A PDF version of this module is available here.
The lead author would like to thank Frédéric Mertens and Marie Eve Rioux‐Pelletier, former students and now collaborators on many research projects linking social networks with health and environment issues. Although they did not contribute directly to this module, our continued collaboration is an inspiration for my work, as has been and continues to be my work with colleagues of Cinbiose and CoPEH‐LAC.
People are embedded in relationships forming patterns and structures that both constrain and enable beliefs and actions. We call these patterns social networks or human communication networks (Monge & Eisenberg, 1987; Monge & Contractor, 2003; Saint‐Charles & Mongeau, 2005). At its most basic, a network is the pattern formed by a set of links uniting a set of “nodes.” Both links and nodes can be of different natures: roads and cities, synapses and neurons, radio waves and routers. The qualifier “social” implies that links and nodes are related to living organisms, most often humans. Although one can study animal networks, human‐animal networks or artefacthuman networks, this module will focus on human networks, the networks created by the relationships between individuals or groups of individuals.
Social network analysis is both a paradigm and a method (Wellman, 1988; Borgatti & Lopez‐Kidwell, in press). As a paradigm, social network can be considered a complex systems perspective on relationships. Nowadays, in many of our societies, the focus is upon the individual, we see numerous studies on “identities,” we blame individuals for their choices, we invite them to become entrepreneurial, we promote healthy behaviours and lifestyles as if this all depended solely on the individual while, paradoxically, talking about social determinants of health.
The social network perspective proposes another lens: a focus on links, on relationships. From this perspective, identities are constructed and modified through our relationships. For example, rather then trying to “prove” that someone is a born leader (something that has been largely disproved – Fisher, 1986, Mongeau & Saint‐Charles, 2005; Stogdill, 1948, 1974) or a born entrepreneur, we look at the position of this person in the social network and see how this gives him or her opportunities for leadership and entrepreneurship (Brass & Krackhardt, 1999; Burt, 2000). As a method, social network analysis (SNA) offers a wealth of mathematical, qualitative and graphing
tools to analyse the patterns formed by relationships (Loblich & Pfaff‐Rudiger, 2011; Wasserman & Faust, 1994). People from many disciplines, such as anthropology, biology, communication, sociology, mathematics, physics, psychology, etc, have studied social networks. Although nourished by all of these disciplines, the perspective proposed here is communicational – meaning it is most interested in the process of sense‐making and self‐making through relationships.
The guiding principles of ecosystem approaches to health imply “dealing” with relationships: when working with transdisciplinarity and participation, there is a need to encourage the development of relationships between people from various contexts; and when integrating gender equity and social justice, one has to understand the underlying patterns of relationships between social groups and the norms that guide them. Using social network lenses as a heuristic, this module explores its potential to contribute to ecosystem approaches to health research and intervention.
The first section sets the scene as to 1) the importance of relationships in the construction of self and society; and 2) the “system view limitations” when looking at social networks. The second section presents some basic concepts used in the study of social networks in non‐technical language. Finally, the second section offers some ideas on how to use social network thinking as a heuristic in a research or intervention project without having to engage in the full (and cumbersome) process of social network analysis. Any combination of these modules can be taught, but module 3 ‐ being essentially conceptual ‐ is not really a stand‐alone; it can, however, be broken down into parts quite easily.
WORKING TERMS
|
Video CapsuleFor a sweeping example of building meaningful relationships to construct a network of allies you can have students watch Ernesto Ráez-Luna’s plenary presentation from the EcoHealth 2014 conference. This presentation also touches on health as a social construct, the importance of communication and the need to build a heuristic to further the field of ecohealth. |
This first session deals with the “definition” of the network that will be observed and studied. Links can be made with systems thinking [see Module 3: Complexity]. It can be conducted as a very short session by removing parts of it and combining it with any of the other sessions in this module.
Histories of individuals are often told in terms of their deeds, sometimes contextualised by social origins or the broader social context. But what if we were to tell our history through the lenses of the network of relationships we are embedded in? We would probably discover how influenced we are and have been by these relationships.
