This paper addresses the problem of estimating the population mean at the current occasion in two occasion successive sampling when non-response occurs on the current (second) occasions. Using the power transformation we have suggested classes of estimators of current population mean and their properties are studied. Optimum replacement strategies for the proposed estimators have been given and empirical studies are carried out to assess the performance of estimators. We have made suitable recommendation to the practitioners on the basis of the empirical study.
In surveys covering human populations it is observed that information in most cases are not obtained at the first attempt even after some callbacks. Such problems come under the category of non-response. Surveys suffer with non-response in various ways. It depends on the nature of required information, either surveys is concerned with general or sensitive issues of a society. Hansen and Hurwitz (1946) have considered the problem of non-response while estimating the population mean by taking a subsample from the non-respondent group with the help of extra efforts and an estimator was suggested by combining the information available from the response and nonresponse groups. We also mention that in survey sampling auxiliary information is commonly used to improve the performance of an estimator of a quantity of interest. For estimating the population mean using auxiliary information in presence of non-response has been discussed by various authors. In this paper, we have developed estimators for estimating the population mean of the variable under interest when there is non-response error in the study as well as in the auxiliary variable. We have studied properties of the suggested estimators under large sample approximation. Comparison of the suggested estimators with usual unbiased estimator reported by Hansen and Hurwitz (1946) and the ratio estimator due to Rao (1986) have been made. The results obtained are illustrated with aid of an empirical study.
In this paper we have suggested a family of estimators of the population mean using auxiliary information in sample surveys. The bias and mean squared error of the proposed class of estimators have been obtained under large sample approximation. We have derived the conditions for the parameters under which the proposed class of estimators has smaller mean squared error than the sample mean, ratio, product, regression estimator and the two parameter ratio-product-ratio estimators envisaged by Chami et al (2012). An empirical study is carried out to demonstrate the performance of the proposed class of estimators over other existing estimators.
Gupta et al (2002) suggested an optional randomized response model under the assumption that the mean of the scrambling variable S is ‘unity’ [i.e. µs = 1]. This assumption limits the use of Gupta et al’s (2002) randomized response model. Keeping this in view we have suggested a modified optional randomized response model which can be used in practice without any supposition and restriction over the mean (µs) of the scrambling variables S. It has been shown that the estimator of the mean of the stigmatized variable based on the proposed optional randomized response sampling is more efficient than the Eicchorn and Hayre (1983) procedure and Gupta et al’s (2002) optional randomized technique when the mean of the scrambling S is larger than unity [i.e. µs > 1]. A numerical illustration is given in support of the present study.
Louise Hjort Nielsen, Sarah van Mastrigt, Randy K. Otto, Katharina Seewald, Corine de Ruiter, Martin Rettenberger, Kim A. Reeves, Maria Fransisca Rebocho, Thierry H. Pham, Robyn Mei Yee Ho, Martin Grann, Verónica Godoy-Cervera, Jorge O. Folino, Michael Doyle, Sarah L. Desmarais, Carolina Condemarin, Karin Arbach-Lucioni and Jay P. Singh
With a quadrupling of forensic psychiatric patients in Denmark over the past 20 years, focus on violence risk assessment practices across the country has increased. However, information is lacking regarding Danish risk assessment practice across professional disciplines and clinical settings; little is known about how violence risk assessments are conducted, which instruments are used for what purposes, and how mental health professionals rate their utility and costs. As part of a global survey exploring the application of violence risk assessment across 44 countries, the current study investigated Danish practice across several professional disciplines and settings in which forensic and high-risk mental health patients are assessed and treated. In total, 125 mental health professionals across the country completed the survey. The five instruments that respondents reported most commonly using for risk assessment, risk management planning and risk monitoring were Broset, HCR-20, the START, the PCL-R, and the PCL:SV. Whereas the HCR-20 was rated highest in usefulness for risk assessment, the START was rated most useful for risk management and risk monitoring. No significant differences in utility were observed across professional groups. Unstructured clinical judgments were reported to be faster but more expensive to conduct than using a risk assessment instrument. Implications for clinical practice are discussed.