actions to influence criminal justice decision-making has been illustrated by research on influences such as victim precipitation (e.g., Gobert, 1977) and the controversial role of the victim impact statement, particularly in capital sentencing (Davis & Smith, 1994; Myers & Greene, 2004; Paternoster & Deise, 2011). A large body of literature has shown the influence of victim race on capital sentencing decisions (for a review of the literature see, Kavanaugh-Earl, Cochran, Smith, Fogel, & Bjerregaard, 2008). Victim demographic characteristics, such as age, race/ethnicity, sex, and character, are not legally relevant; yet, like defendant characteristics, there is the potential for them to influence jury decision-making.

Although the influence of victim characteristics is becoming more central to questions of sentencing impartiality, research that examines capital jury sentencing decisions and the role of these characteristics remains underdeveloped. For instance, Williams, Demuth, and Holcomb (2007) argued that the lack of interest in understanding the relationship between victim gender and the death penalty (evidenced by the traditional focus on offender characteristics) is unfortunate because this line of inquiry “not only provides insight into the decision-making process of criminal justice actors and to the relevance of particular crime characteristics associated with the imposition of death sentences, but also it provokes important questions about how societal stratification and inequality [. . .] influence public perceptions of crime seriousness and official punishment responses” (p. 866). The presence of victim sex effects, particularly in conjunction with other victim characteristics, may be evidence of gender disparities in the administration of capital punishment. This issue is explored further in the current study by examining the impact of victim characteristics on capital-sentencing decisions, specifically victim conduct at the time of the murder and victim–defendant relationship, and how these may influence the context in which victim sex matters.

Capital Sentencing and the Role of the Victim

The death penalty is one of the more polarizing criminal justice issues in social, political, and academic forums. Responding to concerns about discrimination in the death penalty’s implementation, a number of studies have focused on the association of offender characteristics, especially defendant race, with sentencing decisions (e.g., Baldus, Woodworth, & Pulaski, 1990). An equivocal finding from these studies is that the race of the defendant is only marginally related to receiving a death sentence, and is more prevalent in some jurisdictions than others (Williams & Holcomb, 2004). However, a more consistent finding to emerge is that race of the victim, specifically the presence of a White victim, is commonly a significant predictor of the defendant receiving the death penalty, particularly when there is a non-White defendant (Baldus, Woodworth, Zuckerman, Weiner, & Broffitt, 1998; Hindson, Potter, & Radelet, 2006; Paternoster, 1984; Williams & Holcomb, 2004). Four primary studies in the past decade attempted to expand this area of inquiry by exploring the interactive effects of multiple victim characteristics and will be discussed here.

Williams and Holcomb (2004) initiated this line of inquiry, arguing that “it is unlikely that decision makers consider the race or gender of a victim independent of one another” (p. 357), echoing the earlier sentiments of Steffensmeier, Ulmer, and Kramer (1998). In other words, meaningful differences may exist when multiple characteristics are examined interactively and these interactions can provide context for understanding decision maker’s considerations. Pursuing this line of reasoning, Williams and Holcomb examined interactive victim effects using Ohio Supplementary Homicide Reports (SHR) data from the years 1981 through 1994, as well as data from multiple state sources on homicides resulting in the death penalty. Their final sample consisted of 5,320 cases, 271 of which resulted in the death penalty. The findings of their analyses indicated that homicides with a White female victim were significantly more likely to result in the death penalty than cases with other victim sex–race dyads. Specifically, compared to cases with White female victims, there was a 65.8% decrease in the likelihood of receiving the death penalty for cases with a White male victim, a 62.4% decrease in the likelihood for cases with a Black female victim, and a 73.7% decrease in the likelihood for cases with a Black male victim. While Williams and Holcomb cautiously emphasized that their study be viewed as a “preliminary examination of a methodologically complex relationship” (p. 352), the findings were a catalyst for the examination of interactive victim effects in capital sentencing.

Holcomb, Williams, and Demuth (2004) replicated Williams and Holcomb’s (2004) findings, extending the same Ohio SHR data through 1997. Holcomb et al. (2004) conducted additional analyses limited to homicides with felony circumstances (i.e., death penalty eligible homicides). These analyses included partitioning the data based on each of the four race–sex dyads in order to explore differences in the factors associated with death sentences for cases with White male victims, Black male victims, Black female victims, and White female victims. While multiple factors were significantly related to the increased likelihood of a death sentence in White male victim cases (age of the offender, age of the victim, multiple victims, urban area, and victim–offender relationship) and multiple factors were related to the increased likelihood of a death sentence for Black male victim cases (age of the offender, multiple victims, urban area, and use of a gun as the weapon), the only variable that increased the likelihood of a death sentence for Black female victim cases was the presence of multiple victims. Similarly, presence of multiple victims was the only significant factor for cases with a White female victim. Based on these findings, Holcomb et al. concluded that victim sex may be more important than victim race for understanding sentencing disparities.

In a further extension of the work conducted by Williams and Holcomb (2004), Stauffer, Smith, Cochran, Fogel, and Bjerregaard (2006) sought to determine if the White female victim effect would emerge in a sample of 953 jury decisions in capital murder trials from the years 1979 through 2002 in North Carolina. In addition to including similar variables that were examined in the previously mentioned studies, Stauffer et al. (2006) modeled additional relevant variables within the sentencing literature, such as prior criminal history, aggravating factors accepted by the jury, attorney type, and victim involvement in illegal activity. When including the original variables that were examined in Williams and Holcomb, Stauffer et al. found that cases with Black male victims and cases with White male victims were significantly less likely to result in the death penalty than cases with White female victims. However, departing from Williams and Holcomb’s findings, cases with White female victims were not significantly more likely to result in the death penalty compared to cases with Black female victims. In essence, with additional control variables that might influence sentencing outcomes, the interaction effects were no longer statistically significant. The authors extended a number of cautions in comparing their results with the results of Williams & Holcomb (2004) and Holcomb et al. (2004), suggesting that both potentially could be valid within the context of their respective jurisdictions.

