Key Economic Time Series Data
Over the past couple of weeks we have been looking at how we can make our time series data easier to find and use, particularly the long runs that allow for comparison over time. We are working to prototype some big ideas about how we can take this forward, but we also want to make things easier for users now by making more use of some of the current functionality on the website and using it in a better way.
With this in mind, we have developed a page to allow quick access to some of our Key Economic Time Series Data, including series covering: GDP, Inflation, Labour Market, Public Sector Finances and Retail Sales. This is a first step that we will look to build on over the coming weeks and months.
By having a static web address this page will provide a useful tool for users who want to bookmark it and come back regularly, make it easier for search engines to find our data and by tying this in with the existing functionality it will automatically update to show the latest data when we publish at 9.30am.
We have brought together what we believe to be some of the top level and key series that our users want and need regularly, but we need you to tell us if there are additional ones missing. We obviously can’t add everything, but please let us know any suggestions at firstname.lastname@example.org or leave a comment.
5 comments on “Key Economic Time Series Data”
I’m afraid I’ve found it REALLY hard to use – unintelligible in fact. I’m not a statistician, but how hard could trends in household expenditure (which was the first one I tried) be? Well – I guess, very hard 🙁
Take a look:
For a start, what does ABJQ mean? Not a statistician… Don’t know. I’d hazard a guess the A starts for “Average” but, well – who knows? I actually tried Googling for it, just for a laugh… It confused Google.
And then what are the numbers? Amount of money in pounds spent per household? What? 991091 in 2012? Er – not in my house.
But does it mean “overall across the UK” then? Well – I know there’s an austerity squeeze on, but it can’t be that bad. So is it in thousands of pounds?
If only I knew the answers to these secret questions I could perhaps get some useful information…
How much effort would it be to put a non-statistician in front of this and find these things out? Just pay a couple of the cleaning staff an extra £50 each for a couple of hours of their time or something.
Oh – and really, urgently, read Don’t Make Me Think by Steve Krug.
David – I totally agree about the labeling – it is a huge problem for us throughout the site not just around the time series data. I’m no statistician myself and the ‘secret questions’ defeat me as well. Resolving it is not as straightforward as you might hope though (the language of statisticians is very precise in its own way and any changing in labeling is a big deal for the organisation) but we are working on it.
Another useful resource might be “Letting Go Of The Words” by Ginny Redish – http://redish.net/books/letting-go-of-the-words – there’s a whole chapter (or more) in there about creating ‘user personae’ to try and encourage people in organisations to start thinking of site design in terms of ‘conversations’ with ‘real’ users – which might be a good way to make people realise the value of explaining things, rather than setting puzzles.
I’ve a gut feeling that a lot of the content on the ONS website would be very valuable to me, and it’s important to use info that has been compiled by professionals, but the lack of usability is a real barrier that’s preventing this at the moment. It’s very frustrating. So please keep chipping away!
Even the full title of ABJQ would be a start… What does it stand for? 🙁
I appreciate the way you’re happy to have this conversation in public, too – many other organisations would fight shy of doing so, but I hope you’ll gain more benefit from being open about it.
I’ve done a bit of asking about and the answer to the ABJQ questions confirms the secret language theory;
“The four digit ID (e.g ABJQ) does not have any meaning itself, it is just a random string used to identify each series.”
We are trying to add some additional context/explanation around each dataset in the very near future (hopefully the first changes will get made today) and will continue to iterate based on feedback so please continue to share your thoughts.