Evaluation of methods for interval estimation of model outputs, with application to survival models

When a published statistical model is also distributed as computer software, it will usually be desirable to present the outputs as interval, as well as point, estimates. The present paper compares three methods for approximate interval estimation about a model output, for use when the model form does not permit an exact interval estimate. The methods considered are first-order asymptotics, using second derivatives of the log-likelihood to estimate variance information; higher-order asymptotics based on the signed-root transformation; and the non-parametric bootstrap. The signed-root method is Bayesian, and uses an approximation for posterior moments that has not previously been tested in a real-world application. Use of the three methods is illustrated with reference to a software project arising in medical decision-making, the UKPDS Risk Engine. Intervals from the first-order and signed-root methods are near- identical, and typically 1% wider to 7% narrower than those from the non-parametric bootstrap. The asymptotic methods are markedly faster than the bootstrap method.     

THE LABOUR MARKET EXPERIENCE OF CAMBODIANS: POLICY IMPLICATIONS FOR SETTLEMENT SERVICES

A proportion of the annual migrant intake now comprises persons accepted under the Refugee and Special Humanitarian Program entry categories, for reasons other than the direct contribution their skills can make to economic growth. This paper examines the labour market experience of Cambodians, and considers the policy implications for resettlement services for refugees. Cambodians experience high rates of unemployment, and have low levels of English language proficiency, low levels of educational attainment, and few possess post-school qualifications. Reluctance to develop settlement services for refugees, to make special provisions for access within existing services, or even to ensure equitable access to existing services has contributed to the long term economic dependency of a sizeable proportion of this group. It is recommended that they be given greater access to basic, advanced and job related English language tuition. They, and other refugees with similar disadvantages also require special access to job search skills programs, vocational training, and wage subsidy and special employment programs.     

Raising multiracial awareness in family therapy through critical conversations

Multiracial families are uniquely affected by racial dynamics in U.S. society. Family therapists must be prepared to meet the needs of this growing population and to support racial equity. This article includes an overview of literature related to being multiracial and offers a framework for working with multiracial identity development in therapy. A critical conversation approach to working with multiracial identity is shared along with case examples. The authors' experiences developing the model via a practitioner inquiry group are highlighted.     

Applications to Social Work Programmes in England: Students as Consumers?

One of the most resounding impacts of the introduction of the new social work degree in England in 2003 has been an increase in the number of applications to social work courses. However, the processes used by applicants to reach decisions about where to study social work are little understood. This article reports data from six preparatory focus groups and four interviews (n = 38), 17 focus group interviews with first year students (n = 112) from nine social work programmes run in six universities, and 2,606 responses to three online surveys administered to first year students, which were collected as part of the national Evaluation of the New Social Work Degree in England funded by the Department of Health. They show that students use a range of sources to find out about social work education, with rising reliance on electronic media as an information resource. Although the majority cites convenience of location as the chief reason for selecting a particular university, academic reputation appears to be growing in importance. Observations from these data are discussed in the context of consumer behaviour in higher education and in social work education at a time when universities are developing marketing strategies to compete for students.     

Therapeutic alliance between youth and staff in residential group care: Psychometrics of the therapeutic alliance quality scale

Therapeutic alliance has been frequently studied in individual counseling sessions; however, research on therapeutic alliance in residential settings for youth with mental health diagnoses has been limited. This may be due, in part, to the presence of multiple service providers often in caregiving roles. The purpose of this study was to examine the psychometric quality of a widely utilized measure of therapeutic alliance used in psychotherapy with youth in residential care where the treatment is provided by a trained married couple. We also compared the relationship between youth ratings of their male and female service providers, as well as examined correlations in ratings between youth and staff on therapeutic alliance. Finally, we investigated the direction, magnitude, and trajectory of change in therapeutic alliance over a 12-month period following admission into residential care. The method was a longitudinal assessment of 135 youth and 124 staff regarding therapeutic alliance over the course of 12 months or discharge from services. Results indicated strong psychometric properties and high correlations for youth ratings of both their male and female service providers. However, the correlation was low between youth and service provider ratings of alliance. Longitudinal analyses indicated that rates of therapeutic alliance changed over time.