Summary statistic: Mean, Variance and Effect size

Statistics designed to answer questions such as central tendency, spread/variability, extreme values are called summary statistics.


This recipe includes the following topics:

  • Mean
  • Variance and Standard deviation
  • Effect Size
  • Cohen’s d

Mean

  • Mean is the most common summary statistic used to describe the central tendency of the distribution.
  • Useful when samples are pretty similar. Eg: Size of an apple in a supermarket are mostly similar.
    sample_mean
    Mean

Variance and Standard deviation

  • Variance is a summary statistic often used to quantify spread.
  • Standard deviation is the square root of variance, also used as a measure of spread.
  • Useful when samples vary widely.
  • Eg: Pumpkin can weigh between 1lb to 400lb. Both mean and variance may be needed to describe the distribution.
    variance
    Variance

Effect size and Cohen’s d

  • Effect size is a summary statistic intended to quantify the size of an effect.
  • Eg: Size of an effect in two groups where Group A was given Medicine A but Group B was given none.
  • Cohen’s d is intended to measure this effect size.
    cohend
    Cohen’s d

Leave a Reply

Your email address will not be published. Required fields are marked *