The role of Social Network Analysis (SNA) of the Mobile Money (MM) Market in promoting Financial Inclusion: A Case Study of M-PESA in Kenya

By Gayathri Ramani

(I will not be taking the second module)

Introduction

Recently, a Kenyan newspaper published an article about the role of social media in promoting financial inclusion. The Chairman and CEO of Equity Bank, one of Kenya’s largest banks Dr. James Mwangi said “Social networks have the potential of empowering the youth economically by enhancing their access to financial products and services.”[i] By diversifying their marketing strategy onto social media platforms like Facebook, Twitter, and Instagram, the bank is confident that they will be able to enhance financial inclusion. However, despite the growing penetration of internet services, reliable electricity continues to be a major challenge for many parts of rural Kenya.  This implies that a large percentage of households in Kenya continue to have limited or no access to Internet and cannot afford to own a smart-phone with readily available data.

In contrast, mobile services have grown exponentially over the last few years due to the introduction of Mobile Money called M-PESA. During my work experience at Innovations for Poverty Action (IPA) in Kenya, I had the opportunity to manage an evaluation study on the long-term impacts of Mobile Money (M-PESA) around 2000 Kenyan households from different regions in Kenya. According to a recent paper published by Tavneet Suri and William Jack, the principal researchers behind this long-term study, M-PESA adoption rates grew to 70% in just 4 years from its inception. Moreover, almost three quarters of the households had at least one user (adoption in this case refers to either being a registered user and those who had access to an M-PESA account). [ii]

Due to its success, the M-PESA Platform has expanded to incorporate “Mobile Banking”. Products under this platform include M-shwari, a savings account linked to an Equity Bank account through which you can also to take out micro-loans. Additionally, other companies have also started using M-PESA as a payment mechanism using services like M-Kopa and Lipa na M-PESA. However, observational data suggests that take-up of such products is relatively low compared to formal savings instruments (e.g. Bank Accounts) and informal instruments (e.g. Merry go round). In this context, understanding and analyzing the social networks around M-PESA will provide crucial information for stakeholders in not only creating a more successful and targeted social media campaign but will also provide robust evidence for future marketing strategies to promote their range of financial products.

Background

The topic of financial inclusion is increasingly becoming a hotbed for development experts. The Center for Financial Inclusion (CFI) defines it as the “access to quality financial services that are convenient, affordable, suitable, and provided within a robust financial structure and clear regulatory framework”. [iii] Having access to cheaper and competitive credit will theoretically allow the poor to be more economically self-sustainable and encourage a culture of entrepreneurship that can potentially increase growth of the country. However, due to lack of adequate infrastructure and other credit constraints, there is a large proportion of Kenyan households that do not have access to such facilities. In order to overcome these challenges, development practioners began implementing innovative models such as the Grameen Bank in Bangladesh. This fed into a digital revolution of the mobile phone technology and opened the gate to solving development problems.

With mobile phones came “Mobile Money”. One of the most successful model is M-PESA. In 2007, the leading cell phone company in Kenya, Safaricom introduced M-PESA. It was a SMS based money transfer system where individuals were allowed to deposit, send, and withdraw funds using their cell phone. After its launch in 2007, M-PESA gained extraordinary momentum, reaching nearly 70% of Kenya’s adult population in 2011. According to Jack and Suri, M-PESA’s rapid uptake was in part due to the growth of network of agents. Agents are small business outlets that provide the cash-in and cash-out services. The agents exchange cash for so-called “e-money” from the individuals and then send the electronic balances from one account to another via SMS. [iv]

M-PESA has dramatically transformed the social networks of Kenyan communities. Apart from the business network, where entrepreneurs conduct economic transactions, social networks have also burgeoned. According to a working paper by Kusimba, Chagger et al, “Most users in Western Kenya use mobile money as a social and economic tool through which they create relationships by sending money and airtime gifts, assisting friends and relatives, organizing savings groups, and contributing to ceremonies and rituals”.[v] Understanding the dynamics of these networks will be crucial for Safaricom and Equity Bank in promoting their financial products and forging ahead in the realm of financial inclusion.

Research Questions

In order to make recommendations pertaining to the marketing strategies of key stakeholders (i.e. Safaricom and Equity Bank), one must formulate and answer the following research questions:

Question 1: What does the Mobile Money (MM) market look like?