The idea of the importance of relationships is far from new: As a net is made up of a series of ties, so everything in this world is connected by a series of ties (Buddha, c 563‐483 BCE). More recently, we can think of the work of symbolic interactionists in the exploration of the importance of interactions and relationships (Blumer, 1969). And, of course, many social network studies show that attitudes, beliefs and behaviours are very much related to how and with whom people are connected (Borgatti & Lopez‐Kidwell, in press).
Along the lines of connectivity is the famous “Small World Theory” which postulates that everybody is connected to everybody on our planet by a mean of 6 interpersonal connections, painting a picture that you can easily have access to anybody and to the resources they have access to. While this mean is substantiated by the results of some studies (Milgram & Travers, 1977; Watts, 2003), it is, as any mean, not representative of the differences between individuals. The length of the path between one individual to another is very much affected by homophily and equity. Homophily is the tendency of humans to form relationships with similar others, such similarity being based on a combination of socio‐demographics attributes and attitudes, values and beliefs (McPherson, Smith‐Lovin, & Cook, 2001) – these relationships, in turn, contribute to the construction of our attitudes, values and beliefs. So, as we all know, there are “social groups” that do not mix easily together and there are status differentiations between these groups. So, in the end, the length of the path is influenced by social differentiations.
If a network is a set of “nodes” and “links,” then defining what constitutes the set, what (or who) are the nodes and what type of links we are talking about should be important (Laumann, 1983; Richards, 1985). Your first tools in doing this are the objectives of the research / action and some knowledge about the network you intend to study or understand. Whether you plan on doing a “real” social network analysis or you intend to be more observant about the relationships in your research / action (see section 3), these considerations are necessary.
Another way of talking about the “set” would be to talk about defining the boundaries of the system. Often times, what comes to mind are the more “formal” boundaries: members of a group or an organisation, inhabitants of a community, stakeholders in a watershed, and so on. This is a very logical place to start, but the questions we end up asking may change this: what is the purpose of understanding the network? Do you want to know how the information about health and environment might circulate in the community? Do you wish to find out who are the most trusted people in the community when it comes to health problems or farming issues? These two questions are easy ones: the name of the boundary – it’s the community. But what if you want to know more about the social capital of the community? Then it becomes trickier. Social capital has to do with how people support one another in a group or community, but it also has to do with the ability of community members to have access to resources outside the community. So where is the boundary here? The whole world? Hum… a bit ambitious.
Once you have defined the boundaries, you need to figure out the nodes. Will they be individuals? Couples? Households? Groups? Organisations? If you select households or organisations, do you need everybody or only representatives? If you opt for representatives, who will you chose: Official representatives? Secret informants? Men? Women?
Finally, which relationships are relevant? People have various types of relationships: work related relationships, friendships, support relationships, advice and influence relationships. Moreover, most people have what is called multiplex relationships combining affective, cognitive and behavioural aspects. So, if we are to map the network of a bounded group, which relationships are relevant for our objectives? The question is pertinent in terms of our objectives, but it cannot be answered “from the outside.” The meaning of “friendship” is very different from one culture to another (Bidart, 1997) and you cannot just ask people “who are your friends?” without having an idea of what it means to them.
All of the above means that you can’t go about doing a network study without having some knowledge of the types of relationship and of their meaning for the people involved. Fortunately, this doesn’t mean you have to launch a full 10‐year anthropological study before doing any kind of network study (see section 3 about “how to use a network perspective without doing a network study”).
Now, let’s talk a little bit about “personal networks.” So far, we have been talking mostly about the social network of a group, an organisation or a community (called “sociocentric networks”). But we can also map the personal network of an individual (an “egocentric network” or “ego network”) and many of the concepts and measures used with sociocentric networks can be applied to ego networks. As we have seen in the introduction, our relationships have a lot to do with who we are, which values and beliefs we hold. For example, knowing the composition and structure of the ego network of various individuals or groups of individuals may help understand if people have social support (Barrera, 1986, Cohen et al., 2000; Wellman et Wortley, 1990).
