The social construction of a possible male gender bias is negatively impacting fathers in child welfare services: Hammering a square peg into a round hole

The social construction of gender dictates the masculine and feminine roles deemed appropriate within a society. Society has traditionally socially constructed women to be mothers and the primary parent, leaving fathers much too often on the sidelines of parenthood. For example, if a parent is a man, they have fewer rights than a parent who is a woman. Gender biases in society are insidious problems. Such biases can lead to the development of adverse attitudes, beliefs and unfair treatment of members of these groups by both individuals and social institutions. This paper will demonstrate the need to investigate the possibility of a systemic male gender bias in CWS. The paper will include the statement of the problem, the rationale and the framework of how gender is socially constructed to help in the understanding of how it negatively impacts fathers. Next will be a discussion of avoiding fathers at all six principle phases of CWS, the lack of parent programs specific to fathers, direction for future research and potential research implications for fathers, children and child welfare workers.     

Themis-ml: A Fairness-Aware Machine Learning Interface for End-To-End Discrimination Discovery and Mitigation

ABSTRACT

As more industries integrate machine learning into socially sensitive decision processes like hiring, loan-approval, and parole-granting, we are at risk of perpetuating historical and contemporary socioeconomic disparities. This is a critical problem because on the one hand, organizations who use but do not understand the discriminatory potential of such systems will facilitate the widening of social disparities under the assumption that algorithms are categorically objective. On the other hand, the responsible use of machine learning can help us measure, understand, and mitigate the implicit historical biases in socially sensitive data by expressing implicit decision-making mental models in terms of explicit statistical models. In this article we specify, implement, and evaluate a “fairness-aware” machine learning interface called themis-ml, which is intended for use by individual data scientists and engineers, academic research teams, or larger product teams who use machine learning in production systems.     

Behavioral economic phenomena in decision-making for others

We examine whether biases identified in the behavioral-economics literature apply in decision-making for others (DMfO). We conduct a laboratory experiment in which subjects make decisions on behalf of themselves and others in eighteen tasks that measure the following biases: present-bias in time preferences, reflection effect in risk preferences, compound risk aversion, decoy effect, anchoring bias, endowment effect, and identifiable-victim bias. In our experiment, DMfO is DMfO simpliciter: unincentivized decisions made by one individual on behalf of another–the individual making decisions faces no direct costs or benefits when engaging in DMfO (as she would in a principal-agent framework or with bequest motives), and DMfO is not framed as giving advice or guessing others’ behavior. Although we find that DMfO is by and large statistically indistinguishable from decisions for oneself, we identify the following self-other discrepancies: (i) willingness to pay (i.e., bids to procure goods and donations to charity) is higher in DMfO than in decisions for oneself in tasks associated with the anchoring bias, end owment effect, and identifiable-victim bias; and (ii) the propensity to give uninterpretable responses is higher in DMfO than in decisions for oneself. We also find order effects, with DMfO more similar to decisions for oneself when DMfO follows decision making for oneself. Lastly, in response to open-ended items soliciting self-reports of subjects’ approach to DMfO, most subjects report having followed some version of the “Golden Rule” (e.g., deciding for others as they would for themselves) or having tried to maximize the other subject’s payment or utility; very few subjects report motivations that can be construed as rivalrous.     

Antiblackness as a Logic for Anti-Immigrant Resentment: Evidence From California

Drawing on data from a novel survey of Californians, this article examines the relationship between anti-immigrant resentment and antiblackness. Over the last decade, there has been a significant increase in research on anti-immigrant resentment, as well as a steady growth in public opinion literature on antiblack bias. Nonetheless, there are almost no studies that examine the relationship between these types of attitudes. Using two different measures of anti-immigrant resentment and three different measures of antiblack bias, we find that antiblackness among those Californians surveyed was significantly and robustly correlated with anti-immigrant resentment. Our findings suggest that even in a state where there is a relatively small black population and where there is little overlap between the categories of “black” and “immigrant,” attitudes toward blacks may be shaping attitudes toward immigrants. The findings suggest a need to move beyond dyadic theories of othering and in doing so, to think more critically about how the racialization of one group can shape the racialization of other groups.     

Twicing Kernels and a Small Bias Property of Semiparametric Estimators

The purpose of this note is to show how semiparametric estimators with a small bias property can be constructed. The small bias property (SBP) of a semiparametric estimator is that its bias converges to zero faster than the pointwise and integrated bias of the nonparametric estimator on which it is based. We show that semiparametric estimators based on twicing kernels have the SBP. We also show that semiparametric estimators where nonparametric kernel estimation does not affect the asymptotic variance have the SBP. In addition we discuss an interpretation of series and sieve estimators as idempotent transformations of the empirical distribution that helps explain the known result that they lead to the SBP. In Monte Carlo experiments we find that estimators with the SBP have mean‐square error that is smaller and less sensitive to bandwidth than those that do not have the SBP.     

