Tracing Power and Influence in Networks:Net-Map as a Tool for Research and Strategic Network Planning

Eva Schiffer and Douglas Waale
Discussion Paper No. 772 June 2008
Related Resource Net-Map Toolbox Blog

Abstract: Believing that complex problems call for complex solutions and that stakeholders should have a say in policies that concern them, policymakers have strongly promoted the development of forums and organizations made up of many stakeholders to address complex governance issues such as water management. Both developing and developed countries have instituted multistakeholder water governance bodies on local, national, and international levels. However, while the belief is strong that these integrated bodies should improve governance, how and to what extent that actually happens is still unclear, not only because of the complexity of the matter but also due to a lack of appropriate research tools for the analysis of complex governance systems.
Continue reading “Tracing Power and Influence in Networks:Net-Map as a Tool for Research and Strategic Network Planning”

4th UK Social Networks conference

Date: Friday 18 July 2008 – Sunday 20 July 2008
Venue: University of Greenwich, London

“The UK Social Network Conference offers an interdisciplinary venue for social and behavioral scientists, sociologists, educationalists, political scientists, mathematicians, computer scientists, physicists, practitioners and others to present their work in the area of social networks. The primary objective of the Conference is to facilitate interactions between the many different disciplines interested in network analysis. The Conference provides a unique opportunity for the dissemination and debate of recent advances in theoretical and experimental network research.”

Editor’s note: I have included this event because I think social network analysis tools are very relevant to the task of describing and evaluating development project plans and progress. See more on this issue here http://mande.co.uk/special-issues/network-models/

Net-Map Training in Washington

Net-Map Tool- Training for Research and Impact Management

Net-Map is a visualization approach to Social Network Analysis that was developed by IFPRI post-doc Eva Schiffer in her research on multi-stakeholder water governance in northern Ghana. Net-Map is a tool for understanding and improving complex social situations with multiple actors, diverse goals and unclear distribution of influence.

“You want to learn how to use Net-Map in your own specific context, get a first idea of the social network measures we use to understand our data, be guided through the process of data entry and ask me all your remaining questions?

Why not participate in our IFPRI seminar:

Net-Map Tool-Pool Training for Research and Impact Management

on the 14th May in Washington DC?

The seminar has two parts:

Part 1: An Introduction to Net-Map (12:30 PM to 2:00 PM)

Part 2: Practical Training on Net-Map (2:30 PM to 4:30 PM)
(GMT-05:00) Eastern Time (US & Canada)

This seminar is aimed both at researchers and practitioners who want to use influence network visualisation in their work. It does not require previous knowledge in Social Network Analysis.

Go here (236 K) to find out more, and contact me at ifpri-netmap@cgiar.org if you want to either participate in person or via the internet.”

Posted on by Eva Schiffer, (on her website, Net-Map Toolbox: Influence mapping of social networks)

SlideShare presentations on Social Network Analysis: A list

Please note: When you visit the SlideShare website to see these slideshows you can view these presentations on screen, or download them (in some but nor all cases). You can also make public comments on any of the slides that you see, using the Comment box under the screen.

If you have a presentation on social network analysis that you think visitors to the MandE NEWS website would like to see, then upload it to www.slideshare.net then send me (Rick Davies) the link to that presentation. I will then list it here, so others can find it.

PS: You can add an audio track to the presentation very easily. Use open-source Audacity to record the track in MP3 format, upload it to www.slideshare.net and then follow instructions to link the right bit of audio to the right part of the presentation. Easy!

VisuaLyzer software: for visualising and analysing networks

There are now many different software packages available that can be used to visually represent networks, and to generate many different statistical measures of their structure. Unfortunately many of these involve a steep learning curve, and involve far more bells and whistles than I need.  VisuaLyzer is my favourite software package because it is very user friendly, and easy to use.

