Disciplinary differences in preferred research methods: a comparison of groups in the Biglan classification scheme

Abstract

The Biglan system of classifying disciplines in groups based on similarities and differences in their subject matter has been validated in numerous empirical studies. The present study sought to expand that validation by comparing two Biglan groups that include disciplines representing the social and behavioral sciences. As a unique point of comparison, preferred research methods were contrasted between Psychology and Sociology forming one group of pure disciplines and Education and Nursing forming another group of applied disciplines. A code sheet was developed to categorize the various components of published research, distinguishing at the most basic level quantitative, qualitative, and mixed methods type articles. Data was collected from research articles published in high impact journals from each discipline. The complete data set consisted of the codes from 150 randomly selected articles from journals in each discipline, or 300 articles per group. These codes were converted into frequencies and analyzed using the Chi-Square statistic. Findings showed there are significant differences in preferences for quantitative, qualitative, and mixed methods research approaches between the two Biglan groups. Significant differences were also found in certain methodological components of the basic research approaches including research designs, sampling methods, and data collection methods. Evidence was also found of the philosophical paradigms underlying methodological choices, and analysis revealed significant differences in the paradigms preferred by the two groups. All of these findings support the validity of the Biglan scheme of classifying disciplines based on differences in the preferred approach to research methodology. These findings also suggest that the Biglan system represents a useful tool for promoting interdisciplinary discourse on research. Directions for future research are indicated that would further confirm the findings of the present study, move towards a wider validation of the Biglan system, and explore more deeply the philosophical underpinnings of the paradigmatic differences founding diverse research methodologies.