hearseeno replied to your post: does anyone know how to work w/ z-scor…
maybe…depends on what you’re trying to do with them.. Taught psych stats at one point in my career. Let’s see if I remember the basics. 😛
crosses fingers that u still remember
I’m supposed to be doing a bunch of different crap with them, but here’s a sample problem: “A set of scores has a mean of 16 and a standard deviation of 3. What score separates the top 30 percent of the population from the lower 70 percent?”
The formula we were given for this set of problems was the one that goes, (‘score of observations’ minus ‘mean’) divided by (‘standard deviation’)
i cant for the life of me figure out if i’m plugging everything in correctly, and after, how to be sure if I’m selecting the right proportions from the z-score table
*pulls out first statistics course book* Oh, look, copyright 1980. XD Oops, damn, ripped a page just turning it. Jussst a little fragile after all these years. Mmmmmmm. Old book smell.
Okay, that’ll take a couple steps. You’re looking at the table in your book of z score associated with proportions of area under the standard curve, yeah?
So, what you’re looking for is the z associated with the 70th percentile:
Remember that right smack dab in the middle of the perfectly balanced normal curve is the 50th percentile. In order to get to the 70th percentile using the wacky table they provide you, you’re going to need to look for the 50th + xth = 70 – so, you’re looking for 20 percentage points above the mean.
So, looking at that table for as close to .20 (20%) as I can get for that purple section, it looks like that’s associated with a z of 0.53. Is that what you get? Are they telling you to take the closest? Round up? down?
Now that you have a z score, you need to use your formula to convert the z score into the observed score scale.
z = (ObservedScaleScore – ObservedScaleMean) / ObservedScaleStandardDeviation
0.53 = (x-16)/3
Hearseeno 