Parent and Peer Pathways to Adolescent Delinquency: Variations by Ethnicity and Neighborhood Context

Effects of ethnicity and neighborhood quality often are confounded in research on adolescent delinquent behavior. This study examined the pathways to delinquency among 2,277 African American and 5,973 European American youth residing in high-risk and low-risk neighborhoods. Using data from a national study of youth, a meditational model was tested in which parenting practices (parental control and maternal support) were hypothesized to influence adolescents' participation in delinquent behavior through their affiliation with deviant peers. The relationships of family and neighborhood risk to parenting practices and deviant peer affiliation were also examined. Results of multi-group structural equation models provided support for the core meditational model in both ethnic groups, as well as evidence of a direct effect of maternal support on delinquency. When a similar model was tested within each ethnic group to compare youths residing in high-risk and low-risk neighborhoods, few neighborhood differences were found. The results indicate that, for both African American and European American youth, low parental control influences delinquency indirectly through its effect on deviant peer affiliation, whereas maternal support has both direct and indirect effects. However, the contextual factors influencing parenting practices and deviant peer affiliation appear to vary somewhat across ethnic groups. Overall the present study highlights the need to look at the joint influence of neighborhood context and ethnicity on adolescent problem behavior.     

The Relative Influence of Different Domains of Social Connectedness on Self-Directed Violence in Adolescence

Previous research has linked greater social connectedness with a lowered risk of self-directed violence among adolescents. However, few studies have analyzed the comparative strength of different domains of connectedness (e.g., family, peers and school) to determine where limited resources might best be focused. Data to address that gap were taken from the Centers for Disease Control and Prevention’s Student Health and Safety Survey, administered to 4,131 7th–12th graders (51.5% female; 43.8% Hispanic; 22.6% African American or Black). Logistic regressions (controlling for age, gender, race/ethnicity, family structure, academic performance, and depressive symptoms) suggest that family connectedness was a stronger predictor than connectedness to peers, school, or adults at school for non-suicidal self-harm, suicidal ideation, suicide plans, and non-fatal suicidal behavior. In some analyses, peer connectedness was unexpectedly a risk factor. Results have implications for prevention of suicide in adolescence, especially in the context of the current trend towards school-based prevention programs.     

Understanding the Relationship between Health PACs and Health Lobbying in the American States

We examine how health PAC activity in the states is connectedto lobbying. Is the political money that health interest organizationsbring to the policy process a powerful, independent means ofinfluence or is it better understood narrowly as a tactic usedto support lobbying? We examine the relationship between campaigncontributions and lobby activity and the limited work that hasbeen conducted on them and raise a number of questions aboutthe process by which they are connected. We utilize 1998 dataon state lobbying and PAC activity, allowing us to answer causalresearch questions generated from existing studies at the nationallevel. We conclude that PAC activity is best viewed as an adjunctof lobbying rather than an independent form of political activity.     

Estimating the Causal Effects of Media Coverage on Policy-Specific Knowledge

Policy facts are among the most relevant forms of knowledge in a democracy. Although the mass media seem like an obvious source of policy-specific information, past research in this area has been plagued by design and methodological problems that have hindered causal inferences. Moreover, few studies include measures of media content, preventing researchers from being able to say what it is about media coverage that influences learning. We advance the literature by employing a simple but underutilized approach for estimating the causal effects of news coverage. Drawing upon a unique collection of cross-sectional survey data, we make within-survey/within-subjects comparisons under conditions of high and low media coverage. We show how the volume, breadth, and prominence of news media coverage increase policy-specific knowledge above and beyond common demographic factors.     

The Consequences of Concurrent Campaigns for Citizen Knowledge of Congressional Candidates

Congressional campaigns do not occur in isolation, but concurrently with many other races. What are the effects of overlapping campaigns for how people learn about congressional candidates? On one hand, intensely fought races on the same ballot may encourage people to extend interest to contemporaneous races. Alternatively, competing campaigns may distract voters and limit learning about down-ticket contests. Using American National Election Studies data from 1996 to 2002, I test the consequences of state level presidential, senatorial, and gubernatorial campaign intensity on citizen learning about congressional candidates. I find that while the local intensity of the presidential race has little effect on knowledge of House candidates, the intensity of the Senate race is positively associated with candidate recognition. These effects vary depending on the nature of the election, where House candidates benefit from intense Senate races on the ballot particularly in midterm election years.     

Issue Framing and Engagement: Rhetorical Strategy in Public Policy Debates

Conventional wisdom and scholarly research indicate that to win a policy debate political actors should frame the issue strategically—that is, selectively highlight considerations that mobilize public opinion behind their policy position. Engaging the opponent in a dialogue (i.e., focusing on the same considerations) is portrayed as a suboptimal strategy because political actors forfeit the ability to structure the debate. Using over 40 public opinion polls and a detailed content analysis of news stories, I examine the use of framing and engagement strategies during the 1993–94 debate over health care reform. The analysis shows that engagement was more effective at increasing support for reform than framing. This study is the first to document the role of engagement in a policy debate, and it extends work showing that this strategy is more common in election campaigns than scholars once suspected.

Canine Population Data Generated from a Multiplex STR Kit for Use in Forensic Casework*

Abstract: Canine biological specimens are often part of the physical evidence from crime scenes. Until now, there have been no validated canine‐specific forensic reagent kits available. A multiplex genotyping system, comprising 18 short tandem repeats (STRs) and a sex‐linked zinc finger locus for gender determination, was developed for generating population genetic data assessing the weight of canine forensic DNA profiles. Allele frequencies were estimated for 236 pedigreed and 431 mixed breed dogs residing in the U.S. Average random match probability is 1 in 2 × 10³³ using the regional database and 1 in 4 × 10³⁹ using the breed dataset. Each pedigreed population was genetically distinct and could be differentiated from the mixed breed dog population but genetic variation was not significantly correlated with geographic transition. Results herein support the use of the allele frequency data with the canine STR multiplex for conveying the significance of identity testing for forensic casework, parentage testing, and breed assignments.     

Caring Letters Project: a military suicide-prevention pilot program

Background: The Caring Letters Project (CLP) is a suicide prevention program that involves sending brief caring letters to discharged inpatients following psychiatric hospitalization. Several studies suggest that repeatedly sending caring messages may reduce suicides and suicide attempts in high-risk populations. Aims: The aims of this study were to (1) evaluate feasibility of use in the military setting, (2) explore trends toward reduction of psychiatric rehospitalizations, (3) assess preference for and test e-mail correspondence, and (4) identify best practices and gather data to inform a randomized controlled study. Methods: A total of 110 psychiatric inpatients at a military treatment facility consented, were interviewed, and then received personalized handwritten letters or e-mails at regular intervals following discharge. Data collected included demographics, clinical characteristics, preference for e-mail versus postal mail, rates of undeliverable and return correspondence, rehospitalizations, and adverse events requiring safety procedures. Results: A total of 436 letters and e-mails have been sent to date. Most participants indicated preference for e-mail versus postal mail. Fifteen participants were readmitted for treatment compared to 20 patients in usual care. Twenty participants sent responses and all were positive statements about the program. There were no adverse events. Conclusions: This program is feasible for use at a military treatment facility. A randomized controlled trial is needed to determine whether the intervention can reduce suicide rates among military and veteran populations.