Elections Data (UK0069)

Objective: To simplify and improve how the UK collects and publishes election data to enable greater use and reuse of structured information by government and civil society.

Status quo: Currently, there is no standard data structure for reporting election results. This means that to aggregate election results requires obtaining non-uniform, often unstructured data from each publishing authority - this is a highly resource intensive process. Local authority returning officers currently have a statutory duty to publish local and national elections on local authority websites. This activity currently takes place in a piecemeal way from one organisation to another with no official guidance or common practice to publish such data in any particular style, format or web location. The Electoral Commission guides that administrators must give public notice of the name of each candidate elected and of the total number of votes given for each candidate (whether elected or not), together with the number of rejected ballot papers as shown in the statement of rejected ballot papers. Whilst this approach allows scrutiny and review at the individual organisational level, much manual effort is required in finding the local published webpages and then to collate data from every publishing source to create a national overview. The current practice is difficult, labour intensive, time consuming and often error prone. Substantial savings, better data discovery and data reuse is possible if electoral administration departments can be encouraged to publish their data to a simple consistent form which can be read by humans and machines.

Ambition: The vision is to work with all interested parties to agree a simple, minimum burden process and data standard to introduce consistency of data availability across the local government sector. Publishing election results in a consistent way will assist those who need to quickly understand the political landscape after an election and encourages other third parties to develop apps and other analysis services to help to inform the public faster about the overarching outcome from elections. It will also promote wider engagement and outreach with innovative application development and scrutiny by the electorate. It is our aim that by 2020, all election results will be reported digitally using a standard, machinereadable and open standard.

7. Elections data

Commitment Text:Working with interested parties from government, Parliament and civil society, we will develop a common data standard for reporting election results in the UK faster and more efficiently and develop a plan to support electoral administrators to voluntarily adopt the standard.

Objective:To simplify and improve how the UK collects and publishes election data to enable greater use and reuse of structured information by government and civil society.

Status quo:Currently, there is no standard data structure for reporting election results. This means that to aggregate election results requires obtaining non-uniform, often unstructured data from each publishing authority - this is a highly resource intensive process. Local authority returning officers currently have a statutory duty to publish local and national elections on local authority websites. This activity currently takes place in a piecemeal way from one organisation to another with no official guidance or common practice to publish such data in any particular style, format or web location. The Electoral Commission guides that administrators must give public notice of the name of each candidate elected and of the total number of votes given for each candidate (whether elected or not), together with the number of rejected ballot papers as shown in the statement of rejected ballot papers.

Whilst this approach allows scrutiny and review at the individual organisational level, much manual effort is required in finding the local published webpages and then to collate data from every publishing source to create a national overview. The current practice is difficult, labour intensive, time consuming and often error prone. Substantial savings, better data discovery and data reuse is possible if electoral administration departments can be encouraged to publish their data to a simple consistent form which can be read by humans and machines.

Ambition:The vision is to work with all interested parties to agree a simple, minimum burden process and data standard to introduce consistency of data availability across the local government sector. Publishing election results in a consistent way will assist those who need to quickly understand the political landscape after an election and encourages other third parties to develop apps and other analysis services to help to inform the public faster about the overarching outcome from elections. It will also promote wider engagement and outreach with innovative application development and scrutiny by the electorate. It is our aim that by 2020, all election results will be reported digitally using a standard, machine readable and open standard.

Milestones:

1. Develop a draft schema and publishing process for consideration, refinement and agreement by interested parties - particularly data publishers, election management system (EMS) suppliers, data consumers

2. Use the draft data standard for real by gathering local elections results as they are announced

3. Develop guidance materials and a support programme to assist Election Services Departments to participate

4. Data consumer groups to trial early use of the standard - even to the extent of manually re-working published data into the standard themselves to demonstrate benefits

5. Adoption by the suppliers of EMS systems to provide auto-extraction of local election content into the standard format

6. Develop online data search, validation, harvesting and aggregation tools to assemble local data into combined regional and national elections results register

7. Encourage an initial pilot of local authorities to trial data output in the standard form - using May 2016 local election results. Aim for 20-30 participants

8. Encourage wider take up of the process in the 2017 local elections. Aim for 100-120 participants

Responsible institution: Local Government Association (LGA)

Supporting institutions: Ministry for Housing, Communities and Local Government, Cabinet Office, Government Digital Service, Electoral Commission, Association of Electoral Administrators, House of Commons Information Services, Plymouth University Elections Centre, Democracy Club, Democratic Audit, Democratise, LGiU, mySociety, The Open Data Institute, suppliers of the key electoral management systems (EMS.

