The transformation of public and private activities and qualities into numerical data, enabling the tracking and prediction of a variety of phenomena and processes, has been greatly enhanced by the latest digital forms of quantification, data mining and big data analytics. The proliferation of quantification and datafication has not been free of controversy; it has been attributed to the expansion of a neoliberal form of governance that ultimately eliminates genuine democratic deliberation. However, long before the digitisation of communication and the internet, the quantification of public opinion through opinion polls in the 1930s demonstrated the important societal effects of quantification in governance and research. Both earlier and more recent debates on the ambivalence of public opinion polls and their important implications, which would either promote or hinder the development of democracy, contribute to the identification and interpretation of wider societal opportunities and threats to publicness arising from quantification and datafication.

Four decades after Bryce expressed the idea of a ‘higher level of public opinion development’ in which ‘the will of the majority could be determined at any time without having to pass it on through a representative body and perhaps even without the need to use the electoral machinery’ (Bryce 1888/1995, 919), his idea came true in opinion polls on representative samples. Bryce was pessimistic at the time, indicating that such a development was completely utopian, mainly because of technical difficulties, as ‘machines for weighing or measuring the will of the people from week to week or from month to month have not yet been and probably never will be invented’ (ibid.). However, history soon proved that his pessimism was unfounded.

Just a few years after the vibrant disputes over public opinion in the early twentieth century in Europe and the United States, discussions have, oddly enough, almost completely died down. This did not happen because the position of the round table of American political scientists would have been generally approved or because a conclusive theoretical definition would have been reached unexpectedly but because of the development and use of methodological and statistical tools in the empirical research of individual and collective attitudes – opinion polling.