Evaluation of statistical methods applied in theses and dissertations in an Open, Distance and e-Learning University

Area of Science:

• Agricultural and Environmental Sciences
• Research Methodology
• Statistical Analysis

Background:

• Appropriate research methods and statistical analyses are crucial for scientific study quality.
• Selecting statistical methods must align with study data and research objectives to avoid errors.
• Common statistical errors in postgraduate research can compromise study validity.

Purpose of the Study:

• To evaluate the appropriateness of statistical techniques in master's and doctoral theses.
• To identify common statistical errors in planning surveys/experiments and data analysis.
• To assess statistical application in agricultural and environmental sciences theses.

Main Methods:

• Review of 139 master's theses and doctoral dissertations (2015-2020).
• Focus on agricultural and environmental sciences disciplines at an African e-learning university.
• Analysis of mixed and quantitative research methods employed in the studies.

Main Results:

• Analysis of Variance (ANOVA), student t-test, and Chi-square were most frequent statistical tests.
• 41.0% of theses/dissertations exhibited at least one significant methodological error.
• Common errors included inappropriate sampling, data conversion for regression models, and incorrect modeling of correlated data.

Conclusions:

• Significant statistical and methodological errors are prevalent in postgraduate theses.
• Findings underscore the need for enhanced statistical training for postgraduate students.
• Recommendations for university management to develop targeted statistical methods training programs.