Some Notes on the Obsession of the Czech Social Sciences with Statistical Significance

The article explains the various errors that occur in the use of the concept of statistical significance. It points to the problem of census, nonprobability sampling, sampling of small populations and small samples. Another topic is the use of statistical methods on aggregated data files, especially from international research, and on weighted data. The authors point out that in many cases the use of statistical significance is not appropriate, and they warn against the incorrect use of traditional statistical methods. The article also presents methods that can be used to avoid the problems to which the authors have drawn attention. Reprinted by permission of the Institute of Sociology of the Academy of Sciences of the Czech Republic