z test and t test

Six Sigma – iSixSigma Forums Old Forums General z test and t test

Viewing 12 posts - 1 through 12 (of 12 total)
• Author
Posts
• #48877

vandehei
Participant

i was confused on what test should i use if the given is 2 samples and the first sample is >30 and the second sample is 30 and the second sample is 30 and the second sample is

0
#165896

Eric Maass
Participant

Hi Ann,
The t-test is more generally applicable – it will work for smaller sample sizes and larger sample sizes and mixed sample sizes – so I’d suggest that you simply stick with it for simplicity’s sake – ie, for consistently using the same test for comparison of means.
Best regards,Eric

0
#165897

vandehei
Participant

hello erici asked that question because our professor told us to determine on what kind of test should we use. thanx!!!

0
#165899

annon
Participant

Your prof is probably trying to get you to understand which conditions dictate the more appropriate test.  It is often taught that with larger sample sizes (ie n >30) it is permissable to use the Z test and below that use of the t test is advisable….although this isnt always the best way to go about it.
But the previous posting was accurate…when you dont know the population std dev or your sample size is small, using the t test is the best advice.  As sample sizes get larger, the central limit theorm (sp) will begin to equalize the t and z distributions and at or above 30, they become quite similar, hence the question using that magic number.  Good luck.

0
#165901

Whitehurst
Participant

There is one more test for means.  At 30 samples or more you should use a Z test.  Blow 30 samples use the “Student’s t test”.  At 6 samples or below you would want to use the “Protected t test” which compensates for the error in small sample sizes.  In all these formulas you should be reasonably sure the data if from a normal distribution or they don’t apply.

0
#165908

Lolita
Participant

Brief  Elaboration

0
#165909

SUDHA
Member

Hi,I heard lot abot Six Sigma and its importance in Quality Generation.I just need to know how can a Six Sigma help the upcoming small BPO’s or Call centers. Can you please guide me or send me some information on Six Sigma and its significance. So that i can go a head and approach ma team regarding this.Regards,
Sudha

0
#165916

Lolita
Participant

What  is  your  email?

0
#165917

annon
Participant

As to what?

0
#165920

JENNIFER
Participant

just curious, what is the ‘protected’ t-test?

0
#165933

vandehei
Participant

so what should i answer? my professor is looking for a proof or reference on what kind of test should we use, i think it’s about comparison between two sample means.. what if the first sample is 60 and the other is 20 isn’t that we can use z test if it is greater than or equal to 30 and t test if it is less than 30?? so what should i do to the two samples?? add the 60 and 30 and use z test? or what?thanx!!

0
#165939

Six Sigma guy
Member

For Z test you need population standard deviation and I assume you dont have one. Go for T test. It is possible to conduct T test( 2 sample t) with one sample having 60 data points and another one with 30 data points – the data points need not be same. However these tests assume normal distribution of the data and before conducting these tests variance test should be conducted- to check whether variances are equal or not.
Cheers

0
Viewing 12 posts - 1 through 12 (of 12 total)

The forum ‘General’ is closed to new topics and replies.