A day of maps and facts

For someone who is curious by nature, Wikipedia holds an endless fascination. I enjoy rooting around it for snippets of information. Similarly maps hold a so much detail in such little space, they’re just ace.

I’m sure my love of maps stems from a Weetabix Wonderwold Atlas I got in 1988/89 in exchange for some cereal packet tokens. I was so interested in this book – and particularly this page (below) about different scales of maps – I proudly produced a primary school project about it. I used the atlas as my main source of world knowledge for as long as I could but the presence of the USSR, Yugoslav Republic and other since disbanded nations meant that eventually it was just too unusable. And there was also that thing called the internet. Continue reading

It’s easy to open data and here’s how to start

On Monday, Sarah Parker and I ran a workshop at the Lasa even “Unlocking the Potential of Open Data”.

The session was called “The Open Data Chocolate Box”.

(I should say from the outset that it was very much Sarah’s session, I just did a turn at the end).

When talking about open data in the sector, enthusiasts have a tendency to jump to how to find open data sets, using government data, running hack days….all the stuff that scares people looking to just get started. What we really wanted everyone to take away from the session was that actually it takes very little to actively open data and you don’t need much technical knowledge.

To illustrate how much you can do with just a small piece of data, Sarah created the ‘open data challenge’. The challenge was for organisations to fill in how many people they worked with in 2010/ 2011 along with their charity number.

We were really excited that 13 organisations took part in the open data challenge. You might think that 13 isn’t a very large number but the voluntary sector is right at the beginning of the open data adventure and 13 is a massive jump in what information is now out there.

*HEALTH WARNING* The following data mashup is only based on one piece of data from 13 organisations and cannot be used to draw ANY conclusions about the sector or its work. *END OF HEALTH WARNING*

Lamplight data wrangler Matt spent a couple of hours taking this single piece of data and combining it with what is already out there in the Index of Multiple Deprivation (IMD) and information on Open Charities.

He was able to represent spend in different areas against numbers of people reached as well as spend in different quartiles on the IMD.

VCS Open data hack

You can see the whole hack at http://www.lamplightdb.co.uk/hacks/vcsopen/.

So, how can organisations get started with opening up their data?

  1. Get a database – you want to be able to report on your activity at the touch of a button (and as an aside it will help you to be more strategic in your work)
  2. Think about what you’re already sharing, such as Charity Commission accounts and annual reports.
  3. Put that information on your website in a machine readable format e.g. CSV or Google spreadsheet.
  4. Give the information an open licence so others know they can reuse it.
  5. Add a /open page and add it to their directory.

The moral of this story is that to get started with opening up data doesn’t take very much, just one piece of data in a spreadsheet on your website. From there the people that want to can start doing things with it and you might start to think of information you want to get to support your own work.

There are issues of trust around this (who is using your data? how do you know what kinds of conclusions they’re going to draw?) which I’m not going to touch on here but Sarah is planning on exploring those further in a blog post. There are resources from the day at http://storify.com/lasaict/unlocking-the-potential-of-open-data.

In an attempt to capture some of the enthusiasm in the room we asked people to make the ‘open data declaration’:

The 'open data declaration'

(Photo taken by @nicktheowl)

I (the undersigned) declare that I will:

  • Complete the VCS Open ‘open data challenge’ at http://vcsopen.wordpress.com/dataset/.
  • Talk to at least one person (preferably a colleague) about open data and the open data challenge.
  • Investigate adding a /open page to my website.
  • Add at least one piece of opened data to my website in a machine readable format.
  • Tell VCS Open what we’ve done no opening up our data.

Signed:

Lasa, Unlocking the Potential of Open Data
17 September 2012
London

[2/10/12 - ETA: VSC Open post "Trust and Open Data" http://vcsopen.wordpress.com/2012/09/26/trust-and-open-data/]

Creating your own data visualisations

At the NCVO Annual Conference on Monday I went to the workshop on visualising data. Now, I love data and I love visualisations so I was pretty much like a pig in the stinky stuff. The workshop was run by Ed and Matt from Reason Digital.

Visualisations are essentially a way to show (often complex) information using graphics.

Whether you have to write reports full of facts and figures or you need to demonstrate the impact of your work to the public then creating a graphical representation of the information might be an effective way of doing it.

I have to say that I don’t agree with visualisations for visualisation’s sake. You as an organisation need to have something to say that you think will be more effective using graphics.

In general there are two types of data visulations. Firstly ones that illustrate a number of different variables in a way that makes complex data easier to understand at a glance. One example of this is the Minard visualisation of Napoleon’s march into Moscow. This visualisation incorporates information about time, distance, soldier numbers and temperature.

Losses of the French Army in the Russian Campaign 1812-1813

The second type are visualisations that use graphics to illustrate a figure, such as on this poster about Mashable users:

Mashable vs Hashable

In the workshop it was also suggested that an infographic (such as the first example) is telling a specific story and leading you to a conclusion. A data visualisation (the second example) puts the facts in front of you and leaves you to make up your own mind.

People enjoy debating these two types and the terms that are used to describe them, you can read more about it in the article “Infoposters are not Infographics: A comparison”.

Whichever type you end up creating there are 12 steps that Reason Digital think you need to work through. Even if you plan to work with a company, volunteer or staff member to create your visualisation then these steps will help to define exactly what it is you want to achieve.

