Mathwords logoMathwords

Measure of Position — Definition, Formula & Example

A measure of position is a number that tells you where one data value stands relative to the rest of the data set. Percentiles, quartiles, and z-scores are the most common examples.

A measure of position (or measure of relative position) locates an individual observation within a distribution. The ppth percentile is a value with about p%p\% of the observations below it. Textbooks differ on whether "below" means "less than" or "at or below," so exact percentile values can vary slightly by method. The quartiles Q1Q_1, Q2Q_2 (the median), and Q3Q_3 are the 25th, 50th, and 75th percentiles. The z-score of xx is the number of standard deviations xx lies above or below the mean.

Key Formula

z=x−μσz = \frac{x - \mu}{\sigma}
Where:
  • zz = z-score (standardized value) of the observation
  • xx = the individual data value
  • μ\mu = mean of the distribution (use xˉ\bar{x} for a sample mean)
  • σ\sigma = standard deviation of the distribution (use ss for a sample standard deviation)

Worked Example

Problem: The daily high temperatures (in °F) in a town over 12 days in April, sorted from lowest to highest, were: 51, 52, 54, 58, 58, 60, 60, 61, 63, 64, 69, 70. Describe the position of the 64°F day in three ways: (a) its percentile rank, meaning the percent of values at or below it; (b) which quarter of the data it falls in, with Q1Q_1, the median, and Q3Q_3 found by the median-of-halves method; (c) its z-score, using the mean and the sample standard deviation.
Step 1: Confirm the data are sorted and count them. There are 12 values. The value 64 is in position 10.
51, 52, 54, 58, 58, 60, 60, 61, 63, 64‾, 69, 70n=1251,\ 52,\ 54,\ 58,\ 58,\ 60,\ 60,\ 61,\ 63,\ \underline{64},\ 69,\ 70 \qquad n = 12
Step 2: Percentile rank: count the values at or below 64. That is 10 of the 12 values. Divide by nn and multiply by 100.
1012×100≈83.3%\frac{10}{12} \times 100 \approx 83.3\%
Step 3: Find the median. With 12 values, the median is the average of the 6th and 7th values.
Median=60+602=60\text{Median} = \frac{60 + 60}{2} = 60
Step 4: Use the median-of-halves method for the quartiles. Split the data into a lower half (51, 52, 54, 58, 58, 60) and an upper half (60, 61, 63, 64, 69, 70). Each quartile is the median of its half, which here is the average of that half's 3rd and 4th values.
Q1=54+582=56,Q3=63+642=63.5Q_1 = \frac{54 + 58}{2} = 56, \qquad Q_3 = \frac{63 + 64}{2} = 63.5
Step 5: Quartile position: 64 is greater than Q3Q_3, so the 64°F day falls in the top quarter of the data.
64>Q3=63.564 > Q_3 = 63.5
Step 6: Find the mean: add all 12 values and divide by 12.
xˉ=72012=60\bar{x} = \frac{720}{12} = 60
Step 7: Find the sample standard deviation. The deviations from 60 are −9, −8, −6, −2, −2, 0, 0, 1, 3, 4, 9, 10. Their squares add to 396. Divide by n−1=11n - 1 = 11, then take the square root.
s=39611=36=6s = \sqrt{\frac{396}{11}} = \sqrt{36} = 6
Step 8: Find the z-score of 64 with xˉ\bar{x} and ss in place of μ\mu and σ\sigma.
z=64−606=46=23≈0.67z = \frac{64 - 60}{6} = \frac{4}{6} = \frac{2}{3} \approx 0.67
Answer: The 64°F day has a percentile rank of about 83.3%, since 10 of the 12 days were 64°F or cooler. It falls in the top quarter of the data because it is above Q3=63.5Q_3 = 63.5°F. Its z-score is about 0.67, so it is 0.67 standard deviations above the mean of 60°F. Each measure locates the same value differently: by percent of the data, by quarter, and by distance from the mean in standard deviations.

Why It Matters

In AP Statistics, z-scores let you compare values measured on different scales, such as SAT and ACT scores. Pediatricians use percentiles on growth charts. A child at the 90th percentile for height is taller than about 90% of children of the same age and sex.

Common Mistakes

Mistake: Reading the 80th percentile as a score of 80% correct.
Correction: A percentile describes rank, not points earned. Scoring at the 80th percentile means about 80% of test takers scored below you. Your raw score could be 65% or 95%.

Related Terms

  • Measure of Variability — Describes how spread out the data values are
  • Cumulative Frequency — Running totals used to locate percentiles and quartiles
  • Percentile Rank — Percent of values at or below a given value
  • Univariate Data — Single-variable data that position measures describe
  • Percentile — Most common measure of position, ranking data by percent below
  • Quartiles — Common measure of position dividing data into four parts
  • Z-Score — Common measure of position using standard deviations from the mean