Mapping the UK election(s) of 2015 - what was done and did it work?

24
BCS - SoC Conference 2015 Mapping T ogether Mapping the UK elec/on(s) of 2015 what was done and did it work? Danny Dorling, Oxford University This talk is on the theme I have looked at now for almost 30 years: cartograms and electoral mapping. However, in this talk I will look at how the UK media chose to portray the general elec/on of May 7th 2015 and, in the event of a hung and irreconcilable parliament, or failing minority government, any subsequent general elec/on. Newspapers, TV, blogs, magazines and other media are likely to approach the mapping of this elec/on in new and novel ways. Apart from anything else they now need to worry about a far wider range of poli/cal colours on the maps. The cons/tuencies being the same as before should help in the mapping of change over /me, if any is aTempted. Interac/ve mapping should also be much improved on previous years, but maybe not? The talk ends with some sugges/ons for the future and will contain only one reference to the "electoral tetrahedron" although one day before I die I hope to see it on TV. I invented it in the 1980s to cope with the SNP becoming a significant party. I was just a liTle before my /me and there is to be a second (but not general) elec3on of 2015….. … which just might need that tetrahedron…

Transcript of Mapping the UK election(s) of 2015 - what was done and did it work?

BCS - SoC Conference 2015 Mapping Together

Mapping  the  UK  elec/on(s)  of  2015  -­‐  what  was  done  and  did  it  work?  

Danny  Dorling,  Oxford  University  This  talk  is  on  the  theme  I  have  looked  at  now  for  almost  30  years:  cartograms  and  electoral  mapping.  However,  in  this  talk  I  will  look  at  how  the  UK  media  chose  to  portray  the  general  elec/on  of  May  7th  2015  and,  in  the  event  of  a  hung  and  irreconcilable  parliament,  or  failing  minority  government,  any  subsequent  general  elec/on.  Newspapers,  TV,  blogs,  magazines  and  other  media  are  likely  to  approach  the  mapping  of  this  elec/on  in  new  and  novel  ways.  Apart  from  anything  else  they  now  need  to  worry  about  a  far  wider  range  of  poli/cal  colours  on  the  maps.  The  cons/tuencies  being  the  same  as  before  should  help  in  the  mapping  of  

change  over  /me,  if  any  is  aTempted.  Interac/ve  mapping  should  also  be  much  improved  on  previous  years,  but  maybe  not?  The  talk  ends  with  some  sugges/ons  for  the  future  and  will  contain  only  one  reference  to  the  "electoral  tetrahedron"  although  one  day  before  I  die  I  hope  to  see  it  on  TV.  I  invented  it  in  the  1980s  to  cope  with  the  SNP  becoming  a  significant  

party.  I  was  just  a  liTle  before  my  /me  J  and  there  is  to  be  a  second  (but  not  general)  elec3on  of  2015…..    

…  which  just  might  need  that  tetrahedron…  

BCS - SoC Conference 2015 Mapping Together

Some  lovely  mapping  before  the  elec/on  

This  map  appeared  in  the  Guardian  in  April  2015.  I  guess  hand-­‐drawn.  Using  poll  data  it  

suggested  Liberals  could  s/ll  win  in  the  SW  of  

England    hTp://www.theguardian.com/poli/cs/ng-­‐  interac/ve/  2015/

apr/20/elec/on-­‐2015-­‐cons/tuency-­‐map  

BCS - SoC Conference 2015 Mapping Together

The  BBC  put  anima/ons  over  Lancashire  

BCS - SoC Conference 2015 Mapping Together

Popula/on  propor/onal  to  volume?  