  1. Have a point and a purpose – What’s the issue you want to get across? What’s the question you’re asking?
  2. Make it relevant and interesting to a large the right audience – This might involve stepping away from what you as an organisation are doing and looking more at the issue or cause in general terms.
  3. Decide if you’re making an infographic (with narrative and pointing to a conclusion) or a data visualisation (something you can explore and people can make their own minds up).
  4. Deconstruct your point into a series of facts or statements that you can back up with data.
  5. Gather data and statistics: You might have the data you need within your organisation but you can also get supporting data from the ONS site, NHS data store, data.gov.uk, What do they know?, COINS, Guardian data blog and local council websites. [ETA - Once you've gathered your data then make sure you reference it in your visualisation. It will give more credibility to what you're saying.]
  6. Choose and order your facts. This stage needs a certain level of statistical literacy to be able to interprate data and the inconsistencies it can throw up.
  7. Work your data into a narrative. Make sure the story you’re telling comes out in your visualisation. The key to a great visualisation is in finding the story.
  8. Edit, refine and make it flow.
  9. Apply visual methods: This could be maps, scale, timeline, bar charts, icons, pictures of objects, word cloud or illustrations.
  10. Apply an interesting visual style.
  11. Choose the right medium: What will work best for your audience? Static image? Interactive graphic?
  12. Jump start the sharing process: Make sure it’s easy to share by putting it on YouTube, Flickr, Pinterest, visual.ly etc.

Some of the tools we looked at for creating visualisations included Google fusion tables, Google refine, Gapminder and visual.ly.

There are of course barriers to creating visualisations:

  • Having the right data – Once an organisation has decided on the story it wants to tell the next step is finding the right data. Even if you don’t have it straight away, once you’ve identified what it is you can build in mechanisms for collecting it in the future.
  • Knowing how to interpret data – As I mentioned above, you need a certain level of statistical literacy to make accurate visualisations. This kind of skill could come from a staff member or volunteer.
  • Confusing personal data with that needed to create visualisations – There is a difference between using personal information (such as religious beliefs, income, sexual orientation etc. about specific people) and generic figures about how many of instances there are of things or events. The latter is what you need to be able to create a visualisation, how many service users are male, how many service users also access other services etc.
  • Graphical skills – Not everyone has the skill to create an infographic. It may be that at the last stage you have to go to an external agency but if you’ve completed these stages then you’ll find the process a lot easier.

Have you tried creating visualisations? Do you think they are an effective way of communicating data? I’d be interested to hear your thoughts.

[ETA - I came across this great blog post and infographic in February 2013 from Shelter: http://blog.shelter.org.uk/2013/02/we-want-economic-growth-so-how-about-more-homes/]

House building, house prices and economic growth in England.

House building, house prices and economic growth in England.

 

Civil society 2.0 (open data)

The NCVO annual conference took place last week and there was a fantastic session about open data and what it means for the voluntary sector. I’m still digesting everything that was covered but something inside is telling me it could be huge for how organisations operate, but not without the right support and information.

There are two elements to this discussion we need to be considering; the opening up of our own organisational data and making use of data that’s available from Government, local authorities and other organisations.

What could charities do with open data?

It’s almost impossible to know the limits to what organisations could do with open data, but David Kane gave a lot of food for thought in his presentation:

One of the things that David said was that open data could help charities by:

  • Informing lobbying and campaigning - Demonstrate how an issue is affecting people by mapping information that exists on need, services etc.
  • Improving services - Ask for feedback on services using post code details to map your service locations.
  • Demonstrating impact – Match the services you are delivering and their impact with known information about need.

What skills/ support do voluntary organisation need to make the most of using open data? 

But in all of this we need to think about how practical it is for small voluntary and community groups to make the most of open data. Are the technical skills out there in the sector?

I think David summed it up really nicely in his presentation by saying that, yes, there is a need for specific technical skills to make the most of open data BUT the tools we would use are free and there are lots of people out there who want to make a difference, have the skills and are connected by social media networks such as Twitter, Facebook and LinkedIn.

Some of the information and skills I think groups need are:

  • An understanding of why Government is releasing its data – link to the transparency and accountability agenda.
  • Some sense of what might happen around this in the future, whether it’s likely to be a requirement for organisations to release their own data.
  • Dedicated time looking at organisational strategy to get to a clear understanding of what they need to prove need, demonstrate impact and/ or campaign more effectively.
  • Knowing how to access the data, where to store it, and the implications for data protection.
  • An overview understanding of how different websites or websites plus data can talk to each other.
  • Practical experience of using APIs, playing with Google maps etc.

What about organisations releasing their own data?

Rich Watts has been very articulate in describing his organisation’s thinking when releasing their data so I won’t bother repeating it here.

I think there is a real benefit for organisations having the conversation now about opening up their data, as Rich alluded to in the session it’s almost certainly going to become a requirement for organisations in the not too distant future, either as a condition of registering with the Charity Commission or as a requirement when receiving public funding.

So what next?

There are already people on the forefront of testing what can be done with open data; visualisation, mapping etc so my own view is what we really need now are real life voluntary sector examples.

Some of this – I hope – will come out of the Open Data Campaigning Camp in a couple of weeks but we also need people giving it a go, perhaps as part of a structured pilot that involves using some of the open data “recipes” and tools on the Voluntary Sector Datastore.

NCVO is looking to run its own voluntary sector hack day, which is a great start. I can’t wait to be involved and see what emerges.