VisuaLyzer is produced by mdlogix, USA. You can download a trial version or buy a copy from this part of their website. For more information contact Allen Tien <allen@mdlogix.com> at mdlogix. If you do contact him, please mention you heard about Visualyser on Rick Davies’s website, MandE NEWS.

My main use of Visualyzer is to draw the organisational networks I am working with, in the course of my work as an M&E consultant on development aid programmes. These are of two types: (a) literal descriptions (maps) of the relationships as known, (b) simplified models of complex networks showing the main types of organisations and the relationships between them. Less frequently, I also import data from Excel to automatically generate network maps. This data usually comes from project documents or online surveys. I also use the combination of UCINET and Netdraw for this task.

Here is an example of a network that I drew by hand directly on screen. It represents the relationships between AMREF’s partners in the Katine project, Uganda. Click on the image to expand it a new window, then click again to get a focused image. You can represent different types of actors by varying the colour, size and shape of nodes. You can represent the different kinds of relationships between them by varying the kind of line used, its colour and thickness. If you click on a node you can enter detailed text or numerical data describing the actor’s attributes, using as many fields as needed. If you click on any link you can enter data about the attributes of that relationships. Both of these sets of data can be exported, on all actors and relationships, as an Excel file.  You can also import the same kind of data, to automatically generate a network diagram.

mdlogix describe it as “an interactive tool for entering, visualizing and analyzing network data. You can create nodes and links directly or import network data from edgelist/edgearray, Excel, or GraphML formats. Once the network is displayed, you can customize visual properties such as the colour, shape, size, and location of nodes and links to create an informative graphic representation. Images of your choice may be used to represent nodes. XY mapping of nodes as a function of node attributes is supported in layered layout. It also provides a number of analysis functions for calculating network and nodal level indices, and for finding sub-groups, partitions, communities, and roles and positions. In addition, VisuaLyzer includes powerful logic programming capabilities that allow you to investigate networks using axioms of classical set theory.”

This all sounds quite complex. But in practice it is the simplest features of Visualyzer which are the most useful. It does have a very good and easy to read Users Guide (5mb), which you may want to look at.

For more on the development of network models / descriptions and their use in monitoring and evaluation go to the Network Models section of this website.

POSTSCRIPT (1st December 2008): See also Overview of Common Social Network Analysis Software Platforms “This report was developed by the Philanthropy and Networks Exploration, a partnership between the Packard Foundation and Monitor Institute. The exploration is an inquiry into how networks can facilitate greater philanthropic effectiveness. For more information, please go to http://www.philanthropyandnetworks.org

PS2 (16th January 2009): The link to the “Overview …” doc no longer works. I have now uploaded the doc HERE, after receiving a copy via the Pakard Foundation. They also sent a link to: “Working Wikily: How networks are changing social change

Networks and evaluation

This page is about two complementary perspectives: the evaluation of networks, and how a network perspective can inform the design and evaluation of development programs (which may not have been designed as networks)

Please note that the contents of this page has been cut and pasted from the old MandE NEWS website. All links go to contents on the old site. The old contents will be moved to the new website as soon as possible.

Social Network Analysis software: A list

A. Software I have some familiarity with:

UCINET & NetDraw ( a combined package)

  • Easy to import data from Excel
  • Has a huge range of abilities to manipulate and edit the raw data
  • Has an online support group (Yahoo Groups)
  • There is a detailed how to use it text
  • Files can be read by many other software packages
  • Not very expensive, and there is a free trial period
  • Undergoing continuous development
  • Widely used
  • Not easy to draw network diagrams on screen
  • Steep learning curve, many more bells and whistles than you may need
  • No easy to use introductory texts
  • Not easy to edit node and link attribute data on the NetDraw screen
  • PS: See Louse Clark’s very useful and detailed guide to working with NetDraw; and “A Brief Guide to Using NetDraw” by Steve Borgatti; and NETDRAW – BASIC A Practical Guide to Visualising Social Networks by ONA Surveys

