On Statistical Methods in Higher Education Studies:

Bibliographic Information

Other Title
  • 高等教育研究と計量分析
  • コウトウ キョウイク ケンキュウ ト ケイリョウ ブンセキ
  • From the perspective of a boundary around the Japan Society of Educational Sociology

Search this article

Abstract

<p>The purpose of this paper is to describe the quantitative methods used in studies of higher education in the field of the sociology of education. We set two research questions: one, “Does higher education research share the theoretical and methodological frameworks of the sociology of education?”, and, two, “What is the unique issue in higher education studies that use statistical methods? In addition, how do we tackle this issue in the future while also referring to other academic fields?”<br><br>The result of our analysis of articles in academic journals shows that, in terms of method, the “generalized linear model” is the most used model, and that, in terms of content, “the systems and policies” of higher education is the most common research topic. At the same time, articles that use a statistical method in the Journal of Educational Sociology” address “Going to College” but do not address “Deviant,” “Learning,” and “Management”. It seems that this is the boundary around the statistical articles in the Journal of Educational Sociology. On the other hand, from the citations, it is clear that these articles not only tend to cite themselves but also cite articles in other journals of sociology. Thus, our answer to the first question is that some points are shared to some extent.<br><br>We should now focus more closely on trends in the relevant academic fields besides those that are apparent in the Japan Society of Educational Sociology. We need to learn from these fields because EBPM practice raises many challenges, such as that of the raison d’etre of academic professions as experts, the refinement of methods from EBPM, etc., especially in higher education research, which is close to policy-making. Also, we can derive fruitful suggestions from the experiences of the Japanese Society of Educational Sociology, which has an “ecosystem”-an arena for discussion on methodology. Therefore, in answer to the second question, EBPM requires the refinement of statistical methods, but beyond that there is a need to discuss causal reasoning accurately. We always have to reflect on how to handle causal reasoning in any quantitative analysis that refers to discussions in related fields such as psychology, economics, and information science as well as the sociology (of education), and bear in mind the development of the causal models produced by Rubin, Pearl, and Campbell in those fields.</p>

Journal

References(10)*help

See more

Related Projects

See more

Details

Report a problem

Back to top