# Bias Study - How can I calculate significant 't' value? Page 87-90 of MSA manual

E

#### Ehsan Heidari

How can I calculate significant t value (look at P87,88,89&90 3th edition of MSA manual)?

E

#### Ehsan Heidari

Re: Bias Study

See the Stability tab in the file attached to my blog entry. Scroll to the bottom and you will find the bias study calculations.
Thank you Minner for your quick response! but as I asked in page 89(3th edition of MSA manual) are written :"t(v,1-@/2) is found using the standard t tables." can you explain it for me?

#### Miner

##### Forum Moderator
Staff member
Re: Bias Study

Thank you Miner for your quick response! but as I asked in page 89(3th edition of MSA manual) are written :"t(v,1-@/2) is found using the standard t tables." can you explain it for me?
v means the degrees of freedom in the sample. This is typically n - 1 where n is the sample size.

@ means the alpha value. This is the amount of acceptable risk that you will make an incorrect conclusion that you have statistical significance when it is not. Alpha is normally set at 0.05.

E

#### Ehsan Heidari

Re: Bias Study

v means the degrees of freedom in the sample. This is typically n - 1 where n is the sample size.

@ means the alpha value. This is the amount of acceptable risk that you will make an incorrect conclusion that you have statistical significance when it is not. Alpha is normally set at 0.05.
Ok, How can I calculate it? Or I saw your file, it is so useful but I have a question:
You calculate the t(bias) and t(critical) and explain for acceptable most t(bias) < t(critical) , but in MSA manual we have different chart(enclosed), what`s the relation between it and your formula?

#### Attachments

• 51.4 KB Views: 230
Last edited by a moderator:

#### Miner

##### Forum Moderator
Staff member
Re: Bias Study

Ok, How can I calculate it? Or I saw your file, it is so useful but I have a question:
You calculate the t(bias) and t(critical) and explain for acceptable most t(bias) < t(critical) , but in MSA manual we have different chart(enclosed), what`s the relation between it and your formula?
Significant t Value (two-sided) and t(critical) mean the same thing.

The 95% confidence limits of the Bias are additional values that are not included in my Excel file. See .

E

#### Ehsan Heidari

Re: Bias Study

Significant t Value (two-sided) and t(critical) mean the same thing.

The 95% confidence limits of the Bias are additional values that are not included in my Excel file. See .
OK Minner,
1- How can I calculate 95% confidence limits of the Bias?(I want t table)
2- therefore in stability, after drawing Xbar-R diagram, I should to calculate t(bias) and t(critical) and if t(bias) < t(critical), it`s good and my work in stability is finished?

Last edited by a moderator:
G

#### Grecu

Hello,
I'm haveing the same problem , I don't Know how to calc. the confidence interval and I can not find this T Table.

#### harry

Super Moderator
Hello,
I'm haveing the same problem , I don't Know how to calc. the confidence interval and I can not find this T Table.
Pending further response from the experts in this field, you may want to go through the similar discussions in this thread: Bias Study Confidence Intervals - Page 88 of the AIAG MSA manual

't' table is given by Miner in post #11

There are a few more threads (good reading) on the same topic in the 'similar discussions' box below the page. Just scroll down this page.

E

#### Ehsan Heidari

Thank you Miner, I saw your post and I understand how can I calculate t(critical) but my last question is : "for setting calibration schedule I should use you formula(in your excel file) or Bias limitation in P88 Manual?:"

Thank Miner