Interestingly, several control variables did remain significantly related to capital-sentencing outcomes in the Stauffer et al. (2006) study. Particularly relevant to the direction of the current analysis, two victim-related variables, victim involvement in illegal activity at the time of the murder and victim–defendant relationship (stranger homicide), significantly decreased the likelihood of a death sentence. The findings of Stauffer et al. suggest that while victim sex and race interactions were not significant in their data after controlling for variables common in the sentencing literature, victim sex may intersect with other factors such as victim–defendant relationship and victim conduct to produce gendered outcomes.

In an attempt to further explore the interactive effects between victim sex and victim race on jury decision-making, Williams et al. (2007) assessed variables representing sexualized victimization context using the seminal Baldus et al. (1990) Charging and Sentencing Study data. The data consist of a stratified random sample of 1,066 cases in which the defendant was indicted for murder or voluntary manslaughter and convicted in Georgia between the years 1973 and 1979. Williams et al. examined the role of victim sex with results indicating joint effects of victim sex and race. More specific, cases with White female victims were treated the most harsh, while cases with Black male victims received leniency. This finding was similar to the findings of Williams and Holcomb (2004) and Holcomb et al. (2004) in that White female victim and Black male victim cases were the most divergent in regards to sentencing outcomes. However, Williams et al. (2007) emphasized that Black male victim cases were uniquely different from White female victim cases. Stated differently, while defendants who murdered a Black male were less likely to receive a death sentence than any comparison group, the outcomes of cases involving a White female or White male victim were not significantly different.

Williams et al. (2007) also examined variables representing characteristics of the victimization in order to try to explain harsher punishments for female victim cases. They found that three victimization variables—rape, forcing the victim to disrobe, and killing an unclothed victim—greatly reduced the impact of the victim sex effect. Cases involving sex-related victimization variables were the most likely to result in a death sentence, particularly cases where the victim was forced to disrobe (surprisingly, more so than defendant rape of the victim). Williams et al. concluded that the sexual nature, and particularly sexual humiliation, of female victimization is an important factor in understanding gendered effects in capital sentencing.

Taken together, these recent examinations of victim characteristics indicate that victim sex has a probable influence in capital-sentencing decisions. Specifically, cases with female victims are more likely to result in assessments of the death penalty. These studies have contributed to our understanding of the role of the victim in capital sentencing and have initiated a new avenue of research. However, the emphasis has largely remained on the intersection of victim sex and race. Although the focus on explaining racial disparities in capital sentencing is necessary, especially in light of our nation’s turbulent history with race relations, the interaction between victim race and sex does not fully account for disparities in capital sentencing. The findings of Williams et al. (2007) regarding sexual victimization illustrate the utility of investigating gendered influences in capital-sentencing studies. Understanding how victim sex may operate as a stratification device beyond its potential interaction with race can help shed light on criminal justice decision-making as well as society’s gendered notions of punishment. Specifically, research on the role of victim conduct and victim–defendant relationship has indicated that these variables may influence sentencing outcomes and have the potential to intersect with victim sex producing gender disparities. The argument made in recent sentencing literature is that violent crimes against females may be viewed as more heinous than crimes against males, and that female victims may be seen as less able to protect themselves and thus their attackers viewed as more blameworthy for the victimization (Baumer, Messner, & Felson, 2000; Curry, Lee, & Rodriguez, 2004; Franklin & Fearn, 2008; Williams & Holcomb, 2004). Blameworthiness, and its potentially gendered interpretations, may also be associated with victim conduct and victim–defendant relationship.

Victim Sex and Victim Conduct

Prior research on the role of the victim in sentencing has indicated that victim conduct at the time of the crime can influence sentencing outcomes (Baumer et al., 2000; Rye, Greatrix, & Enright, 2006; Spohn & Spears, 1996; Sundby, 2003). If the victim was engaged in illegal conduct immediately prior to victimization, jury members may view the offender as less blameworthy than an offender who perpetrates a crime against an “innocent” victim (Baumer et al., 2000; Rye et al., 2006; Sundby, 2003). Research examining the extent to which jurors draw distinctions between “worthy” and “unworthy” victims indicates that jurors do consider such factors in the process of determining the defendant’s outcome (Sundby, 2003). Findings from the Capital Jury Project show that jurors are more inclined to assign death sentences in cases that have what are perceived to be “innocent” victims. Thus, jurors may make qualitative distinctions between victims who were in the “wrong place at the wrong time” and victims who are construed as having precipitated their victimization (Sundby, 2003). Furthermore, killings of disreputable or stigmatized victims have been found to result in more lenient punishments for their perpetrators (Baumer et al., 2000). For example, Sundby (2003) found that juries recommending life sentences were more likely to discuss victim characteristics during their deliberations than juries recommending death sentences. Since stigmatization based on behavior may be influenced by gender expectations, it is predicted that interactions between victim sex and victim conduct will result in gendered outcomes. Theoretical connections have been made between victim conduct and victim sex using interpretations of chivalry, attribution, and focal concerns theories (Baumer et al., 2000; Curry et al., 2004; Franklin & Fearn, 2008; Williams & Holcomb, 2004). Victim conduct, such as involvement in illegal activity, may mitigate blame placed on the perpetrator resulting in lighter sentences. However, it is not firmly established from previous research whether this impact is more pronounced in cases involving male versus female victims.

Victim Sex and Victim–Defendant Relationship

Research examining victim–defendant relationship and sentencing has produced divergent findings. Some studies have indicated that stranger offenders receive longer sentences or are punished more harshly than nonstranger offenders (Gross & Mauro, 1984; Simon, 1996a) while others have found that victim–offender relationship is not associated with increased sentence length (Simon, 1996b). Yet others have concluded that the relationship is more nuanced, depending on the degree of intimacy between victim and offender (Dawson, 2004). In support of the former perspective, qualitative research has indicated that capital jurors are more likely to sympathize with victims of stranger violence and react more harshly in these cases. As Sundby (2003) speculates, “an individual who preys upon randomly chosen victims poses the starkest image of the dangerous individual, and future dangerousness consistently has emerged as one of the strongest factors for predicting a death sentence” (p. 359). Thus, jurors may identify with victims who were “in the wrong place at the wrong time” because they can put themselves in the victim’s place. In support of the latter perspective, Dawson (2004) concluded that perceptions of the victim–offender relationship are more nuanced than a simple stranger/nonstranger dichotomy, and are influenced by the degree of intimacy between victim and offender. Dawson believed that the degree of intimacy between victim and offender may impact juror considerations of the victim’s role in the crime which in turn may color the perceived blameworthiness of the offender. Thus, as the degree of intimacy between the victim and offender increases, perceptions of increased emotionality involved in the crime may reduce the defendant’s culpability (less premeditation or intent) resulting in less severe sentences.

