The median value is the middle value of
the distribution (half the values of the sample are lower and half are
higher).
If the values are ranked from highest to lowest (or vice versa), the rank of the median is given by the formula: (n+1) / 2, where n = number of values.
If the values are ranked from highest to lowest (or vice versa), the rank of the median is given by the formula: (n+1) / 2, where n = number of values.
Example: we have collected the age of 15 men and 14 women listed below
Conclusion: half the men in the sample are younger than 36, while half the women in the sample are younger than 52.5. The sample of men therefore appears younger, based on the simple comparison of the medians.
Run a Similarity test to see if you can infer that the entire populations are actually different, based on these samples.
- Men: there are 15 values, so the rank of the median is (15+1)/2 = 8. The median is therefore the 8th value, which is 36 yrs: half the sampled men are younger than 36, and half are older than 36.
- Women: there are 14 values, so the rank of the median is (14+1)/2 = 7.5. Because the 7.5th rank doesn’t exist, we must use the average value between the value located just below (7th rank) and the one just above (8th rank): (48+57)/2=52.5 yrs. Half the sampled women are younger than 52.5 yrs, and half are older.
Conclusion: half the men in the sample are younger than 36, while half the women in the sample are younger than 52.5. The sample of men therefore appears younger, based on the simple comparison of the medians.
Run a Similarity test to see if you can infer that the entire populations are actually different, based on these samples.