1 Sample Median

1 Continuous Dependent Variable with non-normal distribution

0 Independent Variables

A one sample median test measures whether a sample median differs significantly from a hypothesized value. It differs from the one sample t-test in that is doesn’t require the data to be normally distributed, the only requirement is that the dependent variable is ordinal or interval.

For example, a Movielens researcher hypothesizes it takes 50 seconds for a new user to add a friend to their buddy list. The researcher conducts an experiment and measures how long it takes several new users to perform the task. 25 subjects participate in the experiment; the researcher plots the results from the sample group and finds they do not appear to follow a normal distribution. The one sample median would be the correct statistical test in this case because it doesn’t require the dependent variable be normally distributed. Like the one sample t-test, it tests whether the mean amount of time it took the experimental group to complete the task varies significantly from the hypothesized 50 second value.

Values to report:

· the mean of the test group

· standard deviation

· t-value (or student t-value if sample size < 30)

· p value