A note on using the F-measure for evaluating record linkage algorithms

Record linkage is the process of identifying and linking records about the same entities from one or more databases. Record linkage can be viewed as a classification problem where the aim is to decide whether a pair of records is a match (i.e. two records refer to the same real-world entity) or a non-match (two records refer to two different entities). Various classification techniques—including supervised, unsupervised, semi-supervised and active learning based—have been employed for record linkage. If ground truth data in the form of known true matches and non-matches are available, the quality of classified links can be evaluated. Due to the generally high class imbalance in record linkage problems, standard accuracy or misclassification rate are not meaningful for assessing the quality of a set of linked records. Instead, precision and recall, as commonly used in information retrieval and machine learning, are used. These are often combined into the popular F-measure, which is the harmonic mean of precision and recall. We show that the F-measure can also be expressed as a weighted sum of precision and recall, with weights which depend on the linkage method being used. This reformulation reveals that the F-measure has a major conceptual weakness: the relative importance assigned to precision and recall should be an aspect of the problem and the researcher or user, but not of the particular linkage method being used. We suggest alternative measures which do not suffer from this fundamental flaw.     

A case study in identifying targeted patients population in major depressive disorder by enhanced enrichment design

Despite advances in clinical trial design, failure rates near 80% in phase 2 and 50% in phase 3 have recently been reported. The challenges to successful drug development are particularly acute in central nervous system trials such as for pain, schizophrenia, mania, and depression because high‐placebo response rates lessen assay sensitivity, diminish estimated treatment effect sizes, and thereby decrease statistical power. This paper addresses the importance of rigorous patient selection in major depressive disorder trials through an enhanced enrichment paradigm. This approach led to a redefinition of an ongoing, blinded phase 3 trial algorithm for patient inclusion (1) to eliminate further randomization of transient placebo responders and (2) to exclude previously randomized transient responders from the primary analysis of the double blind phase of the trial. It is illustrated for a case study for the comparison between brexpiprazole + antidepressant therapy and placebo + antidepressant therapy. Analysis of the primary endpoint showed that efficacy of brexpiprazole versus placebo could not be established statistically if the original algorithm for identification of placebo responders was used, but the enhanced enrichment approach did statistically demonstrate efficacy. Additionally, the enhanced enrichment approach identified a target population with a clinically meaningful treatment effect. Through its successful identification of a target population, the innovative enhanced enrichment approach enabled the demonstration of a positive treatment effect in a very challenging area of depression research.     

Getting back in the DeLorean: modernization vs. anti-modernization in contemporary British politics

ABSTRACT

This article highlights a growing clash between mainstream modernizing and populist anti-modernizing forces in the UK. Whilst scholarship on UK party politics has, for the past three decades, focussed on processes of party modernization, little attention has been paid to the countervailing processes of resistance towards modernization. This contrasts with comparative studies, which show that throughout much of Europe modernization processes have worked to produce populist backlashes from anti-modernizing forces seeking to reassert various types of traditional values and practices. Drawing on the comparative literature on modernization, our argument here is that a similar populist backlash against modernization is occurring across the political spectrum in the UK and has been a factor in: the rise of nationalism associated with UKIP and the SNP; the 2016 BREXIT vote and the efforts of both Labour leader Jeremy Corbyn and Conservative PM Theresa May to distance their parties from the legacies of their modernizing predecessors. Empirically, we show how this increasingly prevalent line of conflict is playing out in the UK, whilst theoretically we argue that a discourse theoretical approach can provide significant advantages over existing approaches for understanding the dynamic interplay between modernizing and anti-modernizing discourses.     

A Hybrid Geometric Phase II/III Clinical Trial Design based on Treatment Failure Time and Toxicity

The problem of comparing several experimental treatments to a standard arises frequently in medical research. Various multi-stage randomized phase II/III designs have been proposed that select one or more promising experimental treatments and compare them to the standard while controlling overall Type I and Type II error rates. This paper addresses phase II/III settings where the joint goals are to increase the average time to treatment failure and control the probability of toxicity while accounting for patient heterogeneity. We are motivated by the desire to construct a feasible design for a trial of four chemotherapy combinations for treating a family of rare pediatric brain tumors. We present a hybrid two-stage design based on two-dimensional treatment effect parameters. A targeted parameter set is constructed from elicited parameter pairs considered to be equally desirable. Bayesian regression models for failure time and the probability of toxicity as functions of treatment and prognostic covariates are used to define two-dimensional covariate-adjusted treatment effect parameter sets. Decisions at each stage of the trial are based on the ratio of posterior probabilities of the alternative and null covariate-adjusted parameter sets. Design parameters are chosen to minimize expected sample size subject to frequentist error constraints. The design is illustrated by application to the brain tumor trial.     

