Risk Society

.....Whether or not the controversial idea of a ‘risk society’ is theoretically coherent or accurate as a historical generalization is much debated, but there has undoubtedly been an avalanche of discussion and literature on risk, hazard, and blame in recent times, and that phenomenon needs some explanation. As well as a ‘risk society’, we are also said to live in a ‘regulatory state’. The idea of the ‘regulatory state’ is that a new institutional and policy style has emerged, in which government's role as regulator advances while its role as a direct employer or property-owner may decline through privatization and bureaucratic downsizing. The two ideas of ‘risk society’ and ‘regulatory state’ could, indeed, be linked in so far as risk and safety is often held to be one of the major drivers of contemporary regulatory growth, for example in the development of EU regulation (see Royal Society 1992). In turn, development of risk regulation is interpreted by many to reflect broader political and cultural change. Building on a ‘grid-group’ analysis of culture that highlights a dynamic of conflict among four fundamentally different sets of beliefs and attitudes, risk is seen as a political weapon used by a society poised between the cultures of individualism and egalitarianism, to blame those who wield power in the state and big corporations for what happens to the rest of us. From this perspective the increased salience of risk and regulation reflects a cultural shift away from ‘hierarchist’ world-views over matters of trust and blame.

Christopher Hood (All Souls College, Oxford), 2003, The Government of Risk:Understanding Risk Regulation Regimes; OUP



"What then is a legislature for? Parliament has multiple roles: to sustain the executive; to scrutinise executive actions; to represent territory; to represent shades of opinion. As before, these objectives are not wholly compatible, and writers with one conception of parliament may deny that a role that according to another conception is central is legitimate at all."

Evidence to the Commission on Electoral Systems
By Iain McLean, Professor of Politics, Oxford University





Selection Bias Modelling

Selection arises in non-randomized studies when individuals self-select into or are non-randomly allocated to a particular group. Selection creates statistical problems when it is based on unobserved individual characteristics. Moreover, selection on the basis of unobserved individual characteristics is very common. Examples include students with high unobserved ability being the most likely to go to university and workers with high unobserved productivity having relatively high probabilities of participating in the labour market. Failure to account for selection leads to biased estimates of the causal effects of explanatory variables on the outcome of interest. Indeed, the estimated effect of education on wages is biased upwards if no account is taken of the higher average ability of individuals with more education.

http://springschool.politics.ox.ac.uk/courses/selection_bias.asp