In order to answer this question, we will need to answer the following sub-questions:

1.  Key players - Who are the key players of the M-PESA Network? We can identify these by looking at several centrality measures (distance, betweeness, high in and out-degrees)
2.  Density of the Network – Is the network really dense or are there several distinct and sparse components? Are there any isolates in the network?
3.  Strength of ties - Have the ties gotten stronger? Do we have more reciprocal ties within members of the network?
4.  Is there any evidence of homophilly – Are there any distinct networks based on gender or other attributes?

Depending on some of these measures, we will be able to identify potential roles that these key players will have in marketing or information campaigns. For example, players that have high out-degrees can be information sources.

Question 2: By narrowing in on the Ego Networks for these key players, what are their characteristics/attributes of all members of this network?

We can filter the data to look at the following characteristics (dichotomized for frequent and very frequent ties:

1. Urban/Rural demographic
2. Gender
3. Occupation
4. Social Media Presence

• Do they have Facebook/Twitter/Instagram?
• How many followers do they have?
• Do they have access to these sites on their phone?
• 5. Function of the money being transacted

By analyzing the characteristics of the Ego Networks, we can segment the target population and identify the M-PESA Users that could have the greatest impact of a targeted marketing model. For example, if we have a rural network that has a “leader” who has a strong social media presence, providing them with information about different products and offers through Facebook or Twitter could eventually trickle down to the rest of the network that might not necessarily be plugged into social media.

Data

There have been only a few studies using social networks analysis in understanding mobile money adoption.  In particular, The Consultative Group to Assist the Poor (CGAP) in collaboration with Real Impact Analytics conducted a study in 2013 on the use of data analytics in understanding take-up of mobile money. They collected and compared data across 3 different countries (including Kenya) on variables such as mobile money transfers, frequency of transfers, and the networks that surround these potential key drivers of adoption.

For purposes of this research design, we can use this data and assume a high level of synonymy between drivers of adoption of Mobile Money or “Technology Leaders” (as defined in their working paper) and drivers of adoption of new financial products within the M-PESA platform.

Given that the data set is large and difficult to interpret through social network analysis, it would be useful for us to take a smaller random sampling of the original sample and conduct a follow-up study on these sampling units (taking into account attrition of the sample). Questions regarding social media usage can be added into the survey.

 More information about this study can be found here:

http://www.cgap.org/sites/default/files/social_networks.pdf

---

[i] http://www.standardmedia.co.ke/?articleID=2000180681&story_title=equity-bank-chief-says-social-media-key-in-expanding-financial-inclusion&pageNo=1

[ii] Jack, William and Tavneet Suri (2011): “Mobile Money: The Economics of M-PESA”

[iii] http://www.centerforfinancialinclusion.org/about/who-we-are/our-definition-of-financial-inclusion

[iv] Jack, William and Tavneet Suri (2011): “Mobile Money: The Economics of M-PESA”

[v] Sibel Kusimba, Harpieth Chaggar, Elizabeth Gross, & Gabriel Kunyu “Social Networks of Mobile Money in Kenya”, 2013, Working Paper for The Institute for Money, Technology, and Financial Inclusion. 

Long Arms of Libya: How Regional Ties Create Libya’s Arms Trafficking Challenges

Long Arms of Libya:  How Regional Ties Create Libya’s Arms Trafficking Challenges

Preamble

Since the Arab Spring began in 2011, Libya has grown into a regional hub for trafficking small arms including guns, shoulder mounted missiles, and mortars.  Weapons have been trafficked in from external nations including the United States of America and the State of Qatar.  Caches of weapons have also been looted from depots that were formerly under control of Libyan leader Muammar Gaddafi before he was killed and overthrown.

Since then, weapons have also been trafficked out of Libya in many directions, including to Egypt in the East and to Algeria, Mali, Niger, and Chad in the West and South.  The networks used to bring these weapons in by land, air, or sea before dispersing them across Libya and North Africa has, thus far, been under-studied and misunderstood.

A Network Analysis

I want to answer the following question: what are key social connections in the North African context that facilitate weapons to be transferred to conflict zones other than the ones they were intended for?  Are tribal or family relationships the strongest indicator of arms networks?  Or are political or religious affiliations better indicators of illicit network channels?

Why is this Important?