Visualyzer

  • Perhaps my favorite, beause it is easy to draw and edit networks on screen, which is very useful in workshop settings
  • Attributes of nodes and links can be easily edited and displayed
  • Can import and export UCINET data
  • Very user-friendly manual
  • Free trial period
  • Now available at a more reasonable price!
  • No online support group
  • Does not seem to be undergoing continuous development

yED Graph Editor

  • Very good for network drawing
  • Many options for layouts
  • Can export files to work as web pages
  • Nodes can include weblinks, allowing quick access to much more information about each node
  • Free
  • Latest version (3.5) can now open data from Excel worksheets, in matrix, edgelist (relationships)  and nodelist (actors) forms. Including as many attributes for the actors and relationships as needed. It seems it will import both one and two mode  (adjacency and affiliation) matrices. This is a major improvement.
  • They are working on capacity to export back to Excel, and ability to search actors and relationships by attribute. Both will be very useful
  • yED is rapidly moving up my list of most favored SNA software packages
  • Now also available as an online version: yED Live
  • Limited analysis capacity

Microsoft NodeXL

  • Free, works as a plug-in to Excel 2007
  • Undergoing continuous development
  • Online support group
  • All node and link attribute data is visible and easy to edit in Excel sheets, which is great
  • Nodes can include weblinks, I think
  • There is a useful users guide here
  • You can’t draw the network direct on the screen,
    • But by using the Excel sheet immediately below the screen you can add nodes and links, and edit their attributes, very easily
  • I have had difficulty in importing yEd (GraphML)
    • PS: They report this is being addressed
  • The layout options (different algorithms) seem quite limited
  • I dont yet know as much about it as the other packages above

C-IKNOW

  • I attended a presentation on C-IKNOW at the 2010 INSNA conference and found this package very impressive, for two broad reasons:
    • User-friendliness
    • Sophisticated range of capacities
  • This is an online service that is open to use by anyone, free of charge
  • Data can be imported, exported and generated by an associated online survey mechanisms
  • There are multiple videos showing how different aspects of the package works, along with a detailed downloadable user guide
  • Development is ongoing and led by Noshir Contactor, a very smart person, and co-author of Theories of Communication Networks
  • “As of September 2012, C-IKNOW Survey is no longer under active development. It can still be used, but no support is available. It should be used at your own risk”

Discourse Network Analyzer

  • software which combines social network analysis and category-based content analysis. After applying categories to text portions, you can automatically extract two-mode networks or one-mode co-occurrence networks in several file formats. There are also some algorithms for longitudinal analysis.”
  • Exports to Excel (in CSV format),  DL files (UCINET), and GraphML files (visone, yEd etc)
  • Free
  • Looks useful but I have yet to try it out on my own data

Visone

  • Patrick Kenis describes this as “very intuitive programme which can be used instantly in consultancy settings”
  • Free
  • Easy to draw networks live on screen
  • Continuous development, but not so often as UCINET

Gephi (last comments added 21 April 2011)

  • Open source (free)
  • Undergoing continuous development, but not so often as UCINET
  • Very sophisticated graphics, the emphasis is on visualisation as a means of exploratory data analysis
  • Capable of visualising very large networks quickly
  • Dynamic views of networks, as they change over time
  • Many filtering options
  • As in NodeXL, has a Data Table view to browse and edit data
  • Drawing networks on the screen is possible, but not so intuitive
  • Imports GraphML files (e.g. as used by yED, NodeXL), vna (as used by Netdraw), csv (used by Excel etc). Exports as csv (for Excel etc) and GraphML.
  • Has Plugins e.g. Social Network Data Import
  • Looks like it could become very good, in time

KUMU

  • Very impressive network diagramming capacity
  • Lots of capacity to annotate network diagrams
  • Online network visualiation tool
  • Free and paid for subscriptions
  • Good tutorial material and technical support

