An Open Data Overview
Should your organization be sharing the information it gathers about programs and services? There are benefits to doing so—and there are risks. Generally, sharing data is not common practice, but as funders increasingly demand information about the effectiveness of the organizations they support, more and more nonprofits are doing so.
Strictly speaking, open data is information that’s accessible, standardized, and reusable, which in this context are defined as follows:
Accessible means the data is free and widely available—most commonly, it refers to data published online at no charge by the people or organizations that gathered it.
Standardized indicates that the data exists in a common format that can be easily manipulated and compared against other datasets. The actual format can vary depending upon the type of data you work with, your specific industry standards, and the audience that will be consuming it.
Reusable means that anyone can use the data regardless of who else might be using it. This refers to the use of an open license. Think of open data as a public good that’s both non-rival and non-excludable, like public radio—just because one person is listening doesn’t mean the utility is diminished for other listeners.
Open data is not the only kind of shared data. Another popular type, public data, is every bit as available—and often more so—but less convenient to deal with. It can still be useful, but users might need to pay for access or submit applications for access, and it may not be standardized in a way that’s conducive to comparison with other datasets.
An example of open data: In 2011, the City of Boston made data about the quality and location of public schools accessible for free online. Standardized in a common, machine-readable format, the data is reusable by anyone who wants to use it—in fact, you can view the data here. An MIT graduate student used that open data to create an updated methodology for assigning public schools in the city.
Without the open data, it would have been more difficult or even impossible to determine the quality of certain schools, limiting or preventing the development of a new methodology.
This example shows one way open data can be important to nonprofits—by providing solutions and answers that were unavailable in the past. It can also make nonprofits more efficient and effective while providing potential donors with more information to use when making decisions about charitable giving.
The National Coalition for Asian Pacific American Community Development used data made open by the U.S. Census Bureau to reveal poverty trends among the portions of the American population represented by the organization in its 2013 report, Asian American and Pacific Islander Poverty. The U.S. Census Bureau’s sharing of information is just one example of government agencies moving toward the open data model.
The nonprofit sector has been slow to follow, and the publication of nonprofit open data remains uncommon. However, there are signs that this might be changing. As more organizations get better at tracking and using data to enhance their own programs, they’re also beginning to share more information about the populations they serve. And funders and individual donors who want to be as informed as possible when giving money are pressuring organizations to open up more data about their effectiveness.
In fact, opening your data might attract more funding. Open data activists believe donors are far more likely to give to organizations that share their data because the donors know exactly where their funding is going and what kind of impact it will have. For example, Splash—a nonprofit that seeks to provide healthy, clean drinking water to kids all over the world—provides status updates on all of its projects via proving.it, a website it created that’s dedicated to monitoring the work the organization does and transparently sharing information and results with investors and donors.
There are plenty of reasons to open your organization’s data, but there are also arguments against it.
Does your nonprofit have data that could be made open for the use of other individuals or organizations? How do you decide whether or not to open it, and what are the risks? Let’s look at some of the issues to take into consideration.
Working With Open Data: Struggles, Barriers, and Other Factors to Consider
The experts we talked to agreed that an organization’s culture is often the most significant barrier to opening data. Everyone knows change is hard—it’s especially hard when it brings extra work with it. From employees that don’t want the extra work to board members who don’t see the need for changes, doing things differently and sharing something you’ve never made public before can be an intimidating prospect.
As an organization, you need to decide how much pushback you’ll get from initiating an open data plan—and whether or not the potential benefits of open data are worth it. But there are additional obstacles beyond organizational attitudes toward opening data.
Some organizations believe they will lose funding if they make their data widely available because that data will reveal flaws or shortcomings in their work, processes, or success rate. For example, if funds were earmarked for a specific project and that project didn’t go well, a donor could see that and choose not to give again. These fears are legitimate, but you may lose more funding in the long-term if your nonprofit doesn’t open its data. Possible failures and weak spots revealed by opening data are still going to exist without openness. By publicizing these weaknesses, organizations can make themselves open to more suggestions about how to fix the problems. Open data is philosophically valuable; by defaulting to transparency, your organization is making a commitment to improvement.
Many nonprofits work in fields where privacy is essential. For example, housing organizations, soup kitchens, and counseling services collect sensitive and private information about clients. If this is true of your nonprofit, make sure you protect this information when making data available. In general, open data should not be about specific individuals and shouldn’t contain any personal information. You should also be clear about what the data you are opening reveals about your constituents, and whether or not it violates their privacy. If you can’t find a way to publish your data in a socially responsible and legal way, don’t publish it at all.
