4 edges 17
2 edges 33
3 apart 4
3 together 4
There was a total 58 experiment. If human being behaved randomly then we would expect
4 edges (.375)*58=21.75
2 edges (.375)*58 =21.75
3 apart (.125)*58 = 7.25
3 together (.125)*58 =7.25
To apply the chi^2 test we first compute the thing we call chi^2 (this was incorrectly done in lecture) which is the sum of the squared differences of the expected and measured frequencies each divided by the expected frequencies, in this case we have the following sum
((21.75-17)^2/21.75)+((21.75-33)^2/21.75)+((7.25-4)^2/7.25)+ ((7.25-4)^2/7.25.75)
which equals about 9.7. Now their are 4 class or 3 degrees of freedom so we can table on page A-107 in the text to see that these frequencies would have between a 1 and 5 percent of occurring if the null hypothesis is true. There for the test is significant and we can safely reject the null hypothesis. The actual value can be found with our chi-square test program. It is seen that there is in fact only a 2 percent chance of seeing this data.