Error in Demographic and Other Quantitative Data and Analyses

Thomas Burch

Published: June 2018

Statistical data consumed, analysed and produced contain errors from more sources than is often recognised and the commercialisation of survey and other statistical research and ‘inventions’ such as ‘big data’ has led to naïve and faulty analysis and propaganda. Oskar Morgenstern has noted that, in contrast to physics, there is no estimate of statistical error within economics and the various sources of error that come into play in the social sciences suggest that the error in economic observations is substantial. It is important to recognise the phenomenon of the propagation of errors; errors in our results may be disproportionate to errors in our input data. Despite documented problems social scientists cannot give up on quantitative data since many of the most important questions in social science are matters of more or less, not either/or.

Download Paper in PDF format