 # Statistical significance - hypothesis testing

I have a dataset consisting of 10 quiz with more than 1000 unique students taking those test one by one. After each test student is given different grades. grade P for pass and F for fail. The one who do not pass the test takes another attempt . quiz at which students are likely to fail most ?

Give statistical significance of the answer for the quiz.

lets suppose i found that 6 th quiz ,maximum fail attempt was there. Now how will i give my hypothesis testing?

You can do the following:
Null Hypothesis (H0) : Most number of students are likely to fail on 6th quiz.
Alternative Hypothesis (H1) : Most no. of students are likely to fail on a quiz other than 6th one.

1 Like

but, which test i should perform to prove this null hypothesis?

one more thing ,the number of attempt for the test for each test follows the geometric distribution.

eg:-

‘ ’ ’

test 1 taken by 800

test 2 taken by 500

test 3 taken by 300

test 4 by 150

test 5 by 75

test 6 by 40

test 7 by 25

test 8 by 10

test 9 by 5

test 10 by 1 student

‘ ’ ’

will our hypothesis change for this distribution.?