Logit models and logistic regressions for social networks: III. Valued relations

This paper generalizes thep* model for dichotomous social network data (Wasserman & Pattison, 1996) to the polytomous case. The generalization is achieved by transforming valued social networks into three-way binary arrays. This data transformation requires a modification of the Hammersley-Clifford theorem that underpins thep* class of models. We demonstrate that, provided that certain (non-observed) data patterns are excluded from consideration, a suitable version of the theorem can be developed. We also show that the approach amounts to a model for multiple logits derived from a pseudo-likelihood function. Estimation within this model is analogous to the separate fitting of multinomial baseline logits, except that the Hammersley-Clifford theorem requires the equating of certain parameters across logits. The paper describes how to convert a valued network into a data array suitable for fitting the model and provides some illustrative empirical examples.This research was supported by grants from the Australian Research Council, the National Science Foundation (#SBR96-30754), and the National Institute of Health (#PHS-1RO1-39829-01).     

The Influence of Close Ties on Depression: Does Network Religiosity Matter?

Despite a growing body of the literature on how features of social networks influence well-being, we know little of how the religiosity of social networks matter. This study addresses three types of religious social network ties and their association with mental health: same (non)-religious ties, religious discussion ties, and ties offering prayers on an individual's behalf. Using ego-centric network data from the 2006 Portraits of American Life Study (N = 2,223), multivariate regression results suggest that a greater number of ties that discuss religion and pray for the respondent are detrimental to the mental health of those of a low religious salience. Taken together, this study demonstrates that religious dimensions of social networks exact an important influence on mental health and highlights the importance of identifying specific features of religion among core network ties.     

Embedding social networks: How guanxi ties reinforce Chinese employees’ retention

Going beyond traditional inquiry into social support from local organizational constituents, this project examined how diverse resources from mutually affiliated contacts within and beyond local work environs boost propensity to stay in firms. We deployed Burt’s (1992) name generator and network closure index to more fully assess guanxi networks in China, which comprise strong, dense, and multiplex ties. Specifically, we tested how closed guanxi networks promote job loyalty among Chinese nationals, while investigating how high-commitment human resource management (HRM) systems moderate network effects. We collected egonet data from 417 employees in four high tech firms in China. Hierarchical linear modeling revealed that guanxi network closure increases propensity to stay, whose effects high-commitment HRM reinforce.     

Social Constraint and Self‐Doubt: Mechanisms of Social Network influence on Resistance to Persuasion

Those around us have a profound influence on our political attitudes and attitude strength, such that people whose social networks include a variety of perspectives have weaker, less deeply entrenched attitudes than those who are surrounded by like‐minded others. In particular, those embedded in attitudinally heterogeneous networks are more open to changing their views. The nature and mechanisms of this network influence on openness to attitude change remain unclear. A survey experiment examines two mechanisms proposed by prior literature: (1) social doubt triggered by network members' dissent and (2) social constraint to maintain similar attitudes. It also provides some data on the more commonly assumed mechanism, (3) information exchange. Results strongly support social constraint and are mixed on social doubt. This contrasts with the theoretical emphasis of much previous interdisciplinary social network research, which has focused primarily on information exchange, to the detriment of other mechanisms. Findings also indicate that like‐minded social network members solidify attitudes at least as much as dissent erodes them, suggesting that prior emphasis on the influence of heterogeneous rather than attitudinally congruent networks is overstated. Implications for political movements are discussed.     

Close Ties,Intercessory Prayer,and Optimism Among American Adults: Locating God in the Social Support Network

Prayer is often an interpersonal phenomenon. It represents not only a form of social support shared between or among people, but also a means of embedding an unobservable actor (God) within a conventionally observable social network. This study considers whether the receipt of intercessory prayer from close network ties is associated with future‐oriented well‐being. Analyses use social network module data from the Portraits of American Life Study (PALS), a nationally representative study of American adults containing a breadth of information not available in prior studies of networks, prayer, and well‐being. Despite experiencing more instances of recent adversity (mental or physical health problem, financial trouble, and unemployment), prayed‐for PALS respondents report the highest levels of optimism. Furthermore, the association between network prayer and optimism is robust to inclusion of individual‐level indicators of religiosity. Finally, other forms of social support that an individual receives from his or her close ties do not explain the benefits of intercessory prayer.     

