This procedure can be used when the:

Outcome (Dependent) variable is a Continuous and skewed / Ordinal measure
Exposure (Independent) variable is >2 independent groups

Present this graphically using a Box and Whisker Plot

In SPSS select: Graphs > Chart Builder - Gallery – Boxplot | Simple Boxplot or Scatter/Dot | Simple Dot Plot

Analyse this in SPSS using:

Analyze >NonParametric Tests > Legacy Dialogs > K Independent Samples
     Use the Exact button and select Monte Carlo, Statistics button and select Quartiles

The Monte Carlo Sig. (2-tailed) is the p-value.

Use Analyze > Descriptives Statistics > Explore to obtain Medians of each group. Select the Statistics button - Percentiles and limit the Display to Statistics.

Report this quoting the:

p-value

Comments

You can use Mann-Whitney procedure to perform post hoc comparisons of groups but apply the Bonferroni correction to avoid Type I errors due to multiple comparisons.

References

Please Report any issues to Brian Yuen