National Student Survey: Reliability and prediction of overall satisfaction scores with special reference to UK Dental Schools

Introduction The aim of this study was to assess the reliability of the UK National Student Survey (NSS) (2018) and explore the predictive values of individual questions and question domains on the overall satisfaction score. Methods UK NSS result data (for 2018) were assessed for all those providers that award an undergraduate degree in dentistry. To analyse the correlation of each question item to the overall satisfaction question (question 27), a bivariate 2-tailed Pearson correlation test was conducted, where a P value of Question 5 ("My course has provided me with opportunities to explore ideas or concepts in depth") had the highest predictive value, explaining 85% of the correlation data to the overall satisfaction question. The strongest positive correlation could be seen at the "teaching on my course" and "learning opportunities" domains, whereby 84% of the data can be explained by each of these variables. The worst predictor domain was assessment and feedback, whereby only 40% of the data could be explained by the correlation line. Cronbach's alpha was 0.952-0.976. Conclusions The NSS is a highly reliable test, both at question and domain levels. The highest predictors of overall satisfaction are teaching, learning opportunities and academic support domains. Assessment and feedback has weak predictive value towards the overall satisfaction score. Question 5 is highly predictive of the overall outcome.