When to Use the Mean, Median and Mode

Mean, Median, and Mode

Mean, Median, and Mode
  • Sometimes you get very similar results with all three.
  • Like when you have a normal distribution.

Mean

Usually, the mean is preferred:

  • It uses all the scores ( so it’s representative of the entire data set ).
  • It’s used to compute the variance and SD.
  • It’s good for inferential statistics.
  • Note that you should have interval or ratio data to compute a mean.

Median

  • Use the median when you have extreme you have extreme scores or a skewed distribution

Median

  • Use the median when you have extreme scores or skewed distribution.
  • Example :
  • X = 10, 11, 11, 11, 11, 12, 12, 13, 13, 100
  • M = 20.3
  • Median = 11.5
  • Median represents most of the distribution best .

Median

  • In psychology, you might encounter an open-ended distribution like this:
  • N = 20
  • Cannot compute a mean.
  • Median = 1.5
  • Use the median!

Number of Children Frequency
5 or more 32
42
32
23
16
04

Median

  • Use the median if you have ordinal data
  • Remember, the mean balances distance
  • With ordinal data, you don’t have equal distances between data points

Mode

  • Use the mode if you have nominal data
  • Example:
    Hair color: 1= brown , 2 = black , 3 = blond 4 = red

Hair color Frequency
42
34
25
17

N=18

Mode

  • If you have a discrete variable like a number of children, you can compute a mean.
  • In this case, means are fractional values that can’t really exist. EXAMPLE: “The average family has 2.5 kids”
  • The mode identifies the typical case: – “The typical family has 2 kids.” – “The modal age for spinal cord injury is 19.”