Recruitment and retention of participants in prevention trials involving family-based interventions.

Family-based preventive intervention trials in the area of child conduct problems face serious challenges regarding the recruitment and subsequent retention of participants. This article focuses on the problems and strategies pertinent to recruitment and retention in prevention research. The issues are discussed first with respect to initiating and maintaining a longitudinal sample, and then engaging and sustaining families in a preventive intervention. General principles are illustrated by EARLY ALLIANCE, a recently launched preventive intervention trial aimed at reducing children's risk for conduct disorder, substance abuse, and school failure. Recommendations for enhancing recruitment and retention are offered.

Promoting intervention fidelity. Conceptual issues, methods, and preliminary results from the EARLY ALLIANCE prevention trial.

Fidelity refers to the demonstration that an experimental manipulation is conducted as planned. In outcome research, an intervention can be said to satisfy fidelity requirements if it can be shown that each of its components is delivered in a comparable manner to all participants and is true to the theory and goals underlying the research. Demonstrating the fidelity of an intervention is a key methodologic requirement of any sound prevention trial. This paper summarizes key conceptual and methodologic issues associated with intervention fidelity, and describes the steps taken to promote fidelity in EARLY ALLIANCE, a large-scale prevention trial currently testing the effectiveness of family, peer, and school interventions to promote competence and reduce risk for conduct disorder, substance abuse, and school failure. The paper presents preliminary results (Trial Year 1) that demonstrate content and process fidelity for two of these interventions, and discusses how the EARLY ALLIANCE methodology may be generalized to address fidelity issues in other prevention studies.

The EARLY ALLIANCE prevention trial: a dual design to test reduction of risk for conduct problems, substance abuse, and school failure in childhood.

This paper describes a preventive intervention trial called EARLY ALLIANCE which is aimed at reducing risk for three adverse outcomes in childhood and adolescence: conduct problems, substance abuse, and school failure. The structure of the prevention trial is unique because two linked designs are being implemented concurrently. The primary design focuses on children at elevated risk for adverse outcomes, and compares a targeted, multicontextual preventive intervention with family, classroom, peer relational, and academic components to a universal, schoolwide preventive intervention that emphasizes peaceful conflict management and serves as a "usual care" control condition. The secondary design focuses on children at lower risk for adverse outcomes and compares a universally administered classroom program to the control condition. The paper describes the theoretical foundation for EARLY ALLIANCE, the goals of the prevention trial, the rationale for design choices, and the methods employed.

Risk factors and correlates of dating violence: the relevance of examining frequency and severity levels in a college sample.

To determine whether categorizing levels of violence along dimensions of frequency and severity would result in informative distinctions among individuals using dating violence, reported use of physical violence, along with variables theorized to be related to use of force in intimate relationships, was assessed in a sample of 617 college students (males = 290; females = 327). When participants' scores were analyzed by dichotomizing them along the lines of ever versus never using dating violence, numerous past findings were replicated. However, participants using a higher frequency of force were distinguished by needing to control their partner and by less inhibition in expressing their anger. Higher severity of force by an individual was predicted only by a need to control his/her dating partner. Implications for these findings were discussed.

Cognitive determinants of quality of life after onset of cancer.

To investigate cognitive coping styles and how they might relate to perceived quality of life for individuals seriously ill with cancer 41 mostly elderly, male patients with a wide variety of cancers were administered the Illness Effects Questionnaire, a quality of life measure, the COPE Questionnaire, which samples different coping strategies, and questions regarding beliefs about illness behaviors, expectations about cancer's effects, comparisons of the participants' lives with those of others, feelings since having cancer, and motivation to resist lifestyle disruptions. Six of the belief factors and two of the coping strategies were related to perceived quality of life. If the effects of cancer were less than expected, quality of life was better; with an expectation of a negative effect from the disease, lowered quality of life was perceived. Two coping strategies were related to quality of life, suppression of competing activities and using religious practices. Other relationships were also discussed.

Contrasting Alternatives to Least Squares in Regression Using Diagnostics for Identifying Influential Data.

The implicit assumptions associated with several alternative methods for estimating regression coefficients are identified using the regression diagnostic measures of leverage and influence. A formulation of these measures for use with the Ridge, Stein, principal component and equal weight estimators is presented and then used to contrast competing assumptions. Based on the implicit priors, situations are denoted where each of the alternatives can be expected to perform best.