Start date: May 2016

End date: June 2018

Commitment Aim:

Election data laws are outdated and there is no consistent or common approach to collecting election data, especially at the local level. Currently, laws governing elections in the UK ask only that notices of results be placed in a public place. This means that election results are placed online in various forms (as PDFs or scanned images) and there is no consistent means of publishing them.^{[Note 60: Interview with Tim Adams, Local Government Association, August 2017.]}

While national elections are relatively well analysed, results are inconsistent and slow for local elections (at the parish, district or county council, as well as for mayoral and police commissioner elections).^{[Note 61: In the UK, local government is divided into parish (village level), district (town or sub-regional), and county council (regional), with other elections for local elected mayors across 25 areas and 31 elected police commissioners – Gov.uk ‘Understand how your council works', https://www.gov.uk/understand-how-your-council-works ]} The publication of consistent data on local elections would mean local elections are reported in a simpler, more efficient and more open way.

Status

Midterm: Limited

The commitment was behind schedule at the end of the first year. Two rounds of consultations led to an agreed common standard for data collection.^{[Note 62: Tim Adams, ‘Consultation No1: Summary of key contributions to the Elections Schema consultation', http://e-sd.org/fmcAY, and Tim Adams, Consultation No2; and Summary of key contributions to the Elections Schema consultation, http://e-sd.org/Rsr9V ]} The different bodies broadly agreed on a publishing approach that already exists. Guidance documents and e-education work has also been done.^{[Note 63: Local Government Association, ‘Local transparency guidance – publishing election results data', http://e-sd.org/vgTJ3; and eLearning modules, Local Government Association ‘Making publishing work for you', http://e-sd.org/zDImh ]} The project team at the Local Government Association (LGA) also manually created and published several other sample datasets and made them available for early access and use. According to the government self-assessment, “the LGA is investigating [whether] sample data can be acquired from a few local authorities”.^{[Note 64: Cabinet Office, ‘Open Government Partnership National Action Plan 2016-18:}

Mid-term Self-Assessment Report' (UK government report September 2017) passed to author pre-publication. ] In November 2017, the Cabinet Office approved the standard's entry into the H.M. Government catalogue of open data standards.

The constant change due to staggered elections in local government has caused some delay, as agreements and new relations needed to be built and re-built if/when new parties came into power. The General Election of June 2017 also slowed the process.^{[Note 65: Interview with Ingrid Koehler, LGIU, 2 August 2017.]} The commitment stalled in October 2016 due to needed updates to election systems. As of March 2018, EMS supplier Democracy Counts confirmed that they planned to implement the new standard into their system at their own cost. ^{[Note 66: Interview with Tim Adams, Local Government Association, 16 August 2017.]} Both the Cabinet Office and LGA were looking into funding.^{[Note 67: Cabinet Office, ‘Open Government Partnership National Action Plan 2016-18:}

Mid-term Self-Assessment Report' (UK government report September 2017) passed to author pre-publication. ] CSOs expressed frustration that the commitment had halted for more than a year after such hard work and agreement.^{[Note 68: Interview with Tim Adams, Local Government Association, 16 August 2017; Interview with Ingrid Koehler, LGIU, 2 August 2017.]}

End of term: Limited

According to the final UK government update ‘progress has continued steadily since the last report' and ‘the LGA and Cabinet Office have been focussing on engaging with Election Management System suppliers to integrate the standard into their systems'.^{[Note 69: UK government (2018), 2016-18 Open Government Action Plan: April 2018 Commitment Progress Updates, https://www.opengovernment.org.uk/resource/2016-18-open-government-action-plan-april-2018-commitment-progress-updates/ ]} Democracy Counts agreed to integrate the standard at their own cost/risk and a sample is currently being worked on.^{[Note 70: Interview with Katie Holder and Thom Townsend, DCMS, 8 August 2018.]} The remaining suppliers have offered quotes for development costs but have indicated they have no capacity to make progress until 2019.

Did It Open Government?

Access to Information: Did Not Change

Civic Participation: Did Not Change

As the full adoption of the standard and publication of the data has not yet happened (and has a longer timeframe than the two-year cycle), it has not resulted in any changes for improving access to information or civic participation. No new information has been published as a result of the commitment so far and there has not been any greater public involvement.

Carried Forward?

This commitment is not being carried forward in future action plans, although the government has suggested other possible commitments around democratic innovation.^{[Note 71: UK Government (2018), Consultation draft of the national action plan for Open Government 2018 – 2020, https://docs.google.com/document/d/1XGUs6X8EHSOm00U-rX2_8cAoq7MnDsBjnetQeW0vnzA/edit#heading=h.y5i6179pcs8d]} Given the longer timeframe, the project will continue past 2019 and into the future.