The Current Study

Using the findings from prior research as a guide, the purpose of the current study is twofold. First, the presence of a female victim effect will be determined, keeping in mind the consistent prior findings concerning a main effect of victim sex. Second, the context in which victim sex matters in capital-sentencing decisions will be explored by testing interaction effects between victim sex and two additional potentially gendered variables, victim conduct and victim–defendant relationship. These analyses should determine if there are additional gender disparities in capital punishment sentencing recommendations not detected in earlier studies.

Based on the findings of Stauffer et al. (2006) it is expected that cases with female victims will be more likely to result in a recommendation of death than cases with male victims. In addition, it is expected that a defendant who murdered a victim not involved in illegal activity would be more likely to receive a death sentence than a defendant who murdered a victim involved in illegal activity. Thus, cases with a female victim not involved in illegal activity are predicted to be the most likely to result in a jury recommendation of death. Therefore, it is hypothesized that cases involving female victims who were not involved in illegal activity at the time of the incident will be more likely to result in the death penalty than (a) cases involving a female victims involved in illegal activities at the time of the incidence, (b) cases with a male victim who was involved in illegal activity, or (c) cases with a male victim who was not involved in illegal activity at the time of the incident.

In terms victim–defendant relationship, it is predicted that as the degree of intimacy between victim and offender decreases, the likelihood of a jury recommending death will increase. Since it is predicted that female victim cases will be more likely to result in death penalty recommendations, the hypothesis regarding the interaction between victim sex and victim–defendant relationship is as follows: cases involving a female victim whose perpetrator was less intimately related will be more likely to result in the death penalty than (a) cases with a female victim whose perpetrator was more intimately related, (b) cases with a male victim whose perpetrator was less intimately related, or (c) cases with a male victim whose perpetrator was more intimately related.

Method

Data

The data utilized in this study are from the North Carolina Capital Sentencing Project. Cases included in this dataset are homicide cases in which (a) a first-degree murder conviction was secured, (b) the state sought the death penalty, and (c) the trial advanced to the sentencing phase whereby the jury recommended either a life sentence or the death penalty. The data used for the current analyses are inclusive of all cases meeting the described criteria for a 32-year period between June 1977 and December 2009, and constitutes a population of 1,356 cases during this era. The initial date marks the beginning of approved capital punishment statutes in North Carolina following the U.S. Supreme Court decision Gregg v. Georgia (1976), and the latter represents the last year for which full information is available. Information about each case was derived from reviews of trial documents contained in North Carolina Supreme Court and Court of Appeals cases and/or from public records obtained from the counties in which the trials were held. Within the dataset, incidents involving multiple offenders tried for the murder of one victim are treated as separate cases; likewise, instances involving one offender and multiple victims are treated as separate cases. In addition to detailed information about the circumstances of the homicide, the data also contain significant information about specific legal factors, including aggravating and mitigating circumstances submitted to and accepted by the jury in each of the cases.¹

Of these, 1,356 cases, 71 did not have complete information regarding aggravating circumstances.² The current study therefore draws on a sample of 1,285 jury decisions from trials in which a complete set of information is available to conduct the multivariate analyses necessary to explore the research questions posed in this study.

Measures

Dependent Variable

In North Carolina capital jurors have only two sentencing options, life in prison (since 1994, without the possibility of parole) or the death penalty. Therefore, the dependent variable of sentencing outcome (jury recommendation) is dichotomous, where 0 = a life sentence and 1 = a death sentence. Although termed a recommendation, jury decisions are binding unless found to be negated by judicial error. Jury recommendations must be unanimous; failure to achieve unanimity results in a default sentence of life in prison.

Independent Variables

The primary focus of this research is to identify the context(s) in which victim sex may have a gendered influence on capital-sentencing outcomes. Thus, the independent variables of interest are victim sex, victim involvement in illegal activity, and victim–defendant relationship. The victim involvement in illegal activity measure was created based on readings of appeals narratives and newspaper coverage that referenced the victim as being engaged in some illegal act at the time of the murder that could have been viewed as contributory or as increasing personal vulnerability (Stauffer et al., 2006). Victim illegal conduct is captured using a dichotomous variable indicating victim illegal activity at the time of the murder (15.4%) or no illegal activity at the time of the murder (84.6%).

Recent examinations of the role of victim–offender relationship in violent crime sentencing has concluded that the relationship between intimacy and criminal justice outcomes are more fluid than traditionally hypothesized, varying across time and situation (Auerhahn, 2007; Dawson, 2004). Therefore, we believed it was prudent to examine relationship in a more nuanced fashion than the traditional stranger/nonstranger dichotomy. Our analysis utilizes a measure of victim–defendant relationship that is categorized as intimate partner/family (21.1%), friends (17.9%), acquaintances (25.9%), and strangers (35.1%). The first category includes current or past intimate relationships (married or unmarried), as well as individuals with a familial relationship; friends includes nonfamilial relationships that were described as closer in nature than casual acquaintances (e.g., roommates); acquaintances were described as those having at least a passing knowledge of one another; and strangers consist of victims and perpetrators who were determined to not have had any prior knowledge or association with one another. While prior research focusing on sentencing and degree of intimacy has used a five-category measure (e.g., Dawson, 2004) the current analysis combines the intimate partner and family category for the purposes of statistical power. Stranger relationship serves as the reference category in the analysis.