For small nonprofits, impact can be a potential hurdle to open data. These organizations sometimes have a tendency to believe they have so little data, opening it up won’t be significant. This may be true, but not because of one organization. A single nonprofit may not have a critical mass of data on its own; but if every small nonprofit provided open data, together they could make a big difference. It’s still important to research who wants your data, though, and make sure that opening data is worth any incurred costs. Opening data isn’t likely to break the bank, but depending upon your organization’s technical capacity and available resources, it may not be worth doing unless you know for sure that there will be a benefit.
Some nonprofits simply don’t think they have the ability to open their data—at least, not to open it well. While they may be collecting data, they either don’t know how to open it or don’t think they have the resources to do so. However, the obstacle of technology for opening data is becoming less significant and the process can be customized to fit any organization’s technical capability. If a nonprofit makes the decision to open its data, open data activists agree that any open data is good data. It may not be clean or easy to use at first, but once people can see the data, suggestions and resources for improvement will present themselves.
How to Open Up Your Data
The first step in opening your data is technical openness. Technically open data is available to anyone online, easy to download, and exists in a common format. This means your data must be machine-readable. While file formats such as PDF are easily read by users, they’re not machine-readable like an Excel spreadsheet and can’t be easily compiled and manipulated. Data is most commonly shared as comma-separated values in Excel or text files.
There are a few options for where to make your data available. Based in part on the options recommended by the Open Data Handbook, an online resource for all things open data, we suggest the following:
Post files for download from your website.
Use FTP servers like FileZilla, or public Cloud servers such as Dropbox, Box.net, or Microsoft SkyDrive.
Utilize a torrent—a transfer process that divides bandwidth between participating computers so large amounts of data can be shared efficiently—through services such as BitTorrent.
Use an Application Programming Interface (API), an automated “port” that allows controlled access to your data.
Work with a project or community explicitly established to provide a means for data distribution, such as the Sunlight Foundation.
Each option has its own strengths and weaknesses, both for your organization and for those who will access it. For example, if you host the data on your own site, you can ask people to fill out a form to download it, which gives you a way to track and follow-up with users. For more information about the different options or how to choose the one that best fits your organization and your data, visit the Open Data Handbook website at opendatahandbook.org.
After technical openness, the second step of open data is legal openness. This means ensuring that people know your data is available and free to be used by anyone without any legal obstacles by applying an open license to it. This is a relatively simple process—for more information, follow the technical directions from the Open Data Commons, an Open Knowledge Foundation project that outlines legal issues surrounding open data.
If you decide to open your data, start small. It doesn’t make sense to immediately open every dataset you have, especially if time or resources are a factor. Creating a culture of transparency in your organization isn’t going to happen overnight, but it’s possible if you make it a habit.
Changing the culture and practices of your organization is never easy. Transitioning into greater transparency and openness with data is no different, especially when questions and uncertainty abound. As you start to think about open data, be sure to weigh challenges against potential payoffs of data sharing and increased effectiveness, not just for your organization, but for everyone in your sector. Opening your data may not produce immediate rewards, but with time and effort, the willingness of your organization and others to take that step can result in great mutual benefit.
Open data can be daunting, especially for an organization that already has its hands full with data collection. However, the following are some great resources to clarify the process and make the most of open data.
Opendatahandbook.org. This site is a great tool no matter where you are on the way to open data. From explaining what open data is to opening up data to using that data to effect change, Open Data Handbook has all the information you need.
Beyondtransparency.org. If you’ve already implemented an open data policy and you’re looking for creative ways to use the data you’re encountering from other organizations, this is an interesting resource. Beyond Transparency focuses on open data from the government, with articles about practical uses for open data and examples of successful data sharing. The collaborations discussed on the site may provide inspiration for a success story of your own.
30 Places to Find Open Data on the Web. This is where you can see open data in existence. Regardless of your organization’s focus, surely one or more of these 30 resources will apply in some way. Divided into seven categories—government and political data, data aggregators, social data, weather data, sports data, universities and research, and news data—this site can give you an idea of where your data will fit in.
Thanks to the open data activists and technology professionals who helped with this article.
Ethan Drigotas researched and wrote this article during his internship with Idealware. Comments or questions? Leave them below or join the conversation on our Facebook page or on Twitter (@idealware) using the hashtag #npopendata.