Efficient likelihood-free Bayesian Computation for?household epidemics

Considerable progress has been made in applying Markov chain Monte Carlo (MCMC) methods to the analysis of epidemic data. However, this likelihood based method can be inefficient due to the limited data available concerning an epidemic outbreak. This paper considers an alternative approach to studying epidemic data using Approximate Bayesian Computation (ABC) methodology. ABC is a simulation-based technique for obtaining an approximate sample from the posterior distribution of the parameters of the model and in an epidemic context is very easy to implement. A new approach to ABC is introduced which generates a set of values from the (approximate) posterior distribution of the parameters during each simulation rather than a single value. This is based upon coupling simulations with different sets of parameters and we call the resulting algorithm coupled ABC. The new methodology is used to analyse final size data for epidemics amongst communities partitioned into households. It is shown that for the epidemic data sets coupled ABC is more efficient than ABC and MCMC-ABC.     

Psychosocial adversities at work are associated with poorer quality marriage-like relationships

The characteristics of work have an impact on individuals?? personal lives, including on health and close personal relationships. Previous research into the negative spillover effects from work to marital quality has predominantly examined structural job characteristics, such as job hours and shift work. This study used data from a large representative community sample of midlife (aged 40?C44 at baseline) employed persons in marriage-like relationships (n?=?2,054) to investigate the relationship between psychosocial job characteristics and relationship quality. Data from three waves of the Personality and Total Health (PATH) Through Life Survey were analysed. The results show that all psychosocial job adversities (high job demands, low job control, and job insecurity) were independently associated with lower levels of positive support from partners for both men and women. In addition, there was a linear relationship between the number of psychosocial adversities reported and lower levels of positive support from partners. These findings are relevant to policy makers in the areas of employment and family services. They identify the potential broader social costs of adverse psychosocial characteristics at work. This study sets the scene for further longitudinal research to examine the causal links between psychosocial job quality and relationship quality.     

Schulleitungscoaching – Konzeption und Ausbildungsevaluation

Since principals in schools are increasingly turning into managers of educational institutions, coaching approaches have an increasing importance. Accordingly, the Bavarian Ministry for Education and Culture in cooperation with the National Institute for Quality in Schools and Educational Research and the Academy of Teacher Education and Personnel Management has introduced the project ?Coaching – an Instrument to Professionalise School-based Leadership“. The project involves exemplary coachings of school management teams and the training of coaches with a focus on school-related issues at the University of Innsbruck. The training of coaches is based on a special curriculum, which was evaluated using a multi-methods approach. The outcomes show that the coaching model is accepted by coaches and that the training is of high quality     

A Survey of Campus Peer Counseling Activities

Abstract

The directors of counseling services at 200 colleges and universities were contacted by mail and asked to complete a survey regarding peer counseling activities on their campuses. One hundred fifty-six responses to this questionnaire were received; one hundred twenty-two indicated ongoing peer counseling activities in a wide variety of settings at their respective campuses. The most common client concerns confronted by peer counselors involved academic difficulties and relationships with friends and lovers. The survey identified a variety of peer counseling training programs, from one-weekend, intensive encounter-type experiences to full-year credit courses in counseling and therapy. Also examined were the funding resources available for peer counseling centers, which often proved quite marginal, and the amount of interaction between student counselors and professional mental health service staff members. Although the authors have been able to uncover a great many peer counseling programs on college campuses, there is a dearth of systematic research evaluating the effectiveness of the services provided by these centers or the appropriateness of various peer counselor training programs. Such work should be encouraged.     

Getting Unstuck: Interdisciplinarity and Aging

Continuing the dialog initiated by Scheff in the March issue of Sociological Forum, we reflect on some of the challenges involved in interdisciplinary research.