You are describing a nested (or hierarchical) sampling plan. The sampling tree might look something like this:
As @P_Bartell suggests, looking at the data with variability charts is a good first step. I'm not sure what you mean by "uniformity"? To understand consistency you can use Range charts. Start at the bottom of the sampling tree (the smallest rational subgroup). If this is stable, roll up the tree and assess the variability at the next layer.
If you want to quantify the variance components (without understanding consistency), make sure the data types are nominal, Analyze>Quality and Process>Variability/Attribute Gauge, enter the response variable in the Y, Response box. Enter the tree from top to bottom in the X, Grouping box. Choose model type Nested. This will produce a variability chart. Select the red triangle next to the Variability Gauge and then select Variance Components. This will give you ANOVA and Variance Components.
"All models are wrong, some are useful" G.E.P. Box