Friendship Dissolution Within Social Networks Modeled Through Multilevel Event History Analysis

A social network perspective can bring important insight into the processes that shape human behavior. Longitudinal social network data, measuring relations between individuals over time, has become increasingly common—as have the methods available to analyze such data. A friendship duration model utilizing discrete-time multilevel survival analysis with a multiple membership random effect structure is developed and applied here to study the processes leading to undirected friendship dissolution within a larger social network. While the modeling framework is introduced in terms of understanding friendship dissolution, it can be used to understand microlevel dynamics of a social network more generally. These models can be fit with standard generalized linear mixed-model software, after transforming the data to a pair-period data set. An empirical example highlights how the model can be applied to understand the processes leading to friendship dissolution between high school students, and a simulation study is used to test the use of the modeling framework under representative conditions that would be found in social network data. Advantages of the modeling framework are highlighted, and potential limitations and future directions are discussed.     

Why do we share aggressive online content? Testing a short cycle model

The present study aimed to observe the impacts of aggressive online content exposure and personality on aggressive content sharing, considering the mediating role of positive and negative affects. A total of 302 Brazilians, equally divided into two groups, participated in a social network simulation, being exposed to aggressive (experimental group) or neutral (control group) posts and choosing what they would like to share on this site (among aggressive or nonaggressive material). A factorial MANOVA showed a significant difference for positive and negative affects and aggressive content sharing (Wilks' Λ = 0.936; F[3, 298] = 6.812; p = .001; η² = 0.064). Through structural equation modeling, an explanatory model was also tested, in which this behavior was directly predicted by aggressive online content exposure and suffered indirect effects from openness, conscientiousness, and extroversion, mediated by positive affects (goodness-of-fit index = 0.997, comparative fit-index = 0.998, Tucker-Lewis index = 0.996, SRMR = 0.028, root-mean-square error of approximation = 0.035, CI = 0.003−0.054). It is possible to conclude that the objectives are met, highlighting the contribution to understanding aggressive online behavior through the proximal processes described in the general aggression model.     

A Unified Nonparametric IRT Model for <Emphasis Type="Italic">d</Emphasis>-Dimensional Psychological Test Data (<Emphasis Type="Italic">d</Emphasis>-ISOP)

The (univariate) isotonic psychometric (ISOP) model (Scheiblechner, 1995) is a nonparametric IRT model for dichotomous and polytomous (rating scale) psychological test data. A weak subject independence axiom W1 postulates that the subjects are ordered in the same way except for ties (i.e., similarly or isotonically) by all items of a psychological test. A weak item independence axiom W2 postulates that the order of the items is similar for all subjects. Local independence (LI or W3) is assumed in all models. With these axioms, sample-free unidimensional ordinal measurements of items and subjects become feasible. A cancellation axiom (Co) gives, as a result, the additive isotonic psychometric (ADISOP) model and interval scales for subjects and items, and an independence axiom (W4) gives the completely additive isotonic psychometric (CADISOP) model with an interval scale for the response variable (Scheiblechner, 1999). The d-ISOP, d-ADISOP, and d-CADISOP models are generalizations to d-dimensional dependent variables (e.g., speed and accuracy of response). The author would like to thank an Associate Editor and two anonymous referees and also Professor H.H. Schulze for their very valuable suggestions and corrections.     