BCS - SoC Conference 2015 Mapping Together

Sadly  the  flying  graphics  fell  out  of  favour  

BCS - SoC Conference 2015 Mapping Together

James  Jacob  Gilchrist  Berry  held  #13  

BCS - SoC Conference 2015 Mapping Together

BCS - SoC Conference 2015 Mapping Together

The  Telegraph    were  quick  but  crude  with  the  actual  result  Oxford  East  gains  neighbour…  

BCS - SoC Conference 2015 Mapping Together

Actual  Result  April  

Poll  

BCS - SoC Conference 2015 Mapping Together

The  Observer  was  garish  

But  the  polls  were  not  

that  far  out  when  it  

came  to  the  map  

BCS - SoC Conference 2015 Mapping Together

Ben  Hennig  started  mapping  while  on  holiday  (Lancashire  again)  

BCS - SoC Conference 2015 Mapping Together

The  Guardian  started  mapping  vote  

shares.    

Oxford  East  is  a  rabbit?  

BCS - SoC Conference 2015 Mapping Together

Ben  mapping  from  the  air  

Using  an  algorithm  first  generated  to  show  US  elec/on  maps  and  used  by  Mark  Newman  and  

colleagues  

BCS - SoC Conference 2015 Mapping Together

BCS - SoC Conference 2015 Mapping Together

US  elec/on  winners    (democrat  =  blue,  republican  =  red)  

BCS - SoC Conference 2015 Mapping Together

Or  show  vote  shares  with  many  shades  of  purple  

BCS - SoC Conference 2015 Mapping Together

To  end  with…  Ben’s  

maps:  to  begin  –  who  came  

2nd?  

A  quick  key  map…  

BCS - SoC Conference 2015 Mapping Together

The  UK  is  looking  more  like  the  USA.  Just  25%  of  the  electorate  can  get  a  majority.  

BCS - SoC Conference 2015 Mapping Together

But  unlike  the  US  we  no  longer  have  two  major  par/es  even  though  the  third  was  decimated  

BCS - SoC Conference 2015 Mapping Together

Compare  the  Liberals  To  UKIP.  

Libs  do  well  near  to  where  UKIP  poll  low  

BCS - SoC Conference 2015 Mapping Together

The  SNP  have  most  support  in  ci/es.  

 Plaid  is  more  rural  

BCS - SoC Conference 2015 Mapping Together

The  cartograms  of  change  may  tell  you  most  you  

need  to  know.    

 Slowly  spa/al  divisions  are  splilng  the  UK  apart  

ever  wider…    

BCS - SoC Conference 2015 Mapping Together

Poverty  and  wealth  remain  

the  best  predictors  of  who  wins  where…  

BCS - SoC Conference 2015 Mapping Together

The Segregation Index of Conservative voters in Britain*, 1885–2015

Election Concentration

1885 7.11%1886 5.53%1892 5.81%1895 4.70%1900 4.39%1906 6.67%

1910 Dec 6.24%1910 Jan 7.91%1918 19.30%1922 14.44%1923 11.57%1924 10.62%1929 9.24%1931 9.23%1935 9.65%1945 7.21%1950 6.74%

1951 6.77%1955 6.93%1959 6.24%1964 6.51%1966 7.69%1970 8.04%1974 Feb 8.01%1974 Oct 10.72%1979 9.17%1983 10.59%1987 11.84%1992 11.88%1997 13.94%2001 15.05%2005 15.69%20102015

16.40%19.89%

Election Concentration

*Northern Irish seats are not included.

0%

6%

8%

12%

18%

20%

4%

2%

10%

16%

14%

1920 1930 1940 1950 1960 1970 1980 1990 2000 2010

Source: Drawn initially in Dorling, D. (2006) ’Class alignment renewal’: The Journal of Labour Politics, vol 41, no 1, p 849, showing the spatial segregation index. Updated in Dorling, D. (2013) Crises and turning points: the pivots of history, Renewal, 21, 4, pp11-20.

Note: The statistic being measured is the segregation index of Conservative votes across all British seats at each general election. The proportion is the minimum number of voters who would have to be moved across constituency boundaries to ensure that within each parliamentary constituency the Conservatives received exactly the same share of the vote.

Figure 13: Concentration of Conservative votes, British general elections, 1918–2015Conclusion  

With  the  first  elec/on  of  2015  one  simple  graph  might  tell  us  more  -­‐à  

The  second  elec/on  result  is  on  September  

12th  

It  needs  a  tetrahedron!  

Thank  you!