Methods behaving differently: The effects of method of data | 17173
International Research Journals

Educational Research

All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Methods behaving differently: The effects of method of data analysis on understanding student satisfaction with their educational experience


Walid El Ansari and Reza Oskrochi

This is a large-scale survey (138 modules, 2650 questionnaires) into student satisfaction at the School of Health and Social Care of a British University in the UK. We assessed the extent and influence of the statistical method employed in the analysis of satisfaction data on the actual understanding of student satisfaction. Satisfaction was computed by four commonly used statistical methods of analyses of satisfaction data selected through literature review. After scrutiny, one was chosen to act as the ‘preferred method’. Each method was used to individually analyse the dataset twice: once without controlling for the effects of variables and clusters that were under investigation; and again after controlling for such effects. Findings of the analyses were compared to those of the ‘preferred method’. After controlling, some initially observed effects of some variables on satisfaction were subsequently lost. Compared to the findings of the ‘preferred method’, different methods exhibited over- or underestimation of satisfaction. Satisfied students were post-qualifying students who attended term one, academic level 2 modules in smaller classes (in terms of student numbers) and whose assessments were 50/50 combinations of coursework/ exams.

Share this article