This is important because the crisis in Libya is likely continue through the coming years.  The relationships between armed groups will be important for containing the conflict and will also be an indicator of how to look for networks of influencers, information, alliances, and money.  The information would also be predictive of where conflict might arise in the future.  For example, if there is an area of Mali that currently has regional conflict but limited weaponry but has tribal ties shared with people in Libya, would this be an indicator that the Mali region might soon experience further conflict due to access to weapons through a tribal network?

Without better understanding the networks that allow for the State of Libya to become a hub of illicit weaponry, this small country of about six million individuals will likely remain a distributor of arms and centerpiece of instability for years to come.

Data Required

The data for this project will rely on several forms of research.  1) ACLED conflict data set which contains information on what kinds of weapons were used and by which groups in North Africa, 2) anthropological and sociological data on where different tribes, political, terror, and religious groups overlap. 3)  Newspaper and terrorism watch journals to identify overlap of weapons traffickers and different identity groups.

Details of the Network Analysis

The network analysis will focus on the following methods:

   Sub-group analysis: What are the historic tribal, political, business, and religious networks in Libya and how do they overlap with surrounding countries?

   Betweeness: Which groups sit in the path of major choke points (cities, ports, roadways, etc)?

   InDirection:  Monitor for groups, like Qatari proxies who exclusively bring weapons into Libya.

   OutDirection:  Monitor trade groups on the Egyptian border that exclusively take arms out of Libya.

   BiDirectional:  Groups that both gather and distribute weapons.

Next Steps for Further Analysis

Core issues that are unlikely to be addressed by this network analysis but that should be followed up in future analysis include:

Further research would include comparing connectivity of networks for weapons in North Africa to networks that trade other good whether illicit like drugs, human trafficking, or legitimate activities like oil and natural resources, commodities, and textiles.

Similarly, it would be useful to see how networks of diaspora Libyans work into the illicit weapons trade in the country and how diaspora networks in the Libyan context compare to illicit arms trafficking in other regions, such as the Irish diaspora for the Irish Republican Army.

International Languages and Online Content Flows: A Comparison of Networks

Preamble

The more I study poverty and inequality, the more I’ve noticed that structural divisions in information sharing between countries, cultures, and people are fundamental to conflict and inequality.  Economists identify information asymmetries as a core market failure in many developing economies.  Reducing information asymmetries requires us sharing information with each other, overcoming structural divisions. Put more simply, we need to communicate and listen more, and more equally.

One potential structural division in information sharing is language.  People without a shared language are simply less able to communicate to one another.  And even when people do share a common language, it may be the case that the flow of information is unevenly balance in favor of native speakers of the core languages, such as English.  One place where asymmetries in information flows might be able to be measured is on the internet, by measuring the directionality of the flow of data between countries.

A Network Analysis

I propose comparing the network of online data flows between countries with the network of shared languages.  In particular, I would like to analyze the directionality of online content flows, and measure how strongly the the information network correlates to the network of shared languages between countries.  The scope of the analysis will be narrowed to text-based online media content and shared languages, and will attempt to identify divisions, isolates, and countries that could be groomed to be connectors in the global network of information sharing and language.

Why this is Important

The key motivation of this analysis is that divisions in the internet community and global language landscape are a key indicator of divisions in communication, and are likely to be a contributing factor to real economic and political divisions between countries, cultures, and people. In addition to increasing the likelihood of conflict, divisions in language and information also reduce the size of the knowledge commons and the slow down the pace of our collective learning in all fields of knowledge.  Some languages are also unequally favored in terms of the content available to speakers of that language, as the map of English Wikipedia content below indicates, as well as the language network map further below. Gaining a better understanding of where the weak points are in the global network of language and information flows will better equip us to address them, by informing us of the languages and countries that are critically isolated or who are potential connectors that could be used to bridge divisions in language and information sharing.

Data Required

The network of online information flows will constitute data from the International Telecommunication Union (ITU), and the Berkman Center for Internet and Society will be approached for the data that they have already published on international data flows, through their Internet Monitor project. The dataset that I hope to obtain will have country level data that records the sum of the total data requested by each country of each other country in the last year.

The data for the language network will be constructed using public records of the languages spoken in each country, and the populations that speak them.  The structure of the language dataset that I intend to build will be symmetrical such that for each country pair in the matrix there will be a measure of the percentage of people in those two countries who share a language. As is illustrated in the table below. In this example, 50% of people in countries A and B share a language, whereas only 10% in B and C share a language:

	Country A	Country B	Country C
Country A	1	0.5	0.33
Country B	0.5	1	0.1
Country C	0.33	0.1	1

A second option would be to map the network of translations between languages. This has been done before, as shown in the image below. However, this method is less ideal for the country comparison between internet flows and language, as is intended in this study.

