The quest for legitimacy and the communication of strategic cross-sectoral partnership on Facebook: A big data study

Nowadays, many wicked problems such as environmental issues require organizations from multiple sectors to form cross-sectoral alliances. Cross-sectoral alliance networks can transfer resources and signal affiliations and value alignment between strategic partners. The communication of cross-sectoral alliances is a form of CSR communication that serves organizations’ strategic goals and objectives. Drawing on the literature on digital CSR communication and legitimacy theory, this article examines which legitimacy needs shape the formation of cross-sectoral ties on Facebook in addressing environmental and sustainable development issues in the United States. Combining data-mining, text-mining, social network analysis, and exponential graph modeling, this research investigates the structure of a network among 3071 organizations across multiple sectors. Findings show that organizations’ cross-sectoral tie formation is mainly driven by social legitimacy and alliance legitimacy needs.     

Composite quantile regression for massive datasets

Analysis of massive datasets is challenging owing to limitations of computer primary memory. Composite quantile regression (CQR) is a robust and efficient estimation method. In this paper, we extend CQR to massive datasets and propose a divide-and-conquer CQR method. The basic idea is to split the entire dataset into several blocks, applying the CQR method for data in each block, and finally combining these regression results via weighted average. The proposed approach significantly reduces the required amount of primary memory, and the resulting estimate will be as efficient as if the entire data set is analysed simultaneously. Moreover, to improve the efficiency of CQR, we propose a weighted CQR estimation approach. To achieve sparsity with high-dimensional covariates, we develop a variable selection procedure to select significant parametric components and prove the method possessing the oracle property. Both simulations and data analysis are conducted to illustrate the finite sample performance of the proposed methods.     

Reversible jump Markov chain Monte Carlo algorithms for Bayesian variable selection in logistic mixed models

In this article, to reduce computational load in performing Bayesian variable selection, we used a variant of reversible jump Markov chain Monte Carlo methods, and the Holmes and Held (HH) algorithm, to sample model index variables in logistic mixed models involving a large number of explanatory variables. Furthermore, we proposed a simple proposal distribution for model index variables, and used a simulation study and real example to compare the performance of the HH algorithm with our proposed and existing proposal distributions. The results show that the HH algorithm with our proposed proposal distribution is a computationally efficient and reliable selection method.     

Crisis perceptions,relationship, and communicative behaviors of employees: Internal public segmentation approach

This study examines when and why employees engage in external communicative behaviors during the periods of a corporate crisis. Combining a cross-situational factor (i.e., pre-crisis relationship quality) and a situational factor (i.e., crisis-specific perceptions), this study segments internal publics to understand employees’ motivations of becoming advocates for or adversaries of their organization. The results of an online survey demonstrated that employees’ pre-crisis relationship to their organization plays a critical role in encouraging them to advocate for their company; meanwhile, their crisis-perception—whether they feel fewer constraints in solving a crisis—is more likely to make them share negative information externally. Segmented groups of employees with high levels of both relationship and activeness are most likely to engage in both positive and negative external communication behaviors. Theoretical and practical implications for public relations and internal crisis communication are suggested.     

Collecting text messages from college students: Evaluating a novel methodology

Social interaction, particularly in older adolescents, increasingly involves computer‐mediated communication. Although studies of public computer‐mediated communication are increasingly common, studies of private text messaging remain rare. As approaches for obtaining such data evolve with technological advances, developmental scientists need designs in which to use such approaches that reduce sampling biases in both participants and text messages. In this study (n = 854; 46% male; 22% African American, 60% European American), we examined selection biases in the participant sample (i.e., factors associated with actual participation), procedural biases in the participant sample (i.e., factors related to failed data capture due to technological or procedural issues), and selection biases in the sample of text messages (i.e., based on self‐reported reasons for texting). Findings from our study suggest that studying human interaction directly through analysis of text message data is not only feasible, but also may be successfully undertaken with minimal biases regarding sample selection and text message selection among those who are engaged in research and engaged in text messaging outside of the study context. However, biases may occur depending on the type of platform (iPhone vs. Android) used by participants for texting.     

Model selection in high-dimensional noisy data: a simulation study

In many practical applications, high-dimensional regression analyses have to take into account measurement error in the covariates. It is thus necessary to extend regularization methods, that can handle the situation where the number of covariates p largely exceed the sample size n, to the case in which covariates are also mismeasured. A variety of methods are available in this context, but many of them rely on knowledge about the measurement error and the structure of its covariance matrix. In this paper, we set the goal to compare some of these methods, focusing on situations relevant for practical applications. In particular, we will evaluate these methods in setups in which the measurement error distribution and dependence structure are not known and have to be estimated from data. Our focus is on variable selection, and the evaluation is based on extensive simulations.     

