Abstract:
The article deals with the problem of using modeling in social statistics
courses. It allows the student-researcher to build one-dimensional and
multidimensional models of the phenomena and processes that are being studied.
Social Statistics course programs from foreign universities (University of
Arkansas; Athabasca University; HSE University, Russia; McMaster University,
Canada) are analyzed. The article provides an example using the education data
set – Guardian UK universities ranking in Social Statistics course. Examples of
research questions are given, data analysis for these questions is performed
(correlation, hypothesis testing, discriminant analysis). During the research the
discriminant model with group variable – modified Guardian score – and 9
predictors: course satisfaction, teaching quality, feedback, staff-student ratio,
money spent on each student and other) was built. Lower student’s satisfaction
with feedback was found to be significantly different from the satisfaction with
teaching. The article notes the modeling and statistical analysis should be
accompanied by a meaningful interpretation of the results. In this example, we
discussed the essence of university ratings, the purpose of Guardian rating, the
operationalization and measurement of such concepts as satisfaction with
teaching, feedback; ways to use statistics in education, data sources etc. with
students. Ways of using this education data in group and individual work of
students are suggested.
