Level of scientific education

On a topic such as GM generally a lack of education can cause people to be daunted and therefore opposed to GMO use. Here, we would be looking at participants with just a high school level of science education. But in this cohort of the participants we saw that 26.5% of those showing no concern over GM crops came from this cohort.

Looking at Figure 4, cumulatively those with a qualification, or studying a qualification at higher education level accounts for 73.3% of all participants. This could also explain the positive response in figure one along with the young participation as, 65.7% of those who showed no qualms to GM have a degree or are studying at higher education level (university).

However, when running statistical analysis (one-way ANOVA) to identify whether the level of education played a role in opinion on GM, we found that this was insignificant (P=0.985). This test was performed by converting ‘yes’ answers to 1, and ‘no’ answers to 0.

Statistical test 3: ANOVA to test the difference between means between groups with different scientific education when asked if they are concerned about GM crops.

To analyse this further, we ran a TukeyHSD test to compare means between every pair of groups. This too found no significant difference between any two groups.

Statistical test 4: TukeyHSD test to show difference in means between any two groups, comparing scientific education level to concerns about GM.

We also looked to try and gain an understanding of the participants knowledge on current GM laws in the EU, GM food labelling and potential GM shopping habits. The results are shown in Figure 5.

At the root of our project, we aim to eradicate the need for harsh chemical pesticides. We were keen to find out if the general public had concerns over their heavy usage in agriculture. These results are shown in Figure 6.

Supporting the cause at the root of our novel strategy, we saw a clear majority of the participants, 86.6%, showing concern over the use of chemical pesticides in agriculture. Furthermore, we also saw great response to GM being a tool to replace or reduce pesticide use shown in Figure 7.

With only 8% protesting for the potential application of GMOs to act as an answer, this falls relatively inline with an overall postsive response to GM, where 21.2% of those who agreed with the replacement or reduction/removal of chemical pesticides by GM application expressed concerns over their use in Figure 1. This shows that even those with concerns would agree that the use of GMOs is tolerable in order to reduce chemical pesticide use.

Location

We decided to see if the country/continent of origin had a significant difference on people’s opinions about GM crops. Our One-way ANOVA found that this did not have a significant effect (P=0.734). This test was performed by converting ‘yes’ answers to 1, and ‘no’ answers to 0.

Statistical test 5: ANOVA to test the difference between means between location groups (continent or country) when asked if they are concerned about GM crops.

There was also no significant difference between means within group pairs as shown below.

Statistical test 6: TukeyHSD test to show difference in means between any two location groups showing whether concern to GM crops changes geographically.

Should GM crops replace pesticides? A statistical analysis

Finally, we decided to see whether people would prefer to eat GM crops than crops that had chemical pesticides used. For this, there were 3 potential answers ‘yes’, ‘no’, and ‘don’t know’. These were converted to 2, 0, and 1 respectively.

Age

Age had a significant effect in determining whether people would prefer to eat GM crops than eat chemical-insecticide treated crops. There was a highly significant difference between means (p=0.000595).

Statistical test 7: ANOVA to test the difference between means between age groups when asked if they would prefer to consume GM crops than chemically treated crops.

Again, we decided to analyse the differences between means of each pair of groups. We find the same pattern as the previous analysis between age and concerns about GM crops. However, this time there is a significant difference between the 41-60 age group and the 18-24 age group (p=0.00108). This suggests that the younger generations would prefer to eat GM crops than chemically treated crops, whereas the opposite is true of older generations. Interestingly, these means are not as different between the youngest and oldest group (under 18 and 61+), where the p-value is only 0.706, which is not significant.

Statistical test 8: TukeyHSD test to show difference in means between any two age groups showing whether they would prefer to consume GM crops rather than chemically treated ones.

Scientific education

When we look at whether scientific education affects whether people would prefer to eat GM crops than chemically treated ones, we find a non-significant difference between groups (p=0.507).

Statistical test 9: ANOVA to test the difference between means between groups with different levels of scientific education when asked if they would prefer to consume GM crops than chemically treated crops.

When we compare means between group pairs, we still find no significant differences.

Statistical test 10: TukeyHSD test to show difference in means between any two groups with different levels of scientific education showing whether they would prefer to consume GM crops rather than chemically treated ones.

Location

Finally, when we look at different locations, we find that this too is insignificant (p=0.104).

Statistical test 11: ANOVA to test the difference between means between location groups when asked if they would prefer to consume GM crops than chemically treated ones.

The difference between group pairs is also insignificant between all groups.

Statistical test 12: TukeyHSD test to show difference in means between any two location groups showing whether they would prefer to consume GM crops rather than chemically treated ones.

Conclusion

Our statistical analysis suggests that only age plays a statistically significant impact on peoples’ opinions and concerns about GM crops, and also in influencing whether they would prefer to eat GM crops over chemically treated ones. Younger generations seem to prefer GM crops over chemically treated crops, whereas the opposite is true for older generations. It is important to note two limitations in these analyses. Firstly, some of the groups have low sample sizes which skews the data and can make it unreliable. This can be improved for more accurate statistical analysis by increasing the sample sizes for each group, and in the study as a whole. Secondly, our method of changing yes answers to high values (1 or 2) and low answers to low values (0), with unsure answers to medium/neutral values (1, where yes is 2), may be sub-optimal, and generate issues when running each statistical test. Thus, to improve our analysis, we would create graded numbers for our surveys, instead of yes/no answers would be questions such as “On a scale of 1-10, how concerned are you about consumption of GM crops, with 1 being not concerned and 10 being highly concerned”. This would create less ambiguous data, with a greater range for more accurate analyses.

References