Scribd
Upload
15 views13 pages

Lecture #4a - Z-Scores - Rev2022-2

The document discusses z-scores and how they describe the precise location of a score in a distribution by measuring its distance from the mean in standard deviations. It provides examples of how to calculate z-scores and uses them to compare scores from different distributions. The key properties of z-scores and standardized distributions are described to allow comparisons across different data sets.

Uploaded by

Nana Doan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
15 views13 pages

Lecture #4a - Z-Scores - Rev2022-2

The document discusses z-scores and how they describe the precise location of a score in a distribution by measuring its distance from the mean in standard deviations. It provides examples of how to calculate z-scores and uses them to compare scores from different distributions. The key properties of z-scores and standardized distributions are described to allow comparisons across different data sets.

Uploaded by

Nana Doan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

z-scores: 

Location of Scores & Standardized Distributions


Jodi Martin, PhD
Explain how a z-score describes the precise
location of a score in a distribution

Compute z-scores for scores from either


LEARNING population or sample data

OUTCOMES Describe the properties of standardized


distributions (i.e., distributions of z-scores)

Use z-scores to make comparisons across


different distributions of scores
Locations of Scores in a Distribution
 Situation #1
 You recently completed a stats exam. The mean for the exam was M =70. You received a
score of X = 76.
 Question: How well did you do on the exam?

Example adapted from Gravetter & Wallnau, 2017


Locations of Scores in a Distribution
 Situation #2
 You and your friend compare stats exam scores – you both earned scores of X = 76 However,
you are in different sections and the exams were slightly different
 The mean for your exam was M = 70 with a standard deviation of s = 3
 The mean for your friend’s exam was M = 70 with a standard deviation of s = 12
 Question: Who earned the higher score on their exam?

Example adapted from Gravetter & Wallnau, 2017


Locations of Scores in a Distribution

 Relative location of a score within a distribution depends on both the mean and the
standard deviation of that distribution of scores

 How can we compare these two test scores in a more standardized way?

Example adapted from Gravetter & Wallnau, 2017


What is a z-score?

 A z-score is a value that specifies the exact location of a score, X, in a distribution


 Measures distance from the mean (above or below) in standard deviation units
 Answers the question → how many standard deviations above/below the mean is this
score?

μ or M

scores below the mean (-) scores above the mean (+)
Figure from Gravetter & Wallnau, 2017
What is a z-score?
 Properties of z-scores:
1) The sign indicates if the X score is above (+) or below (-) the mean
2) The value indicates distance between the score X and the mean in standard deviation units
3) The mean of a distribution of z-scores will always equal 0 (μ or M = 0)

μ or M

scores below the mean (-) scores above the mean (+)
Figure from Gravetter & Wallnau, 2017
Computing z-scores

Population Formula  Numerator is a deviation score


 Measures the distance between the score
𝑋−𝜇 and the mean in original unit of
𝑧=
𝜎 measurement
 Determines if z is above (+) or below (-) the
mean

Sample Formula  Denominator is the standard deviation


 Puts value in standard deviation units
𝑋−𝑀
𝑧=  “__ standard deviations away from the
𝑠
mean”
Computing z-scores: An Example

 The mean of a national exam is μ = 500 with a standard deviation of σ = 100. Results for
three students on this exam were 650, 400, and 500. Where are their scores located
relative to the mean?

𝑋−𝜇 650 − 500 150


Student #1 𝑧= = = = 𝟏. 𝟓𝟎 1.50σ above the mean
𝜎 100 100

400 − 500 −100


Student #2 𝑧= = = −𝟏. 𝟎𝟎 1.00σ below the mean
100 100

500 − 500 0
Student #3 𝑧= = =𝟎 Equal to the mean
100 100
Why would you use a z-score?

 Three uses for z-scores


 To compare scores from different distributions by creating (at least partially) a
standardized distribution (e.g., SAT vs. GRE, different IQ tests)

 To compute relative frequency of scores within a distribution (z-scores and proportions)

 For hypothesis testing to interpret the probability of obtaining specific value in a


distribution of sample means
Standardized Distribution
 Properties of a standardized distribution:
 Same shape as original distribution of scores
 A given score will have same location in standardized distribution as it had in original
distribution
 Mean (μ or M) is always 0
 Standard deviation (σ or s) is always 1

Figure from Gravetter & Wallnau, 2017


Making Comparisons with z-scores

 You and your friend compare stats exam scores – you both earned scores of X =
76 However, you are in different sections and the exams were slightly different
 The mean for your exam was M = 70 with a standard deviation of s = 3
 The mean for your friend’s exam was M = 70 with a standard deviation of s = 12

 Question: Use z-scores to determine who earned the higher score on their exam
Your score

𝑋−𝑀 76 − 70
Your score 𝑧= = = 2.00
𝑠 3 z

𝑋−𝑀 76 − 70
Your friend’s score 𝑧 = = = .50
𝑠 12 Your friend’s score
What you oughta know…
 z-scores describe the exact location of scores in a distribution of data by
measuring their distance above/below the mean

 An entire distribution of scores can be standardized by calculating z-scores


for each score in the original distribution

 Standardized distributions have the same shape as the original


distribution, have a mean = 0 and a standard deviation = 1

 z-scores and standardized distributions allow us to make


comparisons across different distributions

You might also like