Details of the Network Analysis

The network analysis will focus on the following methods:

• Sub-group analysis: The first step for conducting the analysis will be to visually compare the distribution of the sub-groups in the two networks under analysis. Within the sub-group analysis, the relative strength of the groups will be measured using the I-E measures.
• Betweeness: Identifying countries/languages with high betweenness will be a key method for finding countries that are connectors within the network, and who facilitate information sharing.  Contrastingly, high betweenness may also indicate that the connection between sub-groups is too dependent on a limited number of countries, and is vulnerable connection.
• InDirection: this will be used to identify countries who are having their online information requested from them by many other countries. It should be noted that directionality will not be measured in the language network due to the symmetry of the data.
• OutDirection: this will be used to identify countries who are requesting online information from many other countries.

Next Steps for Further Analysis

Core issues that are unlikely to be addressed by this network analysis but that should be followed up in future analysis include:

• Test to see if there is a correlation between the internat and language network maps and political divisions, such as by looking at treaty networks. Similarly, it would be interesting to measure the correlation with the real economy, such as by looking at trade flows and trade agreements.
• Identify key languages and countries that should be groomed for being connectors and diplomatic links between otherwise disconnected sub-groups. These will be countries that have high potential for betweeness.
• Investigate technology based methods for reducing internet and language divisions. This might include investing in improved automated translation services between languages of critical interest.

SNA of Lobbying Organizations

Takumi Onuma

(I am not taking the second module in this year.)

Background

Public policy influences business activities. For example, government officials can facilitate renewable energy business by introducing policy support such as subsidy, tax-breaks, and feed-in-tariff mechanism. This is an example where politics and business are interacted.

In the United States, unlike Japan, many lobbying organizations seem to act eagerly to create optimal business context. Such activities involve a lobbying coalition, and the chances of a successful outcome are usually greater when organizations pool their material and political resources into a joint policy campaign[i]. Although understanding this mechanism is important, it may be difficult, in a traditional academic context, to demonstrate how lobbying groups persuade policy makers to implement a specific policy, and how policy makers attract lobbying groups to win an election, because we face difficulty in finding experimental and objective evidences. However, we can gather data about social movements, such as how people come together, and how they support an influential candidate. Here is the chance for applying social network analysis (SNA), because such collective action can be seen as a kind of network forms.

Research Questions

1. How the structure of lobbying organizations is described?
2. Does a lobbying organization form coalition with others? If true, how it do? Do external social media contribute to this process significantly? Is it just a result of internal interest?

Data Collection

1. In order to map the existing lobbying organizations, I will quote secondary data on demographics of lobbyists, such as OpenSecrets.Org[ii]. This data will indicate numbers, spending, relationship between political candidates.
2. For the sake of grasping coalition process, I will search adequate number of organizations’ HP, and check their public relations, such as social media.
3. To measure the influence, I will get statistics about how many bills are submitted and approved/denied by each supported candidate. I will also extract an interesting political/business theme to make my analysis simple.
4. I will get deep insight by interviewing small samples.

Methodology

First, I will draw two-mode network about lobbying organizations and political candidates by using UCINET. I will process the secondary data to visualize;

• Which organization has an experience about contacting other organizations?
• Which organization supported a particular candidate?

Second, into the data, I will add attribute such as;

• Organization itself; how much budget it had? How many times it took action?
• Organization and other organizations; what relationships, such as collaboration or conflict, existed? How many times the interaction could be observed? What was agenda?
• Organization and candidates; how many times an organization supported a candidate? How much did they support? What was agenda?

In order to measure the tie strength, I will distinguish and order such quantitative data as a variable.

Third, I will analyze the network. My plan is;

• Analyze the centrality in the two-mode network. I will identify the influence by measuring betweenness and eigenvector of each organization. Because betweenness indicates node on the most paths between other nodes (which infers acquiring excellent know-how and evidences), and because eigenvector indicates node connected to other well-connected nodes (which infers having big influence and power), measuring these degree would specify the most influential organization.
• Analyze the tie strength between organizations. By reviewing the frequency in which they interact with others via social media, I will identify the correlation of intimacy with others and these interactions.

