Predicting the structure of a communications network from recalled data

In a series of papers on informant accuracy in social network data, Bernard, Killworth, and more recently, Sailer, have concluded that “what people say, despite their presumed good intentions, bears no useful resemblance to their behavior” (Bernard, Killworth, and Sailer 1982: 63). In this paper we reanalyze one of the data sets (the technical group) utilized by Bernard, Killworth and Sailer in arriving at their conclusions. Unlike Bernard et al. we find that the observed behavior data corresponds closely to the recalled data. Using different methods of analysis we find that the verbal recall data can be used to predict structural aspects of the observed data. Two major findings emerge from our analysis: first, the more similarly two people judge the communication pattern of others, the more they interact with each other, and, second, the more two people share accurate knowledge of others, the more they interact with each other. Implications of our findings for the assertions of Bernard, Killworth and Sailer are discussed.     

Predicting informant accuracy from patterns of recall among individuals

This paper examines the question of informant accuracy in reporting patterns of communication in face-to-face groups. We are attempting to establish the extent to which it is possible to predict individual differences in accuracy from the patterns of recall among informants. We use data from a series of studies by Bernard, Killworth, and Sailer (Killworth and Bernard, 1976, Killworth and Bernard, 1979Bernard and Killworth 1977;Bernard et al., 1980, Bernard et al., 1982) in which they collected observed behavior interaction frequencies and subsequently asked informants to recall and rate the degree of previous communication. In this paper we attempt to predict the accuracy of recall, i.e. how well each individual's ranking corresponds to the overall observed interactions for the group as a whole, by looking solely at the recall ranking. Using this method, we are able to account for a major share of the variance in accuracy among the subjects. We outline a theory and method for predicting accuracy based upon recall data that may be generalized to a variety of situations beyond social interaction data.     

The politics of Canadian social scientists: A comment on Baer and Lambert

À partir d'une analyse secondaire d'un sondage national mené en 1983, Baer et Lambert (1990) soutiennent que les études universitaires ont un effet idéologique unidirectionnel allant dans le sens d'un renforcement de l'idéologie dominante. Us affirment de plus avoir démontré de façon «convaincante» que les études en sciences sociales n'ont pas un effet opposé; que les spécialistes des sciences sociales au Canada, loin d'être radicaux, sont probablement plus conservateurs que la population en général, et que leurs résultats contredisent les conclusions de Guimond, Palmer et Bégin (1989). Nous présentons une brève analyse de l'étude de Baer et Lambert (1990) démontrant qu'en raison de certains problèmes dans I'interprétation de leurs résultats en plus de la négligence d'un ensemble de recherches sur le plan national et international, ces auteurs présentent une conclusion fallacieuse. Il existe des démonstrations solides de la libéralisation des attitudes des étudiants sous l'influence des sciences sociales, démonstrations qui ne sont aucunement remises en question par les résultats de Baer et Lambert. On the basis of their reanalysis of 1983 Canadian survey data, Baer and Lambert (1990) argue that the ideological effects of university education are unidirectional, serving to reinforce the dominant ideology. They further claim to have ‘convincingly demonstrated’ that university studies in the social sciences do not have the opposite effect; i.e. that Canadian social scientists, far from being radicals, are probably more conservative than the population at large, and that their findings contradict the conclusion of Guimond, Palmer and Bégin (1989). We present a brief analysis of the Baer and Lambert (1990) study, showing that through a combination of problems in the interpretation of their results, along with their neglecting a substantial body of national and international research evidence, those authors' conclusions are misleading. There is solid evidence that studying the social sciences is related to a liberalization of students' attitudes, evidence that is in no way contradicted by the Baer and Lambert (1990) results.     

Advances in Exponential Random Graph (p*) Models Applied to a Large Social Network

Recent advances in statistical network analysis based on the family of exponential random graph (ERG) models have greatly improved our ability to conduct inference on dependence in large social networks (Snijders 2002, Pattison and Robins 2002, Handcock 2002, Handcock 2003, Snijders et al. 2006, Hunter et al. 2005, Goodreau et al. 2005, previous papers this issue). This paper applies advances in both model parameterizations and computational algorithms to an examination of the structure observed in an adolescent friendship network of 1,681 actors from the National Longitudinal Study of Adolescent Health (AddHealth). ERG models of social network structure are fit using the R package statnet, and their adequacy assessed through comparison of model predictions with the observed data for higher-order network statistics.For this friendship network, the commonly used model of Markov dependence leads to the problems of degeneracy discussed by Handcock (2002, 2003). On the other hand, model parameterizations introduced by Snijders et al (2006) and Hunter and Handcock (2006) avoid degeneracy and provide reasonable fit to the data. Degree-only models did a poor job of capturing observed network structure; those that did best included terms both for heterogeneous mixing on exogenous attributes (grade and self-reported race) as well as endogenous clustering. Networks simulated from this model were largely consistent with the observed network on multiple higher-order network statistics, including the number of triangles, the size of the largest component, the overall reachability, the distribution of geodesic distances, the degree distribution, and the shared partner distribution. The ability to fit such models to large datasets and to make inference about the underling processes generating the network represents a major advance in the field of statistical network analysis.     

