I’ve never been good at statistics. Fortunately, I’ve never really done any research that has required any high-powered stats beyond what Excel can do. But now I am, so….

Can anyone explain for me, or point me to an online explanation of, the meaning of Z’ value or Z’-factor value?

Cheers. http://hfboards.com/images/smilies/cheers.gif

Are you doing an HTS assay? Sorry, can't explain those statistics.

No, what I’m doing doesn’t really qualify as a HTS assay. I want to look at changes in concentration of downstream second messenger molecules as a result of muscarinic receptor activation.

You are normalising your data for ?

Z score is usually used to normalise or standardise data.

e.g. you're grading apples into A, B and C by weight.

More than 2 ounces = A grade; less than 0.75 ounces is C grade.

If 15.2% are A grade and 6.62% are C grade, then z values are calculated like this:

z=[2.0-population mean]/standard deviation and z=[0.75-population mean]/standard deviation;

For x=2.0 the upper 15.2% is divided from the lower (100-15.2) of the normal distribution (where 100% is the total population) so we need to look up the z value that corresponds to 84.8 (from 100-15.2) or 0.848

Here is a normal distribution table

http://www.math.unb.ca/~knight/utility/NormTble.htm

if you search for 0.848 in the table you can see it is in the row where the far-left column is 1.0, and it is in the column where the top row is 0.03. Therefore the z value is 1.03.

Basically, it is a measure of how different your data score is from the populaton mean. Of course, the population mean and standard deviation is usually an estimate. :)

You can then calculate the z score for the second one (C grade apples) and compare the two.

The standard score indicates how many standard deviations an observation is above or below the mean. It allows comparison of observations from different normal distributions, which is done frequently in research.

Here is a basic guide on the z score
http://www.stats4students.com/Essentials/Standard-Score/Overview.php

Wikipedia has a good page with the math on it:
http://en.wikipedia.org/wiki/Standard_score


The basic guide sam provides does a good job in explaining what makes the z statistic useful.

[edit]
Actually, that guide goes through it really thoroughly. Didn't see all the pages.

[editx2]
Since in most cases you won't actually have all the population data, and just be working with a sample, would you just use the sample data and assume that, if it's a random sample, it's a good estimator of the true mean and standard deviation, then go with the Z-statistic for that?

[editx2]
Since in most cases you won't actually have all the population data, and just be working with a sample, would you just use the sample data and assume that, if it's a random sample, it's a good estimator of the true mean and standard deviation, then go with the Z-statistic for that?

Yeah, see my example for estimation from random sample.

edit: you can actually use the two raw scores to estimate the SD and mu.

Using the same example:

z1s = 2 - mu; z2s = 0.75 - mu

Solve the two equations to get SD or standard deviation.

then mu = 2 - (z1*s); mu =0.75(z2*s)

will give mu or population mean

Btw, my math is pretty crappy, so it would be best to cross check this.:o

No, what I’m doing doesn’t really qualify as a HTS assay. I want to look at changes in concentration of downstream second messenger molecules as a result of muscarinic receptor activation.

Why not use a two tailed t-test instead?

Thanks for the info and links. :D


Why not use a two tailed t-test instead?


Maybe I will. At the moment I am simply reading through references for my intended experiments and many of them mention Z’ values for the different levels of induction of assay targets in different samples.

I’ve never been good at statistics. Fortunately, I’ve never really done any research that has required any high-powered stats beyond what Excel can do. But now I am, so….

Can anyone explain for me, or point me to an online explanation of, the meaning of Z’ value or Z’-factor value?

Cheers. http://hfboards.com/images/smilies/cheers.gif

Go and buy "Choosing and Using Statistics - a biologists Guide" by calvin Dytham

http://www.amazon.co.uk/Choosing-Using-Statistics-Biologists-Guide/dp/1405102438/ref=sr_1_1?ie=UTF8&s=books&qid=1208820159&sr=8-1