Fourth, I will also advance the analysis. In this step;

• I will transform this network into a directed network. For example, if a candidate acted in favor of a particular organization (i.e. agreed with a favorable bill, stated a favorable manifesto), then define it as a directed network from a candidate to an organization.
• I will dig into revealing relationships among organizations by adding in-degree and out-degree into the previous analysis.

Hypothesis and Summary

Through this SNA project, I will find out that how the structure of lobbying organizations is described, and how they form coalition with others. One hypothesis which is tested is that social media plays a significant role to urge this coalition process. Also, I will test another hypothesis that lobbying organization which frequently uses social media has greater influence for both organizations and policy makers than the other organizations which do not use frequently. Although further research will be needed, I hope this project offers a hint to understand a part of lobbying mechanism.

---

[i] John Scott and Peter J. Carrington, “The SAGE Handbook of Social Network Analysis”, p211

[ii] http://www.opensecrets.org/lobby/

Open Data and Civil Society Networks in Kenya

Alex Kostura
(I am not taking the second module)

RESEARCH QUESTION:

What is the effect of the Kenya Open Data Initiative on networks of anti-corruption civil society organizations in Kenya? Can social network analysis help explain why the Open Data portal is under-utilized?

CONTEXT:

Open data is data that is made available by organizations, businesses and individuals for anyone to access, use, and share – The Open Data Institute [i]

Open data as an idea represents both a new trend in international governance and an emerging field of study across disciplines as diverse as democracy, economic development, and urban planning. For my capstone project, I am specifically looking at the idea of open government data and the growth in “open data initiatives” where local, regional, and national governments create a web portal to publish data for public consumption. Two distinct civil society movements have evolved around open data: one that focuses on human rights and the other on accountability and government transparency. I am looking at the latter through a case study into Kenya’s Open Data Initiative.

In 2011, the national government of Kenya launched the Kenya Open Data Initiative (KODI) with praise from the international community and broad government support. This web portal promised its data to be the “key to improving transparency; unlocking social and economic value; and building Government 2.0.” [ii] Three years later, a 2014 study titled “Open Data in Developing Countries” found that Kenyans have little knowledge of the open data portal and it is underutilized as a result.

MOTIVATION and HYPOTHESIS:

This analysis will be part of a larger case study looking at the effect of open data initiatives on transparency/accountability in countries where corruption has historically been an issue. My hypothesis is that a network of non-governmental and civil society organizations (CSOs) builds up around the implementation of the open data initiative. And the Open Data Initiative itself is only as effective as this network’s ability to analyze and distribute data, garner public support, and advocate to the government. Thus, I am hoping a social network analysis will reveal the strength and connectedness of this CSO network.

DATA:

Based on open sources, I will construct the network of CSOs working nationally on transparency and accountability issues. The nodes for this network will be organizations. I will potentially create three networks:

1)   Ties determined by whether or not they participated in the coalition advocating for the creation of the KODI. 0 =  Did not participate; 1 = Did Participate

2)   Ties determined by partnerships or working relationships. Organizations would respond to the question “How often have you collaborated or shared information with another organization in the past year?”   Answers would be valued 0-Never, 1-Not very often, 2-Frequently

3)   Ties determined by sources of information. Organizations would respond to the question, “Which organizations do you go to as sources of information?” Answers would binary with 1 indicating a source of information.

I would seek the following attribute data:

·      International or National Organization

·      Parent/Affiliate Organization

·      Membership size

·      Staff size

·      How often does organization access the KODI data portal?

·      Number of reports produced

·      Website?

·      Social media presence?

·      Location of office

·      Working languages

Data would be collected via open sources and a survey of target organizations. Depending on the number of organizations identified with transparency and accountability missions, data collection could be unwieldy. Data collection would likely be incremental, and the list of organizations included in the network will likely grow with each group surveyed.

ANALYSIS:

Because I am analyzing a newly constructed network, I would run multiple cohesion measures and look at density, connectedness, average degree, components, diameter and average distance. These initial whole-network measures would inform next steps for analysis. For example, I might conclude that the network is fragmented by cliques or subgroups. It would also be interesting to look for structural holes where organizations with very similar missions are not connected. I would likely run this analysis on a network that only includes the coalition of those who advocated for the KODI’s creation initially.

