Find Z-Scores Needed for Top 16% & 2.5% Graduation Honours

  • Thread starter Thread starter 60051
  • Start date Start date
Click For Summary

Homework Help Overview

The discussion revolves around determining the z-scores corresponding to the top 16% and top 2.5% of students for graduation honours, given a mean of 2.7 and a standard deviation of 0.5. The original poster attempts to apply the z-score formula but expresses uncertainty about the correct approach.

Discussion Character

  • Exploratory, Assumption checking, Mathematical reasoning

Approaches and Questions Raised

  • Participants discuss the need to reference a z-score table to find specific values for z1 and z2 that correspond to the desired percentiles. Questions arise about how to interpret the table and locate the necessary probabilities.

Discussion Status

Some participants provide guidance on using the z-score table and the transformation formula to convert z-scores back to x-scores. There is an ongoing exploration of the relationship between the z-scores and the corresponding x-scores, with participants noting similar results for different z-scores.

Contextual Notes

There is mention of confusion regarding the correct formula for z-scores and the distinction between population and sample parameters. Participants are also navigating the limitations of the z-score table and its interpretation.

60051
Messages
16
Reaction score
0

Homework Statement



Students who are in the top 16% and top 2.5% will graduate with special honours. Where should the limits be set in terms of z-scores?

Mean (xbar) = 2.7
Standard deviation (s) = 0.5


Homework Equations



z = x - xbar / s

and

-1 to 1 = 68% of all grades
-2 to 2 = 95% of all grades
-3 to 3 = 99.7% of all grades

(don't know if that's ^ relevant)


The Attempt at a Solution



I tried multiplying 3 (the highest z score) by 0.84 (100 - 16) but that didn't seem to work. I know there's one thing I have to do before I use the z-score formula, but it's just not clicking.
 
Physics news on Phys.org
You need to look at a table of z-scores. You want the numbers z1 and z2 for which P(z < z1) = .84 and for which P(z < z2) = .975.

After you get these numbers, you need to use the transformation formula to convert to x scores. The one you gave converts x-scores to z-scores. To get the transformation that goes from z-score to x-score, solve that formula for x.

BTW, the formula you gave should be written as z = (x - mu)/sigma; i.e., you need parentheses, and it involves the population mean and population standard deviation, not the sample mean and standard deviation.
 
Mark44 said:
You need to look at a table of z-scores. You want the numbers z1 and z2 for which P(z < z1) = .84 and for which P(z < z2) = .975.

I searched for a z-score table, but I don't know how to read it. On one axis they have z-scores and along the other axis are decimal values, but what are these values? How do I find 0.84 and 0.975 if they're not listed?
 
One row will have 0.8. Look for the column with 4 in it. In the cell in the row with 0.8 and the column 4 is the probability that z < .84. The table I'm looking at has 0.7995 at that position.
 
Alright, so 0.7995 corresponds to 0.84 and 0.83523 corresponds to 0.9775. But when I plug these values in and solve for x, I get virtually the same answers (3.1 and 3.11). How can that be?
 
60051 said:
Alright, so 0.7995 corresponds to 0.84 and 0.83523 corresponds to 0.9775. But when I plug these values in and solve for x, I get virtually the same answers (3.1 and 3.11). How can that be?
0.7995 is the probability that corresponds to a z-score of .84. Look in the body of the table for .9775 and find the z-score that corresponds to it?

After you get the two z-scores, solve for x in the formula z = (x - mu)/sigma, and then use that new formula to calculate the two x-scores.
 

Similar threads

Replies
2
Views
5K
Replies
1
Views
5K
  • · Replies 2 ·
Replies
2
Views
4K
  • · Replies 13 ·
Replies
13
Views
5K
  • · Replies 3 ·
Replies
3
Views
4K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
6
Views
648
  • · Replies 11 ·
Replies
11
Views
4K
  • · Replies 5 ·
Replies
5
Views
4K