WHAT’S NEW DASHBOARD – JUNE 2016 UPDATE
Progress has been made in the following areas this quarter (April to June 2016). Please see Annex A for who to contact in each Department for more information and Annex B for the detailed returns.
Publishing
HSE (template for tables)
DWP (gov.uk changes)
ONS (enhanced table builder)
HSCIC
SG (content strategy)
VOA
DEFRA
MOD (host for apps & visuals)
BIS / Content/Commentary
DfT (online quiz)
DWP (fact sheets)
ONS (new style bulletin)
HSCIC
NHS Scotland
VOA
HMRC
ORR (factsheet) / Training and workshops
DfT (user personas, Python)
SG (communicating analysis, data vis)
DCLG
ORR (data science proposal)
BIS
Ofsted
Social Media
DfT (launched account)
ONS (GDS social media playbook)
SG (over 1,000 followers)
NRS
DEFRA (protocols)
ORR / Infographics & charts
HSE
DWP (visualising uncertainty)
NHS Scotland
NRS
MoJ
Ofsted / Reviews/Strategies/Groups
HSCIC (output consultation)
NISRA (peer review and communication strategy)
DCLG (peer reviews)
Guidance
HSE (infographics)
DEFRA (infographics, social media)
NISRA / Other
HSE
DfT
SG (communicating analysis awards, blogging)
HMRC (cross-govt working) / Software
DWP (Stat-Xplore, open data app)
SG (R Shiny Server)
DCLG (open data app)
DFID
Ofsted (Python)
WG (open data platform)
DH (data science hub)
CURRENTLY TRENDING
· HSE analysis of user personas
· NISRA user engagement best practice guide / Data science (DfT, DWP, HSCIC, ORR, BIS, Ofsted, DH)
· DfT hackathon
Tableau (MoJ, SG, NHS Scotland, DFID)
· MOJ first interactive tools / Data visualisation (DCLG, MOJ, DFID, SG, NRS, BIS)
· DCLG exploring Inkscape and Scapetoad
/ CHAMPION OF CHAMPIONS
Is awarded to DfT for their online quiz on electric cars
/
“Highly commended” goes to DWP for reducing bulletin from 150 pages to 11 pages!!!!
ANNEX A – List of Presentation Champions and PDC Representatives
Below is a list of each Department’s Presentation Champions and PDC Representatives. If you would like to find out more about the work being carried out that is highlighted on the dashboard, please contact the colleagues listed below.
If there are any changes required to the list, please let the Good Practice Team know and keep the information in your “What’s New?” return up-to-date. A list is also kept at the following link:
https://gss.civilservice.gov.uk/statistics/presentation-and-dissemination/commentary-champions/
Response rate was 73% (22 out of 30 departments). Responding departments:
Department / Presentation Champion / PDC RepBIS / /
Ofsted /
DCLG / /
DEFRA / /
DWP / /
HSE /
MoD /
DfT /
/
(group Chair)
ORR /
MoJ /
DH/NHS England / /
HSCIC / /
DFID /
ONS / Elizabeth.McLaren.gsi.gov.uk
/
Welsh Government /
Scottish Government /
/
NHS Scotland /
NRS /
VOA / /
HMRC /
OME /
NISRA / /
Non-responding departments:
Department / Presentation Champion / PDC RepDfE /
/
Ofqual /
Home Office / /
DCMS / /
HMT /
Cabinet Office / tbc
DECC / /
Public Health England / tbc /
ANNEX B – Full “What’s New” returns – June 2016
Response rate = 73% (22 out of 30 departments)
Department for Business, Innovation and Skills (BIS)
· Our CBAS analyticalplatform has now been rolled out to 160 users
· Training: Dozens of colleagues have been trained internallyin- excel best practice, R, and HTML 5 and we have expert external contactors providing training in SQL. there is a comprehensive programme of training events scheduled
· Training: In addition to having a dedicated mapping role the data science team has held over20 individualtraining sessions in the QGIS mapping tools with a number of group training sessions scheduled over the coming months.
· Workshop: The data Science and Visualisation Team ran a cross government workshop in conjunction with GDS and DWP on interactive data visualisation using JavaScript, in particular D3.js.