Activity 1: construction ofself through relationships: personal communication network1This exercise is self‐reflexive and helps seeing oneself as constructed through relationships. It invites participants to reflect upon how the relationships they have had throughout their life have influenced their way of being. All they have to do is follow the step‐by‐step instructions below. Copies can be made and students can be asked to complete the exercise on their own. An open discussion, of varying lengths, about what has emerged can be held soon after participants have completed the exercise. |
MATERIALS– Copies of the Past and Present Relationship Table – Copies of Circle no. 1 and Circle no. 2 diagrams |
INSTRUCTIONSSTEP 1: Past relationships
|
STEP 2: Present relationships
|
STEP 3: Keywords
|
STEP 4: Circles |
1Translated from Mongeau P (1982) Les réseaux d'influence. Psychologie 151: 43‐45. Adapted for CoPEH-Canada by Johanne Saint‐Charles, with the author’s permission.
TOTAL TIME: 60 minutes
From a 2‐3 page list of names (taken from a phonebook, for example), ask people to list every person they know or have known with any of those names or surnames. “Knowing” here is being able to recognize this person on the street (provided she or he has not changed too much ) and she or he would be able to recognize you. Stop the exercise after 10‐15 min. The result should be impressive for most. According to Degenne & Forsé (1994), our relationships are organised in “concentric circles” where there are close to 5000 acquaintances, 200 people in our entourage, 20 people with whom we have regular interactions and 2 or 3 confidents. These are means that do not take into account the context in which live (e.g. urban vs. rural), the social group to which we belong and the mobility possibilities we have. Have people examine their results in terms of gender, region and even name similarities (Mateos et al. 2011) for 15 minutes and discuss them (30 minutes).
Barrera MJ (1986) Distinctions Between Social Support Concepts, Measures, and Models. American Journal of Community Psychology 14:413‐445
Blumer H (1969) Symbolic interactionism: Perspective and method. Englewood Cliffs, NJ: Prentice Hall Borgatti SP, Lopez‐Kidwell V (in press) Network theory In: The SAGE Handbook of Social Network Analysis, Scott J and Carrington PJ (editors) London: Sage Publications
Cohen S, Underwood LG, Gottlieb BH (2000) Social support measurement and intervention. New York: Oxford University Press
Degenne A, Forsé M (1994) Les réseaux sociaux. Paris, Armand Colin
Laumann EOA‐M (1983) The Boundary Specification Problem in Network Analysis. Applied Network Analysis (p. 19‐34). Beverly Hills: Sage Mateos P, Longley PA, O’Sullivan D (2011) Ethnicity and Population Structure in Personal Naming Networks. PloS one 6:e22943
McPherson M, Smith‐Lovin L, Cook JM (2001) Birds of a Feather: Homophily in Social Networks 27:415‐444
Milgram S, Travers J (1977) An Experimental Study of the Small World Problem. Social networks: a developing paradigm (p. 465). New York: Academic Press
Mongeau P (1982) Les réseaux d'influence. Psychologie 151: 43‐45
Richards WDJ (1985) Data, Models, and Assumptions in Network Analysis. Organizational Communication: Traditional Themes and New Directions (p. 109‐128). Beverly Hills: Sage
Wasserman S, Faust K (1994) Social Network Analysis – Methods and Applications. Cambridge: Cambridge University Press
Watts DJ (2003) Six Degrees ‐ The Science of a Connected Age. New York: W. W. Norton & Company
Wellman B, Wortley S (1990) Different Strokes from Different Folks: Community Ties and Social Support. American Journal of Sociology 96:558‐88
Wellman B, Carrington PJ, Hall A (1988) Networks as Personal Communities. Social Structures: A Network Analysis (pp. 130‐184). Cambridge, UK: Cambridge University Press
The whole idea of mapping the social network of a group or a community is to elicit the patterns formed by the relationships and to try to understand how these patterns are related to some aspects of the community. For example, this could be its resilience (Mertens et al. 2008), its potential for the diffusion and appropriation of scientific research results (Saint‐Charles et al. 2012), its social capital (Krishna & Uphoff, 1999; Krishna, 2002), the successful completion of its goal (Pagliccia et al., 2010), natural resource management (Bodin and Crona, 2009), etc. Throughout the years, researchers have identified quite a few concepts and measures that help to “read” a network. Here we will present some of the basic ones which will help people to understand data from social network analysis.
