R. Barry Ruback

Saliva testosterone and criminal violence among women

Free testosterone concentrations were measured in the saliva of 84 female inmates and 15 female college students. Testosterone differed among inmates convicted of unprovoked violence, defensive violence, theft, drugs, and a set of other crimes. It was highest with unprovoked violence and lowest with defensive violence, where inmates had reacted violently after being physically assaulted. Testosterone was also related to number of prior charges and to Parole Board decisions about length of time to serve before being released on parole. Mean testosterone levels were similar for inmates and college students.

Dimensions of Group Process: Amount and Structure of Vocal Interaction

Publisher Summary This chapter brings together the work of Hackman and Moms, Bales, and Jaffe and Feldstein into a model that is both manageable and intricate enough to capture the subtle details of ongoing group process. It describes a conceptual model that speaks directly to questions related to the dimensions of amount and structure and discusses data from three studies that are suggestive of the importance of these two dimensions to an understanding of group process. It also presents an automated data collection system that requires no human observers. The chapter defines process in terms of talking (Shaw, 1964), including both the amount of talking and the patterns of talking among group members. The chapter considers the amount and structure of content-free measures and the way they both can be affected by factors about the task facing the group, the overall group, and the individual members. The amount of vocal activity can range from silence to everyone speaking at once. Although much of the research on brainstorming has contrasted nominal groups with real groups, group interaction can range from none at all (in nominal groups) to various levels (in various real groups).

Comparative distress levels of inner-city family members of homicide victims

This study investigated the distress levels of 150 family members of homicide victims, as well as how pre-event, peri-event, and postevent variables were related to distress. Distress levels were very high, with 26% of the sample reporting clinical distress. Because it was not possible to say if this distress resulted from the homicide itself or from the fact that people who lose family members to homicide generally have lives rooted in stressful contexts, we compared the homicide sample to two sociodemographically comparable groups of 108 other trauma victims and 119 nonvictims selected from a larger epidemiological dataset. Homicide survivors were significantly more distressed than either group, suggesting that loss of a family member to homicide has definite clinical implications. Although event-related variables were somewhat predictive of distress, pre-event and postevent variables selected for this study had greater predictive utility.

Repeat and multiple victimizations: The role of individual and contextual factors.

The present research uses hierarchical modeling to examine the relative contributions of factors about the person, factors about the context, and, most important the interaction of factors about the person and factors about the context in models of both repeat victimization (more than one of the same type of crime) and multiple victimization (two or more different types of crime). Using telephone survey data from a multistage sample of Seattle residents, we estimate separate hierarchical models for repeat property, repeat violent and multiple victimization. Results indicate that repeat victimization of both types varies substantially by neighborhood, whereas multiple victimization seems more determined by individual-level factors. Implications for social disorganization theory, routine activity/lifestyle exposure theory, and future work on repeat victimization are discussed.

Individual and Contextual Influences on Sentence Lengths: Examining Political Conservatism

This study examined the impact of legal, extralegal, and contextual variables on prison sentence lengths for violent felons sentenced in Georgia from 1981 to 1989. Multiple linear regression analyses were conducted for all violent crimes and separately for four types of violent crime: murder and manslaughter, rape, aggravated assault, and robbery. Results indicated that the legally relevant factors—seriousness of the crime and number of convictions—had the strongest influence on sentence lengths. Across most violent crimes, male, older, and better-educated offenders received longer sentences than those without such characteristics. Political conservatism had a positive effect on sentence lengths for overall violent crime, robbery, and aggravated assault. Interaction effects for political conservatism and the number of convictions were significant, indicating that sentence length increased disproportionately as a courts conservatism and the felons number of convictions increased. Findings suggest that political conservatism is an important contextual feature affecting prison sentence length.

Rural-Urban Differences in Sexual Victimization and Reporting - Analyses Using UCR and Crisis Center Data

In this study, the authors analyzed data from Uniform Crime Reports and from rape crisis centers in all 67 counties in Pennsylvania. Although absolute numbers of sexual victimization cases reported to rape crisis centers were higher in urban counties, rates of sexual victimization were higher in rural counties, particularly those with above median proportions of female officeholders in the county. Subsequent analyses indicated this effect was not related to the provision of victim services but could be related to a better educated population being more likely to use such services. In terms of reporting to the police, urban counties and counties with higher levels of assaults by strangers had higher rates of reporting to the police, although these effects were somewhat qualified by an interaction of the two variables. Additional analyses suggest that whether the county is urban or rural is related to the amount of funding allocated to victim services programs.

Prior relationship, resistance, and injury in rapes: an analysis of crisis center records

Information about the rapes of 2,526 adult females was coded from the records of a rape crisis center to test the hypothesis that physically resisting a stranger would be more strongly related to injury than would physically resisting someone known to the victim. Among other differences, attacks by strangers were more likely to involve a weapon and to occur outdoors than were attacks by nonstrangers, and victims were less likely to physically resist strangers than nonstrangers. Across the entire sample, multivariate analyses revealed that physical resistance was significantly related to injury, even when other factors were held constant. Consistent with the hypothesis, physical resistance was more strongly related to injury when the rapist was a stranger than when the rapist was known to the victim.

Normative advice to campus crime victims: effects of gender, age, and alcohol

Three studies investigated the appropriateness of calling the police as a function of crime, victim, and subject factors. In particular, the studies focused on whether and how the victim’s consumption of alcohol affected normative advice to report the crime, as opposed to other options. Across the three studies, subjects viewed reporting as more appropriate for female victims, for victims who were 21 or older, and for victims who had not been drinking. In addition, females were more likely than males to believe reporting to the police was appropriate whereas males were more likely than females to favor some type of private action. Subjects viewed reporting as particularly inappropriate when the victim was underage and had been drinking. Results suggest that, because of the perceived stigma attached to victims who have been drinking, even serious victimizations may go unreported.

Vocal Patterns in Male and Female Groups

Ten male and ten female five-person groups spent 20 minutes talking and getting acquainted. Subjects wore lavaliere microphones, and their voices were monitored continuously by a microcomputer system that detected patterns of speech and silence. Raw vocal data were transformed into codes representing conversational states that can occur in a group discussion (e.g., individual turn, pause, simultaneous speech, group turn), and these codes were related to peer ratings and to overall ratings of the group. Talking was generally associated with positive evaluations, though the pattern varied with different speech codes and differed for males and females. Females reported enjoying group interaction more than males did, and females vocalized more within their turns and paused less between their turns.

DECISION MAKING BY CRIME VICTIMS: A MULTIMETHOD APPROACH

Since most crimes investigated by police result from notification by victims, crime victims can be viewed as the gatekeepers of the criminal justice system. This paper describes the results of a series of studies which employed multiple methodologies to investigate the decisions of property crime victims to notify the police. The four methods used in this research included (a) an archival analysis of police records, (b) interviews with crime victims, (c) simulation studies using college students, and (d) experiments in a field laboratory. Two convergent findings emerged from the data. Social influence was found to be an important determinant of both the decision to call the police and the delay in such notification. If others advised the victim to call the police, the crime was more likely to be reported, but, ironically, the greater the number of people consulted, the longer the delay in reporting the crime. The second convergent finding was that situational factors played a stronger role in affecting delay in notification than did characteristics of the victim. Situational factors included the type of crime and the time of day when it was discovered, as well as the number of others talked to and the type of advice received from them. Several divergent findings across methods raised questions about the relative adequacy of each method for the problems being investigated, and helped identify sources of similarities and differences in the conceptual and operational definitions employed.