统计数据的使用与误用

不要依赖统计分析,除非你了解其中的陷阱缺陷比率正在下降吗?顾客满意度在提高吗?你查看数字,却并不确定是否该相信它们。你并不担心存在欺骗或操控,而是不知道在多大程度上信任统计数据。谨慎从事,是有道理的。哈佛商业院     

The minimum regularized covariance determinant estimator

Statistics and Computing - The minimum covariance determinant (MCD) approach estimates the location and scatter matrix using the subset of given size with lowest sample covariance determinant. Its...     

Youth Justice

This paper illustrates how mindfulness and reflection exercises can be integrated into the teaching and learning of active listening skills, an integral component of relationship skills training. This pedagogy was first trialled on a small cohort of 15 students and the improved version was applied with a second cohort of 53 students. Through the mindfulness and reflection exercises which were built in throughout the module, students identified the three most prominent ‘bad habits’ that hindered active listening as ‘mind wandering’, ‘multi-tasking’ while listening and ‘thinking ahead’. The mindfulness and reflection exercises appeared to bring the usually-unnoticed ‘bad habits’ into awareness so that a conscious effort could be channelled into preventing them from interfering with relationship building. This expanded awareness facilitated the strengthening of ability to observe oneself in interactions. This paper proposes a combination of exercises that expand students' awareness of their own habitual listening styles, together with skills training through conventional methods of role play, video-taping and skills drilling, in order to strengthen motivation to change and bring about deeper learning.     

What do we mean by evaluation?

The popularity of evaluation research has increased markedly in recent years. Although initially associated with experimental and quasi‐experimental methods, current practices and methods have become much more diverse. This paper provides a brief overview of the recent history of evaluation research – its meanings and purposes – and asks whether the current emphasis on ‘evidence‐based policy’ heralds a brighter future. Copyright © 2001 John Wiley & Sons, Ltd.     

A comparison of subgroup identification methods in clinical drug development: Simulation study and regulatory considerations

With advancement of technologies such as genomic sequencing, predictive biomarkers have become a useful tool for the development of personalized medicine. Predictive biomarkers can be used to select subsets of patients, which are most likely to benefit from a treatment. A number of approaches for subgroup identification were proposed over the last years. Although overviews of subgroup identification methods are available, systematic comparisons of their performance in simulation studies are rare. Interaction trees (IT), model‐based recursive partitioning, subgroup identification based on differential effect, simultaneous threshold interaction modeling algorithm (STIMA), and adaptive refinement by directed peeling were proposed for subgroup identification. We compared these methods in a simulation study using a structured approach. In order to identify a target population for subsequent trials, a selection of the identified subgroups is needed. Therefore, we propose a subgroup criterion leading to a target subgroup consisting of the identified subgroups with an estimated treatment difference no less than a pre‐specified threshold. In our simulation study, we evaluated these methods by considering measures for binary classification, like sensitivity and specificity. In settings with large effects or huge sample sizes, most methods perform well. For more realistic settings in drug development involving data from a single trial only, however, none of the methods seems suitable for selecting a target population. Using the subgroup criterion as alternative to the proposed pruning procedures, STIMA and IT can improve their performance in some settings. The methods and the subgroup criterion are illustrated by an application in amyotrophic lateral sclerosis.     

Closing the Loophole: A Case Study of Organizing for More Equitable and Affordable Access to Health Care in San Francisco

This paper presents in-depth case study of a successful hybrid political and community organizing campaign to ensure equitable access to health care through the perspective of a grassroots San Francisco community-based organization, the Chinese Progressive Association (CPA), which has been organizing low-income Chinese immigrants for over four decades. First, it outlines the Health Care Security Ordinance (HCSO), which, since its passage in 2006, has established a near-universal health care access program, helping to make health care accessible and affordable to individuals living and working in San Francisco. Then it presents the campaign to save the HCSO, focusing on CPA’s participation in the HCSO coalition. Finally, it discusses health care as it relates to the San Francisco’s affordability crisis and the political economic context in which it is taking place. Despite the limitations inherent in small case studies like this one, it nevertheless provides a valuable opportunity to better understand how one politically progressive city attempted to address the problem of grossly inequitable health care access through the lens of community organizing, advocacy, and coalition building. San Francisco, like many major American cities today, is being confronted with rapid gentrification and growing economic inequality—the backdrop to the HCSO. Through innovative experiments in social responsibility like the HCSO, however, the city has made leaps in health care access. It concludes with lessons learned from local organizing and advocacy to save the HCSO as these may inform other local efforts to promote health care for all.     

The Message in Daily Exchange Rates: A Conditional-Variance Tale

Formal testing procedures confirm the presence of a unit root in the autoregressive polynomial of the univariate time series representation of daily exchange-rate data. The first differences of the logarithms of daily spot rates are approximately uncorrelated through time, and a generalized autoregressive conditional heteroscedasticity model with daily dummy variables and conditionally t-distributed errors is found to provide a good representation to the leptokurtosis and time-dependent conditional heteroscedasticity. The parameter estimates and characteristics of the models are found to be very similar for six different currencies. These apparent stylized facts carry over to weekly, fortnightly, and monthly data in which the degree of leptokurtosis and time-dependent heteroscedasticity is reduced as the length of the sampling interval increases.