Additional logistic regression tables for the Constitutional reform chapter
Independent variables (on each, values 8-9 set to missing)
RAgeCat recoded into 6pt scale (1=18-24, 2=25-34, 3=35-44, 4=45-54, 5=55-64, 6=65+)
HedQual recoded into 6pt scale (1=Degree, 2=HE below degree, 3=A level or equivalent, 4=O level or equivalent, 5=CSE or equivalent, 6=None/Foreign)
Political interest (Politics1):
Politics copied into 5pt scale (1=Great deal of interest, 2=Quite a lot, 3=Some, 4=Not very much, 5=No interest at all)
Party ID (PartyID):
PartyID2 recoded into 5pt scale (1=Conservative, 2=Labour, 3=Liberal Democrat, 4=Other/Green, 5=None/DK)
Summed scale (TrstParl, MP, Gov) recoded into 5pt scale (1-1.67 = 1 (High trust), 1.6% (N=35), 2-2.67 = 2, 3-3.67 = 3, 4-4.67 = 4, 5- = 5 (Low trust) 7.5% (N=161))
- Download chapter
- This picture of low trust in politicians relative to other professional actors, such as the police, is confirmed by data from other polling organisations, such as MORI's 'Trust in Professions' surveys (Ipsos-MORI, 2011).
- Readings are indicated by data marker; the line indicates an overall pattern but where there is no data marker the line cannot be taken as a reading for that year.
- These figures have been collated from various House of Commons Library papers supplemented by data from the New Local Government Network.
- Nor have attitudes to elected mayors improved since their introduction in 2000. The same questions about speaking up for the area, getting things done and giving too much power to a single person were also asked on British Social Attitudes in 1998 and 2000. The proportions agreeing that mayors speak up for the area and help gets things done were no higher in 2011 than in 1998, while the proportion agreeing that mayors give too much power to a single person fell by only 10 percentage points, from 45 per cent in 1998 to 35 per cent in 2011.
- Two other of England's largest cities, Leicester and Liverpool, had previously decided to introduce a directly elected mayor without holding a referendum. In four other referendums on directly elected mayors held since the 2010 general election, the proposal was approved in two cases (Salford and Tower Hamlets) and rejected in a third (Great Yarmouth). Doncaster voted in May 2012 to keep its elected mayor.
- The Coalition's proposal is that a referendum should be held when a council wishes to increase the level of council tax by more than a limit specified by the government. To simplify matters, we couched this as referring to an above inflation increase. Note though that voters are not necessarily keen that decisions about the council tax should routinely be referred to them. Only 43 per cent say that decisions about the council tax should be made by voters in a referendum, while 52 per cent would prefer the decisions to be made by their elected council. It would appear that, while voters are happy to have a referendum as a potential bulwark against a particularly large increase in council tax, they are not sure they trust their fellow citizens to make decisions about the tax on a regular basis.
- Strictly speaking this provision would apply to custodial sentences of 12 months or less, as longer sentences already result in automatic disqualification from membership of the Commons.
- The full question wording was:
It has been suggested that sometimes voters should be able to force their local MP to resign and fight a by-election. First of all, say that the MP has broken the rules. How much do you agree or disagree that in those circumstances voters should be able to force their MP to resign?
Respondents were invited to answer using a five point scale ranging from "strongly agree" to "strongly disagree".
- The full question wording was:
And what if the MP had not broken any rules, but voters thought he or she was not doing a very good job? Should voters be able to force their MP to resign?
Again respondents were invited to answer using a five point scale.
- The scale was created by adding the scores (ranging from 1 to 4) across the three items and dividing the resulting total by three. Multi-item measures of complex concepts like trust are usually held to be more reliable and valid than single item measures (Zeller and Carmines, 1980: 48-52; Heath and Martin, 1997). Cronbach's alpha for this particular scale is 0.90.