Longitudinal data provides novel insights into voting patterns

Scoop.it ShareThis


Moving to a new house can have a profound effect on your friendships and relationships. But it can also affect your political views too – if you move to a Tory area, that is.

A study analysing the political beliefs of the 10% of Britons who move home each year has found that you are more likely to become economically rightwing if you move to an area regarded as a safe Conservative seat.

Based on data from close to 10,000 respondents moving home between 1991 and 2008, one of the the study’s authors, Patrick Sturgis of the University of Southampton, analysed people’s political preferences five years up to the move and five years after it.

The report found that:

  • Movers to safe Conservative seats became more economically rightwing and more likely to vote Conservative after the move whatever their political views before.
  • The longer someone lives in a Conservative constituency, the more likely an individual is to support the Conservatives. However, there is no equivalent effect on those moving to Labour seats and staying there.

The paper by Aina Gallego, Franz Buscha, Patrick Sturgis and Daniel Oberski, Places and Preferences: A Longitudinal Analysis of Self-Selection and Contextual Effects, confirmed what research methodologists have long known:

“It has long been recognised that cross-sectional studies do not provide convincing evidence of contextual effects because it fails to adequately distinguish contextual influence from self-selection of individuals into areas,”
says Patrick Sturgis.

“The longitudinal approach safeguards against the strongest criticism of cross sectional designs, which is that people choose which areas they wish to move to and that these choices are themselves related to pre-existing political preferences. Overall, contextual effects are found to be weak and dominated by the larger effect of non-random selection into areas.

“The combination of appropriate data and methods enables us to produce robust estimates of contextual effects and has generated novel insights – both for the research community and for the wider policy arena.”

Related links

this news has been re-posted from the Understanding Society website