Generalized estimator of population mean by using conventional and non-conventional measures in the presence of measurement errors

In survey research, it is assumed that reported response by the individual is correct. However, given the issues of prestige bias, self-respect, respondent's reported data often produces estimated values which are highly deviated from the true values. This causes measurement error (ME) to be present in the sample estimates. In this article, the estimation of population mean in the presence of measurement error using information on a single auxiliary variable is studied. A generalized estimator of population mean is proposed. The class of estimators is obtained by using some conventional and non-conventional measures. Simulation and numerical study is also conducted to assess the performance of estimators in the presence and absence of measurement error.     

Bayesian nonparametric modelling of the link function in the single-index model using a Bernstein–Dirichlet process prior

This paper proposes the use of the Bernstein–Dirichlet process prior for a new nonparametric approach to estimating the link function in the single-index model (SIM). The Bernstein–Dirichlet process prior has so far mainly been used for nonparametric density estimation. Here we modify this approach to allow for an approximation of the unknown link function. Instead of the usual Gaussian distribution, the error term is assumed to be asymmetric Laplace distributed which increases the flexibility and robustness of the SIM. To automatically identify truly active predictors, spike-and-slab priors are used for Bayesian variable selection. Posterior computations are performed via a Metropolis-Hastings-within-Gibbs sampler using a truncation-based algorithm for stick-breaking priors. We compare the efficiency of the proposed approach with well-established techniques in an extensive simulation study and illustrate its practical performance by an application to nonparametric modelling of the power consumption in a sewage treatment plant.     

The sociolinguistics of late modern publics

This issue examines struggles over public discursive power in late modern society, understanding that late modern publics are characterized by a destabilization of formerly unmarked public authorities, in other words, an unsettling of standardized “voices from nowhere” (Gal & Woolard, 2001). The contributions to the issue illustrate traditional social and academic elites’ discursive reactions to such developments. In the introduction, we outline different theoretical frameworks of modern and late‐modern publics, identify factors that contribute to a destabilization of modern certainties of social public order, and argue that a weakening of traditional public norms not only implies emancipation but also, as the contributions to the issue demonstrate, may bring along new forms of public dominance. For the positioning of contemporary sociolinguistic theory and practice, we conclude that it is the crucial task of the current public moment to critically self‐engage with our own relations with, and ideologies of, the public. Dieses Themenheft befragt Strukturen öffentlicher Diskursmacht in spätmoderner Gesellschaft und zeigt auf, dass spätmoderne Öffentlichkeit durch eine Destabilisierung von vormals unmarkierten Autoritäten ist, in anderen Worten, von standardisierten “voices from nowhere” (Gal & Woolard, 2001). Die Beiträge illustrieren Reaktionen traditioneller sozialer und akademischer Eliten hierauf. In dieser Einleitung skizzieren wir verschiedene theoretische Zugänge zu moderner und spätmoderner Öffentlichkeit, identifizieren Faktoren, die zu einer Destabilisierung von sozialen Gewissheiten der Moderne beitragen und legen dar, dass eine Schwächung traditioneller öffentlichen Normen nicht nur soziale Emanzipation impliziert sondern, wie die Beiträge zu diesem Heft zeigen, auch neue Formen öffentlicher Dominanz hervorbringt. In Bezug auf zeitgenössische soziolinguistische Theorie und Praxis schlussfolgern wir daraus, dass es im aktuellen öffentlichen Kontext eine zentrale Aufgabe der Soziolinguistik ist, sich kritisch mit den eigenen Beziehungen zu und Ideologien von Öffentlichkeit befassen. Das Themenheft soll zu dieser Auseinandersetzung beitragen.     

Symposium on political communication and social movements: audience,persuasion, and influence

ABSTRACT

This article argues that an important contribution that political communication can offer social movement studies is a more variegated understanding of social movement audiences and their role in social movement strategy and processes. Specifically, the article introduces a coarse typology of social movement audiences and discusses the importance of understanding differences in the goals of these audiences and what kinds of influence, from messaging or other forms of pressure, may be important to affect different audiences in movements’ favors. The article also examines the ways in which audiences are active, shaping what messages they are exposed to, consume, believe, and act upon. The call of the article is to bring a concern for audiences into social movement studies in the hopes of wedding these more media and communication-focused concerns with the kinds of structural and material influences social movement studies is so accomplished in investigating.