From locality to continent: A comment on the generalization of an experimental study

Kocher et al. (2008) find that conditional willingness to contribute to a public good is considerably stronger at a U.S. research site, Appalachian State University, than at sites in Europe and Asia. I find that the willingness at Brown University, in Rhode Island, is not significantly different from that in Europe and Asia in Kocher et al. (2008). Clear evidence on significantly different demographic statistics between these two U.S. research sites is provided. This implies that in the experimental study of societies, or more generally in empirical studies, we need exercise care to generalize from results at a specific research site to geographically broader conclusions.     

The Politics of Canadian Social Scientists: A Reply to Guimond and Palmer

TWO VERY DIFFERENT PICTURES regarding the relationship between the education system and socialization into dominant ideology emerge from the work of Guimond et al. (1989a; 1989b) and ourselves (Baer and Lambert, 1990). Using data from a single Quebec university, Guimond et al. argued that the conservatizing effect of education in the case of some programmes of study (e.g., commerce) was offset by a radicalizing effect in social science education. In their own words, "it can hardly be argued that education increases support for the status quo when one looks at the data relating to social science students.… [Our research program] clearly suggest[s] that studying the social sciences is a radicalizing experience" (Guimond et al., 1989b: 207; our emphasis). In our case, we used national, cross-sectional data to make a case for the substantive insignificance of any such social science effect in the long run, and to support our contention that overall, the education system in Canada has probably had a conservatizing influence.     

Social Support: A Mixed Blessing for Women in Substance Abuse Treatment

Using a personal social network framework, this qualitative study sought to understand how women in substance abuse treatment describe their network members' supportive and unsupportive behaviors related to recovery. Eighty-six women were interviewed from residential and outpatient substance abuse treatment programs. Positive and negative aspects of women's social networks were assessed via open-ended questions. Analysis was guided by grounded theory techniques using three coders. The findings extend classic social support concepts such as emotional, tangible, and informational support. Practice implications are presented in light of the potential roles network members may play in substance use and recovery.     

The Real Controversy About Child Sexual Abuse ResearchContradictory Findings and Critical Issues Not Addressed by Rind, Tromovitch, and Bauserman in Their 1998 Outcomes Meta-Analysis

This article presents a review of all types of child sexual abuse research ignored by Rind, Tromovitch, and Bauserman in their 1998 meta-analytic study. Eight major findings are addressed. Altogether, these findings demonstrate the narrow focus of the Rind et al. meta-analysis. By restricting a supposedly broad meta-analysis to only some of the research and population in question, the conclusions Rind et al. drew regarding this complex topic (primarily, that adult-child sex is not necessarily harmful to children) are invalid.     

Models,methods and network topology: Experimental design for the study of interference

How should a network experiment be designed to achieve high statistical power? Experimental treatments on networks may spread. Randomizing assignment of treatment to nodes enhances learning about the counterfactual causal effects of a social network experiment and also requires new methodology (ex. Aronow and Samii, 2017a, Bowers et al., 2013, Toulis and Kao, 2013). In this paper we show that the way in which a treatment propagates across a social network affects the statistical power of an experimental design. As such, prior information regarding treatment propagation should be incorporated into the experimental design. Our findings justify reconsideration of standard practice in circumstances where units are presumed to be independent even in simple experiments: information about treatment effects is not maximized when we assign half the units to treatment and half to control. We also present an example in which statistical power depends on the extent to which the network degree of nodes is correlated with treatment assignment probability. We recommend that researchers think carefully about the underlying treatment propagation model motivating their study in designing an experiment on a network.     