Additional control variables were selected based on their use in previous capital punishment sentencing studies to examine the interaction of multiple victim characteristics on capital-sentencing decisions (see especially Stauffer et al., 2006). These include demographic characteristics of victims and offenders (sex, race, age), other extra-legal factors that could impact sentencing decisions (county urbanicity, attorney type, number of victims, weapon type), and legally relevant control variables (most notably, aggravating factors accepted by the jury). Demographic information about offenders was obtained from the North Carolina Department of Corrections website while information about victims was provided by the North Carolina medical examiner’s office or through use of a commercial CD, North Carolina Vital Records: Deaths 1968-1996 (Ancestry.com, 2000). Dichotomous measures of defendant sex (male = 96.1%, female = 3.9%), defendant race (White = 42.6%, non-White = 57.4%), and victim race (White = 62.6%, non-White = 37.4%) are included, as well as continuous measures of victim age (M = 38.84) and defendant age (M = 28.49).

Measures of additional extra-legal factors were derived from various sources. Counties were classified as urban or rural based on the N.C. Rural Economic Development Center, Inc. (2012) categories and indicate whether the capital trial was conducted in one of 15 urban counties (47.3% of trials) or one of 85 rural counties (52.7%). The type of attorney representing the defendant at trial (private = 6.2%; public = 93.8%) was identified from the case record. The multiple victim variable indicates whether there was a single homicide victim (65.4%) or more than one victim (34.6%) and is included due to its importance in prior research (e.g., Holcomb et al., 2004). Whether the victim died as the result of a gunshot(s) (58.4%) or as the result of another method such as stabbing, asphyxiation, bludgeoning (41.6%) is also included to remain consistent with prior capital sentencing studies.

Finally, the effects of three legally relevant variables serve as controls. The total number of aggravators accepted refers to the number of aggravators that were accepted by the jury as indicated per their responses on the Issues and Recommendations as to Punishment form (range = 0-8; M = 1.88; see endnote 1). Prior record of the defendant was determined by whether the state submitted an aggravating factor indicating the defendant had a prior violent or capital conviction (yes = 29.3%, no = 70.7%). Finally, given the gendered nature of rape, a jury’s acceptance of rape as an aggravating factor is included as a separate measure from the total aggravators accepted (and is not included in the measure of total aggravators accepted). Rape was accepted as an aggravator in 6.8% (n = 88) of cases and not accepted and/or not submitted in the remaining 93.2% (n = 1,197).

To assist readers in navigating the scope of the analysis, a list of variables utilized and their distributions among the sample is shown in Table 1. Although not directly pertinent to this study, the proportion of death sentences associated with those variables has been included as additional information for readers.

Table 1.

Descriptive Statistics of Variables (N = 1,285).

Variable	n (%)	Death sentence* n (%)
Sentence
Life	711 (55.3)
Death	574 (44.7)
Victim sex
Male	750 (58.4)	299 (39.9)
Female	535 (41.6)	275 (51.4)
Victim illegal activity
No	1,087 (84.6)	507 (46.6)
Yes	198 (15.4)	67 (33.8)
Victim–defendant relationship
Intimate, family	271 (21.1)	116 (42.8)
Friend	230 (17.9)	91 (39.6)
Acquaintance	333 (25.9)	166 (49.8)
Stranger	451 (35.1)	201 (44.6)
Defendant sex
Male	1,235 (96.1)	560 (45.3)
Female	50 (3.9)	14 (28.0)
Defendant race
White	548 (42.6)	253 (46.2)
Non-White	737 (57.4)	321 (43.6)
Victim race
White	804 (62.6)	387 (48.1)
Non-White	481 (37.4)	187 (38.9)
Homicide in urban area
Nonurban	677 (52.7)	324 (47.9)
Urban	608 (47.3)	250 (41.1)
Attorney type
Public defender	1,205 (93.8)	551 (45.7)
Private attorney	80 (6.2)	23 (28.8)
Multiple victim homicide
No	840 (65.4)	352 (41.9)
Yes	445 (34.6)	222 (49.9)
Weapon
Other	534 (41.6)	272 (50.9)
Gun	751 (58.4)	302 (40.2)
Defendant prior record
No prior record	909 (70.7)	368 (40.5)
Prior record	376 (29.3)	206 (54.8)
Rape aggravator accepted
No	1,197 (93.2)	505 (42.2)
Yes	88 (6.8)	69 (78.4)

Note: Defendant age: M = 28.49. Range = 15 to 77. Victim age: M = 38.84. Range = 0 to 100. Total number of aggravators accepted: M = 1.88. Range = 0 to 8.

n is the number of death sentences recommended within that category; the percentage shown is calculated as the number of death sentences recommended divided by the n for that category.

Analytic Procedure

Logistic regression analysis is used to examine the interactive effects of victim sex with victim conduct and victim sex with victim–defendant relationship, controlling for relevant independent variables. Interactions can help explain when certain independent variables have an effect on the outcome of interest. Modeling interaction effects in logistic regression is the most appropriate way to examine the hypotheses given that the focus of this investigation is whether the effect of certain independent variables (victim conduct, victim–defendant relationship) on a binary outcome (jury recommendation) differs based on a third variable (victim sex).

Results

The main effects of the independent and control variables on jury sentence recommendations are presented in Table 2. Notably, the relationship between victim sex and jury sentence recommendation is not significant when controlling for other factors. (The victim sex effect [or lack thereof] will be explored further by the analyses and discussion below associated with Table 5.) However, there is a significant main effect of victim involvement in illegal activity indicating a 48.3% decrease in the odds of the jury recommending death when the victim was involved in illegal activity compared to cases where the victim was not involved in illegal activity (B = −0.661, p < .001). Turning to type of relationship, offenders who were family and friends are not significantly different from offenders who are strangers in regards to sentencing recommendations. However, there is a 38.6% increase in the odds of the jury recommending death in cases where the victim and offender were acquaintances (B = 0.313, p = .061) compared to stranger cases. In addition, several of the control variables exhibited statistically significant main effects (i.e., defendant sex, victim race, victim age, defendant age, county rurality, type of attorney, total aggravators accepted, and rape accepted as an aggravator) in the expected directions based on prior studies including the same or similar controls (Stauffer et al., 2006; Williams & Holcomb, 2004).