Node centrality measures would also be useful in identifying organizations who may serve critical roles in ensuring the open data ultimately strengthens civil society networks. These include:

·      Nodes with high in-degree centrality in the “information network” may be identified as sources of information. These organizations are likely ones that have the technical capacity to download, analyze, and interpret government data. They might be targeted to lead data training.

·      Nodes with high betweenness in this network may be identified as brokers who connect nodes with organization with different expertise.

·      In the collaboration network, nodes with high in-degree and out-degree eigenvector centrality may be identified as potential leaders who will be key to strengthening the network.

 Additionally, it would be useful to analyze the network based on who currently accesses the KODI and who does not. Looking at this network would ideally reveal how open data is making its way through the network; and if it is not, how could Kenya's civil society improve that?

PREVIOUS RESEARCH:

David J. Marshall and Lynn Staeheli, “Mapping Civil Society with Social Network Analysis: Methodological Possibilities and Limitations.” Geoforum. Volume 61, May 2015, pp.56-66

Social Network Map of Civil Society Organizations with Partnership Ties and Donor Ties:

---

[i] http://theodi.org/guides/what-open-data (accessed October 22, 2015)

[ii] https://www.opendata.go.ke/ (accessed October 22, 2015)

A Good Free Survey Service

Per discussion in today's class, those of you conducting a survey may wish to explore Typeform as a free alternative to Survey Monkey.

I have used Typeform in the past and found it to have a very nice user interface (not to be underestimated!) and have similar if not superior tools to other services. Also, if you're willing to pay a bit (even if just for a month), their conditional response options (i.e. if/then questions) were quite useful for my purposes at the time.

Not taking a commission on this, of course - you should certainly decide what will work best for you!

Mapping for Change: A Network Analysis of the Association for Women's Rights in Development
Medha Basu
(I am not taking the second half of the module)

BACKGROUND:

The Association for Women's Rights in Development (AWID) is a global membership and advocacy organization, with its member base of women’s rights activists, organizations and movements increasing to an all time high of 565 institutional and 3496 individual members in 2013, coming from a mix of 159 countries.[i] 

AWIDs core mission is rooted in, and driven by strengthening collective social networks across issues, regions and member constituencies, making the study of their active and growing network essential to any strategy and future plans of the organization.

AWID's mission is to strengthen the collective voice of their members to  influence and transform structures of power and decision-making and advance human rights, gender justice and environmental sustainability worldwide.[ii] AWID believes that working together is key for women’s rights and gender justice to be a lived reality, and therefore a significant amount of their work is directed towards supporting women’s rights organizations and movements to form meaningful connections with each other and collaborate effectively. Their projects fall under the following umbrellas:

• Mobilizing members to strengthen collective action in solidarity with women’s rights causes
• Organizing and facilitating constructive spaces for members and donors, development agencies and other CSOs to explore and strengthen connections
• Bringing together groups in the gender equality space that have not yet found common ground.
• Including and engaging groups who have an underrepresented voice but are stakeholders in the issue
• To serve as clearing house for information to and from our members and broader women’s rights movements.

This SNA will specifically focus on the connections between the institutional member base of AWID. While AWID undoubtedly keeps lists of their member base dis aggregated by country, issue area, date of joining (break-up by region is below), there is no mention on their website, or in any of their literature, about the quality, quantity and nature of connections among their network.

Fig 1: Source: AWID 2013 Annual Report

RESEARCH QUESTION:

What is the current level and nature of connections between institutional member organizations of AWID, and how can this information be leveraged by AWID to strengthen the network for collective and representative action on women's rights?

More specifically, each sub-question/strand of this SNA umbrella question would draw insights which would help AWID make more nuanced and effective decisions while planning and implementing their strategy over the next few years, as articulated below: 

A) Sub-Question: How dense are the connections between members in each country, and between members in each region (in the categories laid out in Figure 1)? Do some countries/regions have a higher number of, denser, stronger or higher reciprocity of ties than other countries/regions, and are some under-connected? Are some countries underrepresented by way of being comparatively less connected than others? 

Decisions this would inform: Does AWID need to provide more opportunities for engagement and convening in some countries and regions, so that they have the chance to become better connected, not only within country platforms, but to the wider network? This could involve organizing a regional conference, a data and research sharing platform, or matching organizations which do similar work. AWID can also note which regions have more outgoing/incoming than reciprocal ties, as a basis to investigate what kinds of projects can be implemented to find common ground and increase chances of collaboration.  

