Change we can believe in: Comparing longitudinal network models on consistency,interpretability and predictive power

While several models for analysing longitudinal network data have been proposed, their main differences, especially regarding the treatment of time, have not been discussed extensively in the literature. However, differences in treatment of time strongly impact the conclusions that can be drawn from data. In this article we compare auto-regressive network models using the example of TERGMs – a temporal extensions of ERGMs – and process-based models using SAOMs as an example. We conclude that the TERGM has, in contrast to the ERGM, no consistent interpretation on tie-level probabilities, as well as no consistent interpretation on processes of network change. Further, parameters in the TERGM are strongly dependent on the interval length between two time-points. Neither limitation is true for process-based network models such as the SAOM. Finally, both compared models perform poorly in out-of-sample prediction compared to trivial predictive models.     

Chlamydia and gonorrhea screening and expedited partner therapy (EPT) practices of college student health centers

Objective: To examine college student health centers' (SHCs) practices related to sexually transmitted disease (STD) screening and treatment over a 5-year period. Participants: College SHCs that completed the ACHA Pap and STI Survey between 2010 and 2014. Methods: Chi-square tests were conducted with Cramer's V providing a measure of association. Results: Chlamydia screening of women under 25 years of age was a well-established practice. Almost one-third of SHCs did not report screening practices of men who have sex with men (MSM) consistent with current chlamydia and gonorrhea guidelines; however, there was improvement over time. Few college SHCs utilized expedited partner therapy (EPT), though fewer had blanket restrictions prohibiting its use in 2014 compared to 2010. Conclusions: Over the 5-year period, more SHCs followed current best practices. Improvement in compliance with guidelines related to MSM STD screening as well as increased usage of EPT is needed to best address the needs of the diverse college student population.     

A DSM-5-Based Online Mental Health Referral Inventory: A Large-Scale Validation Study

In 2007, a comprehensive mental health referral inventory based on the 4th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) was posted online as an alternative to the informal tests the public now uses to self-diagnose mental health problems. A validation study was published in 2011 (Epstein &; Muzzatti, 2011 Epstein, R., &; Muzzatti, L. (2011). Preliminary validation of an online DSM-based mental health referral inventory. Journal of Technology in Human Services, 29, 284–295. doi:10.1080/15228835.2011.638421[Taylor &; Francis Online] , [Google Scholar]). In 2013, a revision of that inventory that was consistent with the DSM-5 was posted. The present study evaluates the new inventory with a diverse sample of 201,625 people from 184 countries (mainly the United States, Canada, and the United Kingdom). The 63-item inventory screens for 21 common problems and typically takes about 5?min to complete. Test scores proved to be good predictors of a variety of self-reported criterion measures, including happiness, personal and professional success, history of hospitalization, history of therapy, current participation in therapy, and employment. Females were found to have slightly more mental health problems than males, and effects were also found for education and race/ethnicity. The revised test is more accessible than the original, reducing the reading level from grade 10.2 to 6.6. It is not designed to diagnose; rather, its primary purpose is to refer people to mental health professionals for further evaluation. It can also be used in clinical settings for quick screening purposes.     

Confidence intervals for a population proportion using group testing in the presence of misclassification

In this study, we consider the construction of the confidence interval for the population proportion while using group testing with misclassification. We propose two confidence intervals based on Cornish-Fisher expansion and a modified Wilson’s interval based on a newly developed estimator. We investigate the performance of these intervals extensively and also apply the methods to real datasets. Our newly derived methods have competitive performance compared with existing methods.     

Test for the mean matrix in a Growth Curve model for high dimensions

We consider the problem of estimating and testing a general linear hypothesis in a general multivariate linear model, the so-called Growth Curve model, when the p × N observation matrix is normally distributed.

The maximum likelihood estimator (MLE) for the mean is a weighted estimator with the inverse of the sample covariance matrix which is unstable for large p close to N and singular for p larger than N. We modify the MLE to an unweighted estimator and propose new tests which we compare with the previous likelihood ratio test (LRT) based on the weighted estimator, i.e., the MLE. We show that the performance of these new tests based on the unweighted estimator is better than the LRT based on the MLE.     

Generalized Holm’s procedure for multiple hypotheses testing problems

A generalized Holm’s procedure is proposed which can reject several null hypotheses at each step sequentially and also strongly controls the family-wise error rate regardless of the dependence of individual test statistics. The new procedure is more powerful than Holm’s procedure if the number of rejections m and m > 0 is prespecified before the test.     

Women’s experiences of antenatal transfer for threatened premature labour in NSW: A qualitative study

BackgroundEach year thousands of pregnant women experiencing threatened premature labour are transferred considerable distances across Australia to access higher level facilities but only a small proportion of these women go on to actually give birth to a premature baby. Women from regional areas are required to move away from their home, children and support networks because of a perceived risk of birthing in a centre without neonatal intensive care facilities.AimThis study examines the experience of women undergoing antenatal transfer for threatened premature labour in New South Wales and the Australian Capital Territory who do not give birth during their transfer admission.MethodsThirteen semi-structured in-depth interviews were held with women across five tertiary referral sites across New South Wales and the Australian Capital Territory, and analysed until saturation for themes.FindingsSeven urban and six rural women were interviewed. Women and their families were all negatively affected by antenatal transfer. Factors that helped enable a positive experience were; enhanced sense of safety in the tertiary unit, and individual qualities of staff. Factors that contributed to negative experiences were; inadequate and conflicting information, and no involvement or choice in the clinical decision-making process to move to another facility.ConclusionsAntenatal transfer is an extremely stressful experience for women and their families. The provision of high quality written and verbal information, and the inclusion of women's perception of risk in the clinical decision making process will improve the experience for women and their families in NSW and the ACT.     

Sequential testing procedures for a class of distributions representing various life-testing models

A class of probability density functions is considered, which covers several life-testing models as specific cases. Sequential probability ratio tests are developed for testing simple and composite hypotheses regarding the parameters of the probabilistic model. Expressions for the operating characteristic and the average sample number functions are derived and their behaviour is studied by means of graph-plotting.     

Gatekeeping procedures in dose-response clinical trials based on the Dunnett test

This paper discusses multiple testing procedures in dose-response clinical trials with primary and secondary endpoints. A general gatekeeping framework for constructing multiple tests is proposed, which extends the Dunnett test [Journal of the American Statistical Association 1955; 50: 1096-1121] and Bonferroni-based gatekeeping tests developed by Dmitrienko et al. [Statistics in Medicine 2003; 22:2387-2400]. The proposed procedure accounts for the hierarchical structure of the testing problem; for example, it restricts testing of secondary endpoints to the doses for which the primary endpoint is significant. The multiple testing approach is illustrated using a dose-response clinical trial in patients with diabetes. Monte-Carlo simulations demonstrate that the proposed procedure provides a power advantage over the Bonferroni gatekeeping procedure. The power gain generally increases with increasing correlation among the endpoints, especially when all primary dose-control comparisons are significant.     

语言测试与第二语言习得研究的接口与合作

尽管二语习得和语言测试属应用语言学的两个不同分支,研究的重点也不尽相同,但二者在一些研究问题和研究方法上仍有不少相通或接口之处。遗憾的是长期以来两领域缺少沟通与合作。文章主要探讨了双方的互补性并且指出今后合作的必要性和可行性。