…about information on Places to Go, Things to Do
RSS icon Home icon
    Print This Post Print This Post
  • Linking up – learning from the Data Quality Working Group

    Posted on December 22nd, 2009 Tim Davies 3 comments

    Plings and PKHDThe first meeting of the Plings/PKHD Data Quality Working Group took place last Thursday – exploring the opportunities to share learning and data between these two local data projects.

    The results of a survey carried out across the local authorities involved in the Plings project highlighted that the teams responsible for collecting and publishing information on positive activities are mostly working separately and independently of the Family Information Service (FIS) staff who are responsible for much of the content in the Parent Know How Directory (PKHD). However the face-to-face meeting last week demonstrated that there are many opportunities for collaboration and building links between the two projects.

    Linking agendas to get prioritised

    Plings is, in the scheme of things, a small budget research project. Some members of the Data Quality Working Group have found that, compared to PKHD, with it’s top-down mandate, and funding to support local authorities to adapt to it, it can be harder to secure management support for work on Plings and promoting positive activities to young people.

    By finding the links between positive activity information and PKHD, there are opportunities to benefit both projects – and secure increased support for positive activity information work.

    Linking learning

    One of the big challenges both Plings and PKHD have been exploring is ensuring data quality. There is a balance to strike between only having information in a database when it has been quality assured, complies with quality standards, and is complete information (which would dramatically cut the amount of information that could be added to the database), and putting all information into a database regardless of whether it can be checked out easily or not – and publishing it with caveats explaining to the end-user how they can verify the information.

    This challenge becomes particularly acute when sharing information between local authorities – where different policies to govern how information is collected and added to the database may be in operation.

    In the PKHD project, a simple field to explain what standards have been applied in the collection and publishing of data by any LA has been used – so policies and processes can differ across authorities – but end users have basic information on how to respond to the data they are given.

    Through the Data Quality Working Group, members were able to discuss learning from Family Information Services about putting in place processes and procedures to ensure the quality of data – and were able to discuss the sorts of standards that would be necessary to allow data from Plings and PKHD to be easily exchanged.

    Linking locally and nationally

    The sorts of information the PKHD may hold for parents, and that Plings contains for young people, can often overlap. It makes sense then, to explore how there could be integration between the two systems to prevent double-entry of information.

    Integration can happen at different levels – and between now and the next Data Quality Working Group meeting the team will be researching into the various possibilities more.

    Integration could be at a national level – with data from PKHD and Plings swapped by the central platforms. This would require the data fields from the two systems to easily map onto one another.

    Or integration could take place locally – with authorities adopting a single local system to feed both platforms. This would require thought about where the responsibilities to manage data fell in the local authority.

    Keeping the focus

    Thinking of integration raises a number of further challenges that the Working Group will be exploring. PKHD is a directory for parents. Plings is focussed on information for young people. Careful thought needs to go into the right places to integrate, and the right places to let each dataset retain it’s own distinctive content and style.

    More to come

    The Data Quality Working Group will be meeting again in January and we’ll have more on the blog soon after they meet…

     

    3 responses to “Linking up – learning from the Data Quality Working Group” RSS icon

    • Cambridgeshire’s data feed to PKHD from its community website Cambridgeshire.net includes positive activities alongside information for younger children. It could be a real bonus if the positive activities could reach Plings without dual entry or another integration piece being needed. Interested to hear if that idea takes off or if other authorities are interested in it.

    • @San Taylor – The point you raise is very relevant, esp when time and resource is limited. I know in Newcastle they have been sending data to Plings via their evince system, which also sends data up to PKHD – so no double entry going on. This has been undertaken through their involvement in the plings pilot project.

      On a wider issue, I think there is still a slight “mismatch” in terms of how the two data models are thought of. One is a “directory” of services (at a basic level) whilst the other is a listing of “events” (which could be ongoing activities). One answers the question “who can help?”, whilst the other addresses “what’s on today?”. This isnt a mismatch however – we can see how data can be pulled from a directory into a listing (every Weds at 7-9pm), but we then need to be careful about quality (what if xmas day were a Wednesday?), which Im sure the group discussed.

      I plan to make it to the next meet, so look forward to it!

    • @stevieflow – Hiya. Thanks for the reply.
      Interesting what you say about data models.

      Sharing news items, for example, is a doddle – they have a simple, widely agreed & understood structure. This contrasts with events, activities etc, where there’s more going on in the data so more scope for divergence & incompatibilities – the mismatches that you talk about. These mean that info that’s meaningful in one system might not fit into another &, even if it does, can easily lose a key part of its meaning (turning high quality info in one context into what seems like poor quality info in another).

      When we were setting up Cambridgeshire.net, we looked for models/standards for event & activity data with a view to inter-operability, but there seemed to be little consensus beyond the obvious. If Plings or PKHD had been further on at the time, this could well have influenced aspects of our approach!

      When designing schemas, another pressure, of course, is that collecting data involves compromise – keeping things quick & simple for your data providers whilst providing useful options for data users. Compromise differently than your neighbour & your data maybe shares badly as a result.

      Data sharing with its obvious benefits will get easier: 1. the more agreement there is on the base entity types in this area (eg. events vs regular activities vs organisations) & 2. the more compatibility there is between the data schemas in use.

      I always think projects like Plings are important in providing the space to work through questions like these, finding practical answers & – most importantly – getting them out there. Good, then, to see today’s Plings blog posting – (Plings.net going national – 30 Jan 10).

      Anyway, just a few thoughts.


    Leave a reply