# Zst

Six Sigma – iSixSigma › Forums › Old Forums › General › Zst

- This topic has 9 replies, 4 voices, and was last updated 12 years, 5 months ago by Darth.

- AuthorPosts
- October 2, 2007 at 1:31 pm #48295
If a spec limit is 4.5 for zst and you get 1.6; isn’t this good news? From what I understand your zst is the number of standard deviations you are from the mean. Is this correct? The way I see it is 1.6 is a tighter variance than 4.5. Also, is it correct to set up a lsl of 4.5 and an upper spec limit of 4.5?

0October 2, 2007 at 1:36 pm #162308What in the world are you talking about?

0October 2, 2007 at 5:12 pm #162323Is 1.6 a good zst score?

0October 2, 2007 at 5:45 pm #162324No – it’s awful

0October 2, 2007 at 5:52 pm #162327If a Z score indicates how many standard deviations a value, x , is from the mean then why is 4.5 better than 1.6?

0October 2, 2007 at 5:55 pm #162330Think about it for a minute!!!! If z is the number of standard deviations that the SPEC is away from the mean then a z value of 1.6 means the spec is close to the mean and lots of stuff will be beyond the spec…that’s not a good thing.

0October 2, 2007 at 7:16 pm #162347Like Stan, I have no clue what you mean by “zst” (there are various programs out there called zst, and the term is used in microwave engineering, but what do you mean by “zst” in the context of z-values and six sigma.

If you look at the conversion of z-values into sigma scores, here is a table that may answer your question (scroll down for the conversion).

https://www.isixsigma.com/library/content/zdistribution.asp

If you mean “zstd” here is a technical definition:

ZSTD is the infit mean-square fit statistic t standardized to approximate a theoretical “unit normal”, mean 0 and variance 1, distribution. ZSTD (standardized as a z-score) is used of a t-test result when either the t-test value has effectively infinite degrees of freedom (i.e., approximates a unit normal value) or the Student’s t-statistic distribution value has been adjusted to a unit normal value. The standardization is shown on RSA, p.100-101. When LOCAL=Y, then EMP is shown, indicating a local {0,1} standardization. When LOCAL=LOG, then LOG is shown, and the natural logarithms of the mean-squares are reported. More exact values are shown in the Output Files.

Ben Wright advises: “ZSTD is only useful to salvage non-significant MNSQ>1.5, when sample size is small or test length is short.”0October 2, 2007 at 7:44 pm #162350I am referring to a Z score – ST (meaning short term) – Zst.

What would cause a score of 1.6 compared to 4.5?

0October 2, 2007 at 7:44 pm #162351I am referring to a Z score – ST (meaning short term) – Zst.

What would cause a score of 1.6 compared to 4.5?

0October 2, 2007 at 9:20 pm #162362Draw a picture of a normal distribution. Put the 1.6 spec away from the mean and see how much is outside of spec aka to the right of the spec. Now draw the spec at 4.5 away from the mean. How much is out of spec now? Which has less out of spec? Which is better assuming that you don’t want to make out of spec.

0 - AuthorPosts

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