Table 2.

Logistic Regression Analysis Testing for Main Effects (N = 1,285).

Variable	B	SE	Odds ratio
Female victim	0.142	0.149	1.153
Victim illegal activity	−0.661***	0.194	0.517
Intimate, Family	−0.042	0.206	0.959
Friend	−0.117	0.197	0.890
Acquaintance	0.313^†	0.168	1.368
Male defendant	0.856*	0.348	2.354
Non-White victim	−0.421**	0.157	0.656
Non-White defendant	−0.011	0.150	0.989
Victim age	−0.010**	0.003	0.990
Defendant age	0.040***	0.008	1.041
Urban county	−0.280*	0.128	0.756
Private attorney	−0.652*	0.282	0.521
Multiple victims	0.091	0.143	1.095
Gun used as weapon	−0.245	0.145	0.782
Prior record	0.022	0.153	1.023
Total aggravators accepted	0.643***	0.068	1.903
Rape aggravator accepted	1.365***	0.300	3.917
Constant	−2.660	0.480	0.070

†

p = .061. *p < .05. **p < .01. ***p ≤ .001.

Note: χ² = 267.052 (df = 17, p < .001), R² = .251, –2 log likelihood = 1,499.702.

The logistic regression results that examine the influence of victim sex in the context of victim conduct are presented in Table 3. The results indicate a 56.4% decrease in the odds of the jury recommending death in cases with a male victim involved in illegal activity (B = −0.830, p < .001), compared to female victim cases where the victim was not involved in illegal activity. The remaining interaction effects are not statistically significant. These findings indicate that, in particular, there is a distinction between female victims who are not involved in illegal conduct prior to their victimization compared to male victims who are involved in illegal conduct prior to their victimization.

Table 3.

Logistic Regression Analysis Testing for Effects of Victim Sex and Conduct Interactions (N = 1,285).

Variable	B	SE	Odds Ratio
FV*VI	−0.545	0.348	0.580
MV*VI	−0.830***	0.247	0.436
MV*VNI	−0.124	0.156	0.884
Intimate, Family	−0.036	0.206	0.965
Friend	−0.116	0.197	0.890
Acquaintance	0.314^†	0.168	1.368
Male defendant	0.860*	0.348	2.363
Non-White victim	−0.419**	0.158	0.658
Non-White defendant	−0.012	0.150	0.988
Victim age	−0.010**	0.003	0.990
Defendant age	0.040***	0.008	1.041
Urban county	−0.281*	0.128	0.755
Private attorney	−0.651*	0.282	0.521
Multiple victims	0.091	0.143	1.095
Gun used as weapon	−0.242	0.145	0.785
Prior record	0.019	0.153	1.020
Total aggravators accepted	0.644***	0.068	1.904
Rape aggravator accepted	1.362***	0.300	3.903
Constant	−2.534	0.495	0.079

Note: FV = female victim, MV = male victim, VI = Victim involved in illegal activity, VNI = Victim not involved in illegal activity; χ² = 267.212 (df = 18, p < .001), R² = .251, –2 log likelihood = 1,499.542.

†

p = .061, *p < .05, **p < .01, ***p ≤ .001.

The results that examine the influence of victim sex in the context of the relationship between victim and defendant are presented in Table 4. The findings indicate a 97.9% increase in the odds of the jury recommending death in cases with a female victim who was an acquaintance of the defendant (B = 0.682, p < .05), compared to female victim cases with a stranger perpetrator. The remaining victim sex-relationship interactions were not significantly related to the jury’s sentence recommendation. This finding illustrates the importance of examining victim–defendant relationship beyond a simple stranger/nonstranger dichotomy. Furthermore, it raises questions about the differences between acquaintance and stranger capital cases involving female victims. In both Table 3 and Table 4 the same eight control variables that exhibited significance in the main effects model remained significantly related to sentencing recommendations.

Table 4.

Logistic Regression Analysis Testing for Effects of Sex and Relationship Interactions (N = 1,285).

Variable	B	SE	Odds ratio
FV*Family	−0.154	0.259	0.857
FV*Friend	−0.082	0.336	0.921
FV*Acquaintance	0.682*	0.288	1.979
MV*Family	0.329	0.359	1.390
MV*Friend	−0.239	0.273	0.787
MV*Acquaintance	0.030	0.249	1.031
MV*Stranger	−0.112	0.237	0.894
Victim illegal activity	−0.658***	0.195	0.518
Male defendant	1.120**	0.371	3.064
Non-White victim	−0.386*	0.158	0.680
Non-White defendant	−0.006	0.151	0.994
Victim age	−0.011***	0.003	0.989
Defendant age	0.041***	0.008	1.042
Urban county	−0.296*	0.129	0.743
Private attorney	−0.671*	0.282	0.511
Multiple victims	0.066	0.144	1.068
Gun used as weapon	−0.206	0.146	0.814
Prior record	0.020	0.153	1.020
Total aggravators accepted	0.653***	0.069	1.921
Rape aggravator accepted	1.294***	0.305	3.648
Constant	−2.850	0.522	0.058

Note: χ² = 273.963 (df = 20, p < .001), R² = .257, -2 log likelihood = 1,492.791.

p < .05. **p < .01. ***p ≤ .001.

The influence of controlling for aggravating factors on potentially gendered variables in capital sentencing decision-making is further explored in Table 5. As indicated in Table 2, victim sex is not significantly related to capital-sentencing recommendations. Given the findings of Williams et al. (2007) regarding the potential for the sexual nature of some (female) victimization to mediate the relationship between victim sex and capital-sentencing outcome, it is important to examine the impact of rape accepted as an aggravating factor on victim sex.³ Table 5 presents three nested models which illustrate the impact of including the total aggravators accepted and rape accepted as an aggravating factor on the independent variables. Model 1 includes all control variables from Table 2, except for total aggravators accepted and rape accepted as an aggravator. The results of this model indicate that there is a 44.7% increase in the odds of the jury recommending death for cases with a female victim (B = 0.370, p < .01) compared to cases with a male victim. Model 2 contains the same variables as Model 1 with the addition of total aggravators accepted. As illustrated, the female victim effect is reduced but still significant, indicating a 39.6% increase in the odds of the jury recommending death in cases with a female victim (B = 0.333, p < .05). Finally, Model 3 contains Model 2 variables as well as rape accepted as an aggravating factor. Once rape is controlled for in the model there is no longer a significant female victim effect.⁴

Table 5.

