question about t-test

#1
Im a little confused about the difference between a t test for one sample, for two independent sample, and two related samples.
For this problem:
You are in charge of quality control for the company that produces gatorade. If all the production machines in the factory are functioning properly, then the amout of high-fructose corn syrup in gatorade should have a mean of 345 mg per bottle. To ensure that no machines are malfunctioning, every month you collect and test a random sample of 30 bottles of gatorade. This onth, the mean amount of fructose in your sample is 320 mg, and s=20. Conduct the appropriate t test at alpha=.05 to determine whether the mean amount of fructose in this month's batch of gatorade differs from 345 mg.

Would this be a t test for two related samples???
 

JohnM

TS Contributor
#2
This is a t-test for 1 sample.

If you're comparing a data set against a known standard, it's 1-sample test.

If you're comparing 2 data sets against each other, it's a 2-sample test.

If the samples in the two groups are not independent random samples, then it's a dependent samples test (i.e. the samples in one group were obtained based upon the samples in the other group, or if they're the same item/object/person in both groups).
 
#3
Would it be a two tailed test?
so Reject Ho if t greater than or equal to 2.045 and if t is less than or equal to -2.045?

and to find df I just subtract 30-1=29 right?
(30 being the # of samples)...