A social network is never random: if you try to generate a series of random networks with the same number of links and nodes, your social network will always be skewed towards certain configurations. This is because there are human tendencies at play. Here we present 3 of these.
Triangle closure: when person « A » has a strong relationship with both persons « B » and « C » then there will probably be, at least, a weak relationship between B and C.
Homophily: People have a tendency to associate with people who resemble them. These people are said to share a homophily link. This bias refers to the “degree” of similarity of the individuals who interact (Rogers, 2003). Homophily is not a new concept (Lazarsfeld and Merton, 1954). Aristotle said people “love those who are like themselves” and Plato claimed, “similarity begets friendship” (McPherson et al., 2001). The first studies of this phenomenon, conducted in the beginning of the 20th century, showed that people in homophile networks resemble each other in terms of sociodemographic and psychological characteristics. Sex, age, ethnicity, membership
to the same group, values/attitudes/beliefs are all attributes which people could share (McPherson et al., 2001; Rogers, 2003). The relative “difference” of two people will translateinto a “distance” in the network (McPherson et al., 2001).
Reciprocity: the probability that a link will be reciprocal. If a link from “a” to “b” exists then the probability that a link from “b” to “a” exists is significantly greater than chance (Fararo & Skvoretz, 1984).
As we have seen, people tend to gather themselves along the lines of homophily, reciprocity and relational proximity (triangle closure). Such tendencies form “pockets of density” in a network and it is not unusual to find in a group, an organisation or a community “areas” where there are noticeably more relationships between a group of people then between these people and the rest of the network. With certain types of relationships (advice relationships, for example), we may even see that the network is not connected – that there are “components” meaning that there is no relational path between certain people (at least for the studied relationships). A good example of this would be the tendency of people in an interdisciplinary research network to seek advice primarily from people from their own research discipline, despite existing connections with people from other disciplines.
Scholars of social networks have created a wealth of measures to capture these groupings: cliques, clans, components, core and so on (Scott, 2000; Wasserman and Faust, 1994). But a simpler way to look at these components without having to go through all of the measures is to measure the density of the network and of its various components and to compare the measure thus obtained. Density is the number of existing relationships divided by the number of possible relationships based on the number of nodes in the network.
A starting hypothesis in resource management is that a higher network density leads to greater communication and trust between actors and, hence, a better outcome in co‐management of resources (Bodin and Crona, 2009). High density has also been shown to increase knowledge flow but only up to a certain point. When the network contains too many ties a homogenization of knowledge can occur which can be counter productive for resource management (Bodin and Crona, 2009).
From an individual point of view, the density of one’s personal network (often called “ego network”) can be a source of social and emotional support but it can also act as a prison (Saint‐Charles et al., 2008).
These various groupings eventually form larger structures, some of which are typically found in social networks. Here we will present two of these: the core‐periphery structure and the smallworld structure.
Core‐periphery structure: The core periphery structure is very frequent (Borgatti and Everett, 1999) and is to be expected in many groups and communities. For example, in natural resource management, core‐periphery structures can facilitate co‐management through 1) creating a centre where diverse ecological knowledge from the periphery can be concentrated (in the core) and 2) creating a hub (the core) of information dissemination to the periphery. The fact that there are not competing subgroups, but rather one coherent group, can also reduce issues related to “us‐vs.‐them” (Bodin and Crona, 2009). Communities of practice also often develop “core‐periphery” structures, a development aided by the fact that most communities of practice emerge from the coming together of a small group of devoted and passionate people (Saint‐Charles and Thoër, in press).
