SEMIPARAMETRIC REGRESSION AND GRAPHICAL MODELS

Semiparametric regression models that use spline basis functions with penalization have graphical model representations. This link is more powerful than previously established mixed model representations of semiparametric regression, as a larger class of models can be accommodated. Complications such as missingness and measurement error are more naturally handled within the graphical model architecture. Directed acyclic graphs, also known as Bayesian networks, play a prominent role. Graphical model-based Bayesian 'inference engines', such as bugs and vibes , facilitate fitting and inference. Underlying these are Markov chain Monte Carlo schemes and recent developments in variational approximation theory and methodology.     

Good health and the bridging of structural holes

Bridges that span structural holes are often explained in terms of the entrepreneurial personalities or rational motivations of brokers, or structural processes that lead to the intersection of social foci. I argue that the existence and use of bridges in interpersonal networks also depends on individuals’ health. Poor health may make it more difficult to withstand the pressures and to execute some of the common tasks associated with bridging (e.g., brokerage). I examine this possibility using egocentric network data on over 2500 older adults drawn from the recent National Social Life, Health, and Aging Project (NSHAP). Multivariate regression analyses show that both cognitive and functional health are significantly positively associated with bridging, net of sociodemographic and life-course controls. The relationship between functional (kinesthetic) health and bridging appears to be partially mediated by network composition, as older adults who have poorer functional health also tend to have networks that are richer in strong ties. Several potential mediation mechanisms are discussed. Cognitive function remains significantly associated with bridging net of network composition, suggesting that the inherent challenges of maintaining bridging positions may be more difficult to cope with for those who have cognitive impairments than for those who have functional impairments such as limited mobility. An alternative explanation is that cognitively impaired individuals have more difficulty recognizing (and thus strategically using) bridges in their networks. Theoretical implications and possibilities for future research are discussed.     

Clustering in weighted networks

In recent years, researchers have investigated a growing number of weighted networks where ties are differentiated according to their strength or capacity. Yet, most network measures do not take weights into consideration, and thus do not fully capture the richness of the information contained in the data. In this paper, we focus on a measure originally defined for unweighted networks: the global clustering coefficient. We propose a generalization of this coefficient that retains the information encoded in the weights of ties. We then undertake a comparative assessment by applying the standard and generalized coefficients to a number of network datasets.     

城乡困难家庭的求助网络及其政策启示

文章使用“城乡困难家庭调查”2014年跟踪数据,对我国城乡困难家庭的求助网络结构和家庭因素进行分析,发现：血缘和亲缘关系仍是我国城乡困难家庭最天然的帮扶网络,但官方社会救助系统的作用已超过民间系统,邻里朋友仅在血亲和官方系统失灵时起补充作用;基层治理结构对求助网络的城乡差异产生影响,流动人口家庭求助网络呈现多元均衡的特征,但面临官方救助系统的制度性缺失;人口规模、低保资格两个家庭因素对城乡困难家庭求助网络有显著影响。文章认为,应进一步发挥社会力量的作用,针对我国城乡困难家庭建立官民并举、多元协作的社会救助政策体系。     

An intuitive risk factors search algorithm: usage of the Bayesian network technique in personalized medicine

The article focuses on the application of the Bayesian networks (BN) technique to problems of personalized medicine. The simple (intuitive) algorithm of BN optimization with respect to the number of nodes using naive network topology is developed. This algorithm allows to increase the BN prediction quality and to identify the most important variables of the network. The parallel program implementing the algorithm has demonstrated good scalability with an increase in the computational cores number, and it can be applied to the large patients database containing thousands of variables. This program is applied for the prediction for the unfavorable outcome of coronary artery disease (CAD) for patients who survived the acute coronary syndrome (ACS). As a result, the quality of the predictions of the investigated networks was significantly improved and the most important risk factors were detected. The significance of the tumor necrosis factor-alpha gene polymorphism for the prediction of the unfavorable outcome of CAD for patients survived after ACS was revealed for the first time.     