B) Sub-Question: AWID aims to build action for gender equity holistically, ie, beyond the gender binary, ensuring they are not only engaging only women activists for women's rights. Therefore, this SNA can reveal: how well embedded and connected are member organizations who work with engaging men and boys for gender equality, deal with LGBTQA rights or transgender perspectives?

Decisions this would inform:
If there is an under-representation of transgender or LGBTQA focussed members who lie at the periphery of the network with few ties to the larger network,  or less heard voices which need to be boosted in the mainstream flow of information in the 'clearing house', AWID staff can reach out to these organizations to understand where they need more support and facilitation. 

C) Sub-Question: How well connected are the members categorized by the issue areas they work on? AWID has 6 issue areas along the lines of which they organize and mobilize collaboration: Economic Justice; Resourcing Women’s Rights; Challenging Religious Fundamentalisms; The Right to Defend Rights: Women Human Rights Defenders and Young Feminist Activism. Are some issue area networks not well-connected globally (or in certain regional blocks), and are some very well connected? 

Decisions this would inform:
The force of an advocacy network to push for change in a specific issue area is affected by the strength of the bonds in the network itself and the level of collaboration. Therefore if a particular issue area network needs attention, AWID can allocate relationship building efforts, or launch joint projects/platforms in those issue areas.

D) Sub-question: Are there any 'success models' of collaboration which emerge, in terms of hubs which are very highly connected and reciprocal, either by region, issue area, or other factors like the type of projects they have worked on together, or the mode of their interactions?

Decision this would inform:
This would shed light on what has worked and what has not worked in bringing together organizations and facilitating strong bonds. AWID could accordingly allocate resources and efforts to those types of joint projects which work best (example- if the SNA showd that organizations seem to make stronger and lasting connections through joint report writing over attending workshops together, then this can inform the balance of activities planned for the coming year). They could also examine the  connections between diverse organizations from different countries and cultural contexts, to assess whether cross-cultural and cross-issue exchange is occurring, and if not, what activities can be planned to encourage this. 

E) Sub-Question: Are there some member organizations which stand out as very well connected and active, with multiple bonds with other members in-country or internationally? Who are the organizations who are isolated?

Decision this would help inform:
This process will help identify existing and emerging leaders. AWID can use this information to invite these high-influence, pro-active organizations, which are well-embedded in their own geographic or issue network to kick-start new advocacy project. They could also act as a conduit to bring on board new or peripheral constituencies (other members, donor agencies, policy bodies, etc).

DATA AND SOURCES: 

Defining 'connected':
'Connected' in the context of this network can be defined and measured in many ways. For the purposes of this SNA, I am defining it the following way:
1) Level 1 ties indicate that 2 members have participated in a workshop together where minimal interaction was necessary, contributing to a pooled data-sharing platform along with other organizations,  or signing the same petition together.
2) Level 2 ties could involve working on a joint project together- from contributing separately run but coordinated pieces to a larger national campaign, being on a break-out group team at a conference and presenting recommendations together, attending a strategy meeting together and collaborating with other organizations.

3) Level 3 ties could involve deeper collaborative ties indicated by co-writing a report, lobbying for policy or legal reform together, jointly applying for a grant, collaborating to implement a long-term programme, becoming organizational partners

Sources of Data:

• Annual Reports for 2013, 2012, 2011, 2010, 2009. These reports list all the achievements and executed projects in the year under the 6 issue areas AWID works on, along with a description of what kind of collaboration and process of each project entailed. (available online).[iii]
• Updated list of all members - disaggregated by country, region, issue areas, date of joining (very likely to be available in internal records).
• List of all activities facilitated by AWID between 2010-14 (available through annual reports), and an accompanying list of all members/partners who were involved in each activity (likely to be available in internal records):

a) Strategy meetings, agreed plans of action, memorandums of understanding signed

b) Policy advocacy groups formed

c) Publications written jointly (best practices, manuals, policy recommendations)

d) Public campaigns carried out (social media, protests, awareness drives)

e) Research (studies conducted and analysed collaboratively, data sharing platforms created)