Risks exist though. Core‐periphery structures have the potential to succumb to power issues between the core and the periphery and these social networks can also be very sensitive to the removal of key individuals, reducing their resiliency (Bodin and Crona, 2009; Mertens et al., 2008). A study evaluating intersectoral action on health determinants in Cuba, which has a decentralized health care system, found that three sectors – health, education and the People’s Power Assembly – accounted for most of the observed links in the network (Pagliccia et al., 2010). The authors point out that to improve the interconnectedness of the network (network density) policy makers should strive for more intersectoral balance. The concentration of links to these three sectors creates a situation where the system is at risk of seeing these sectors control information and resources.
Small World: We have already talked about the limitations of the “Small World Theory” in terms of equity (or the lack of). That doesn’t mean that the theory doesn’t hold, only that we need to nuance its interpretation. Behind the theory, is a pattern frequently found in human networks (and other types of networks as well, see: Watts, 2003). Small World Networks have high local clustering and low path‐length. Figure 5 illustrates clearly that there are small densely connected groups themselves connected by a few “transversal” lines creating short‐cuts between the groups. The small groups are generally connected by strong ties, which tend to fragment the network into non‐connected small groups, while intergroup connections tend to be made by weak ties (Granovetter, 1973, 1983; Hansen, 1999; White & Houseman, 2003).
As we have seen earlier, who we are is strongly influenced by the relationships we hold. These relationships are not only influential in themselves (for example being influenced by the opinion of loved ones), but they are also embedded in a structure that can facilitate or hinder our actions. Social network analysts have developed concepts and measures to better understand the role of the structural position of a node (Borgatti and Lopez‐Kidwell, in press; Erickson, 1988). Here we will explore three concepts of “centrality.” These three concepts have been chosen because they are easily observable without having to do a comprehensive social network analysis.
When Moreno (1934) created “sociometry,” one of the ancestors of quantitative social network analysis, “popularity” was among the first concepts to be explored. “Popularity” (now more often called prestige) was the number of nominations received by an individual. In social network analysis this concept was expanded to include both received and sent nominations and was formalised as degree centrality. Indegree centrality is the part of degree centrality expressing popularity.
Degree centrality is easy to grasp as a concept and easy to measure as well. It is also easy to observe. There are other ways of being central which are equally important for the understanding of the network. In a seminal article in 1979, Freeman uncovered “three distinct intuitive conceptions of centrality” (Freeman, 1979 : 215): degree centrality, betweeness centrality, closeness centrality. Subsequent studies have supported these “intuitions” and
demonstrated their importance. It is far beyond the scope of this module to review all of these studies, but an exploration of these three basic notions should contribute to the development of a “network lens.”
Aside from the “popularity idea,” which is self‐explanatory, degree centrality, as measured by the number of links of a node, expresses the “communication activity” of an individual. Betweeness centrality, is the number of times an individual is on a path between two others who are themselves not directly connected. It has to do with the control of communication in the network. When these links connect two subgroups, rather than two individuals, they are referred to as “bridging ties.” Finally, closeness centrality measures the mean distance of a node to all other nodes in the network and has to do with the independence or efficiency of a node. Individuals who occupy a central position in a social network often have substantial influence. In natural resource management, “bridging ties” between subgroups allow the central individuals to have access to the specialized ecological knowledge of each group (Bodin and Crona, 2009). They can then facilitate the transfer of tacit knowledge between groups, if they are willing, that is.
Let us note in conclusion that these three measures have all been shown to be related to influence and to the diffusion of ideas or practices (Bodin and Crona, 2009; Brass, 1992; Ibarra and Andrews, 1993; Valente, 2010).
