TABLE 3.

Student agreement with statements from exploratory survey.

Statement responses were strongly disagree (0), disagree (1), neutral (2), agree (3), and strongly agree (4)	Pre-score		Post-score
	Mean	SD	Mean	SD	Wilcoxon W	P
1. Eating foods or supplements with certain types of bacteria can have a positive impact on human health.	3.15	0.795	3.27	0.761	24.00	0.430
2. In order to have any impact on human health, microbes must be alive at the time they are consumed.	2.45	1.063	2.85	1.093	49.00	0.055
3. In order to have any impact on human health, microbes must stay alive for at least part of the time they spend in the digestive system.	2.70	0.847	2.73	1.039	88.50	0.909
4. Commercially produced yogurt is safer than homemade yogurt.	2.15	0.906	2.12	1.083	133.50	0.825
5. Bacteria are used to produce yogurt.	3.33	0.777	3.88	0.331	8.50	< 0.001*
6. All yogurt sold in stores contains bacteria.	2.73	1.098	3.48	0.906	8.00	< 0.001*
7. Some yogurt sold in stores contains living bacteria.	3.18	0.0.584	2.91	1.208	135.00	0.251
8. All yogurt sold in stores contains living bacteria.	1.94	1.029	2.82	1.211	60.00	0.002*
9. We use this statement to discard the surveys of people who are not reading the questions. Please select “Agree” for this question to preserve your answers.
10. I am comfortable eating food developed in a (food-grade) laboratory.	3.06	0.747	3.30	0.728	26.50	0.084
11. I am comfortable eating genetically modified foods.	3.03	0.770	3.18	0.769	28.00	0.178
12. Natural foods are safer than commercially produced foods.	2.45	0.833	1.97	0.728	170.00	0.011
13. Scientists have a good understanding of how the human gut microbiome influences health.	2.30	0.984	2.06	1.144	107.50	0.334

Deidentified student response data was analyzed as follows. Only students who completed both the pre- and post-survey on time were included, and students who did not respond to all items were removed. Students who did not correctly respond to the control statement (item 9) were removed, and this item was not used in subsequent analysis. For each item, students selected “Strongly Disagree,” “Disagree,” “Neutral,” “Agree,” or “Strongly Agree;” these were converted to scores of 0, 1, 2, 3, or 4, respectively. Pre- and post- scores for each student were compared using a Wilcoxon signed-rank test (non-parametric, paired, n = 33). Statistics were computed in jamovi (The jamovi project, 2020). Mean scores and standard deviation are shown, and items with a significant difference in response distribution (Bonferroni corrected α of p < 0.004) are starred (*).