26 May 2010

Last Research Assesment Results Analysed - Russell group vs. Non-Russell

Using figures from the last RAE, one can estimate the number of academics judged to be of each kind: "world-class" (4*), "international" (3*), etc.) in each RAE unit of assessment in each university.  Using this we can see that 54% of the highest category are in the Russell group university and 46% of the "international" standard.  The rounded figures are as follows.
 So roughly half of the top academics are not in Russell group universities.  Also it shows that the Russell group entered a lot of academics who were judged to be in the lower categories.  Below is a histogram of the number of 4* academics in all universities.
There is a "long tail" of non-Russell universities with some "World Class" academics.  Thus, although many of the best academics are concentrated in Russell group univsersites, there are many spread around in a variety of universities.

25 May 2010

Votes in UK General Elections 1945-2010

Out of interest I plotted the raw number of votes cast for Conservative, Liberal/Lib-Dem, and Labour parties in each general election from 1945-2010, using Wikipedia as the source of data.

Its interesting that For most of the time Labour and Liberal(-Dem) are anti correllated - a gain for one is a loss for the other.  The Liberals have generally failed to attract conservative votes.  The Conservative vote has been fairly constant, except immediately after WWII and after the scandals in the early 1990's after which many of the Conservative voters simply did not vote (or voted for a fringe party).

Raw figures are:
        year      Labour    Conservative     Liberal 
1945 11,967,746 8,716,211 2,177,938
1950 13,226,176 11,507,061 2,621,487
1951 13,948,385 13,724,418 730,546
1955 12,405,254 13,310,891 722,402
1959 12,216,172 13,750,875 1,640,760
1964 12,205,808 12,002,642 3,099,283
1966 13,096,629 11,418,455 2,327,457
1970 12,208,758 13,145,123 2,117,035
1974 11,645,616 11,872,180 6,059,519
1974.5 11,457,079 10,462,565 5,346,704
1979 11,532,218 13,697,923 4,313,804
1983 8,456,934 13,012,316 7,780,949
1987 10,029,270 13,760,935 7,341,651
1992 11,560,484 14,093,007 5,999,384
1997 13,518,167 9,600,943 5,242,947
2001 10,724,953 8,357,615 4,814,321
2005 9,562,122 8,772,598 5,981,874
2010 8,604,358 10,683,787 6,827,938

12 May 2010

2 Science papers on social dilemma's and punishment

Lab Experiments for the Study of Social-Ecological Systems
Marco A. Janssen,1,* Robert Holahan,2 Allen Lee,1 Elinor Ostrom1,2
Science 30 April 2010:
Vol. 328. no. 5978, pp. 613 - 617
DOI: 10.1126/science.1183532
Governance of social-ecological systems is a major policy problem of the contemporary era. Field studies of fisheries, forests, and pastoral and water resources have identified many variables that influence the outcomes of governance efforts. We introduce an experimental environment that involves spatial and temporal resource dynamics in order to capture these two critical variables identified in field research. Previous behavioral experiments of commons dilemmas have found that people are willing to engage in costly punishment, frequently generating increases in gross benefits, contrary to game-theoretical predictions based on a static pay-off function. Results in our experimental environment find that costly punishment is again used but lacks a gross positive effect on resource harvesting unless combined with communication. These findings illustrate the importance of careful generalization from the laboratory to the world of policy.
http://www.sciencemag.org/cgi/content/abstract/328/5978/613

Coordinated Punishment of Defectors Sustains Cooperation and Can Proliferate When Rare
Robert Boyd,1,2,* Herbert Gintis,2,3,4,* Samuel Bowles2,5,*
Science 30 April 2010:
Vol. 328. no. 5978, pp. 617 - 620
DOI: 10.1126/science.1183665
Governance of social-ecological systems is a major policy problem of the contemporary era. Field studies of fisheries, forests, and pastoral and water resources have identified many variables that influence the outcomes of governance efforts. We introduce an experimental environment that involves spatial and temporal resource dynamics in order to capture these two critical variables identified in field research. Previous behavioral experiments of commons dilemmas have found that people are willing to engage in costly punishment, frequently generating increases in gross benefits, contrary to game-theoretical predictions based on a static pay-off function. Results in our experimental environment find that costly punishment is again used but lacks a gross positive effect on resource harvesting unless combined with communication. These findings illustrate the importance of careful generalization from the laboratory to the world of policy.
http://www.sciencemag.org/cgi/content/abstract/328/5978/617