ACTIVITY 1: CONSTRUCTION OF THE GROUP'S SOCIAL NETWORKTOTAL TIME: 45‐90 minutes DESCRIPTION: This exercise is a group activity and helps to see oneself and others as constructed through relationships. This activity can follow a session on social networks to help grasp the concepts, but it is also a nice activity to build a feeling of community. If you are going through the exercise for this purpose it might be helpful to briefly demonstrate a few of the social network concepts before embarking. It can be a nice activity to end a course or workshop on as it gives a feeling of something that is growing. MATERIALS: several large pieces of paper, tape, a large empty wall, markers, a chair. BEFORE THE ACTIVITY: tape together as many pieces of paper as will fill your wall. Tape the |
STEP 1: Constructing the group’s social network prior to the course (45 minutes) |
STEP 2: Constructing the group’s social network after the course (30 minutes) |
ACTIVITY 2: CHANGING THE FOCUS OF ONE'S PROJECTTOTAL TIME: 60-90 minutes |
STEP 1: Analysis of one’s own project (20 minutes) |
STEP 2: Small group discussion (40 minutes) |
STEP 3: Visualizing the network (30 minutes) |
Bodin Ö, Crona BI (2009) The role of social networks in natural resource governance: What relational patterns make a difference? Global Environmental Change 19:366‐374
Borgatti SP, Everett M (1999) Models of Core‐Periphery Structures. Social Networks 21:375‐395 Borgatti SP, Lopez‐Kidwell V (in press) Network theory In: The SAGE Handbook of Social Network Analysis, Scott J and Carrington PJ (Editors) London: Sage Publications
Brass DJ (1992) Power in Organizations: A Social Network Perspective. Research in Politics and Society 4:295‐323
Erickson B (1988) The Relational Basis of Attitudes. Social Structures: A Network Approach (pp. 99‐121). Cambridge: Cambridge University Press
Fararo TJ, Skvoretz J (1984) Biased networks and social structure theorems: Part II. Social Networks 6:223‐258. doi:10.1016/0378‐8733(84)90012‐1
Freeman LC (1979) Centrality in Social Networks : Conceptual Clarification. Social Networks 1:215‐239
Granovetter MS (1983) The Strength of Weak Ties: A Network Theory Revisited. Sociological Theory 1:201‐233
Granovetter MS (1973) The Strength of Weak Ties. American Journal of Sociology 78:1360‐1380
Hansen MT (1999) The search‐transfer problem: The role of weak ties in sharing knowledge across organization subunits. Administrative Science Quarterly 44:82‐111
Ibarra H and Andrews SB (1993) Power, social influence and sensemaking: effects of network centrality and proximity on employee perceptions. Administrative Science Quarterly 38:277‐303. doi:0001‐8392/93/3802‐0277
Krackhardt D (1990) Assessing the Political Landscape: Structure, Cognition and Power in Organizations. Administrative Science Quarterly 35:342‐369
Krishna (2002) Active Social Capital: tracing the roots of development and democracy. New York: Columbia University Press
Krishna A, Uphoff N (1999) Social Mapping and Measuring Social Capital: A conceptual and Empirical Study of Collective Action for Conserving and Developing Watersheds in Rajasthan, India. Washington DC: The World Bank
Lazarsfeld PF, Merton RK (1954) Friendship as a Social Process. Freedom and Control in Modern Society (pp. 18‐66). New York: Nostrand D. Van
McPherson M, Smith‐Lovin L, Cook JM (2001) Birds of a Feather: Homophily in Social Networks 27:415‐444
Mertens F, Saint‐Charles J, Lucotte M, Mergler D (2008) Emergence and robustness of a community discussion network on mercury contamination and health in the Brazilian Amazon. Health Education and Behavior 35:509‐521
Moreno JL (1934) Who Shall Survive? Washington, Dc: Nervous and Mental Disease Publishing Company
Pagliccia N, Spiegel J, Alegret M, Bonet M, Martinez B, Yassi A (2010) Network analysis as a tool to assess the intersectoral management of health determinants at the local level: A report from an exploratory study of two Cuban municipalities. Social Science & Medicine 71:394‐399
Rogers EM (2003) Diffusion of Innovations. New York: Free Press
Saint‐Charles J, Mongeau P, Biron J‐F (2008) A Communication Perspective on Video Lottery Terminals. International Gambling Studies 8:233‐247 doi:10.1080/14459790802405848
Saint‐Charles J, Rioux‐Pelletier M‐E, Mertens F, Mongeau P (2012) Diffusion of environmental health information: the role of sex‐ and gender‐differentiated pathways. What a Difference Sex and Gender Make: A Gender Sex and Health Research Casebook. Can be downloaded on the Web : http://www.cihr‐irsc.gc.ca/e/44734.html [French version: Saint‐Charles, J., Rioux‐Pelletier, M.‐E., Mertens, F., & Mongeau, P. (2012b). Diffusion d’informations en santé environnementale : le rôle des chemins différenciés selon le sexe et le genre. Recueil de cas sur la recherche liée au genre, au sexe et à la santé.]
