Littler Britain: the 14 council areas with falling populations

The Office for National Statistics has just published its annual population estimates for mid-2018. I have updated a map I first produced two years ago, showing how each area’s population has changed between 2008 and 2018. As well as the colour-coding, you can see the percentage change by hovering over an area. The data is derived from sheet MYE5 of this spreadsheet.

There are 14 council areas whose populations have fallen over the decade (shown in blue), while the UK’s population as whole grew by 7.5%. Most of them are rural and coastal areas: in England they include Cumbria, the Isles of Scilly and (an exception to the rural bit) Blackpool; in Wales, Ceredigion, Powys and Blaenau Gwent. In Scotland, seven council areas have fallen in population, mostly on the west coast: Argyll and Bute, Dumfries and Galloway, Na h-Eileanan Siar (the Western Isles), Inverclyde, North Ayrshire, South Ayrshire and West Dunbartonshire.

Continue reading “Littler Britain: the 14 council areas with falling populations”

Ministry fails to do justice to data protection

When I pitched a data journalism project to PublicTechnology.net and Civil Service World on government departments and personal data breaches, I didn’t expect the biggest perpetrator to be the Ministry of Justice, or that its 3,184 incidents in 2017-18 would be 27 times the number of the second-placed Ministry of Defence.

The ministry has two reasonable arguments for such high numbers. Firstly, it reports every data breach while some departments do not. Secondly, it directly runs the justice system in England and Wales through HM Courts and Tribunal Service, which was responsible for 70% of its breaches. If the Department of Health and Social Care ran health and social care directly, it would have a much higher figure too. Continue reading “Ministry fails to do justice to data protection”

Data in journalism training at the NUJ on 17 May

I am again running an introductory day on using data in journalism at the National Union of Journalists in London on Friday 17 May. The course, which assumes little or no knowledge of the subject, covers risk, quality of data, assessing sources including surveys, government and open data, Freedom of Information and graphing.

The course, organised by the NUJ’s London Freelance Branch, costs £55 for NUJ members, £45 for student and temporary members and £110 otherwise.

More information and booking here.

Reinventing higher education: ‘Can we use AI to give the lecturer superpowers?’

Article by SA Mathieson, Guardian Labs, (Transforming the student experience series paid for by Jisc), 9 April 2019

Technology sometimes makes people’s lives more impersonal. But in higher education it is leading in the other direction, as large group lectures are replaced with tutorials, digital resources and software-based coaching.

“We believe the technologies that are coming will reinvent higher education teaching,” says Paul Feldman, chief executive of Jisc, a membership organisation that provides digital solutions for UK education and research. “We think it will turn the whole thing on its head.” Continue reading “Reinventing higher education: ‘Can we use AI to give the lecturer superpowers?’”

Wikipedia gets it wrong even as a source of sources

I am currently teaching a term-long data journalism course at Birkbeck, University of London for MA students. Aside from seeing their overall projects develop, it is fun to see how students respond to challenges, such as to find data visualisations. (They came up with fine examples on gender pay gaps, China’s Uighur prison camps and trade discrepancies.)

One particularly interesting set of responses came when I asked students to find something wrong on Wikipedia (without editing in their own errors). Several found basic factual mistakes in pages on their home towns and other things they know well. Having said that, in data journalism courses I always advise using Wikipedia as a source of sources, to follow links and footnotes to good primary material. Continue reading “Wikipedia gets it wrong even as a source of sources”