Official statistic-making as a social practice : the UK 'Measuring National Well-being' programme

Abstract

The thesis examines the construction of official statistics using the qualitative case study of the UK Office for National Statistics' 'Measuring National Well-being' programme. Its major original contribution is to critically engage with the making of official statistics, theorising their construction as a social process. This provides novel ways of explaining the form and content of official statistics. It also furthers debates on 'wellbeing' through an examination of the concept's theoretical and institutional history. The research argues that official statistic-making is an activity conducted by actors with agency. This agency is not taken into account by existing accounts of the nature of official statistics, which are more abstract and which do not engage with the statisticmaking process. It is argued that attention to the social processes of official statisticmaking make a fuller understanding of the form and content of statistics possible. This argument is supported by an examination of the 'Measuring National Well-being' programme. The institutional and theoretical context for this programme presents a number of challenges to existing accounts, highlighting the need for agential action. This context is explored through semi-structured interviews with those involved in the creation of the programme, triangulated against secondary material such as meeting minutes produced by the Office for National Statistics and published statistical outputs. In the specific case of the 'Measuring National Well-being' programme, the research finds a high degree of autonomy for those constructing the programme. This freedom was used to position the programme within wider European and international statistical contexts, rather than domestic political ones. Despite being a programme in a new area of social research, the final content of the programme fitted within established ways of thinking partly as a result of path dependency created by the statisticians' preestablished ways of working. More generally, it is argued that approaching the construction of official statistics as a social process helps explain the form and content of the statistic. The research demonstrates that it is possible to trace linkages between the features of the final statistic and social interactions which gave rise to them. It is argued that this is applicable beyond the case study used and beyond statistic-making in the UK.