What is a Z-test?

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

What is a Z-test?
Message posted by Harsey on February 1, 2001 at 12:00 AM (ET)

Please help!

READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: What is a Z-test?
Message posted by Bill on February 1, 2001 at 12:00 AM (ET)

A z-test most commonly refers to a test for determining if the observed proportion of individuals (or objects) in one group is equal to the proportion in a second group. For instance, is the cure rate (successful treatment - yes vs no) equal to (or greater) the treatment group versus the no-treatment group. The chi-square test is more commonly used in this situation.

Re: What is a Z-test?
Message posted by Nancy Diehl on February 2, 2001 at 12:00 AM (ET)

Another application of the Z-test is when you have a continuous
variable, e.g. height, weight, and you wish to test if the
sample mean comes from the parent population of that variable.
For example, let's say I'm in manufacturing and I'm putting a
coating on a part. The requirements state that this coating
should have a hardness rating of 15. It's impractical to
100% inspect the hardness for all parts coming through, so
instead I pull a representative sample (for a Z-test you should
have a sample size of at least 30) and calculate the sample
average. Let's say the average is 13.5. The Z-test is then
used to compare the value of 13.5 to the population mean of 15
to see if a sample average of 13.5 could come from this
parent population. By the way, the sample standard deviation
(amount of scatter in the sample) is also taken into consideration
in this test.

Your $5 contribution helps cover part the $500 annual cost of keeping this site online.