Data Sources

You have two options for finding data to use in your advocacy work: use existing data or collect it yourself.  Data that already exists has the benefit of not needing to be collected, but it may not be formatted, organised or aggregated in a way that is helpful.  Data that you collect yourself means that you will be in control of how it is formatted, organised and aggregated, but it may take more effort and time to get enough of it to use effectively in a campaign.

Regardless of how you get your data, what is important is that you work closely with your allies and stakeholders to assure that the data is useful, can possibly be reused by them, and that it won’t ultimately undermine their long-term goals. You will also need to incorporate a process of verification for the data you are using.

Remember that finding data will be the start of a long journey that will impact your advocacy efforts in many ways.  It is likely that:

• Steps to reach your goals will change and evolve.
• Your understanding of users/stakeholders and their needs will change.
• You will constantly need to evaluate your data and the tools you use to connect that data to citizens.

Existing Data Sources

Data sources include governments, civil society organizations, academic institutions, think tanks and ‘open data’, data that can be freely used and shared, on the Internet. It is also increasingly easy to crowdsource data from individuals; you just need to give them a reason and be willing to share ownership. There are also lots of Internet tools that make it easy to collect data. The Internet also makes it easy for everyone to collect data across geographically dispersed regions, so in this age of globalization we can get a clear picture of what is going on around the world.

• Government Sources

FairPlay Alliance takes Slovakian government data and makes it accessible to citizens.

In recent years, governments have begun to release some of their data to the public. Many governments host special government data platforms that openly provide the data they create. For example, the United Kingdom started data.gov.uk. Similar data portals exist in the United States, Brazil, and Kenya and also in many other countries. If you want to find out if your country has an open data portal visit, datacatalogs.org. If you are trying to find data about a country that doesn’t have open data, you can try looking at another country’s data source to see if they have data on the country in which you’re interested.

• Organizations and Institutions

There is a growing number of initiatives that aim to provide public access to published data. Organizations such as the World Bank and the World Health Organization release large amounts of data available to the public. Scientific institutions such as NASA also release data to their own repositories, some of which are open to the public. To help people find data, projects like the Open Access Directory’s Data Repository List or the Open Knowledge Foundation’s datahub.io have been started. They aim either to collect data sources, or collect different data sets from various sources.

• Search Engines

Another very good source for data is the Internet. You can search for CSV on Google by typing +filetype:csv in the search bar. CSV files hold plain text as a series of values (cells) separated by commas (,) in a series of lines (rows). For example, searching for “South Africa” +filetype:csv will result in CSV files mentioning South Africa, in the name or description. You can try searching different file types as well, such as xls for excel spreadsheets or pdfs.

Google Tables search also allows users to search within open tables available on the Internet for the data that they are looking for in tabular format, such as geographical data in the form of maps. This differs from searching for spreadsheets online, as it results in tables that are on websites or readily available to fuse with Google Fusion tables to make compelling results.

• Academic Institutions and Think Tanks

Colleges and universities are hubs of research activity.  It’s always good to make inquiries about current research projects and to see if they might have useful data sets–many are made available online.

For more on existing sources, see School of Data’s Finding Data (which inspired this post!)

Crowdsourcing

Another option is to collect your own data and make it available to others who might find it useful. This allows you to control the type of data you are collecting and involve others in your data collecting efforts, which can help you build allies. If you are crowdsourcing data from citizens, it is critical that you explain value or benefit to the people you are collecting information from. Always make sure people understand why they are participating and what the goal is, so that you are very transparent and inclusive. Most importantly, listen and pay attention to your users’ needs and interests . You need to understand them, and any outreach you do should be an engagement opportunity for them to come back to you.

If you are crowdsourcing and collecting your own data, a good practice is to make it open data. Below are some key characteristics of good open data to consider when crowdsourcing and collecting your own data:

• It can be linked to and easily shared so it can be used by others.

• It is available in a standard, structured format so that it can be easily processed.
• It has guaranteed availability and consistency over time so others can come to rely on it.
• It is traceable so that it goes right back to where it originates and others can work out whether or not to trust it.
• It has a license that says it is open data.

Learn more on how to assure your data is open at the Open Data Institute’s What Makes Data Open

Try a map

Mapping platforms are a great way to start a crowdsourcing effort.  Maps immediately translate information about individuals into information about communities.  Mapping platforms such as Ushahidi’s Crowdmap make it easy to collect geo-tagged data from a large group of people.  Just remember that a map may not necessarily be the best way to visualise information when you are preparing it for use in your advocacy campaign.

Examples of Crowdsourcing:

The Land Matrix takes crowdsourced data and makes it reusable.

The goal of Women Against Violence is to reduce police violence against women who use drugs.

Harrassmap’s effort to stop sexual harrassment in Cairo.

Nabz-Iran collects reports of human rights abuses in Iran.

Verification

Verification is an important component of collecting data. An important question to ask when collecting data is “What sort of scrutiny will your data be put under?” Courts of law or international tribunals, for example, require very rigorous verification processes, so it is important to make sure that your data can be verified by legitimate sources. At a minimum you want to verify data to the point where your reputation remains credible; if you are putting data out in the world, you need to make sure it is reliable. It is very important to always review and fact-check crowdsourced data.

About Data Advocacy Basics

FabRiders provides workshops, webinars and advice on how to use data effectively for advocacy, social change and social justice. This series of blog posts on Data Advocacy Basics pulls together a lot of the information we use. Much of this knowledge has been openly shared by many others, such as School of Data, Tactical Technology Collective and the Transparency and Accountability Initiative. A big thanks to Sarvenaz Fassihi for compiling and editing Data Advocacy Basics and Beatrice Martini who gave great comments, feedback and support.

See also:

• Data Advocacy Basics, Pt I: Data Action Cycles and Stakeholders
• Data Advocacy Basics, Pt II: The Data Pipeline
• Data Advocacy Basics, Pt IV: FAQs on Visualising Data