Saint‐Charles J, Thoër C (in press) Les communautés de pratique virtuelles en santé In: Internet et Santé, Lévy J and Thoer C (Editors) Presses de l’université du Québec
Scott J (2000) Social network analysis a handbook. London: Sage
Valente TW (2010) Social Networks and Health. New York: Oxford University Press
White DR, Houseman M (2003) The Navigability of Strong Ties: Small Worlds, Tie Strength andNetwork Topology. Complexity 8:72‐81
This short section is intended to highlight some ways one can integrate a social network perspective into one’s ork without having p w to learn the all the tricks of the trade.
What if Uncle Jim and Aunt Alice were not the enemies you thought them to be but rather secret lovers? It is not uncommon to “discover” such surprising connections between people we thought we knew well (Casciaro et al, 1999). This points out that it is not easy to have a perfectly clear picture of a social network by observation alone. This is not pointed out simply to discourage you from using your observational skills to better understand a social network. Rather, it is to invite you to refine those skills. For example:
When doing ecohealth research or conducting an intervention, we are often outsiders and having informants tell us about the network might be a good idea. But, how to choose a “good” informant? There is no perfect recipe here, but you might want to remember that, among other things, positive affectivity and central position in the network have been shown to positively influence the accurate perceptions of other’s relationships in a network (Casciaro et al, 1999; Krackhardt, 1990). Care should be taken here though: the very positive, very central individual in a community is also well placed not to see discontent or not to be aware of factions.
Inspired by the activity “Construction of the group’s social network” proposed in Section 2 of this module, you could devise an activity for the group you are working with to have them draw a picture of their network. Be mindful of sensitivities – don’t ask people to say publicly who they like or dislike – rather choose a more “public” relationship. Have people talk about the emerging picture they are creating.
If your research or intervention involves using a questionnaire, you may find it interesting to add one network question to it. For example:
Casciaro T, Carley KM, Krackhardt D (1999) Positive Affectivity and Accuracy in Social Network Perception. Motivation and Emotion 23:285‐306
Clark L (2006a) Network Mapping as a Diagnostic Tool. La Paz, Bolivie: Centro Internacional de Agricultura Tropical
Clark L (2006b) Manual para el Mapeo de Redes como una Herramienta de Diagnóstico. La Paz, Bolivie: Centro Internacional de Agricultura Tropical
Krackhardt D (1990) Assessing the Political Landscape: Structure, Cognition and Power in Organizations. Administrative Science Quarterly 35:342‐369
Waltner‐Toews D, Kay J, Murray TP, Neudoerffer C (2003) Adaptive Methodology for Ecosystem Sustainability And Health (AMESH) : An Introduction. Community Operational Research: OR and Systems Thinking for Community Development (Contemporary Systems Thinking). Kluwer Press
Adamsa AM, Madhavanb S, Simon D (2002) Women’s social networks and child survival in Mali. Social Science & Medicine 54:165‐178 Borgatti SP, Lopez‐Kidwell V (in press) Network theory In: The SAGE Handbook of Social Network Analysis, Scott J and Carrington PJ (editors) London: Sage Publications
Boulay M, Valente TW (2005) The Selection of Family Planning Discussion Partners in Nepal. Journal of Health Communication 10:519–536
Brass D, Krackhardt D (1999) Social Capital for Twenty‐first Century Leaders. In: Out‐of‐the Box Leadership Challenges for the 21st Century Army, Hunt, JG and Phillips, RL (Editors), Emerald Group Publishing Limited (pp. 179‐194) http://www.andrew.cmu.edu/~krack/academic/papers.html
Burt RS (2000) The network structure of social capital. In: Research in Organizational Behavior, Sutton RI and Staw BM (editors), JAI Press.