A Note on the Frequency and Sources of Close Interreligious Ties

As the religious landscape of the United States becomes increasingly diverse, understanding how individuals of different faiths relate to each other and the consequences of those relationships presents a significant issue for research. Research studying such faith‐bridging interactions has been limited and primarily focused on interfaith marriage. Using social network data from the 2004 General Social Survey we examine the dynamics underlying the frequency of interfaith ties of all types. We then analyze how interreligious ties differ in strength from intrareligious ties.     

The Strength of Weak and Strong Communication Ties in a Community Information Program1

A study is described that investigated the influence of a target group's communication network on information diffusion and the impact of a community-based information program. It was expected that the process of information diffusion would be related to the quantity of ties. The diffusion of information brought into a clique by external sources that did not belong to any of the target group's cliques was expected to be related to the existing number of weak ties, whereas the diffusion of specific information sent out by internal sources within a clique was expected to be primarily related to the number of strong ties. Furthermore it was expected that the impact of the program would be related to the perceived quality of network ties. Results support the hypotheses and confirm Granovetter's “strength-of-weak-ties” hypothesis. In addition, existing direct ties between the internal sources of information and the target group appeared to have played a crucial role in the process of information diffusion.     

Plagues,Pagans, and Christians: Differential Survival,Social Networks,and the Rise of Christianity

In his book, The Rise of Christianity, Rodney Stark argues that the early Christian church indirectly benefitted from the plagues that struck the early Roman Empire in the 2nd and 3rd centuries of the Common Era. In particular, he argues that the early church's doctrines concerning love, charity, and social service would have led Christians to enjoy higher survival rates than pagans, which would have left the social networks of the former relatively intact, while those of the latter in disarray. Moreover, since recruitment to new religious movements occurs largely through social ties, the probability that an average pagan would have converted to Christianity would have been greater after the plagues than before. In this article, we use computer modeling to test Stark's hypothesis. Not only do we find support for his hypothesis, we also show that higher survival rates for either group not only would have benefitted their respective social networks but also would have benefitted their competitor's social networks, highlighting how network effects often flow in multiple and unexpected ways. We conclude with suggestions for future research, in particular how scholars can build upon our analyses and explore alternative assumptions.     

Logit models and logistic regressions for social networks: I. An introduction to Markov graphs andp

Spanning nearly sixty years of research, statistical network analysis has passed through (at least) two generations of researchers and models. Beginning in the late 1930's, the first generation of research dealt with the distribution of various network statistics, under a variety of null models. The second generation, beginning in the 1970's and continuing into the 1980's, concerned models, usually for probabilities of relational ties among very small subsets of actors, in which various simple substantive tendencies were parameterized. Much of this research, most of which utilized log linear models, first appeared in applied statistics publications.But recent developments in social network analysis promise to bring us into a third generation. The Markov random graphs of Frank and Strauss (1986) and especially the estimation strategy for these models developed by Strauss and Ikeda (1990; described in brief in Strauss, 1992), are very recent and promising contributions to this field. Here we describe a large class of models that can be used to investigate structure in social networks. These models include several generalizations of stochastic blockmodels, as well as models parameterizing global tendencies towards clustering and centralization, and individual differences in such tendencies. Approximate model fits are obtained using Strauss and Ikeda's (1990) estimation strategy.In this paper we describe and extend these models and demonstrate how they can be used to address a variety of substantive questions about structure in social networks.This research was supported by grants from the Australian Research Council and the National Science Foundation (#SBR93-10184). This paper was presented at the 1994 Annual Meeting of the Psychometric Society, Champaign, Illinois, June 1994. Special thanks go to Sarah Ardu for programming assistance, Laura Koehly and Garry Robins for help with this research, and to Shizuhiko Nishisato and three reviewers for their comments. INTERNETemail addresses:pattison@psych.unimelb.edu.au (PP);stanwass@uiuc.edu (SW). Affiliations: Department of Psychology, University of Melbourne (PP); Department of Psychology, Department of Statistics, and The Beckman Institute for Advanced Science and Technology, University of Illinois (SW).     