Logistic Regression Nested Models Illustrating Impact of Including Total Aggravators Accepted and Rape Aggravator Accepted on Victim Sex, Victim Involvement in Illegal Activity, and Victim–Defendant Relationship (N = 1,285).

Variable	Model 1: Base model without aggravators accepted [Odds ratio coefficient (SE)]	Model 2: Model 1 with total aggravators accepted [Odds ratio coefficient (SE)]	Model 3: Model 2 with rape aggravator accepted [Odds ratio coefficient (SE)]
Female victim	1.447**	1.396*	1.153
	0.370	0.333	0.142
	(0.137)	(0.143)	(0.149)
Victim illegal activity	0.519***	0.519***	0.517***
	−0.657	−0.655	−0.661
	(0.182)	(0.191)	(0.194)
Intimate, Family	0.586**	0.777	0.959
	−0.534	−0.252	−0.042
	(0.191)	(0.200)	(0.206)
Friend	0.782	0.852	0.890
	−0.246	−0.160	−0.117
	(0.184)	(0.194)	(0.197)
Acquaintance	1.256	1.342^†	1.368^††
	0.228	0.294	0.313
	(0.159)	(0.166)	(0.168)

Note: Model 1 includes all of the control variables from Table 2 with the exception of total aggravators accepted and rape accepted as an aggravating factor. Model 2 adds total aggravators accepted in addition to the variables included in Model 1. Model 3 contains both total aggravators accepted and rape accepted as an aggravating factor, in addition to the variables included in Model 1.

†

p = .075. ^††p = .061. *p < .05. **p < .01. ***p ≤ .001.

In addition, the three models in Table 5 indicate the influence of controlling for aggravating factors on victim conduct and victim–defendant relationship: victim involvement in illegal activity significantly reduces the odds of the jury recommending death regardless of whether aggravating factors are controlled; and in terms of victim–defendant relationship, the odds of the jury recommending death for intimate/family relationship cases is significantly reduced if aggravating factors are not controlled, and the influence of acquaintance relationship on the odds of recommending death is only significant once aggravating factors are controlled. The implications of these findings are discussed in detail below.

Discussion and Conclusion

This study builds on the extant research in capital punishment sentencing disparities by exploring the context in which victim sex may have a gendered influence on capital-sentencing recommendations. Several important findings emerged from these results. First, and initially the most unexpected finding, was the lack of a significant main effect of victim sex. Prior research examining victim sex and victim race had found consistent evidence of a female victim effect (Holcomb et al. 2004; Williams & Holcomb, 2004; Williams et al., 2007). Using the same dataset as the current study, Stauffer et al. (2006) found a significant main effect of victim sex. However, closer examination reveals that Stauffer et al.’s main effects model only includes those variables replicated from Williams and Holcomb (2004), and not the additional control variables (e.g., total aggravators accepted, rape accepted as an aggravating factor) that were utilized in their interaction effects analyses. Thus, it was speculated that certain factors, specifically rape accepted as an aggravator, may account for the lack of an individual female victim effect.

This hypothesis was examined through additional analyses presented in Table 5. Results indicate that victim sex does exert a significant effect on the odds of the jury recommending death when rape accepted as an aggravating factor is not controlled for in analyses. It is important to note that all of the cases in which rape was accepted as an aggravator involved a female victim and a male defendant. In other words the rape aggravator, which is the strongest predictor of a capital sentence in all models, is representative of gendered victimization. Based on the findings in Table 2, one might argue that prior studies omitting aggravating factors were misspecified and that the “female victim effect” was an artifact of those models. However, the results of Table 5 indicate that the relationship between victim sex and capital-sentencing recommendations is mediated by controlling for acceptance of the rape aggravator.

Given the consistency of an independent female victim effect in prior research (Holcomb et al., 2004; Stauffer et al., 2006; Williams et al., 2007; Williams & Holcomb, 2004), it was expected that potentially gendered variables might moderate a victim sex effect. However, the results reported here indicate that victim sex moderates the effect of victim conduct and victim–defendant relationship on sentencing recommendations. In certain contexts victim sex does intersect with these characteristics, indicating disparities in the odds of receiving capital-sentencing recommendations for defendants that murder certain victims. Specifically, findings illustrate that cases with a male victim who was involved in illegal activity have significantly lower odds of resulting in a death sentence recommendation, compared to cases with a female victim who was not involved in illegal activity. In other words, a sentencing disparity is evident between the two most theoretically distinct dyads. This finding lends some support to the application of chivalry or focal concerns to understanding gendered sentencing disparities (Baumer et al., 2000; Curry et al., 2004; Franklin & Fearn, 2008; Williams et al., 2007; Williams & Holcomb, 2004). Based on these theoretical contexts, we would expect that females with unquestionable conduct would be extended the privilege of chivalry (harsher punishment for their perpetrators) or that their perpetrators would be seen as the most blameworthy. It would also be expected that the perpetrators of male victims engaged in questionable conduct might be seen as less blameworthy and therefore may receive less severe punishments. Generally, the significance of the victim conduct variable supports the findings of Baumer et al. (2000) and Sundby (2003) regarding the importance of victim behavior in influencing sentencing decisions. Evidence that victim conduct matters more for certain sex-conduct dyads is indicative of potential gendered disparities in capital punishment sentencing recommendations.