· Training: We have held training sessions on Inkscape which is a free open source tool to producing dashboards using Scalable Vector Graphics (SVG).
· Data visualisation: We have set up an online data visualisation gallery: http://cbas/publications/DataVisualisationGallery/
· Seminar: On the 4th July Charles of the ONS will be holding a seminar in BIS on ʺEffective tables and Chartsʺ and attendanceon this is now ʺsold outʺ.
Office for Standards in Education, Children’s Services and Skills (Ofsted)
· Ofsted upskilled representatives on infographics and piktochart which is the main website Ofsted uses for creating infographics. An infographics schedule is in place for our upcoming statistical releases.
· The data science group has begun work on several projects, including scraping school websites using a Python web scraping package. A briefing was published on findings of a project on coastal schools, which used R and mapping packages to calculate closest distances between schools and the coast.
· One member of the data science group has been accepted for the next round of the data science accelerator, working on schools risk assessment and web scraping.
· The Data View tool was successfully relaunched in December 2015. More work will take place over the course of 2016 to improve the tool.
· Ofsted met with colleagues from ONS in order to link into the publishing/GDS work being taken forward.
Department for Communities and Local Government (DCLG)
· A successful season of GSS led training sessions and more informal visualisation 'teach-ins' has concluded. All sessions have been well attended and received positive feedback, driving the potential to run more classes in the near future. Other colleagues within the Department are also keen to use the 'teach-in' platform to present other work related to the innovative presentation of data and statistics.
· Groups of statisticians are exploring new visualisation software acquired by the Department (Inkscape, Scapetoad) and working to demonstrate wider to colleagues in near future.
· DCLG Open Data App now available across Android and IOS
· Peer reviews of data visualisations in DCLG stats publications to commence later in summer.
Department for Environment, Food and Rural Affairs (DEFRA)
· HTML: Working to develop a small number of rural statistical releases to be published in html format to make it easier to access content alongside more common publication formats such as pdf/A. Html is particularly suited to mobile access and we are exploring the use of bespoke software packages to enable easy transcription to html rather than using the GDS tool which is not ideal for this purpose.
· Identifying user needs: We circulate to statistics team a 6 monthly update on web usage figures for statistical content with the ability to undertake more detailed analysis if required. This information gives the statistician more detailed insight into users and user needs as well as valuable data on overall usage and developing trends. The information comes from our Statistics Dashboard. Based on Google Analytics, this provides a real time view of website usage. We are now being asked by the GPT to promote our dashboard to others. We circulate detailed information annually about queries received in the Enviro-Inbox broken down by topic. This can provide further intelligence on user need and requirements regarding our environment statistics.
· Standardisation of data: We are investigating implementing a more standardised approach to formats for data being published by the department. This will make it easier to re-use, to access and to share information both internally, across government and for external users.
· Working with comms: Food and Trade stats team published an ‘ad hoc stats release’ entitled ‘Food and Farming at a Glance’. This was a pioneering project for Defra – conceived by Strategic Comms and specifically produced as a briefing document for the media.
· Open data:We have a target of unlocking 8,000 data sets by 25 June 2016. By mid-May we had already published 5366 new open datasets. On 28 June we are holding a #DataMarket event for data users and custodians.
· Social media and infographics protocols:We are looking to develop, in conjunction with our comms teams, protocols for the use of social media and infographics
Department for Work and Pensions (DWP)
· Template for releases: The new template for our statistics has begun to be used by lots of teams across the department. For the poverty statistics releases this is having a real positive impact: it’s reduced the publications from around 150 pages, making the timeliness of the publication better and the risk of typos far lower. We spoke to some of our key “critical friends” about moving to the new release in the next couple of months and received very positive feedback. We have mandated the use of the template – ad hoc analysis is also now released in the new format, ensuring consistency across the department’s outputs, not just the regular statistics releases.
· Visualising uncertainty: Our Fraud and Error statistics were released in May, using a good visualisation of the uncertainty around the estimates: there was no media coverage of a spurious difference between two of the series, due to this clear message.