Others not yet examined in any detail

Inflow: [Not yet tested, but looks good]

Social Networks Visualizer (SocNetV) [Not yet tested, but used by Valdis Krebs]

Cytoscape Thomas Delahais says: “I’ve been using consistently Cytoscape, which was designed for neuro-biological analysis but works very well for social sciences! Cytoscape is free, open source and you should complete it with the Max Planck Analyser Plugin, which includes all or most of the usual indicators (diameter, shortest path, etc.) in a unique interface (free for non-commercial use if I remember well). Cytoscape needs some formatting first but then it is very easy to use, very easy to draw on screen too. As a sidenote this is the software I picked when I decided that Ucinet was too complicated for transferring this competency to my colleagues”

SocioWorks “is an innovative set of web tools for the online application of Social Network Analysis (SNA) methods to collect and analyze data regarding social relationships, from individual to institution to national levels.” (posted 2013 05 02)

B. Lists of software most of which I dont know about, maintained by:

  1. KM4DEV list
  2. Wikipedia list
  3. International Network for Social Network Analysis list
  4. Top 10 Open-Source Social network  Development Platforms
  5. Mark Round’s  “SNA Tools and Formats diagram – updated”showing how different software packages are linked by use of the same data formats

The number of social network analysis packages is exploding, a bit like the Cambrian explosion of organic life. No software package has yet achieved dominance because of its ability to meet a wide variety of needs.

C. Online SNA software

  • IdeaTree was not developed as SNA software, but in practice provides many of the same functions, in terms of visualisation. Key features: (a) it supports online collaborative development of network diagrams, (b) it seems quite user friendly, (c) data can be exported in XML, which can be converted elsewhere into graphml, and as pdf documents

PS April 2011: GraphML is a format for storing network data, used by yED, Gephi, and others. The GraphML Primer provides a simple introduction to its use.

Social Network Analysis and Evaluation: A List

Preface to the list

There are however three challenges in promoting the use of network models. One is to get people thinking in terms of networks as a kind of a base metaphor, in the same way that in the past people may have seen clockworks as a base metaphor for how the world works. The second is to sift through what often appears to be a surfeit of technical capacity to analyse networks, in order to focus in on the simplest and often most useful basics. The third challenge is to develop more participative and interpretative approaches to the description and analysis of networks, in contrast to the number crunching nature of much social network analysis in the academic world. This is all very much a ‘work in progress‘.” Rick Davies, in newsletter article for American Evaluation Association, 2008

The list

  • Using Social Network Analysis in Evaluation: A Report to the Robert Wood Johnson Foundation. By Kimberly Fredericks, PhD, MPA, RD, Associate Professor
    The Sage Colleges School of Management, 2013  “…Using data gathered from telephone interviews with the program officers, email surveys from project directors, and supplemental information from nine project summaries (see Appendix 1 for the summaries), this report: ? Describes how the Foundation took deliberate steps to create an emerging learning
    community to support using social network analysis to understand and evaluate its funded work ? Illustrates how social network analysis can be used in many applications, ranging
    from a one-time study to longitudinal studies, with a summative purpose (to evaluate the success of a particular intervention) or formative purpose (for learning and improvement)
    ? Identifies the challenges associated with using social network analysis, in terms of understanding the methodology’s limits or constraints, the implications presented by
    the size and scope of the projects, and the need to develop the technical expertise and capacity to manage the projects and use the data”
  •  