Victim–defendant relationship has previously evinced a significant effect on capital-sentencing outcomes, but the direction of the relationship (measured dichotomously) has varied. Stauffer et al. (2006) indicated that cases where the victim and offender were strangers were less likely to result in a death penalty recommendation while Williams and Holcomb (2004) and Holcomb et al. (2004) found that stranger cases were more likely to result in the death penalty. The dichotomization of the relationship variable may have contributed to differences in directionality. As mentioned earlier, Dawson (2004) asserted that dichotomizing the measure of victim–offender relationship (i.e., stranger and nonstranger) makes the assumption that all cases in the two categories are treated similarly when in actuality there is variation in outcomes even within intimate and nonintimate categories. The current study improves upon the assessment of victim–defendant relationship in capital sentencing by using a measure of relationship with four categories based on degree of intimacy. This measure provides a more nuanced understanding of how certain victim–defendant relationships operate, particularly with victim sex. Supporting Dawson’s (2004) conclusion that the effect of intimacy on criminal justice decision-making may have attenuated over time, the current study finds no significant differences between stranger cases and intimate/family or friend relationship cases. However, acquaintance relationship cases were significantly more likely to result in a recommendation of death compared to stranger relationship cases, but only for female victims.

Prior research has also found a distinction between acquaintance and stranger cases indicating that at the charging phase acquaintance killers were more likely to be charged with first-degree murder than those who killed strangers (Dawson, 2004). While all defendants in the current study’s analyses were charged with first-degree murder, it may be that certain characteristics increase the likelihood of harsher treatment of acquaintance cases at several stages in the criminal justice process. Cursory examination of this data indicates that cases with a victim involved in illegal activity are most heavily concentrated among acquaintance relationship cases (38.4%, n = 76). Thus, victim conduct may be influencing the findings with regard to victim–defendant relationship. In light of these results, further examination of the differences between acquaintance cases and stranger cases may prove useful in understanding sentencing disparities.

The current study contributes to research exploring the context(s) in which victim sex matters in influencing capital punishment sentencing recommendations. The finding that the victim sex effect is explained by controlling for the rape aggravator supports the argument made by Williams et al. (2007) that the sexualized nature of some (primarily female) murders may have important implications for capital punishment sentencing. Exploring the role of victim sex also assists in our understanding of the context in which other victim characteristics, namely victim conduct and victim–defendant relationship, influence capital-sentencing decisions. Victim characteristics can influence sentencing decisions in a gendered fashion, such as recommending harsher penalties (death) for the perpetrators in cases with female victims not engaged in illegal conduct compared to the perpetrators in cases with male victims engaged in illegal conduct. Future research should consider the theoretical implications of gendered disparities in the implementation of capital punishment.

Limitations and Future Research

The current study is not without limitations. First, the study was limited in some ways by restrictions inherent to secondary data analysis. In regards to victim conduct, the data utilized offered a variable that indicated whether the victim was involved in illegal activity at the time of the victim’s murder. While this measure is a cogent representation of victim precipitation or “mitigating conduct,” the emphasis placed on the victim’s behavior during the trial process is not known. In other words, the presence of knowledge about victim activities does not necessarily translate to jury discussion or emphasis of this factor. Furthermore, this variable refers to victim conduct at the time of the homicide event but not victim conduct or characteristics that may have been discussed during trial (e.g., was the victim a loving and caring stay-at-home mother or a drug abuser with a criminal record). However, the findings of Baumer et al. (2000) indicate that conduct at the time of the homicide may be more important than past conduct. The specific effect of victim portrayal on juror decision-making may be difficult to discern but is vital for policy implications regarding identification of extra-legal factors in capital sentencing.

A second limitation inherent to the topic of focus is the limited sample size of some of the interactive pairs. Keeping this in mind, the current study is intended to provide an exploratory analysis into victim and case characteristics that may be gendered. While larger samples would improve confidence in the results, the current findings are still viewed as an important contribution to the literature due to the fact that the data constitute a statewide population of capital-sentencing decisions during a 30-year span. Furthermore, the data contain vitally important control variables that are often not readily available in studies of murder and capital sentencing.

Future research should attempt to replicate these findings with data from different states in addition to data that provide information on stages prior to the sentencing phase. The interactive effects of victim characteristics may have differential impacts at various stages of the criminal justice process. Furthermore, future research aimed at understanding or establishing sex effects in capital sentencing should remain cognizant of the role of gendered victimization (e.g., rape) and legally relevant variables (e.g., aggravating factors) as these factors may reduce or account for sex effects at the sentencing stage.

The processing of capital cases has evolved over time in response to legal, ethical, and constitutional concerns. The most prominent of these concerns is rooted in social inequality and has raised questions about equity in the administration of capital punishment. The relatively recent emphasis on understanding the impact of the victim in capital-sentencing decisions is rooted in the broader debate pertaining to the legitimacy of capital punishment in our society.

Victim–defendant relationship and victim conduct are extra-legal factors that can be influenced by gendered notions and expectations and thus may be particularly pertinent in explaining sex disparities in sentencing decisions. Using comprehensive statewide data the current analysis adds to an understanding of the role that the victim plays in capital sentencing by highlighting factors that may contribute to the previously identified female victim effect as well as identifying sentencing disparities that may be gendered. These findings support further exploration of the multidimensional influence of victim characteristics and behavior in future capital-sentencing research.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Lane Kirkland Gillespie is an Assistant Professor in the Department of Criminal Justice at Boise State University. Her research interests include intimate partner homicide, rural domestic abuse, corporate victimization, and victim effects in criminal justice processing. Her recent published work appears in Feminist Criminology and Violence Against Women.

Thomas A. Loughran is an Associate Professor in the Department of Criminology and Criminal Justice at the University of Maryland. His research interests include deterrence and offender decision-making, individuals’ responses to sanctioning, and methods to infer treatment effects from non-experimental data.

M. Dwayne Smith is a Professor in the Department of Criminology at the University of South Florida where he currently serves as the institution’s Senior Vice Provost. He has published numerous articles and book chapters on crime and criminal justice topics, and has co-edited two books, Homicide: A Sourcebook of Social Research and Studying and Preventing Homicide: Issues and Challenges.

Sondra J. Fogel is an Associate Professor of Social Work at the University of South Florida and Director of Special Programs for the Honors College. She earned her PhD at the University of Illinois Urbana-Champaign.Her research interests include poverty, homelessness, and capital punishment.

