Interval estimation for log-linear models with one variable subject to non-ignorable non-response

Summary.  Log-linear models for multiway contingency tables where one variable is subject to non-ignorable non-response will often yield boundary solutions, with the probability of non-respondents being classified in some cells of the table estimated as 0. The paper considers the effect of this non-standard behaviour on two methods of interval estimation based on the distribution of the maximum likelihood estimator. The first method relies on the estimator being approximately normally distributed with variance equal to the inverse of the information matrix. It is shown that the information matrix is singular for boundary solutions, but intervals can be calculated after a simple transformation. For the second method, based on the bootstrap, asymptotic results suggest that the coverage properties may be poor for boundary solutions. Both methods are compared with profile likelihood intervals in a simulation study based on data from the British General Election Panel Study. The results of this study indicate that all three methods perform poorly for a parameter of the non-response model, whereas they all perform well for a parameter of the margin model, irrespective of whether or not there is a boundary solution.     

Self‐Perceived Gender Typicality and the Peer Context During Adolescence

This research examined adolescents' gender identity in relation to the peer context and their self‐concept. Participants were 229 adolescents who completed questionnaire measures of self‐concept and multidimensional gender identity. Regression analysis indicated peer acceptance partially mediated the relation between self‐perceived gender typicality and self‐worth. Cluster analysis revealed four groups of adolescents with differing profiles of self‐perceived gender typicality, felt peer pressure for gender conformity, and peer acceptance. Findings highlight the inherently social and contextual nature of gender identity. Also, the pathologizing of gender‐nonconforming youth is discussed.     

Physical fitness effects on substance abuse risk factors and use patterns

In this study, seventy-four adolescents in either a school, community or hospital based drug intervention program received and eight to nine week structured fitness activity class as an integrated element of their respective prevention or treatment program. Prepost assessments indicated significant gains in the field fitness tests of one mile run, 1 minute situp, 1 minute pushup, percent fat and flexibility for the total sample. Based upon prepost change on the one mile run time (less than 1:00) subjects were categorized as improvers (n = 38) and non-improvers (n = 36). Improvers demonstrated a significant increase in the self concept risk factor (Piers Harris Self Concept Scale) and a significant decrease in anxiety and depression risk factors (General Well-Being Scale) compared to the non-improvers. At posttest, the improvers self reported substance use patterns were significantly lower compared to the non-improvers for percentage of the sample who were multiple drug users and alcohol uses per week and were significantly higher for the percentage of the sample demonstrating total abstinence. These findings are suggestive of the usefulness of physical training as a supplemental intervention for adolescent substance abusers.     

Bayesian Analysis of DNA Profiling Data in Forensic Identification Applications

The utilization of DNA evidence in cases of forensic identification has become widespread over the last few years. The strength of this evidence against an individual standing trial is typically presented in court in the form of a likelihood ratio (LR) or its reciprocal (the profile match probability). The value of this LR will vary according to the nature of the genetic relationship between the accused and other possible perpetrators of the crime in the population. This paper develops ideas and methods for analysing data and evaluating LRs when the evidence is based on short tandem repeat profiles, with special emphasis placed on a Bayesian approach. These are then applied in the context of a particular quadruplex profiling system used for routine case-work by the UK Forensic Science Service.     

Trauma, coping resources, and family functioning in emergency services personnel: A comparative study

In this study a sample of male, shift-working, operational ambulance officers were compared with a group of male shift-workers from a range of occupations in which exposure to traumatic events was not inherent in the role ( N = 71). Three dimensions of family functioning were examined: intimacy, conflict, and parenting styles, with respect to occupational type (i.e. work-related exposure to trauma) and accounting for personal resources (coping). Personal resources were found to have a significant impact on the dimensions of family functioning. Furthermore, differences emerged in the pattern of coping between the two groups. Ambulance officers demonstrated a more varied repertoire of personal resources than the control group. Social support was found to be the sole predictor of intimacy and conflict levels in the control sample. However, in the ambulance group social support was predictive of intimacy levels, rational-cognitive strategies were predictive of conflict levels, and self-care, social support, and rational-cognitive strategies were all significantly correlated with the three dimensions of family functioning.     

Religion and Giving in Australia

This article draws on a household survey of giving and volunteering undertaken as part of the Giving Australia project to explore the relationship between religion and the giving behaviour of adult Australians. We find that people who identify themselves as having a religion are more likely to give and to give more on average than people who do not but that this relationship is produced by a subset of this group; namely, people who regularly attend religious services. Indeed, the likelihood of giving and average amounts given by givers over a year increase with levels of attendance at religious services. The relationship holds even after allowing for other factors that affect the likelihood of a person giving and the amount given. It also holds for giving to nonreligious causes (ie when giving to religion is omitted). However, when we also omit giving to charities and look at giving to civic causes alone we find that the frequency of attendance at religious services has an ambiguous relationship with giving. A suggestion in overseas literature that it is participation in religious groups, rather than attendance at religious services that is associated with giving is not supported.     

What Works in Strategic Partnerships for Children: a Research Review

Strategic partnerships have a key role to play in delivering the government's ‘Every Child Matters’ agenda. There is a developing body of research and other evidence and practice‐informed literature about what works in relation to partnership working. Drawing on a research review commissioned by Barnardos on strategic partnership working for children, this study summarises research findings from a number of sectors relating to the development, delivery and review of effective partnerships. Although there is widespread consensus within the research on the factors contributing to effective partnership processes, there is, currently, relatively little evidence that partnerships lead to positive outcomes for children and young people. © 2006 The Author(s). Journal compilation © National Children's Bureau.     

Taking a risk and playing it safe: the use of actors in interagency child protection training

Interagency training in child protection is an important part of engaging professionals in a common approach to their collective responsibility for safeguarding children. Training is a key method for empowering professionals to carry out their complex role in child protection with appropriate knowledge, skills and confidence. In Buckinghamshire, experiential learning involving actors has met with some success as part of the interagency child protection training programme. The following paper proposes the argument for using actors, describes how this is done and evaluates its effectiveness for participants, trainers and actors. It is suggested that this method has ample rewards if properly planned, particularly for practitioners in lessons for practice. A close working relationship must be developed between actors and trainers. Scenarios that seem real can be portrayed and skills can then be practised in the knowledge that no one is really being harmed but valuable lessons for real‐life situations can be learned. A central aspect is learning together in a supportive environment away from a blame culture. Copyright © 2005 John Wiley & Sons, Ltd.     

Multiple Public Service Performance Indicators: Toward an Integrated Statistical Approach

Public service organizations usually produce multiple outputs,measured on different scales, giving rise to a suite of performanceindicators. The traditional approach to statistical analysisof organizational performance has been to develop a separateregression model for each performance indicator. This piecemealapproach, the article argues, may discard valuable information,as it ignores potentially important relationships between individualperformance measures. We therefore propose modeling an organization'sperformance measures simultaneously, using the methods of seeminglyunrelated regressions. The approach implicitly introduces alatent organizational variable into the regressions and maytherefore economize on the need to assemble explicit measuresof organizational characteristics. The method is illustratedusing an example from English public hospitals.