Predicting Support for Arabs' Autonomy from Social Dominance: The Role of Identity Complexity and Dehumanization

The present study extends previous research on the influence of social dominance orientation (SDO) in international affairs by investigating the role of sociocognitive processes such as social identity complexity (SIC) and intergroup dehumanization in explaining the relationship between SDO and support for Arabs' autonomy. An Italian heterogeneous sample (N = 123), in terms of gender, age, political orientation, religiosity, and income level, was considered. We expected that those low on SDO would be more likely to support Arabs' autonomy, would have higher SIC, and would be less likely to dehumanize Arabs. We also expected that SIC and dehumanization would sequentially mediate the relationship between SDO and support for Arabs' autonomy. Results revealed that Italians low on SDO have higher SIC (i.e., a more inclusive social identity) which endorses lower dehumanization of Arabs. This in turn explained the influence of SDO on support for Arabs' autonomy. These findings integrate the literature about SDO and SIC and enhance the understanding of the sociocognitive processes underlying people's support for international egalitarian movements.     

Predicting Chinese international students’ acculturation strategies from socio‐demographic variables and social ties

Different acculturation strategies (i.e. integration, assimilation, separation and marginalization) tend to contribute to different adaptation outcomes for international students. The current study examines Chinese international students’ acculturation strategies under the influence of sociodemographic variables and social ties they developed in the host country. A group of Chinese international students in Belgium (N = 183) participated in an online survey. The results indicate that integration was the most commonly adopted strategy among Chinese international students in Belgium, followed by separation, marginalization and assimilation. The results from three separate multiple regressions show that English proficiency, local language proficiency, prior adaptation experience and female were significant predictors of Chinese students’ social ties (i.e. host‐national ties, international ties and co‐national ties). The discriminant analysis identified host‐national ties, international ties, co‐national ties, local language proficiency and prior adaptation experience as important variables that can distinguish Chinese students’ acculturation strategies.     

The effects of social comparison,emotion induction,and parent vaccination status on parental intentions for COVID-19 vaccine uptake in children

Parental vaccine hesitancy—delays in vaccine uptake for children—is a significant public health concern. Using an online adult sample of U.S. parents (N = 183), the current research experimentally examined how exposure to cautious or risky social comparison models on social media (in terms of their COVID-19 beliefs and behaviors) influenced parental intentions to vaccinate their children against COVID-19 in the early stages of the pandemic. Additionally, we examined whether the influence of social comparison models was moderated by emotional state (fear or contentment) and parental vaccination status. Overall, we found that parents exposed to cautious (vs. risky) comparison models and vaccinated (vs. unvaccinated) parents reported greater vaccine intentions for their children. We further found that vaccination intentions were highest among unvaccinated parents after exposure to cautious (vs. risky) comparison models, whereas intentions were highest among vaccinated parents after exposure to risky (vs. cautious) comparison models (but only when induced to feel content). Overall, our findings highlight the importance of understanding the additive and interactive impact of psychological and situational factors in shaping parental vaccine hesitancy.     

Two alternative models of health behaviour and recovery from activity limitations due to acute injury: A prospective study

Cognitions influence recovery from activity limitations. In this study we aimed to independently test, compare and combine the common sense self-regulation model (CS-SRM) and social cognitive theory (SCT) in predicting recovery from activity limitations due to acute injury. Measures were gathered at two time points 5–6 weeks apart. The sample consisted of 146 university students (Mean age: 21.86, SD: 4.83, 62% female) with a heterogeneous range of injuries that limited their participation in physical activity. The dependent variable was recovery from activity limitations (Physical Functioning?–?Short Form-36). The predictor variables were measured using the Brief Illness Perception Questionnaire and SCT items designed according to theoretical recommendations. Time-line (TL) and self-efficacy (SE) were significant predictors of recovery in a multivariate analysis, controlling for reported pain at Time 1. A combined model including the best predictors from both models, TL (β = ?0.25, p < 0.05, R ² change = 0.17, p < 0.01) and SE (β =0.31, p < 0.05, R ² change = 0.05, p < 0.05), accounted for a significant amount of the variance in recovery from activity limitations. A combination of key variables from both models may be particularly useful for understanding the cognitive factors that influence recovery from activity limitations.     

