Kruskal-Wallis

Test of the Equality of Medians - Individual Statistics

  

Use the table of individual statistics to assess the following properties of your data:

·    N - Number of observations for each level of the factor.

·    Overall - Total number of observations

·    Median - Median of the observations for each level, which provides an estimate of the population medians for each level

·    Ave Rank - Statistic that ranks the levels of data and is used to determine the Kruskal-Wallis statistic

Example Output

Kruskal-Wallis Test on Beds    

 

Hospital    N    Median    Ave Rank         Z

1          11     16.00        14.0     -1.28

2          11     31.00        23.3      2.65

3          11     17.00        13.7     -1.37

Overall    33                  17.0

 

H = 7.05  DF = 2  P = 0.029

H = 7.05  DF = 2  P = 0.029 (adjusted for ties)

Interpretation

The results of the beds data analysis indicate that:

·    The median number of unoccupied beds is the smallest in the hospital 1 (16.00) and the largest in hospital 2 (31.00).

·    Two hospitals have ranks (14.0 and 13.7) close to the overall rank (17.0) while hospital 2's rank is 23.3, which may indicate that hospital 2's occupancy rate is different from the other two.

 

Related Charting Tools

Free online chart generators for statistical analysis:

Useful tools: