Ragnar Karlsson, Education and cultural statistics, Statistics Iceland. Ragnar.Karlsson@statice.is

Until more recently culture statistics were neither high on the agenda of National Statistical Institutes nor in international statistical comparison. Statistics about culture were mostly confined to rather modest documentation of supply and national governments spending. Otherwise cultural activities were reputed to be not amendable to statistics. This negligence has recently been reversed and measuring culture has become almost an `industry¿. This shift has much ado with increasing commodification and commercialization of culture and increased international traffic and trade in cultural artefacts and meaning. It is the firm belief of policy makers that cultural industries are to become more and more important part of future economies as the key new growth sector of the economy, both nationally and globally, and the key source of future employment growth and export earnings. Hence, culture is frequently claimed to be the most visible manifestation of globalization.

Quantitative cross-national comparison of culture is difficult, even between countries of close cultural proximity. It is extremely rare for cultural data to be gathered in exactly the same way in different countries, because from differences in institutional arrangements in the cultural policy arena, in definitions, classification and methodologies. Empirical measures fall often short as data are spotty and often inconsistent. Notwithstanding rising standardisation in cultural statistics exemplified recently with UNESCO¿s revised framework for cultural statistics and initiatives on behalf of international bodies like the EU and OECD to produce more robust indicators and tools for a harmonized measure of cultural activities, there are inherent difficulties in quantifying culture. This paper highlights some of the challenges, problems and pitfalls associated with quantitative cross-national comparison of culture by citing examples from Nordic, European and international culture statistics. It is argued that the trap of cross-national comparison by consensus of harmonized indicators and compromises in definitions through active avoidance of the epistemological issues at the heart of comparative research is prone to the danger of producing measurement out of context.