Example of Runs Test

A researcher for a pharmaceutical company wants to compare the effects of three treatments on 35 patients. The researcher assigns a treatment to each patient. The researcher wants to determine whether the treatment assignments are random.

To determine whether the treatment assignments are random, the researcher performs a runs test.

  1. Open the sample data, TreatmentAssignments.MTW.
  2. Choose Stat > Nonparametrics > Runs Test.
  3. In Variables, enter Treatment.
  4. Click OK.

Interpret the results

The p-value is 0.930, which is greater than the significance level of 0.05. The researcher does not have enough evidence to conclude that the data are not in random order.

Descriptive Statistics



Number of
Observations


NK≤ K> K
3522312
K = sample mean

Test

Null hypothesisH₀: The order of the data is random
Alternative hypothesisH₁: The order of the data is not random
Number of Runs
ObservedExpectedP-Value
1716.770.930