The Student Room Group

Help - I don't understand significance levels

Can someone please explain the significance levels? For b and c they give the same reason for rejecting the null hypothesis but i thought for c you should reject null hypothesisScreenshot 2023-03-30 6.49.48 PM.pngScreenshot 2023-03-30 6.49.28 PM.png
(edited 1 year ago)
Original post by leoishush
Can someone please explain the significance levels? For b and c they give the same reason for rejecting the null hypothesis but i thought for c you should reject null hypothesisScreenshot 2023-03-30 6.49.48 PM.pngScreenshot 2023-03-30 6.49.28 PM.png


Error in the solution for (b). Should be - reject H0H_0.
(edited 1 year ago)
Reply 2
Screenshot 2023-03-30 7.09.57 PM.png
Original post by ghostwalker
Error in the solution for (b). Should be - reject H0H_0.


is this right then?
Original post by leoishush
Screenshot 2023-03-30 7.09.57 PM.png

is this right then?


Yes.
Reply 4
Original post by ghostwalker
Yes.


I still dont get why is the p value is less than the significance level u reject null, pls explain
Original post by leoishush
I still dont get why is the p value is less than the significance level u reject null, pls explain


There should be decent explanations in your textbooks, but cobbling something together off the top of my head (so hope it's reasonable):

The p-value is the probability of the test statistic taking the value it does or something more extreme, given the null hypothesis is true.

So, if for example, if the p-value was 0.4, then the chances of getting the given value of something more extreme is 40%, i.e. it's not unusual, given H0H_0.

If the p-value was 0.01, then the chances of getitng that value or something more extreme is 1%. I.e. it's very unlikely this would occur given H0H_0.

The smaller the p-value, the more unlikely it is the test statistic will take the given value or something more extreme, or putting it another way, the smaller the p-value the more significant it is.

So, if you set the significance level to 5% say, then if you get a p-value less than that, it's saying it's more significant, or less likely to occur given H0H_0, so reject H0H_0.
(edited 1 year ago)

Quick Reply

Latest