Mixture structure analysis using the Akaike Information Criterion and the bootstrap

Given i.i.d. observations x1,x2,x3,...,xn drawn from a mixture of normal terms, one is often interested in determining the number of terms in the mixture and their defining parameters. Although the problem of determining the number of terms is intractable under the most general assumptions, there is hope of elucidating the mixture structure given appropriate caveats on the underlying mixture. This paper examines a new approach to this problem based on the use of Akaike Information Criterion (AIC) based pruning of data driven mixture models which are obtained from resampled data sets. Results of the application of this procedure to artificially generated data sets and a real world data set are provided.     

Thinking Critically

This study evaluated a ten-week course designed for social work field instructors interested in developing a critically reflective approach to field instruction. The course trained participants to think critically about their role in preparing professional social workers. A pretest-posttest nonequivalent comparison group design was utilized. The Watson-Glaser Critical Thinking Appraisal was used as the measurement instrument. Using t-tests, the findings indicate that the participants significantly increased their overall ability to think critically when compared to the control group. The implications for practicum instruction are addressed.     

Conflict in Divorcing and Continuously Married Families: A Study of Marital,Parent–Child and Sibling Relationships

ABSTRACT

Family members in both separated or divorcing and married families completed a range of questionnaires about the conflict in 3 relationships—couple, parent–child, and sibling—across 2 phases of the study approximately 12 months apart. In addition, the adolescents completed measures of adjustment: self esteem, depression, and anxiety. Adolescents in a subset of the divorcing families were interviewed about the conflict in their families across the 2 phases (107 at Phase 1 and 35 at Phase 2). Analyses showed that conflict was higher in the separated or divorcing families across all 3 relationships, and that the highest levels of conflict occurred for the high-conflict divorcing families. Qualitative data from interviews illustrated the nature of the conflict that was occurring.     

Should There Be Mentoring Efforts to Help Retain and Recruit More Males Into Baccalaureate and Master's Social Work Programs?

Social work programs need to actively set mechanisms in place to attract more men into a predominately female academic major through mentoring, recruitment, and retention efforts. This article conducts a review of the literature to examine whether mentoring efforts should be directed at male social work students. It is inconsistent for the social work profession to embrace the concept of providing culturally competent practices to clients but fail to address and combat a significant gender gap in the delivery of services to male clients.     

“Koe wai hoki koe?!”, or “Who are you?!”: Issues of trust in cross‐cultural collaborative research

Abstract

Cross‐cultural research designs are increasingly employed in the New Zealand context to investigate a range of complex social and environmental issues. However, there is little guidance for researchers who work for mainstream organisations and hapū (subtribe(s)) and iwi (tribe(s)) on how to undertake cross‐cultural collaborative research. Trust is a key issue that makes cross‐cultural research possible and it is surprising that issues relating to trust have not featured widely in the literature. In this paper, we reflect on our experiences working cross‐culturally and collaboratively with a hapū in Northland, New Zealand. We discuss how a mainstream research organisation worked with a community health trust and developed a working relationship with a hapū, forming the basis of a community‐driven collaborative evaluation of a central government‐funded project. Key findings were the establishment of a tuākana‐tēina relationship (where teacher and learner are interchangeable) between the members of the collaborative research team and the employment of appropriate research methods to develop a research agenda that ensured everyone would benefit from the research.     

Adaptive changes in gait of older and younger adults as responses to challenges to dynamic balance

The study proposed to identify balance strategies used by younger and older adults during gait under proprioceptive, visual, and simultaneous proprioceptive-visual challenges. Participants ambulated under 4 conditions: consistent, noncompliant surface; inconsistent, compliant surface (C); consistent, noncompliant surface with vision obscured (NCVO); and inconsistent, compliant surface with vision obscured (CVO). Balance adaptations were measured as changes in gait velocity, cadence, and gait-stability ratio (GSR). Participants were 5 younger (mean age = 27.2) and 5 older (mean age = 68) healthy adults. Significant age differences were found for GSR (p = .03) on all surfaces. Older adults adopted a more stable gait pattern than younger adults regardless of the challenge presented by surface. Significant condition differences were found for velocity (p < .001) and cadence (p = .001). All participants exhibited significantly decreased velocity and increased cadence on surfaces C and CVO. Gait speed and cadence did not significantly change in NCVO. Younger and older adults exhibited similar adaptive balance strategies, slowing and increasing steps/s, under proprioceptive and proprioceptive-visual challenges to dynamic balance.     

Parameterizing age patterns of demographic rates with the multiexponential model schedule

"For nearly 200 years actuaries, statisticians, and demographers have sought to summarize the age pattern of mortality rates by means of a limited number of parameters. Such 'model schedules' have also been useful in representing schedules of rates other than mortality....This paper illustrates a particular general functional form for such model schedules: the multiexponential function. It discusses the changing behavior of this function as its parameters take on different values and examines the quality of the fits of this function to observed data on mortality, fertility, and migration." This is a revised version of a paper originally presented at the 1993 Annual Meeting of the Population Association of America. (SUMMARY IN FRE)     

Age patterns of elderly migration: An international comparison

Age-specific rates of migration exhibit remarkably persistent regularities in age profile. These regularities seem to hold all over the world and across time. This article identifies some of the principal antecedents of such regularities, focusing especially on the age patterns of migration exhibited by the elderly. It examines the differentials introduced by gender, distance, and marital status.     

Estimating the regional migration patterns of the foreign-born population in the United States: 1950-1990

"In this paper, we apply model schedules to graduate data on the internal and external regional [U.S.] migration patterns of the foreign-born population for the 1950-1990 period.... To find estimates of the unrecorded migration flows in-between for four census-defined periods in our study (that is, for 1950-1955, 1960-1965, 1970-1975, and 1980-1985) we interpolate between the data of adjacent census time periods. Finally, we combine the estimated migration data with the corresponding mortality data to calculate and analyze the multiregional life tables and projections associated with each five-year time interval." (EXCERPT)