Sharon L. Weinberg

Canonical Variate Analysis and Related Techniques

This paper discusses statistical methods for studying relations between two sets of variables, when each set contains more than one variable. The number of methods discussed is about 20 or 25, depending on how one counts. About a third of the methods are old methods criticized here, another third are old methods mentioned favorably, and the rest are new methods published here for the first time. It has not been adequately recognized that problems involving two sets of variables arise frequently in almost every area of the behavioral sciences, as the following list of examples will attest:

Sexual Behavior Problems and Psychopathology Symptoms in Sexually Abused Girls

OBJECTIVE This study contrasted a group of sexually abused girls, aged 6 to 12 years, with two demographically comparable control groups, girls from a child psychiatry outpatient department, and girls from a general pediatric clinic to determine whether differences in sexual behavior and psychopathology symptoms could be demonstrated. METHOD All girls and their mothers underwent an evaluation protocol composed of two parent-report inventories, the Child Behavior Checklist and the Child Sexual Behavior Inventory. RESULTS Sexually abused girls and psychiatric controls manifested more psychopathology symptoms, including internalizing and externalizing behaviors, than the nonpsychiatric controls. Relative to both control groups, sexually abused girls manifested more sexual behavior problems: masturbating openly and excessively, exposing their genitals, indiscriminately hugging and kissing strange adults and children, and attempting to insert objects into their genitals. Abuse by fathers or stepfathers involving intercourse was associated with particularly marked sexual behavior disturbances. There was a subgroup of sexually abused girls who tended to force sexual activities on siblings and peers. All of these girls had experienced prolonged sexual abuse (more than 2 years) involving physical force which was perpetrated by a parent. CONCLUSIONS Findings suggest that sexual abuse in preadolescent girls is associated with sexual behavior problems.

Confidence regions for INDSCAL using the jackknife and bootstrap techniques

Bootstrap and jackknife techniques are used to estimate ellipsoidal confidence regions of group stimulus points derived from INDSCAL. The validity of these estimates is assessed through Monte Carlo analysis. Asymptotic estimates of confidence regions based on a MULTISCALE solution are also evaluated. Our findings suggest that the bootstrap and jackknife techniques may be used to provide statements regarding the accuracy of the relative locations of points in space. Our findings also suggest that MULTISCALE asymptotic estimates of confidence regions based on small samples provide an optimistic view of the actual statistical reliability of the solution.

Classroom-Based Surveys of Adolescent Risk-Taking Behaviors: Reducing the Bias of Absenteeism

OBJECTIVES This investigation examined the effectiveness of intensive efforts to include frequently absent students in order to reduce bias in classroom-based studies. METHODS Grade 10 students in 13 New York City high schools (n = 2049) completed self-administered confidential surveys in 4 different phases: a 1-day classroom capture, a 1-day follow-up, and 2 separate 1-week follow-ups. Financial incentives were offered, along with opportunities for out-of-classroom participation. RESULTS Findings showed that frequently absent students engaged in more risk behaviors than those who were rarely absent. Intensive efforts to locate and survey chronically absent students did not, however, significantly alter estimates of risk behavior. Weighting the data for individual absences marginally improved the estimates. CONCLUSIONS This study showed that intensive efforts to capture absent students in classroom-based investigations are not warranted by the small improvements produced in regard to risk behavior estimates.

Low response rate schools in surveys of adolescent risk taking behaviours: possible biases, possible solutions

Study objective: To examine the potential biases introduced when students in low response rate schools are dropped from classroom based surveys of adolescent risk taking behaviour. Design: Self administered confidential surveys were conducted in classrooms, with follow up visits to each school to survey students absent during the initial survey administration. Data on students in schools that achieved a 70% response rate are compared with data on students in schools that did not achieve this level of response. Setting: New York City, United States. Participants: 1854 10th graders in 13 public (state supported) high schools. Main results: Students in schools with low response rates resulting from high rates of absenteeism have different demographic characteristics and engage in more risk behaviours than students in schools with low absenteeism and high response rates. Excluding schools with low rates of response can have an effect on estimates of risk behaviour, even after data are weighted for individual absences. The potential for bias is greatest when, in sampling schools, the proportion of schools with low response rates is large, and when such schools represent a large share of the students in the area under study. Conclusions: Excluding schools with poor response rates from survey samples using a classroom based approach does not improve, and may, under some circumstances, underestimate risky behaviour among adolescent populations.

