Perturbed robust linear estimating equations for confidentiality protection in remote analysis

National statistical agencies and other data custodians collect and hold a vast amount of survey and census data, containing information vital for research and policy analysis. However, the problem of allowing analysis of these data, while protecting respondent confidentiality, has proved challenging to address. In this paper we will focus on the remote analysis approach, under which a confidential dataset is held in a secure environment under the direct control of the data custodian agency. A computer system within the secure environment accepts a query from an analyst, runs it on the data, then returns the results to the analyst. In particular, the analyst does not have direct access to the data at all, and cannot view any microdata records. We further focus on the fitting of linear regression models to confidential data in the presence of outliers and influential points, such as are often present in business data. We propose a new method for protecting confidentiality in linear regression via a remote analysis system, that provides additional confidentiality protection for outliers and influential points in the data. The method we describe in this paper was designed for the prototype DataAnalyser system developed by the Australian Bureau of Statistics, however the method would be suitable for similar remote analysis systems.     

Sensitive topics in social research involving children

While there is broad consensus that involving children in research is critically important for understanding their lives, there is considerable caution around their participation in social research on ‘sensitive’ issues. Such caution is further amplified by a lack of agreement about what constitutes a ‘sensitive topic’. This article draws on data from interviews with a range of Australian stakeholders, including children, parents, researchers, ethics committee members, government and non-government organisation representatives, and other decision-makers, specifically related to the question of ‘what is a sensitive topic?’. While participants identified a range of sensitive topics, findings point to sensitivities being closely linked with the contexts of children’s lives and experiences. This requires researchers to approach research in ways that reflect more nuanced understandings of these sensitivities, help address potential concerns and facilitate the development of research relationships that promote ethical conduct of research with children.     

Analysing the effects of tax-benefit reforms on income distribution: a decomposition approach

To assess the impact of tax-benefit policy changes on income distribution over time, we suggest a methodology based on counterfactual tax-benefit simulations. Changes in inequality/poverty indices are decomposed into three contributions: changes in the tax-benefit structure, changes in nominal levels of market incomes and tax-benefit money parameters, and all other changes, including shifts in market income inequality and demographic composition. The policy effect can be evaluated conditionally on base-period data or end-period data; it is also possible to average the two measures, which corresponds to an application of the Shapley value method as reinterpreted by Shorrocks (Decomposition Procedures for Distributional Analysis: A Unified Framework Based on the Shapley Value, University of Essex and Institute for Fiscal Studies, Wivenhoe Park, 1999). The decomposition is used to quantify the relative role of policy changes on inequality/poverty trends in France and Ireland in the 1990s. When end-period data are not available, e.g., for forward looking analysis of possible reforms, the base weighted decomposition helps to extract an absolute measure of the impact of tax-benefit changes on income distribution as evaluated against a distributionally neutral benchmark; in our application, it is not significantly different from the policy effect stemming from the Shorrocks-Shapley decomposition. Estimates of this type are derived to assess recent policy changes in twelve European countries.     

Care proceedings and parents with learning difficulties: comparative prevalence and outcomes in an English and Australian court sample

This paper presents and contrasts the findings from two independent but comparable court studies in England and Australia of care proceedings involving parents with learning difficulties. Significant differences between the two jurisdictions were found in both prevalence and outcomes, with a much higher proportion of cases showing in the England sample, with many more children being returned home to live with their parents in Australia, and with far more children being placed out‐of‐home and outside the family network in England. We explore the reasons for these differences and discuss the implications of the findings.     

Optimal Experiments for Time-Dependent Mineral Processes

Time dependence is an important characteristic of mineral processing plant data. This paper finds that the time dependence in the recovery data for an experiment at Bougainville Copper Limited (BCL) (Napier-Munn, 1995) can be described by an autoregressive-one process. The paper investigates the optimum form of experimental design for such data. Two intuitive approaches for the design of experiments involving time-dependent data have been disproved recently. Cheng & Steinberg (1991) showed that in some circumstances systematic experiments are preferable to replicated randomized block designs, and Saunders & Eccleston (1992) showed that rather than sampling at a frequency which ensures independent data, in some circumstances sampling intervals should be as small as possible. A third issue, raised in this paper, concerns the use of standard statistical tests when the data are serially correlated. It is shown that the simple paired t -test, suitably modified for time dependence, is appropriate and easily adapted to allow for a covariate and a sequential analysis. The results are illustrated using the BCL data and are already being used to design major experiments involving another mineral process.     

Going with the Grain in African Development?

In view of the disappointing performance of the Good Governance agenda in Africa, policy‐makers are asking whether better results could be obtained with approaches that attempt to ‘work with the grain’ of African societies. This article explores what this might mean. It identifies a core set of beliefs and values – concerning power, accountability and social morality – that have proved extremely durable and remain powerful drivers of behaviour across sub‐Saharan Africa. It finds that, in general, Western institutions sit ill with these traditions. The question is, therefore, how to redirect development efforts so that they stop working against, and start to build upon, the extant notions of moral obligation and interpersonal accountability in the region.