  • Using Social Network Analysis to Advance Traditional Qualitative Methods in Evaluation and Program Design. Caroline J. Wilson, Anne E. Dougherty, Mary Sutter, Jennifer Mitchell-Jackson,  Pamela Wellner,  Nick Hall, ABSTRACT Social network analysis (SNA) is a technique used to study relationships between actors, such as people or organizations. It has been applied to a wide array of disciplines, on topics ranging from destabilizing Al-Qaeda to explaining campaign fundraising success. While these applications may seem far removed from the efforts of AESP members, this paper will provide insights into the innovative ways social network analysis can be used in energy-related evaluations. We start by providing parameters for the use of this technique for evaluation efforts and articulate a number of researchable issues central to evaluation that may be answered through social network analysis. Our evaluation team then demonstrates how this method was employed to conduct a more thorough process evaluation as a complementary approach to other, more traditional research techniques. We also demonstrate the value of the social network analysis approach by comparing a network map with a more traditional organization and implementation chart. We then outline other ways it was utilized, including discussion of social network analysis measures such as centrality and density. This paper aims to add value to current evaluation methods and introduce social network analysis to the best practices of program evaluation and design.

 

  • Social Network Analysis: A Useful Tool for Visualizing and Evaluating Forestry Research. N.L. Klenk1, G.M. Hickey1, J.I. MacLellan2, R. Gonzales3 and J. Cardille3 International Forestry Review 11(1):134-140. 2009 SUMMARY “One of the foundational studies of social network analysis produced a depiction of scientific collaboration by tracing a network of scientific papers linked by co-authorships and co-citations, which has since spurred numerous studies on the typology, organization and dynamics of scientific research networks. This paper introduces social network analysis and its analytical measures of network structure. It then demonstrates the utility of social network analysis in forestry, in the evaluation of large research networks such as the Sustainable Forest Management Network (SFMN), and suggests other important uses of network visualization to facilitate exploring, discovering and selecting resources in a database.
  •  

  • Social Network Analysis and the Evaluation of Leadership Networks. Bruce Hoppe, Ph.D., Claire Reinelt, Ph.D. April 2008 . Abstract: “Leadership development practitioners have become increasingly interested in the formation of leadership networks as a way to sustain and strengthen relationships among leaders within and across organizations, communities, and systems. This paper offers a framework for conceptualizing different types of leadership networks and identifies the outcomes that are typically associated with each type of network. One of the challenges for the field of leadership development has been how to evaluate leadership networks. Social Network Analysis (SNA) is a promising evaluation approach that uses mathematics and visualization to represent the structure of relationships between people, organizations, sectors, silos, communities and other entities within a larger system. Core social network concepts are introduced and explained to illuminate the value of SNA as an evaluation and program tool.”
  •  

    Conference evaluation and network mapping Glenn O’Neil (2008) “Often we attend conferences where one of the stated objectives is “increase/build/create networking” and I always found it odd that there is never any attempt to measure if networking really took place. A possible solution is to map networks created by participants at conferences – and compare these networks to those that existed before the conferences. This is exactly what I have done recently in a network mapping study that you can view here (pdf – 1 MB) and the above image is from. From the LIFT conference of 2007, we mapped the networks of 28 participants (out of 450 total participants) before and after the conferences. We found some quite surprising results:…”

  • Evaluating Performance of Project-Centred Research Networks: PhD thesis by Camille Ryan, 2007. Abstract” A socio-economic network analyses of federally funded research projects* Managing knowledge in new technological realms, such as genomics and nanotechnology, involves the collaboration of geographically dispersed actors across multiple disciplines from both the public and private sectors. However, evaluating performance of collaborative activity is lagging the adoption of the model. This paper adapts social networks analysis (SNA) to the task of evaluate not only the government-funded collaborative research projects themselves but to the circumstances under which they are funded as well.”
  •  

  • Social Network Analysis and Non-Profit Organizations. 2007 “Non-profits each exist in a social network. While many other sectors have at least the possibility of performing their services in a vacuum, non-profit organizations very rarely operate without a large constituency of donors, volunteers, community partners, and also exists with an unusual central sense of being ‘in the public eye’, operating as they do as guardians of a public trust. With in mind, let us examine some of the major concepts of Social Network Analysis as it relates to non-profits….(16/06/07)