Validation of a scale for network therapy : A technique for systematic use of peer and family support in addiction treatment

Substance abuse treatments are increasingly employing standardized formats. This is especially the case for approaches that utilize an individual psychotherapy format but less so for family-based approaches. Network therapy, an approach that involves family members and peers in the patients relapse prevention efforts, is theoretically and clinically differentiated in this paper from family systems therapy for addiction. Based on these conceptual differences, a Network Therapy Rating Scale (NTRS) was developed to measure the integrity and differentiability of network therapy from other family-based approaches to addiction treatment. Seven addictions faculty and 10 third- and fourth-year psychiatry residents recently trained in the network approach used the NTRS to rate excerpts of network and family systems therapy sessions. Data revealed the NTRS had high internal consistency reliability when utilized by both groups of raters. In addition, network and nonnetwork subscales within the NTRS rated congruent therapy excerpts significantly higher than noncongruent therapy excerpts, indicating that the NTRS subscales measure what they are designed to measure. Implications for research and training are discussed.

The influence of integrated tuberculosis and human immunodeficiency virus service delivery on patient outcomes.

SETTING Public health clinics in Cape Town, South Africa. OBJECTIVE To examine the influence of integrated tuberculosis (TB) and human immunodeficiency virus (HIV) service delivery on mortality, TB cure and successful treatment completion and loss to follow-up of TB-HIV co-infected patients on concurrent anti-tuberculosis and antiretroviral treatment (ART). DESIGN A survey instrument was used to measure the degree to which TB and HIV services were jointly delivered, and patient data were collected retrospectively from clinic sites and the Department of Health. Six domains measuring integrated TB and HIV service delivery were modelled to assess their relationship with patient outcomes. RESULTS Two domains, integrated TB and ART service delivery and the delivery of TB and HIV care by one clinical team, were associated with lowered odds of death. Care by the same clinical team was also associated with reduced loss to follow-up. CONCLUSION Overall, these findings show that the organization and delivery of health services are important factors that influence health outcomes. These findings strongly support efforts by local governments to integrate TB and ART services, and may help to alleviate concerns that restructuring of TB programs could have a negative impact on long-standing gains.

Measuring the degree of integrated tuberculosis and HIV service delivery in Cape Town South Africa.

To address the considerable tuberculosis (TB)/HIV co-infected population in Cape Town, a number of clinics have made an effort of varying degrees to integrate TB and HIV services. This article describes the development of a theory-based survey instrument designed to quantify the extent to which services were integrated in 33 clinics and presents the results of the survey. Using principal factor analysis, eight factors were extracted and used to make comparisons across three types of clinics: co-located TB and antiretroviral therapy (ART) services, clinics with TB services only and clinics with ART only. Clinics with co-located services scored highest on measures related to integrated TB/ART service delivery compared to clinics with single services, but within group variability was high indicating that co-location of TB and ART services is a necessary but insufficient condition for integrated service delivery. In addition, we found almost all clinics with only TB services in our sample had highly integrated pre-ART services, suggesting that integration of these services across a large number of clinics is feasible and acceptable to clinic staff. TB clinics with highly integrated pre-ART services appear to be efficient sites for introducing ART given that co-infected patients are already engaged in care, and may potentially facilitate earlier access to treatment and minimize loss to follow-up.

Early mother–child attachment and behavior problems in middle childhood: the role of the subsequent caregiving environment

The current study investigated associations between early mother–child attachment, as well as mother–child and teacher–child relationships, and internalizing and externalizing behaviors in middle childhood. Data from the NICHD Study of Early Child Care and Youth Development were used. Findings from a series of individual growth curve analyses revealed that attachment security was negatively related to internalizing and externalizing behaviors, while insecure/other and avoidant attachment were positively related to internalizing behaviors. In addition, longitudinal associations were found between mother–child and teacher–child relationships and internalizing and externalizing behaviors across middle childhood. Implications for attachment theory are discussed.

THE SUGAR SCORING SYSTEM FOR THE BENDER-GESTALT TEST: AN OBJECTIVE APPROACH THAT REFLECTS CLINICAL JUDGMENT

Although several objective scoring systems have been developed for use with the abbreviated form of the Bender-Gestalt test of visuomotor development (Figures A, 1, 2, 4, 6, and 8), each has been criticized as being inadequate in some way or other. This paper reviews the inadequacies of these systems as described in the literature and presents the results of a small study designed to investigate the psychometric properties of a new objective scoring system, recently proposed by Sugar, which incorporates both quantitative and qualitative perspectives. The new system is shown to be psychometrically sound in validity and reliability. Additional advantages are described. The new system is recommended as a good, if not better, alternative to other, existing methods for scoring the Bender-Gestalt test.