Networks for Welfare Provision: Getting a Grip on Processes of Social Exclusion by Evaluating Network Effectiveness

It has been repeatedly stated that the topic of network effectiveness has often been neglected. This has led to a situation of ‘network euphoria’ amongst practitioners and policymakers with regard to the creation of networks as a means to (re‐)organize welfare provision. Starting from a case study of networks for welfare provision in two Belgian cities, this article focuses on the effectiveness of service delivering networks at the client level, which is understood as the contribution of the network in improving the conditions of life of those being served or targeted by the network. Therefore, we focused on the collective efforts via these networks (e.g. by creating a night shelter or by conducting case consultations) to look after a population of homeless people who face multiple problems and are not able yet to benefit from welfare provision. Our findings revealed that these networks were able to realize a ‘collaborative advantage’ by filling in service gaps and by overcoming fragmentation of care. Nevertheless, these networks equally maintained criteria that restricted the accessibility or usefulness of welfare services provided to homeless people. In the concluding section, we highlight some of the factors that could help us to explain our findings and highlight the tension between the effectiveness of networks at the client level and the effectiveness at the organizational level.     

Strength matters: Tie strength as a causal driver of networks’ information benefits

Studies of social networks have often taken the existence of a social tie as a proxy for the transmission of information. However, other studies of social networks in the labor market propose that the likelihood of information transmission might depend on strength of the tie; and that tie strength is a potentially important source of the tie's value. After all, even if job seekers have social ties to those who have valuable job information, the seekers will gain little information benefit when the ties do not actually transmit the information. This paper clarifies the conditions under which social ties might provide information benefits. We use a survey vignette experiment and ask MBA students about their likelihood of relaying job information via strong ties (to friends) or weak ties (to acquaintances), holding constant the structural locations spanned by the tie and job seekers' fit with the job. The results support the claim that strength of tie has a causal effect on the chances of information transmission: potential referrers are more likely to relay job information to their friends than to acquaintances. The larger implication of these findings is that whatever benefits there might be to using weak ties to reach distant non-redundant information during job search, these benefits need to be considered against the likely fact that people connected via weak ties are less likely to actually share information about job opportunities than are people to whom the job seeker is strongly tied.     

Who gets evicted? Assessing individual,neighborhood, and network factors

The prevalence and consequences of eviction have transformed the lived experience of urban poverty in America, yet little is known about why some families avoid eviction while others do not. Applying discrete hazard models to a unique dataset of renters, this study empirically evaluates individual, neighborhood, and social network characteristics that explain disparities in displacement from housing. Family size, job loss, neighborhood crime and eviction rates, and network disadvantage are identified as significant and robust predictors of eviction, net of missed rental payments and other relevant factors. This study advances urban sociology and inequality research and informs policy interventions designed to prevent eviction and stem its consequences.     

Reconsidering “community liberated”: How class and the national context shape personal support networks

The “community liberated” thesis has been influential in describing contemporary social support systems. Specifically, “community liberated” argues that people do not seek support in their immediate neighborhood but rather entertain a network of far-flung ties to support-providing alters. This paper uses personal network data from six countries – Australia, Germany, the US, Austria, Hungary and Italy – to evaluate this argument and shows that the degree of liberation of one's community is strongly linked to one's socioeconomic status – specifically, one's education level. Additionally, we describe strong country-level heterogeneity in the spatial dynamics of personal support networks and find national contexts to be moderating the effect of education on community liberation, especially in Italy and Hungary, thus suggesting network geographic dispersion to be linked to national economic structures and labor markets. The paper thus elucidates the effect of two different, yet related social contexts on personal networks: the class context and the national context.     

The association between social support networks and maternal employment: a comparison of western German,eastern German,and migrant mothers of preschool-aged children

A lack of adequate childcare can delay mothers’ return to the labor market after childbirth. This paper examines whether social support with childcare by kin and friends facilitates maternal employment in the first 72 months after childbirth. Using data from the German Socio-Economic Panel (SOEP) 1993–2009, a comparison of natives (n?=?1409) and migrants (n?=?411) in corporative-conservative western Germany with mothers in former socialist eastern Germany (n?=?528) shows that kinship support is positively associated with maternal employment when public childcare is limited. Western German and migrant mothers return to work sooner if they are surrounded by kin. But kin do not provide support for maternal employment in eastern Germany, where public childcare is more easily accessible and continuous female employment is a prevalent social norm. Friendship networks, by contrast, are most valuable for maternal employment if they complement public childcare.