I am doing Hypothesis testing (how many hooks are in a package of blah blah hooks?)
My claim and null are M = 150 (M being mu) and my alternative is M does not equal 150.
sample size = 15
P > alpha (so I accepted the null)
BUT! I still do not quite understand how to determine weather or not the claim is supported, and is it is one tailed or two tailed.
I need to Know:
one tailed or two tailed?
What value do I use to determine if the claim is correct or false.
What table do I use.
I know i did not describe this very well, but if you can help, please do! I need this project to go well. Finals week :(
Thank you!!
My claim and null are M = 150 (M being mu) and my alternative is M does not equal 150.
sample size = 15
P > alpha (so I accepted the null)
BUT! I still do not quite understand how to determine weather or not the claim is supported, and is it is one tailed or two tailed.
I need to Know:
one tailed or two tailed?
What value do I use to determine if the claim is correct or false.
What table do I use.
I know i did not describe this very well, but if you can help, please do! I need this project to go well. Finals week :(
Thank you!!
-
According to this question, it tells you the alternative hypothesis is not equal. Therefore, it is two-tailed test.
For next question, p>alpha, you will fail to reject the null, so you don't have enough evidence to reject that M=150. (PS: because the claim is null hypothesis, your final conclusion is about rejecting or failing to reject the null.)
Since you don't know the population standard deviation, you have to use Student T table.
I hope my answer will make sense to you.
For next question, p>alpha, you will fail to reject the null, so you don't have enough evidence to reject that M=150. (PS: because the claim is null hypothesis, your final conclusion is about rejecting or failing to reject the null.)
Since you don't know the population standard deviation, you have to use Student T table.
I hope my answer will make sense to you.
-
You say that your alternative does not equal 150. This means that the alternative would be selected if you found from your sample that the count was significantly LESS or MORE than 150.
You would use a one tail if your alternative from the sample was that the count was significantly LESS. That is you are only testing for a LESS THAN hypothesis .
I would think that a two tail test would be the one you would use. Most commonly, significance tests are done at the 0.95 level of significance and in your case you would use a Z table. The table areas are probabilities that the standard normal random variable lies between 0 and Z.
You would use a one tail if your alternative from the sample was that the count was significantly LESS. That is you are only testing for a LESS THAN hypothesis .
I would think that a two tail test would be the one you would use. Most commonly, significance tests are done at the 0.95 level of significance and in your case you would use a Z table. The table areas are probabilities that the standard normal random variable lies between 0 and Z.
-
OK LucyFur---
It appears you have a population mean M = 150 but you don't mention a standard deviation which should be specified....as well as the hooks thing is normally distributed.
Now, you have taken a sample of 15 packages and your Null Hypothesis Ho is: sample average = population mean. The Alternative Hypothesis would be Ha: sample average is not = pop. mean.
That means it could be higher or lower which says you must use a two-tailed test.
You cannot accept the Ho you just cannot reject Ho based on the data.
But you need to calculate the statistics.....xbar and sigma/n for the sample and generate the z-statistic. Then using the tables you can determine if you should use an alpha/2 of 0.05 or 0.025...those are customary. Then you find out if your sample average is outside the +/- z value for the alpha level. If it is then you reject the Ho, If it is not then you cannot reject Ho.
Read about Type I and Type II errors in your stats book
It appears you have a population mean M = 150 but you don't mention a standard deviation which should be specified....as well as the hooks thing is normally distributed.
Now, you have taken a sample of 15 packages and your Null Hypothesis Ho is: sample average = population mean. The Alternative Hypothesis would be Ha: sample average is not = pop. mean.
That means it could be higher or lower which says you must use a two-tailed test.
You cannot accept the Ho you just cannot reject Ho based on the data.
But you need to calculate the statistics.....xbar and sigma/n for the sample and generate the z-statistic. Then using the tables you can determine if you should use an alpha/2 of 0.05 or 0.025...those are customary. Then you find out if your sample average is outside the +/- z value for the alpha level. If it is then you reject the Ho, If it is not then you cannot reject Ho.
Read about Type I and Type II errors in your stats book