  • Network Mapping as a Diagnostic Tool [A Manual], by Louise Clark. 2006. (Also in Spanish) “This publication was made possible with support from the FIT-DFID Programme, Bolivia to the FIT 3 (RedCampo) Project – Boosting the Production and Marketing of High-Value crops through ICT-enabled Information Networks – A project implemented by InforCom of the Centro Internacional de Agricultura Tropical – CIAT, to support the Bolivian Agricultural Technology System – SIBTA (Sistema Boliviano de Tecnología Agropecuaria). The results of this action-research project have demonstrated the utility of social network analysis as a diagnostic tool to improve understanding of how information flows among the different actors involved in agricultural supply chains.” Contents include: 1. Who this manual is for.,2. Introduction to Social Network Analysis (SNA)., 3. SNA as a diagnostic tool.., 4. Social Network Analysis: Step by step. Step 1: Survey design. Step 2: Data collection. Step 4: Preparing the database to be transferred to Netdraw..Step 5: Using Netdraw., Step 6: Try for yourself.. 5. The 2-mode network. Step 1 Survey structure. Step 2: Preparing the database.. Step 3: Transferring the data to Ucinet, Step 4: Visualising 2 mode networks with Netdraw..6. A flexible tool.” (16/07/07)

  •  

  • Investigating the Potential of Using Social Network Analysis in Educational Evaluation William R. Penuel, Willow Sussex, Christine Korbak, Christopher Hoadley American Journal of Evaluation, Vol. 27, No. 4, 437-451 (2006) This article describes results of a study investigating the potential of using social network analysis to evaluate programs that aim at improving schools by fostering greater collaboration between teachers. The goal of this method is to use data about teacher collaboration within schools to map the distribution of expertise and resources needed to enact reforms. Such maps are of great potential value to school leaders, who are responsible for instructional leadership in schools, but they also include information that could bring harm to individuals and school communities. In this article, the authors describe interview findings about concerns educators have with collecting and sharing social network data. A chief finding is that although the majority of teachers consider collecting social network data to be problematic but feasible, some teachers report concerns about privacy and the effect on their school’s goals to foster community if the data are shared with their schools.

  •  

  • New Directions in the Use of Network Analysis in Research and Development Evaluation Jonathon Mote, Jerald Hage, University of Maryland, Gretchen Jordan, Sandia National Laboratories, PowerPoint presentation at the 2006 American Evaluation Association Conference: The Consequences of Evaluation
  • Does your strategic planning make a difference? Andrew Rixon. 2005. This brief paper outlines how social network analysis (SNA) was used to assess the change resulting from the use of “Open Space Technology” as a strategic planning tool by a research organisation. A SNA survey was sent out to participants before and after the meeting. The focus of the survey was on who the participants thought “felt passionate about the theme under discussion” The paper identifies and analyses the changes that were observed via the follow up-survey.
  •  

  • A network approach for researching partnerships in health. Jenny M Lewis Department of Political Science, University of Melbourne, Parkville, 3010, Australia. Australia and New Zealand Health Policy 2005, 2:22 “The last decade has witnessed a significant move towards new modes of governing that are based on coordination and collaboration. In particular, local level partnerships have been widely introduced around the world. There are few comprehensive approaches for researching the effects of these partnerships. The aim of this paper is to outline a network approach that combines structure and agency based explanations to research partnerships in health. Network research based on two Primary Care Partnerships (PCPs) in Victoria is used to demonstrate the utility of this approach. The paper examines multiple types of ties between people (structure), and the use and value of relationships to partners (agency), using interviews with the people involved in two PCPs – one in metropolitan Melbourne and one in a rural area.”
  •  