Care management and social work: A case study of the role of social work in hospital discharge to residential or nursing home care

The role of social work with vulnerable adults has changed markedly in many countries, including England and Wales following the 1990 NHS and Community Care Act. The basis for this paper is a small qualitative study of the social work role in helping people entering residential or nursing care in the independent sector following an emergency hospital admission in England (Phillips and Waterson 1997). This double transition is an under-researched area (Bywaters 1993; Downs and Crossan 1999). Furthermore, the literature on hospital discharge to date relates to developing policy objectives (Department of Health 1989, 1992, 1994a) or a critique of them (Pearson 1994; Cresswell 1994; Nazarko 1994) or the impact on practice, highlighting issues of inter-disciplinary cooperation (Rachman 1993; Higgins et al. 1994; Davies and Connolly 1995a; Clark et al. 1997; Healey et al. 1999). Until very recently, there has been much less about the users and carers perspective of the effects on the social work role in this area (Hardy et al. 1999; Henton et al. 1999; MacDonald 1999; Chesterman et al. 2001; Hellstrom and Hallberg 2001). The research findings review the core tasks of care management and point to a devaluing of the 'social' dimension, which had previously been the mainstay of hospital social work, and which is precisely what users and carers continue to value. Whilst this paper is written from a UK perspective, and focuses on a small study of an atypical area of work, this finding raises issues of wider interest to those concerned with the health and social care of vulnerable adults in a range of transnational contexts. It raises fundamental questions about the nature of contemporary European social work.     

The Real Controversy About Child Sexual Abuse Research: Contradictory Findings and Critical Issues Not Addressed by Rind,Tromovitch, and Bauserman in Their 1998 Outcomes Meta-Analysis

SUMMARY

This article presents a review of all types of child sexual abuse research ignored by Rind, Tromovitch, and Bauserman in their 1998 meta-analytic study. Eight major findings are addressed. Altogether, these findings demonstrate the narrow focus of the Rind et al. meta-analysis. By restricting a supposedly broad meta-analysis to only some of the research and population in question, the conclusions Rind et al. drew regarding this complex topic (primarily, that adult-child sex is not necessarily harmful to children) are invalid.     

Effects of competition on collective learning in advice networks

This paper looks at the effect of identifying alters as direct competitors on their selection as advisors. We differentiate between two kinds of competition: cut-throat vs friendly. We argue that, unlike cut-throat competition, friendly competition makes collective learning possible as a social process: when knowledge is built in interactions that are able to mitigate the negative effects of status competition and take place in homophilous social niches; and when the quality of this knowledge is guaranteed by members with epistemic status in these niches. Social niches and status facilitate advice seeking and collective learning because advice seeking between direct competitors is not obvious even when members have a common interest in sharing advice – a learning-related dilemma of collective action. We apply this reasoning to a network dataset combining identification of direct competitors and selection of advisors among the elite of cancer researchers in France. We use a procedure of multiplex stochastic block-modeling designed by Barbillon et al. (2015) to measure the effect of these identifications of direct competitors on the structure of the advice network. Results obtained with this dataset support our theory.     

Vietnam as a class war: Myth or reality?

This article reports an extension of the analyses of the fatalities in the Vietnam War presented by Barnett, Stanley, and Shore (1992) as printed in the September‐October 1992 Operations Research. The additional analyses involved an examination of results by Barnett et al. in numeric as well as percentage form and a consideration of how the revised analyses generalize to the population of all American fatalities in Vietnam. These analyses yielded a different conclusion about the role of economic class in the fatalities in the Vietnam War than those drawn by Barnett et al. They suggested that the term class war was not supported by their data analyses. Our analyses indicate that their conclusion is incorrect. The lower class sacrificed considerably more lives than the upper class.     

Social and geographic contexts of adolescent substance use: The moderating effects of age and gender

This study investigates whether age and gender moderate the effects of social network and neighborhood contexts on adolescent substance use using a spatially embedded, egocentric social network data set comprised of 254 urban adolescents. Results indicate that substance use is enhanced by being older and male, as well as the presence of neighborhood concentrated disadvantage and interaction with substance using peers at adolescents’ perceived risky places. Older adolescents are more strongly influenced by social and neighborhood contexts than younger adolescents, where ages 14–16 appear to be a key transition age for the emergence of contextual effects on substance use.     

The Peer Context of Adolescent Substance Use: Findings from Social Network Analysis

To examine the peer context of adolescent substance use, social network analysis was used to measure three domains of attributes of peer networks: social embeddedness, social status, and social proximity to substance users. The sample was a panel of 5,104 sixth, seventh, and eighth graders in three public school systems surveyed every 6 months for five assessments. Hierarchical generalized linear models showed that adolescents less embedded in the network, with greater status, and with closer social proximity to peer substance users were more likely to use substances. Also, adolescents in less dense networks and networks with higher smoking prevalence were more likely to smoke and use marijuana. Results establish the utility of social network analysis for measuring peer context and indicate that conventionality of relationships—having friends in the network, being liked but not too well liked, and having fewer friends who use substances—is most beneficial.     

