EDITING A LARGE-SCALE DEMOGRAPHIC SURVEY–THE EXPERIENCE OF THE MELBOURNE FAMILY SURVEY1

This paper discusses the editing procedures for a survey on family formation conducted in Melbourne during 1971. Examples are given of the incidence and types of inconsistencies revealed during the editing stage in the mainly quantitative variables concerning childbearing, numbers of children, and the demographic characteristics of the children. The principal point emphasized is the need for automatic detection and manual correction of coding deficiencies in a survey of this type, i.e. an intensive, “once-off” survey, with a sample size of a few thousand, from which detailed and complex analyses are planned.     

Comparison of basic data for the National Survey of Residential Group Care Facilities: 1966-1982

Major findings from this analysis of the data gathered in preparation for the 1966 and 1981 surveys of children's residential group care facilities are as follows below. In the field of residential care of pregnant adolescents, it appears that, despite an increase in the number of births to adolescent mothers since 1966, the preferred mode of serving this population is no longer residential group care. While there has been a small increase (6 facilities) in the number of residential facilities for pregnant adolescents that operate under public auspices, the 49% decrease in facilities under private auspices is the real indicator of the direction of residential services to pregnant adolescents since 1965. In the Child Welfare Stream, the most notable change between 1965 and 1981 was the decrease in the relative proportion of residential group care facilities for dependent, neglected, or abused children and youth. Overall, child welfare facilities accounted for 55% of all residential facilities listed for survey in 1965; in 1981 this percentage had fallen to 37%. As in 1965 the Child Welfare Stream was dominated by the private sector in 1981, with 83% of all child welfare facilities operating under private auspices. This pattern was found for most of the 50 individual states as well. In the Juvenile Justice Stream, the overall increase of 154% in the number of residential facilities since the 1965 listing is due in part to a notable increase in the number of private facilities in this stream of care. Although the majority of juvenile justice facilities are still public, there has been an increase of 17% since 1965 in the proportion of facilities under private auspices. An even greater increase in total number of facilities than was seen in the Juvenile Justice Stream is seen in the mental health field between 1965 and 1981. There were almost four times as many residential mental health facilities listed for survey in 1981 as in 1965. From representing about one-eighth of the total number of all residential group care facilities for children and youth with special problems or needs, the Mental Health Stream increased to a quarter of all such facilities by 1981. As in the earlier study, the Mental Health Stream was dominated by the private sector in 1981. The proportion of mental health facilities under private auspices had increased by 16% since 1965, indicating an even greater involvement of the private sector in the delivery of residential mental health services to children and youth.(ABSTRACT TRUNCATED AT 400 WORDS)     

The reflecting team as a reflection of second order therapeutic ideals

The reflecting (R.T.) team is examined within its theoretical, therapeutic and training contexts. A distinction is made between the reflecting team technique (R.T.T.) and the reflecting team process (R.T.P.), in order to explore whether or not the practice of team reflection is consistent with the requirements of the second order family therapies, the theoretical development which supported the R.T.'s emergence and subsequent popularity. The critique of the R.T.P. suggests that although this therapeutic process does represent attempts towards increasing acceptance of complexity, reduced attempts to control, collaboration, interconnectedness and contextual respect, there is little room for complacency either theoretically or practically.     

Household Debt and Marital Instability: Evidence from the Korean Labor and Income Panel Study

Korean society has recently experienced rapid increases in household debt and divorce rates. This study investigates whether household debt and debt ratios lower marital stability and increase the probability of divorce among Korean families. Six-year panels from the Korean Labor and Income Panel Study (KLIPS) household surveys were used for analysis. The effect of household debt upon the probability of divorce was estimated through pooled and population-averaged Logit models. The results suggest that household financial strain measured by the amount of household debt and the debt-to-income ratios does not significantly affect the probability of divorce. This finding contradicts the widespread notion that households’ excessive borrowing has been partially responsible for the recent increase in divorce rates in Korea.