· Using APIs: Our Data Science colleagues have created "Churchill". It’s a way to make the communication between policy analysts and policy makers more efficient – by giving policy makers access to published statistics, in an easy-to-use front end they can quickly answer more simple questions or delve into more detail, helping them to better formulate more complicated questions they’ll need help from the analysts about.
· Open data on Stat-Xplore: We are drafting a consultation document about ceasing the production of our Tabulation Tool, moving more data onto the Stat-Xplore platform. Stat-Xplore supports the open data agenda far better than the Tab Tool because it offers users the ability to create their own tables, define filers and variables and it has an API. We hope to consult publicly, to inform our prioritisation, within the next quarter.
· Gov.uk: We have made changes to our gov.uk statistics landing page. This was informed by some user testing, where we set people tasks to find out information about specific stats and found that the organisation of the page was causing much confusion. We will be presenting this work at the June PDC meeting.
· Posters not emails: Our Insight statisticians are trialling a different dissemination technique to support advertising campaigns. Instead of emailing tables of numbers that are unlikely to be read or understood, they are creating fact sheets with commentary about the effectiveness of each campaign, printing them as posters and talking campaign managers through the results. This is helping to engage people and raise the accountability. They are planning to roll this out to all of their Management Information reporting.
Health and Safety Executive (HSE)
· User personas: Analysis of the results from our User Engagement Survey suggest that our user base fits with the current user persona model and that HSE statistics are accessed by a diverse range of users across the three main personas (Inquiring Citizen, Information Forager and Expert Analyst). This suggest that our current statistical release communication strategy of developing a variety of products for use across the available media channels (Twitter, HSE statistics website etc.) to appeal to different users is the correct approach and allows us to focus on refining those products.
· Table formatting: As part of our data warehouse work programme, we have developed a new template for our data tables. This style allows for a more customisable user experience in terms of selecting the data they are interested in while also removing some of the more complex elements that made comparisons difficult for users.
· User feedback: As a follow up with some of the users who took part in our User Engagement Survey, we will be asking for feedback on the guidance for using the new table template with a view to improving the user experience before they launch on our statistics page later this year.
· Infographics: HSE statisticians have been working with experts within HSE’s Agriculture team to provide advice and guidance on the use of infographics to communicate statistics as part of an upcoming report they will be publishing.
Ministry of Defence (MoD)
We have launched tools.mod.uk (which Adam Beirne demo’d at the symposium). It is designed to host apps and visualisations, which can’t be hosted on gov.uk.
Department for Transport (DfT)
· User engagement/social media: We launched our DfTstats twitter feed in March 2016, and since then have sent out 74 tweets, regarding the availability of our statistical series. We now have 261 followers and we’re still pushing to achieve our goal of 1000 followers by next March. In total, the twitter page has received 3200 profile visits. Our top follower in May was NatCen.
· User engagement/relationship with DfT strategic coms: We’ve started attending DfT communications directorate weekly Forward Look Grids meetings. We are using these meetings to identify all news/ events / announcements which might be related to our statistics and which we might want to follow up with policy and press colleagues, to raise the profile of statisticians and statistics with coms colleagues and to identify opportunities to work collaboratively with coms colleagues to produce high quality graphics & data analysis in support of DfT’s strategic narrative.
· Dissemination/User personas: Our Presentation champions have run a workshop for our J-Stats group (below G7) promoting awareness of the ONS user personas and have started some evaluation of how our products align (which will be followed up in next yrs priorities). They are also working with MoJ to offer a similar session there.
· Dissemination/ New approaches to dissemination: We have been developing our HTML skills and have used this to produce an online quiz to support the Go Ultra low policy campaign, which aimed to change people’s perceptions about electric cars. The quiz drew on a variety of statistical sources to challenge users to test how much they knew about electric vehicles.
· Dissemination/ Capability & Data science Our data science community participated in the departments first ever Hackathon. Dubbing themselves the DfT Hack Club, the group’s remit was to come up with creative and technical solutions to some knotty DfT questions - ranging from ‘how to best interrogate Hansard data’ to ‘modernising the way we do BICC reporting’. We also been running some popular Python training sessions.