THE RUN TEST
- The run test is used to test the randomness of a series of observations.
- We can use this to find out whether our sample data has been collected by a randomised procedure or not.
- Randomisation is important to give credibility and validity to our observations.
- The RUN test is a simple non parametric test.
- When the size of the sample is small, it may be done simply, by counting the number of "runs" in a series of observations and comparing the number of runs with a table value. .
- When the sample is large, it may be carried out by calculating a Z value and comparing it with a critical value.
- A run is a sequence of one or more identical symbols representing a common property of the data.
- It is a sequence of letters of one kind surrounded by a sequence of letters of another kind.
- The number of elements in a run is called the length of a run.
- The observations may be quantitative such as the weights of tablets being punched out by a tablet machine, or drug content of a randomly taken sample of bottles of suspension.
- The observations may be qualitative, such as male or female, defective or not defective, head or tail or above or below the median.
- Whether the observations are qualitative or quantitative, the runs test divides the data into two mutually exclusive categories.
- So a series is always limited to two distinct symbols.
The runs test is based on the order in which the sample observations are drawn.
It tests the null hypothesis H0 that the observations are indeed drawn in a random manner