  • New Directions for Evaluation Volume 2005, Issue 107 , Special Issue: Social Network Analysis in Program Evaluation Issue Edited by Maryann M. Durland, Kimberly A. Fredericks. Published Online: 2 Feb 2006 Contents: An introduction to social network analysis (p 5-13); The historical evolution and basic concepts of social network analysis (p 15-23); Exploring and understanding relationships (p 25-40); An evaluation of communication among high school faculty using network analysis (p 41-53); Network analysis of a demonstration program for the developmentally disabled (p 55-68); Application of network analysis in evaluating knowledge capacity (p 69-79); A formative evaluation of the integration of two departments (p 81-94); The value of social network analysis in health care delivery (p 95-98); Next steps for nodes, nets, and SNA analysis in evaluation (p 99-101)
  •  

  • Mapping the Distribution of Expertise and Resources in a School: Investigating the Potential of Using Social Network Analysis in Evaluation William R. Penuel Willow Sussex Christine Korbak SRI International Paper presented at the Joint Conference of the Canadian Evaluation Society and theAmerican Evaluation Association, Toronto, Ontario, Canada, October 2005.“This paper describes results of a study investigating the potential of using social network analysis to evaluate the capacity ofa school to undertake a schoolwide educational reform. The goal of this method is to use data about teacher collaboration within schools to map and explain the distribution of expertise and resources needed to enact reforms. Such maps are of great potential value to school leaders, who are responsible for instructional leadership in schools; but they also include information that can potentially bring harm to individuals and school communities. In this paper, we describe interview findings about potential concerns teachers might have for collecting and sharing social network data. In addition, we describe some of the procedures we undertook to protect participants’ rights and minimize potential harm that could arise from sharing information about collegial interactions with evaluation researchers in a subsequent study in our project” (28/04/06)
  •  

  • Network Approaches to Global Civil Society. Helmut Anheier and Hagai Katz. “…our focus is on global civil society as a transnational system of social networks and, methodologically speaking, on analysing global civil society through the lens of network analysis. We are interested in finding out how useful the various approaches and tools of network analysis are for describing, analysing and understanding global civil society.” Being Chapter 4 of Global Civil Society 2004/5 Anheier, Helmut, Marlies Glasius and Mary Kaldor (eds.). London: Sage, 2004 (Posted 05/12/05)
  •  

  • Network Perspectives In The Evaluation Of Development Interventions: More Than A Metaphor. Rick Davies, for the EDAIS Conference November 24-25, 2003 New Directions in Impact Assessment for Development: Methods and Practice. “In this paper I argue the case for the use of a network perspective in representing and evaluating aid interventions. How we represent the intentions of aid activities has implications for how their progress and impact can be assessed. Because our representations are by necessary selective simplifications of reality they will emphasise some aspects of change and discourage attention to others. The benchmark alternative here is by default the Logical Framework, the single most commonly used device for representing what an aid project or programme is trying to do. Five main arguments are put forward in favour of a network perspective as the better alternative, along with some examples of their use. Firstly, social network analysis is about social relationships, and that is what much of development aid is about. Not abstract and disembodied processes of change. Secondly, there is wide range of methods for measuring and visualising network structures. These provide a similarly wide range of methods of describing expected outcomes of interventions in network terms. Thirdly, there is also a wide range of theories about social and other networks. They can stimulate thinking about the likely effects of development interventions. Fourthly, network representations are very scalable, from very local developments to the very global, and they can include both formal and informal structures. They are relevant to recent developments in the delivery of development aid. Fifthly, network models of change can incorporate mutual and circular processes of influence, as well as simple linear processes of change. This enables them to represent systems of relationships exhibiting varying degrees of order, complexity and chaos. Following this argument I outline some work-in-progress, including ways in which the conference participants may themselves get involved. Finally I link this paper into its own wider web of intellectual influences and history. ” Posted 29/10/03)

 