A Nash bargaining model for simple exchange networks

Starting from exogenously given negotiation networks, sociological exchange theories explain bilateral divisions of fixed surpluses (e.g., cake, dollar) as consequences of the partners’ structural embeddedness. In accordance with the available experimental evidence, we focus on simple exchange networks and present a formal model for predicting profit splits from such structures. In contrast to other approaches, we combine the generalized Nash bargaining solution from game theory with the assumption that both relational features and network positions affect exchange outcomes. The resulting point predictions for profit splits correspond closely with experimental results obtained by Cook et al. [Cook, K.S., Emerson, R.M., Gillmore, M.R., Yamagishi, T., 1983. The distribution of power in exchange networks: theory and experimental results. American Journal of Sociology 89, 275–305], Lovaglia et al. [Lovaglia, M.J., Skvoretz, J., Willer, D., Markovsky, B., 1995. Negotiated exchanges in social networks. Social Forces 74, 123–155], Markovsky et al. [Markovsky, B., Willer, D., Patton, T., 1988. Power relations in exchange networks. American Sociological Review 53, 220–236], Simpson and Willer [Simpson, B., Willer, D., 1999. A new method for finding power structures. In: Willer, D. (Ed.), Network Exchange Theory. Praeger, Westport, CT, pp. 270–284], Skvoretz and Fararo [Skvoretz, J., Fararo, T.J., 1992. Power and network exchange: an essay toward theoretical unification. Social Networks 14, 325–344], Skvoretz and Willer [Skvoretz, J., Willer, D., 1993. Exclusion and power: a test of four theories of power in exchange networks. American Sociological Review 58, 801–818] as well as Yamagishi et al. [Yamagishi, T., Gillmore, M.R., Cook, K.S., 1988. Network connections and the distribution of power in exchange networks. American Journal of Sociology 93, 833–851].     

News and information leadership in the digital age

This paper examines information networks on social media to draw conclusions about influence relationships among members of the mass media. The project considers social networks and information patterns using Twitter data, first at the newspaper level and second at the journalist level. Using a computational approach, we look for evidence of elite-directed information flows, as well as exploring whether we find evidence of an increase in the democratization of newsmaking. This study finds that elite voices continue to dominate information networks in the digital age; however, it also finds evidence that information can move expeditiously from journalists in local and regional outlets to elite ones, and vice versa. We move further to explore the content of tweets among the journalist network, finding that there are substantial, direct interactions among elite and regional and local journalists. Our results taken together uncover new network patterns and provide a novel insight on the role of information technologies in newsmaking in the digital age.     

Does self-assessment in a group help students to learn?

'One of the greatest misconceptions about self-assessment is that it can be undertaken in isolation from others' [Boud (1995) Enhanced Learning through Self-assessment (London, Kogan Page), p. 200]. Social work practice - the experience from which we learn in social work education [Boud et al. (1985) Reflection: Turning Experience into Learning (London, Kogan Page)] - operates in the context of social relationships. This means that learning develops within its social environment, be it a student group, workshop, team or project. If self-assessment is to be an effective aspect of reflective learning in social work education, then it is important that social work students develop skills in self-assessment in group contexts. This paper charts an exploration of self-assessment in a group setting in order to reach some conclusions about how self-assessment can be encouraged amongst social work students as they carry out group assignments.     

The ecology of social networks in four urban neighborhoods

This paper re-examines the question of the social “fabric” of urban neighborhoods on the basis of residents' personal networks. Data were collected on the number, relative intimacy, and spatial distribution of social relationships among residents of two ethnically homogenous and two ethnically heterogenous neighborhoods in a medium-sized city in the midwestern United States. The analysis focused on spatial distributions and variables associated with differences in the average number or intimacy of neighborhood network ties. Herbert Gans had predicted that in heterogenous neighborhoods residential proximity would be a less important factor in social network formation than has previously been reported for socially homogeneous residential settings (especially Festinger et al. 1950). The results from this study indicated that the effects of proximity were more, rather than less, reflected in the spatial distribution of social relationships in the ethnically heterogenous neighborhoods. The face-block was identified as an important socio-spatial unit in all four neighborhoods.     

Parental substance misuse and child care social work: research in a city social work department in England

This research set out to establish evidence about the scale and impact of and response to parental substance misuse in child care social work teams in a city social services department in England. The article draws on some aspects of the data collected in the research, which includes: a snapshot survey of all child care social work caseloads in the city; group interviews with practitioners and parents in recovery; individual interviews with parents using a pilot project that focused on parental substance misuse. The research provides evidence of parental substance misuse as a key factor that needs greater consideration within child care social work assessments and as an issue to target in developing preventative responses to child welfare concerns. Child care social workers are shown to need specialist support in undertaking this task to best effect. Copyright © 2004 John Wiley & Sons, Ltd.