Use the table of individual statistics to assess the following properties
of your data:
· N<=
- The number of observations for each level
of the factor that are less than or equal to the overall median.
· N>
- The number of observations for each level
of the factor that are greater than the overall median.
· Median
- The median of the observations for each level.
These sample medians provide estimates of the population medians for each
level.
· Q3-Q1
- The interquartile range for each level of
the factor (which is, like the variance, a measure of the spread of data).
Mood's median test assumes that the population variances for all levels
are equal. Thus, if the sample interquartile ranges differ by a lot, you
might want to test the data for equality of variances.
· Overall median
- The median of all observations.
Example Output |

|
|
Mood median test for Weight
Chi-Square = 1.44 DF = 3 P = 0.697
Individual 95.0% CIs
Temp N≤ N> Median Q3-Q1 ---------+---------+---------+-------
38 4 3 19.0 4.0 (---*---------)
42 3 3 19.0 9.5 (---------*------------------)
46 2 4 22.0 7.3 (-----------------*------------)
50 4 2 18.0 4.3 (---------*-------)
---------+---------+---------+-------
17.5 21.0 24.5
Overall median = 19.0 |
Interpretation |

|
The results of the fish data analysis indicate that:
· The
median weight for fish is the smallest at 50 degrees F (18.0) and the largest at 46 degrees F (22.0).
· The
interquartile ranges for the different temperatures do not appear to vary
enough from each other to be cause for concern.