  • Organizational Network Analysis as a Tool for Program Evaluation, Merrill Eisenberg, Nancy Swanson, University of Connecticut Health Center, Evaluation & the Health Professions, Vol. 19, No. 4, 488-506 (1996) Abstract: “Health program evaluation is generally focused on an examination of individual program characteristics and accomplishments, yet many programs are part of a broader service system. Evaluation of the role a program plays in that system is an important evaluation question to address. A network analysis of program referral patterns was used to evaluate Connecticut’s Healthy Start program. Network analysis showed that Healthy Start played a “broker” role in 4 case study communities, sending and receiving referrals of pregnant women to a higher than average number of other programs. Further, in the urban area case study, competing market players providing services to pregnant women resulted in subsets of services with dense referral patterns within the subsets, but little referral between subsets. Healthy Start was found to be instrumental as an integrator of these otherwise disconnected service subsets”

 

  • [Readers: Feel free to propose additions to this list]

 

A digression on complexity and networks…

….a side argument from the Rick on the Road post: Cynefin Framework versus Stacey Matrix versus network perspectives

In that post I said

PS1:Michael Quinn Patton’s book on Developmental Evaluation has a whole chapter on “Distinguishing Simple, Complicated, and Complex”. However, I was surprised to find that despite the book’s focus on complexity, there was not a single reference in the Index to “networks”. There was one example of a network model (Exhibit 5.3) , contrasted with a Linear Program Logic Model…” (Exhibit 5.2), in the chapter on Systems Thinking and Complexity Concepts. [I will elaborate further]

One interpretation: Complexity arises through the interaction of many agents having some degree of autonomy. With no autonomy there is simple order (complete predictability), with complete autonomy there is chaos (no predictablity). How do we define autonomy? One view: Autonomy = The number of possible relationships an actor can have with others. When realised, this can be measured in terms of  network density (a Social Network Analysis (SNA) measure). Two cariacature examples of the extremes: 1. An army, with a hierarchical chain of command,  is highly ordered. Here the network structure is  sparse (i.e.  a tree structure) and low in density. 2. “Economic man” , who is free to interact with anyone, in order to maximise his/her utility. Here all possible relationships can be realised, as everyone interacts with everyone. Complexity is the territory in between where actors have some degree of choice of who they interact with. And where there is some degree of predictability. When realised, those choices can also be described in terms of different kinds of network structures. So if we want to explore complex systems we need to look at the structure of networks of actors, both as “initial conditions” affecting what happens next and as “final states”, reflecting what has happened over a given period of time. I.e. an empirical approach, not mysticism :-)

PS: The concept of autonomy could probably be further differentiated, in terms of relationship choices, as follows : (a) the range of relationships available to an actor, already discussed above (b) the freedom to choose amongst those that are available, (c) the range of behaviors available within a given relationship. But how do you measure freedom (b) ? One measure might be the degree to which any choices made are uncorrellated with other events. The diversity of choices made could also be important. Diversity suggests freedom from constraint (more on this theme here).

Horizontal Evaluation: Fostering Knowledge Sharing and Program Improvement within a Network

Authors: Thiele, Graham; Devaux, Andre; Velasco, Claudio; Horton, Douglas
American Journal of Evaluation, v28 n4 p493-508 2007

Abstract: Horizontal evaluation combines self-assessment and external evaluation by peers. Papa Andina, a regional network that works to reduce rural poverty in the Andean region by fostering innovation in potato production and marketing, has used horizontal evaluations to improve the work of local project teams and to share knowledge within the network. In a horizontal evaluation workshop, a project team and peers from other organizations independently assess the strengths and weaknesses of a research and development (R&D) approach being developed and then compare the assessments. Project team members formulate recommendations for improving the R&D approach, and peers consider ways to apply it back home. Practical results of horizontal evaluation have included strengthening the R&D approaches being developed, experimenting with their use at new sites, improvements in other areas of work, and strengthened interpersonal relations among network members. (Contains 2 tables.)”

Also available as ILAC Brief: http://www.cgiar-ilac.org/files/publications/briefs/ILAC_Brief13_Horizontal_Evaluation.pdf

And a  Spanish version of the same Brief