Teaching: the role of active manipulation of three-dimensional scatter plots in understanding the concept of confounding

Table 2 Results of evaluation with students

	Mean score (% with a score of 4 or 5)
Evaluation question*	BSc course (n = 100)	MSc course (n = 44)	International PhD course (n = 19)
1. The 3D plots help me to understand confounding.	3.6 (60)	4.4 (92)	4.2 (89)
2. It was useful to work with the 3D plots in addition to the lectures and textbook.	3.7 (68)	-^†	-^†
3. I enjoyed studying confounding using the 3D plots.	3.4 (53)	4.6 (100)	4.7 (100)
4. Active handling the 3D plots helps me to understand confounding.	3.5 (52)	4.5 (100)	4.2 (100)
5. The self-tests were useful.	- ^‡	4.6 (100)	- ^‡
6. Overall rating of the 3D plots (1 = poor to 5 = excellent).	3.7 (64)	4.5 (100)	4.2 (95)

*All questions were Disagree – Agree questions with a five-point Likert scale. As indicated an average score of 3 is considered satisfactory while an average higher than 4 is considered excellent.
^† In the MSc and PhD course this question was not included on the evaluation form because there was no additional learning material provided about confounding.
^‡ Self-tests were only available in the MSc and PhD course.