Teaching statistics to social science students demands an in-depth look at the assumptions needed to design a valid study using statistical inferential methods. The hidden assumptions relate to problems in measurements in the social science research that are not addressed in the usual study in the physical sciences. These include measures of non-preciseness and fuzzy data, measurements of low reliability and validity, use of Bayesian statistics. Although even modern textbooks usually do not mention these problems, discussions on these topics can be found on the Internet. Course in research design thus are heavily depended on actual development of the mentioned difficulties on the Internet. There is also a need to discuss assumptions which are used for statistical inference, social science research is probably not as robust as physical science claims to be. The usual non-randomness of the sample seems to be the most difficult problem to deal with; that is where Bayesian statistics might be enabling a solution.
|Keywords:||Proximate Measurements, Non-Precise Data, Fuzzy Logic, Bayesian Statistics|
Associate Professor, University of Calgary, University of Calgary, Calgary, Alberta, Canada
There are currently no reviews of this product.Write a Review