Fast robust feature screening for ultrahigh-dimensional varying coefficient models

In this paper, we propose a new partial correlation, the so-called composite quantile partial correlation, to measure the relationship of two variables given other variables. We further use this correlation to screen variables in ultrahigh-dimensional varying coefficient models. Our proposed method is fast and robust against outliers and can be efficiently employed in both single index variable and multiple index variable varying coefficient models. Numerical results indicate the preference of our proposed method.     

Joint GEEs for multivariate correlated data with incomplete binary outcomes

This study considers a fully-parametric but uncongenial multiple imputation (MI) inference to jointly analyze incomplete binary response variables observed in a correlated data settings. Multiple imputation model is specified as a fully-parametric model based on a multivariate extension of mixed-effects models. Dichotomized imputed datasets are then analyzed using joint GEE models where covariates are associated with the marginal mean of responses with response-specific regression coefficients and a Kronecker product is accommodated for cluster-specific correlation structure for a given response variable and correlation structure between multiple response variables. The validity of the proposed MI-based JGEE (MI-JGEE) approach is assessed through a Monte Carlo simulation study under different scenarios. The simulation results, which are evaluated in terms of bias, mean-squared error, and coverage rate, show that MI-JGEE has promising inferential properties even when the underlying multiple imputation is misspecified. Finally, Adolescent Alcohol Prevention Trial data are used for illustration.     

Linearity,symmetry, and equidistance in semantic differential scales for measuring images of self and images of others

ABSTRACT

Violations of linearity, symmetry, and equidistance of scale points in semantic differential scales may be due to respondents failing to figure out antonyms, to a positivity bias, or to respondents curtailing their reading the options. Multiple correspondence analysis on data provided by a randomized between-subjects experiment (split-ballot), using a web survey of 537 German residents, shows that bipolar semantic differential scales allow for linear measurement and conceal no positivity bias, and that using not suitable adjective pairs in semantic differential scales destroys symmetry.     

“I'm Not Supporting His Kids”: Nonresident Fathers' Contributions Given Mothers' New Fertility

The authors examined whether nonresident fathers provide informal support to their children and whether support stops if their ex‐partner goes on to have a child with a new man. A logistic regression analysis of longitudinal survey and administrative data for 434 women who received welfare in Wisconsin showed that fathers are less likely to provide informal support when their ex‐partner has a child with a new partner. Alternative models that control for unobserved characteristics suggest somewhat different results, providing stronger evidence of declines in support that can be shared across family members than in support that can be directed to a particular child.     

A Qualitative Study of Middle Eastern/Arab American Sexual Minority Identity Development

The development of one's sexual minority identity is often stunted by a heterosexist society. For individuals with multiple minority oppressions, this process becomes even more complicated. As such, there has been a call among researchers for more empirical research on the experiences of lesbian, gay, bisexual (LGB) individuals from racial/ethnic minority groups. This study uses qualitative methods to fill gaps in the literature related to identity development among same-gender attracted Middle Eastern/Arab individuals living in the United States. From 12 interviews, 13 themes associated with intersectionality, race/ethnicity, sexual identity development, discrimination, stigma, oppression, and invisibility. Themes, subthemes, and their implications are discussed.     

论汉代政权对民间信仰的多重政策

民间信仰是中国传统文化的重要组成部分,也是俗文化的核心。针对民间自发形成的信仰与崇拜,汉代政权采取多重政策,或引导支持,或借鉴利用,或排斥打击,主要取决于其是否与官方信仰及主流文化相抵触,是否从根本上影响到社会秩序和百姓生产生活。不管官方态度如何,作为一种强大的民间力量,民间信仰以其特有的方式参与政治生活,与官方政权相互影响、相互补充,形成既配合又抗衡的互动局面,并以顽强的生命力传承、延续和发展。     

多元智能理论在《思想道德修养与法律基础》课教学中的应用

多元智能理论认为,人的智能是多元的,每个正常的人都在一定程度上拥有其中的多项技能,人类个体的不同在于所拥有的技能的程度和组合不同。借鉴这一理论,有利于充分发挥高职院校《思想道德修养与法律基础》课的思想政治教育作用,形成科学的教学观、学生观、师生观和评价观,促进学生的全面发展,增强教学实效性,提高教学质量。     

全球化背景下的中国现代性建构

全球化并不预示着一个全球大同时代的来临，反而在很大程度上表现了文化之间的差异性甚至对抗性。在这样的时代背景下，对于各民族文化尤其是对非西方社会来说，如何理解现代性与追求自身的现代化，成为一个紧迫的任务。在中国现代性的建构过程中，既要求加深对现代性的理解，也要求更好地把握自己的文化传统。     

岩体变形参数的应力状态效应研究

岩体力学参数存在随应力状态不同而变化的特性，称为应力状态效应。岩体力学特性的实测值具有空间场分布特征，分析表明地应力（或围压）的有规律变化是造成岩体力学参数空间变化的主要原因。考虑到实测地应力多具有较大的离散型，尝试利用反演方法获取地应力场信息，通过多元回归分析得到实测岩体变形参数与地应力的经验关系，以研究岩体变形参数的应力状态效应。通过一工程实例研究：得到地下厂房区域的岩体变形模量为16 GPa左右，与反演分析结果也比较符合；通过应力状态效应研究得到的岩体变形参数物理场，是一个渐变的物理场模型，是更为符     

Who by Accident? The Social Morphology of Car Accidents

Prior studies in the sociology of accidents have shown that different social groups have different rates of accident involvement. This study extends those studies by implementing Bourdieu's relational perspective of social space to systematically explore the homology between drivers’ social characteristics and their involvement in specific types of motor vehicle accident. Using a large database that merges official Israeli road‐accident records with socioeconomic data from two censuses, this research maps the social order of road accidents through multiple correspondence analysis. Extending prior studies, the results show that different social groups indeed tend to be involved in motor vehicle accidents of different types and severity. For example, we find that drivers from low socioeconomic backgrounds are overinvolved in severe accidents with fatal outcomes. The new findings reported here shed light on the social regularity of road accidents and expose new facets in the social organization of death.