In the fall semester of 2008, about 23 Tufts students used a beta version of YouthMap to produce three interesting and creative projects in just a few class periods and some homework assignments. They also began to build an impressive Boston-area map, with hundreds of nodes and links.

Ultimate purposes of the mapping project

• Recruiting people and organizations. One can search for individuals who are connected, even indirectly, to a given issue and then ask them to participate in events or projects.
• Finding opportunities. One can search for places to volunteer, give money, or organize politically. The search can be by key-word. More interesting is to search for organizations that are linked to other organizations.
• Analysis and deliberation. One can link two issues together, or link an issue to an organization, and then debate the connection. Is homelessness worsened by zoning? That hypothetical connection can be discussed on the map itself.
• Broadening sources. Journalists, government agencies, foundations, and researchers tend to ask the same people for information and opinions. They often rely on formal credentials as evidence of knowledge. The network map can lead them to overlooked citizens who are useful sources because of the social roles they play.
• Investigations. One can look for inappropriate or problematic connections, or lack of connections.

The three Tufts groups decided to map:

1. Immigrant-rights organizations: This group used web research (only) to identify organizations that provide legal services and support to immigrants in Somerville and Cambridge. They originally wanted to be comprehensive. They struggled a bit with how to define the limits of their topic, since many nonprofits belong to immigration-support coalitions but may not actually provide legal services. Also, most immigrants-rights groups that serve Cambridge and Somerville are located in Boston. Their research yielded a map with numerous nodes and links. (This is a portion of their map shown as an image.)

They were able to identify some gaps, e.g., for the Nepali community. They hypothesize that there are many organizations working on their topic but a lack of coordination among organizations.

2. Healthy food options. This group began with one member's personal connection to a particular nonprofit in Waltham. They defined their scope as Waltham, Roxbury, and Jamaica Plain. The Waltham nonprofit turned out to provide good services but was not well linked to other nonprofits. (One student was "frustrated" that the map does not demonstrate the value of such a group, because the map is all about network ties.) Further research (and a different student's personal connections) led them to Shape Up Somerville, an impressive nonprofit that happens to collaborate a lot with Tufts and that is very well networked. They report that it was interesting to use YouthMap to depict the connections of one very well networked organization.


This group found YouthMap's current software version confusing enough that they actually diagrammed the network separately and then tried to recreate the diagram on the YouthMap page, but the nodes moved around.

3. Student governments. This group sought to identify links among student governments in the Greater Boston area. They began with a very well-designed, short online survey. They emailed the survey link to all the student governments in the city but got relatively few responses. The fragmentary data they collected suggested that student governments do not collaborate in the Boston area, which is entirely possible. However, they also tried mapping out links between Tufts' students organizations and student organizations at other campuses. They were able to find many such links. This suggested that (a) student organizations other than student governments are better networked; (b) Boston-area schools are more networked than their preliminary research suggested; and (c) it's helpful to have detailed knowledge of an institution (such as one's own college) rather than rely on a survey. By means of a few conversations with fellow Tufts students, they identified more links than their whole survey had yielded.



They could not find student organizations at some of the area community colleges and music/arts colleges. As the team noted, there may be scarcity of such organizations, but it is also possible that they simply couldn't find them.

Major recommendations

Depth before breadth: The students learned (in various ways) that it was better to try to map out the relationships of a few organizations that they can get to know well, rather than try to map a whole category. There were several reasons for this:

• It's hard to get data for broad categories. The student government group, for instance, obtained a low response rate from a survey.
• If you quickly identify links among many organizations, you get too many links too fast and can't really make sense of the information. Perhaps visualization tools (see below) would help. But part of the problem is that you can't tell which links are meaningful and important instead of superficial.
• It's tough to define the target population when you treat it broadly. All kinds of organizations, for instance, are connected in various ways to healthy food issues. Where should you stop?
  
• It's hard to know what to do with a broad map, whereas you can share a focused map with the organizations that are on it.

Over time, we can achieve comprehensiveness by building up detailed pieces of the map from many in-depth projects conducted across the city. That is better than trying to be comprehensive at first.

Partnerships are important:  Students found they could learn much more if they knew organizations well. For instance, they learned much more about connections among student groups by asking which Tufts groups have connections to other institutions, instead of asking strangers at other colleges to complete a survey. Likewise, they did a great job mapping the networks of one Somerville NGO (Shape Up Somerville), which one student was connected to. They recommended that organizations should map themselves. We need students to be at the center of this work. But we can achieve that by partnering students with other groups.

Issues that students brought up

• Privacy: Whose information should go on the map, and who decides that?
• Chilling effects: Would people be discouraged from linking to controversial organizations and causes if their links could be mapped?
• Spam and other bad stuff: Inappropriate content can be added to the map
• Marketing: Instead of recruiting volunteers or activists for a social cause, a company could use the map to find influential customers.
• Sustainability: It's fun for me and my colleagues to build the map. But other people who contribute need to know that it will still be there (and kept current) in five years.
• Limits: This is a Boston area map. That geographical definition gives it useful density. But Darfur could belong on the map, since Boston-area students work on Darfur. Former Bostonians could place themselves on the map. Is there any reason to have geographical limits?

The software

We are just beginning to develop the software, and the students were politely but consistently critical about how it functions. Some technical issues:

• Map often loads slowly
• Sometimes nodes are lost after being entered
• Text can't be formatted, e.g., with spaces between paragraphs
• No "undo" button
• Can't scroll to see the whole map or a large portion of it

Some more substantial software issues:

• Visualization/simplification tools are essential. Note that students did not use tags, because I didn't tell them to do so. Tags are available and would have helped. Other visualization tools are in development.
  
• It would be good to have more tabs for each node, so that different kinds of information could be separated (e.g., the mission statement and contact info could be on separate pages within each node.)