To begin, we distributed copies of the Blood Type Survey to a mixed population of students, staff and faculty at Carleton College. From the initial surveys we selected for both A and O bloodtypes who had a strong aversion to either Diet 1 or Diet 4 (n=94). The observed food aversions for both blood types are shown below in Table 1.

After we tallied the observed diet aversions, we calculated the expected food aversions for A and O blood types based upon our sample size. The expected food aversions are shown in Table 2. We used the following formula to derive the expected food aversion values:

We used the observed and expected values to perform a chi-squared statistical test on our data. The following formula was used to derive the numbers shown below in Table 3:

Using the values presented in Table 3 we calculated the chi-squared and p values for our data set. The formulas we used are shown below and the derived values are shown in Table 4.

A 1 degree of freedom along with the chi-squared value of 12.87 revealed that our data is statistically significant (p < .0005). In other words, there is a > 99% chance that our data means something.