Prevalence and Diagnostic Correlates of DSM-IV Pathological Gambling in Psychiatric Outpatients

Studies of the prevalence of pathological gambling (PG) in psychiatric and substance abusing patients suggest that the disorder is not rare. Most studies have been of substance abusers in treatment, and the rate of PG has been found to be several times higher than the rate found in community based epidemiological surveys. However, only one study has examined the prevalence of PG in a heterogeneous sample of patients, and this was a study of psychiatric inpatients. We are not aware of any prior study of the prevalence of PG in a psychiatric outpatient sample. In the present report from the Rhode Island Methods to Improve Diagnosis and Services (MIDAS) project we examined the current and lifetime prevalence of PG in 1,709 psychiatric outpatients interviewed with a semi-structured diagnostic interview that included a module to diagnose DSM-IV PG. Forty (2.3%) patients had a lifetime history of DSM-IV PG, all of whom had at least one other DSM-IV axis I disorder. Patients with PG had significantly more axis I disorders than patients without PG, and had significantly higher rates of bipolar disorder, social phobia, panic disorder with agoraphobia, alcohol use disorder, and other impulse control disorders. Possible reasons for the low prevalence of PG in our sample are discussed.     

Ridge Regression Estimation for Survey Samples

This paper describes procedure for constructing a vector of regression weights. Under the regression superpopulation model, the ridge regression estimator that has minimum model mean squared error is derived. Through a simulation study, we compare the ridge regression weights, regression weights, quadratic programming weights, and raking ratio weights. The ridge regression procedure with weights bounded by zero performed very well.     

Decomposition of Scatter Ratios Used in Monitoring Multivariate Process Variability

Wilks’ ratio statistic can be defined in terms of the ratio of the sample generalized variances of two non-independent estimators of the same covariance matrix. Recently this statistic has been proposed as a control statistic for monitoring changes in the covariance matrix of a multivariate normal process in a Phase II situation, particularly when the dimension is larger than the sample size. In this article we derive a technique for decomposing Wilks’ ratio statistic into the product of independent factors that can be associated with the components of the covariance matrix. With these results, we demonstrate that, when a signal is detected in a control procedure for the Phase II monitoring of process variability using the ratio statistic, the signaling value can be decomposed and the process variables contributing to the signal can be specifically identified.     

Detection and Interpretation of a Multivariate Signal Using Combined Charts

A control procedure is presented in this article that is based on jointly using two separate control statistics in the detection and interpretation of signals in a multivariate normal process. The procedure detects the following three situations: (i) a mean vector shift without a shift in the covariance matrix; (ii) a shift in process variation (covariance matrix) without a mean vector shift; and (iii) both a simultaneous shift in the mean vector and covariance matrix as the result of a change in the parameters of some key process variables. It is shown that, following the occurrence of a signal on either of the separate control charts, the values from both of the corresponding signaling statistics can be decomposed into interpretable elements. Viewing the two decompositions together helps one to specifically identify the individual components and associated variables that are being affected. These components may include individual means or variances of the process variables as well as the correlations between or among variables. An industrial data set is used to illustrate the procedure.     

Quantile Regression with Left-Truncated and Right-Censored Data in a Reproducing Kernel Hilbert Space

Li et al. (2007 Li, Y., Liu, Y., Zhu, J. (2007). Quantile regression in reproducing kernel Hilbert spaces. J. Amer. Statist. Assoc. 102:255–268.[Taylor & Francis Online], [Web of Science ®] , [Google Scholar]) developed an estimation method for quantile functions in a reproducing kernel Hilbert space for complete data, and Park and Kim (2011 Park, J., Kim, J. (2011). Quantile regression with an epsilon-insensitive loss in a reproducing kernel Hilbert space. Statist. Probab. Lett. 81:62–70.[Crossref], [Web of Science ®] , [Google Scholar]) proposed an estimation method using the ε-insensitive loss. This article extends these estimation methods to left-truncated and right-censored data. As a measure of goodness of fit, the check loss and the ε-insensitive loss were used to estimate the quantile function. The ε-insensitive loss can shrink the estimated coefficients toward zero; hence, it can reduce the variability of the estimates. Simulation studies show that the estimated quantile functions based on the ε-insensitive loss perform slightly better when ε is adequately chosen.