Fisher AB (1986) Leadership. When Does the Difference Make a Difference, In: Communication and Group Decision‐making, Hirokawa, RY and Poole MS (Editors), (pp. 198‐215). Beverly Hills: Sage
Loblich M, Pfaff‐Rudiger S (2011) Network analysis: A qualitative approach to empirical studies on communication policy. International Communication Gazette 73(7):630‐647. doi:10.1177/1748048511417159
Mertens F, Saint‐Charles J, Mergler D (In press) Social communication network analysis of the role of participatory research in the adoption of new fish consumption behaviors Social Science & Medicine
Mertens F, Saint‐Charles J, Demeda K, Castro M, Passos C J S, Lucotte M, Guimarães JRD, et al. (2006) Community Network Analysis For Addressing Gender, Equity And Participation In Ecohealth Research. Proceedings of IDRC’s participation in the 11th World Congress on Public Health. Rio de Janeiro, Brazil: CRDI
Mertens F, Saint‐Charles J, Lucotte M, Mergler D (2008) Emergence and robustness of a community discussion network on mercury contamination and health in the Brazilian Amazon. Health Education and Behavior 35:509‐521
Mertens F, Saint‐Charles J, Mergler D, Passos CJ, Lucotte M (2005) A network approach for analysing equity in community involvement in participatory research. Ecohealth 2:1‐15
Monge PR, Contractor NS (2003) Theories of Communication Networks. New York: Oxford University Press
Monge PR, Eisenberg EM (1987) Emergent Communication Network. Handbook of Organizational Communication: an Interdisciplinary Perspective. Beverly Hills: Sage
Mongeau P, Saint‐Charles J (2005) Communication et émergence du leadership dans les groupes, In: Communication : horizons de pratiques et de recherche, Saint‐Charles, J., and Mongeau, P. (editors). Montréal, Presses de l’Université du Québec
Saint‐Charles J, Rioux‐Pelletier M‐E, Mertens F, Mongeau P (2012) Diffusion of environmental health information: the role of sex‐ and gender‐differentiated pathways. What a Difference Sex and Gender Make: A Gender Sex and Health Research Casebook. Can be downloaded on the Web : http://www.cihr‐irsc.gc.ca/e/44734.html [French version: Saint‐Charles, J., Rioux‐Pelletier, M.‐E., Mertens, F., & Mongeau, P. (2012b). Diffusion d’informations en santé environnementale : le rôle des chemins différenciés selon le sexe et le genre. Recueil de cas sur la recherche liée au genre, au sexe et à la santé.]
Saint‐Charles J, Mongeau P (2005) L’étude des Réseaux Humains de Communication. In : Communication : horizons de pratiques et de recherches (pp. 73‐99). Québec: Presse de l’Université du Québec
Stogdill RM (1948) Personal factors associated with leadership. Journal of Psychology 25:35‐71
Stogdill RM (1974) Handbook of leadership. New York: Free Press
Valente TW (2010) Social Networks and Health. New York: Oxford University Press
Wasserman S, Faust K (1994) Social Network Analysis – Methods and Applications. Cambridge: Cambridge University Press
Wellman B (1988) Structural Analysis: from Method and Metaphor to Theory and Substance. In: Social Structures ‐ A Network Approach, Wellman B and Berkowitz SD (editors), Greenwich, Connecticut: JAI Press, pp. 19‐61