Just banter? Friendship,teasing and experimental aggression in adolescent peer networks

Aggressive behaviour serves many useful social functions, yet can also have damaging consequences. In line with evidence showing adolescent development in social cognitive abilities, we hypothesized that the use of aggression would become more sophisticated with age. We investigated adolescent aggression towards peers using an experimental, hypothetical aggression paradigm, the hot sauce paradigm, in a school‐based social network setting. Participants (N = 162 aged 11–17, 98 male) indicated which strength of imaginary hot sauce they would allocate to each of their classmates. A Social Network Questionnaire quantified participants’ perceived dyadic social tie strength with each classmate, and the incidence of mutual or unilateral dyadic real‐world aggression (e.g. teasing). Participants allocated weaker hot sauce to peers with whom they reported strong, positive social ties and an absence of self‐reported unilateral real‐world aggression. With increasing cross‐sectional age, there was a decrease in the impact of social tie strength and an increase in the extent to which hot sauce allocation was predicted by self‐reported mutual real‐world aggression. This pattern of findings is consistent with young (vs. late) adolescent use of experimental, hypothetical Hot Sauce aggression to reflect real‐world animosity, while late adolescents’ behaviour is more subtle. These findings extend our understanding of the dyadic social context of adolescent aggressive behaviour using a novel experimental aggression paradigm.     

Social Support and the Religious Dimensions of Close Ties

Despite the influence of social support on physical and mental health, few studies have examined why some close ties are more supportive than others. Though religion provides a rich context for social interaction and a meaningful social identity, it has received little attention in the social support literature. A growing literature on religion and health offers insight into how religion affects social support processes. Using dyadic network data derived from the nationally representative 2006 Portraits of American Life Study, I examine how the religious dimensions of close, nonhousehold ties relate to provision of social support. Results from logistic regression analyses indicate that (controlling for a range of other social tie characteristics) same‐faith ties are significantly more likely to be sources of help “in times of need,” while religious discussion is a strong predictor of receiving both help and advice. The effect of religious homophily is strongest for evangelical Protestants and African‐American Protestants. My findings underscore the need for social support researchers to consider the role of religion in shaping support processes in close relationships.     

Social network ties and organizational citizenship behavior: evidence of a curvilinear relationship

ABSTRACT

We examine the relationships between employee friendship and advice network centrality and organizational citizenship behavior (OCB). Using social exchange theory as a basis, we argue that centrality within the workplace friendship network will be positively and linearly related to employee OCB. Further, we draw on conservation of resources theory and role theory to predict a curvilinear (inverse U) relationship between employees’ advice network centrality and OCB such that employees will engage in higher levels of OCB at moderate levels of advice centrality than at low and high levels. We test the theoretical model in an academic setting (using other reports of network ties and employee ratings of OCB) and a US-based organization (using other reports of network ties and supervisor ratings of OCB). The results support the idea of a curvilinear relationship between advice network centrality and OCB in both studies and a linear relationship between friendship centrality and OCB in Study 2. Practical implications and directions for future research are addressed.     

Social cognition models and health behaviour: A structured review

Abstract

A number of social cognition models have been developed to account for socio-demographic variations in health behaviour. This paper distinguishes between: (a) motivational, (b) behavioural enaction, and (c) multi-stage models of health behaviour. The models are evaluated in terms of advancement of existing knowledge and - where appropriate - predictive utility. Common themes that appear within- and between- these categories are discussed, with consideration of ways in which theory may be advanced by future research. Each approach has associated strengths and weaknesses, suggesting that a “consensus” approach to the study of health behaviour may prove fruitful. Identification of the key constructs across different model types would allow coherent integration and promote further understanding of the psycho-social determinants of health behaviour.