Beth Bjerregaard is a Professor and Chair in the Department of Criminal Justice at the University of North Carolina at Charlotte. She received her Ph.D. in Criminal Justice from the State University of New York at Albany. Her research interests include the capital punishment, gang membership, and delinquency among other things. Her most recent publications have appeared in Justice Quarterly, Crime and Delinquency, Homicide Studies, Police Quarterly, and the Journal of Criminal Justice Education.

References

Ancestry.com. (2000). North Carolina vital records: Deaths 1968-1996. [CD-ROM]. Provo, UT: MyFamily.com, Inc.

Auerhahn

(2007). Adjudication outcomes in intimate and non-intimate homicides. Homicide Studies, 11, 213-230.

Baldus

D. C.

Woodworth

Pulaski

C. A.

(1990). Equal justice and the death penalty: A legal and empirical analysis. Boston, MA: Northeastern University Press.

Baldus

D. C.

Woodworth

Zuckerman

Weiner

N. A.

Broffitt

(1998). Racial discrimination and the death penalty in the post-Furman era: An empirical and legal overview, with recent findings from Philadelphia. Cornell Law Review, 83, 1638-1770.

Baumer

E. P.

Messner

S. F.

Felson

R. B.

(2000). The role of victim characteristics in the disposition of murder cases. Justice Quarterly, 17, 281-307.

Curry

T. R.

Lee

Rodriguez

S. F.

(2004). Does victim gender increase sentence severity? Further explorations of gender dynamics and sentencing outcomes. Crime & Delinquency, 50, 319-343.

Davis

R. C.

Smith

B. E.

(1994). The effects of victim impact statements on sentencing decisions: A test in an urban setting. Justice Quarterly, 11, 453-469.

Dawson

(2004). Rethinking the boundaries of intimacy at the end of the century: The role of victim-defendant relationship in criminal justice decisionmaking over time. Law & Society Review, 38, 105-138.

Franklin

C. A.

Fearn

N. E.

(2008). Gender, race, and formal court decision making outcomes: Chivalry/paternalism, conflict theory or gender conflict? Journal of Criminal Justice, 36, 279-290.

10.

Gobert

J. J.

(1977). Victim precipitation. Columbia Law Review, 77, 511-553.

11.

Gregg v. Georgia, 428 U.S. 153 (1976).

12.

Gross

S. R.

Mauro

(1984). Patterns of death: An analysis of racial disparities in capital sentencing and homicide victimization. Stanford Law Review, 37, 27-153.

13.

Hindson

Potter

Radelet

M. L.

(2006). Race, gender, region and death sentencing in Colorado, 1980-1999. University of Colorado Law Review, 77, 549-594.

14.

Holcomb

J. E.

Williams

M. R.

Demuth

(2004). White female victims and death penalty disparity research. Justice Quarterly, 21, 877-902.

15.

Karmen

(2010). Crime victims: An introduction to victimology. Belmont, CA: Wadsworth.

16.

Kavanaugh-Earl

Cochran

J. K.

Smith

M. D.

Fogel

S. J.

Bjerregaard

(2008). Racial bias and the death penalty. In Lynch

M. J.

Patterson

E. B.

Childs

K. K.

(Eds.), Racial divide: Racial and ethnic biases in the criminal justice system (pp. 147-196). Monsey, NY: Criminal Justice Press.

17.

Kremling

Smith

M. D.

Cochran

J. K.

Bjerregaard

Fogel

S. J

. (2007). The role of mitigating factors in capital sentencing before and after McKoy v. North Carolina. Justice Quarterly, 24, 357-381.

18.

McKoy v. North Carolina, 494 U.S. 433 (1990).

19.

Myers

Greene

(2004). The prejudicial nature of victim impact statements: Implications for capital sentencing policy. Psychology, Public Policy, and Law, 10, 492-515.

20.

N.C. Rural Economic Development Center, Inc. (2012). Rural/urban counties in North Carolina. Retrieved from http://www.ncruralcenter.org/index.php?option=com_content&view=article&id=75&Itemid=126

21.

Paternoster

(1984). Prosecutorial discretion in requesting the death penalty: A case of victim based racial discrimination. Law & Society Review, 18, 437-478.

22.

Paternoster

Deise

(2011). A heavy thumb on the scale: The effect of victim impact evidence on capital decision making. Criminology, 49, 129-161.

23.

Rye

B. J.

Greatrix

S. A.

Enright

C. S.

(2006). The case of the guilty victim: The effects of gender of victim and gender of perpetrator on attributions of blame and responsibility. Sex Roles, 54, 639-649.

24.

Sebba

(1996). Third parties: Victims and the criminal justice system. Columbus:Ohio State University Press.

25.

Simon

L. M. J.

(1996a). Legal treatment of the victim-offender relationship in crimes of violence. Journal of Interpersonal Violence, 11, 94-106.

26.

Simon

L. M. J.

(1996b). The effect of the victim-offender relationship on the sentence length of violent offenders. Journal of Crime & Justice, 19, 129-148.

27.

Spohn

Spears

(1996). The effect of offender and victim characteristics on sexual assault case processing decisions. Justice Quarterly, 13, 649-679.

28.

Stauffer

A. R.

Smith

M. D.

Cochran

J. K.

Fogel

S. J.

Bjerregaard

(2006). The interaction between victim race and gender on sentencing outcomes in capital murder trials: A further exploration. Homicide Studies, 10, 98-117.

29.

Steffensmeier

Ulmer

Kramer

(1998). The interaction of race, gender, and age in criminal sentencing: The punishment cost of being young, black, and male. Criminology, 36, 763-798.

30.

Sundby

S. E.

(2003). The capital jury and empathy: The problem of worthy and unworthy victims. Cornell Law Review, 88, 343-376.

31.

Williams

M. R.

Demuth

Holcomb

J. E.

(2007). Understanding the influence of victim gender in death penalty cases: The importance of victim race, sex-related victimization, and jury decision making. Criminology, 45, 865-891.

32.

Williams

M. R.

Holcomb

J. E.

(2004). The interactive effects of victim race and gender on death sentence disparity findings. Homicide Studies, 8, 350-336.