Kelvyn Jones

  • Permalink Gallery

    General election polling goes geographical: the accuracy and value of constituency-level estimates

General election polling goes geographical: the accuracy and value of constituency-level estimates

The 2017 general election saw a largely unremarked geographical extension to opinion polling, with three analysts publishing estimates of which party was likely to win in each of the country’s constituencies. Ron Johnston, Kelvyn Jones, David Manley, Charles Pattie, Todd Hartman, and David Rossiter have analysed their accuracy and considered the implications of that development for the conduct of […]

  • Permalink Gallery

    Predicting the Brexit vote: getting the geography right (more or less)

Predicting the Brexit vote: getting the geography right (more or less)

In April 2016 in two contributions to this blog Ron Johnston, Kelvyn Jones and David Manley predicted the likely geography of support for Brexit in the EU referendum. In this concluding piece they compare their predictions to the result. The general pattern of their predictions turned out to be very accurate, but regional differences were more pronounced than anticipated, with variations in both […]

  • Permalink Gallery

    The geography of the Brexit vote – what difference will turnout make?

The geography of the Brexit vote – what difference will turnout make?

In their recent analysis, Ron Johnston, Kelvyn Jones and David Manley used a large body of YouGov polling data to explore which social groups are most likely to vote Leave in the upcoming EU Referendum and where they live, producing a clear geography of support for Brexit. Here, they explore how differential turnout rates across those groups could alter […]

  • Permalink Gallery

    Can we really not predict who will vote for Brexit, and where?

Can we really not predict who will vote for Brexit, and where?

In a recent Guardian article, Simon Jenkins suggested that voter decisions regarding the EU referendum will be made on the basis of gut instinct alone, and that personal characteristics and previous party support provide no guide. Using a new modelling strategy applied to a large body of YouGov opinion poll data, Ron Johnston, Kelvyn Jones and David Manley address Jenkins’ claim, and find it wanting. The young and the well-qualified […]

  • Permalink Gallery

    Statistically flawed ‘evidence’ has been used to scrap AS-levels

Statistically flawed ‘evidence’ has been used to scrap AS-levels

Michael Gove and David Laws have changed the examination system for post-16 students at English schools and colleges, and in doing so have altered the basis on which universities can make their conditional offers to applicants. To justify that change, they commissioned in-house research – published in May 2013. That research appears to be statistically flawed, however; Ron Johnston, […]