f) Grants jointly applied for by members

g) Public, 'access to information' platforms collaboratively worked on

• An end-of year survey of all members, where they are asked to list the ten organizations they collaborated most with, or approached the most in a given year (much like the Women to Women post-survey, where girls were asked to list the 3 girls they interacted the most with). This survey could also ask which mode of AWID activity members found most beneficial to increasing their knowledge, capacity, and engagement with the broader network. This information does not seem to be available or mentioned in the online literature, but could be conducted along with end-of year catch-ups or other meetings with members through regional offices of AWID.
• The attendance list at various workshops and events at AWIDs flagship international forum, which is held every 2-4 years and garners attendees from members and other stakeholders in the range of 1000-2000. The last one took place in 2012 in Turkey, and the next one will be in 2016 in Rio, Brazil. [iv]       

CRITICAL SNA MEASURES (with each member organization as a node)

Density: This means the number of actual connections over the number of potential ones between member organizations. This 'whole network measure' will be critical in identifying the success models of a bonded network and areas of untapped potential for collaboration, to be facilitated by AWID.  Density can be compared and analyzed of networks categorized by region (MENA, South and South-East Asia, etc).

Strength of ties: This will analyze the network at Level 1,2 and 3 ties. It would  be interesting to compare the whole network density at the 3 levels, and also if layering the 2009-2013 data/matrices, to compare which ties have gone up a level over the years. When looking at the joint activity/platform which initiated and built the tie, this could indicate which efforts at collaboration have been effective catalysts for deepening bonds.  

Arc  reciprocity: to what extent the existing ties in a whole network are reciprocal (0-1). If arc reciprocity within a given issue area or geographic network are low, AWID as a membership organization could identify sub-networks high on one-way links, and devise ways to find mutual interest or convene members around common issues.

Centrality Measures:

Degree: In-degree indicates how many nodes that link to a given organization 'a'. This could be gauged from the proposed survey where at the end of the year- all members indicate the top 10 organizations who they have collaborated with or reached out to, dichotomized by level 2 and 3 ties. Therefore those organizations with very high in-degree could be identified as potential leaders/driving partners for new initiatives

Eigenvector : This, from the survey and activity lists combined, will measure which organizations are connected to the most important/influential nodes. This measure is crucial to know when facilitating a joint advocacy opportunity, where there is an effort to bring about reform, and coalesce support from the most influential and credible stakeholders in the network. The organizations with high eigenvector could be assigned as the coodinators and partnership management leads on projects to maximize support around lobbying/backing policy recommendations for a given cause.

Betweenness : This reveals organizations who are on the path between most other nodes. These organizations would be approached as the key nodes for passing on important information to the rest of the member network, including updates on AWID activities, changes in relevant domestic and international law or policy, calls for applications for grants, etc. They could also be useful for information gathering and feedback to AWIDs team on who the main players are in their network, and ideas on who to engage further for which project.

EI Index: This indicates (on a scale from -1 to 1), whether a given geographical, issue area or activity based network is homophilous or heterophilous. Since one of the aims of AWID is to encourage bonding over common ground, but also idea exchange across members who have different core interests and constituencies, this measure will help identify where a bridging effort is needed to create platforms for diverse organizations.

Cliques: clique is a sub-set of a network in which the organizations are more closely and intensely tied to one another than they are to other members of the network. Identification of cliques would be helpful for AWID to assess where there are strong communities of collaboration within the network and then further investigate why this works, and what kinds of joint activities or interests cause these strong bonds to occur. Perhaps success stories can be replicated. If a clique is very insular and not connected to the rest of the network, efforts can be made to provide opportunities for broader interaction.

CONCLUSION:

A comprehensive Social Network Analysis of AWIDs organizational member network would give them a wealth of information about why, and through what modes their members connect, and subsequently build strong ties and execute joint projects. The gaps, evidence of weak ties and comparatively sparsely connected issue area or geographic networks can inform organizational decisions on where to direct new initiatives to facilitate collaboration. Emergent leaders can be identified to spearhead advocacy movements, and those who seem less connected can be approached to ask if they need support. Overall, this would feed back into helping AWID have a more targeted and nuanced approach to reaching its goals.

---

[i] Annual Report 2013, http://www.awid.org/annual-report-2013

[ii] AWID, About Us, http://www.awid.org/who-we-are; Accessed 22 Oct 2015

[iii] AWID Annual Reports, http://www.awid.org/annual-reports

[iv]International Forum, http://www.awid.org/awids-international-forum