The Role of Performance Assessment in the Regulation of Underground Disposal of Radioactive Wastes: An International Perspective

Much has been written about the development and application of quantitative methods for estimating under uncertainty the long-term radiological performance of underground disposal of radioactive wastes. Until recently, interest has been focused almost entirely on the technical challenges regardless of the role of the organization responsible for these analyses. Now the dialogue between regulators, the repository developer or operator, and other interested parties in the decision-making process receives increasing attention, especially in view of some current difficulties in obtaining approvals to construct or operate deep facilities for intermediate or high-level wastes. Consequently, it is timely to consider the options for regulators' review and evaluation of safety submissions, at the various stages in the site selection to repository closure process, and to consider, especially, the role for performance assessment (PA) within the programs of a regulator both before and after delivery of such a submission. The origins and broad character of present regulations in the European Union (EU) and in the OECD countries are outlined and some regulatory PA reviewed. The issues raised are discussed, especially in regard to the interpretation of regulations, the dangers from the desire for simplicity in argument, the use of regulatory PA to review and challenge the PA in the safety case, and the effects of the relationship between proponent and regulator. Finally, a very limited analysis of the role of PA in public hearings is outlined and recommendations are made, together with proposals for improving the mechanisms for international collaboration on technical issues of regulatory concern.Until August 13, 1997, the author was employed as a Principal Pollution Inspector by the Environment Agency (EA) of England and Wales. Responsible for managing the development and application of the regulatory postclosure assessment capability of Her Majesty's Inspectorate of Pollution (HMIP) between 1982 and 1996, when HMIP was merged with the National Rivers Authority and the Waste Regulatory Authorities to form the EA on April 1, 1996. Represented HMIP/EA on a number of international committees related particularly to PA; 20 Bonser Road, Twickenham, Middlesex TW1 4RG, U.K.     

A statistical framework for ecological and aggregate studies

Inference from studies that make use of data at the level of the area, rather than at the level of the individual, is more difficult for a variety of reasons. Some of these difficulties arise because frequently exposures (including confounders) vary within areas. In the most basic form of ecological study the outcome measure is regressed against a simple area level summary of exposure. In the aggregate data approach a survey of exposures and confounders is taken within each area. An alternative approach is to assume a parametric form for the within-area exposure distribution. We provide a framework within which ecological and aggregate data studies may be viewed, and we review some approaches to inference in such studies, clarifying the assumptions on which they are based. General strategies for analysis are provided including an estimator based on Monte Carlo integration that allows inference in the case of a general risk–exposure model. We also consider the implications of the introduction of random effects, and the existence of confounding and errors in variables.     

Local sensitivity approximations for selectivity bias

Observational data analysis is often based on tacit assumptions of ignorability or randomness. The paper develops a general approach to local sensitivity analysis for selectivity bias, which aims to study the sensitivity of inference to small departures from such assumptions. If M is a model assuming ignorability, we surround M by a small neighbourhood N defined in the sense of Kullback–Leibler divergence and then compare the inference for models in N with that for M . Interpretable bounds for such differences are developed. Applications to missing data and to observational comparisons are discussed. Local approximations to sensitivity analysis are model robust and can be applied to a wide range of statistical problems.     

Properties of Model-Averaged BMDLs: A Study of Model Averaging in Dichotomous Response Risk Estimation

Model averaging (MA) has been proposed as a method of accounting for model uncertainty in benchmark dose (BMD) estimation. The technique has been used to average BMD dose estimates derived from dichotomous dose-response experiments, microbial dose-response experiments, as well as observational epidemiological studies. While MA is a promising tool for the risk assessor, a previous study suggested that the simple strategy of averaging individual models' BMD lower limits did not yield interval estimators that met nominal coverage levels in certain situations, and this performance was very sensitive to the underlying model space chosen. We present a different, more computationally intensive, approach in which the BMD is estimated using the average dose-response model and the corresponding benchmark dose lower bound (BMDL) is computed by bootstrapping. This method is illustrated with TiO(2) dose-response rat lung cancer data, and then systematically studied through an extensive Monte Carlo simulation. The results of this study suggest that the MA-BMD, estimated using this technique, performs better, in terms of bias and coverage, than the previous MA methodology. Further, the MA-BMDL achieves nominal coverage in most cases, and is superior to picking the "best fitting model" when estimating the benchmark dose. Although these results show utility of MA for benchmark dose risk estimation, they continue to highlight the importance of choosing an adequate model space as well as proper model fit diagnostics.     

关于布迪厄反思性的几个问题

对当代法国著名社会学家皮埃尔·布迪厄反思性的目的、对象和特点 ,其反思性与其他学者的反思性的区别 ,其反思性的实现方式及实现的障碍 ,其反思性所针对的三种偏见等四个问题进行了分析和探讨 ,以期对其反思性有一个准确的理解和把握