☐ cool new feature
☑ could help many users!
☑ removes something that feels like a „bug“
☐ nice to have
☐ nobody needs it
What inspired this wish list request?
In Graph Builder, if there are nominal values on an axis, you can use the 'Order by' option to select another column or count to specify an order.
This option is available if multiple columns are nested on the axis, but it's unclear how the values are sorted.
the examples below are from
with main category "age" and subcategory "student"
JMP19: values are sorted, ignoring the categories on the axis. Then the axis labels are added.
Therefore one age group (e.g. 14) can show up multiple times:
< JMP19: The values are sorted for each category (ignoring the other categories).
- age sorted by age, ignoring student names
- students sorted by name, ignoring age
-> Robert is twice in the data set. The position of Rober#1 and Rober#2 is determined by mean(Rober#1 / Rober#2)
-> Rober#1 comes later than expected, Rober#1 comes earlier than expected / height jumps
What is the improvement you would like to see?
1) Please add an option to specify how sorting works if multiple columns are nested on an axis.
2) Please add a new mode:
- main categories are sorted by averaging the respective entries (like in JMP 18)
- subcategories are sorted by their order WITHIN the parent category.
e.g.
dt = Open( "$SAMPLE_DATA/Big Class.jmp" );
New Column( "myOrderBy",
Formula( Col Mean( :height, :age ) * 10000000 + :height ),
) << run formuals();
Graph Builder(
Variables(
X( :age, Order By( :myOrderBy, Ascending, Order Statistic( "Mean" ) ) ),
X(
:age_name,
Position( 1 ),
Order By( :myOrderBy, Ascending, Order Statistic( "Mean" ) )
),
Y( :height )
),
Elements( Bar( X( 1 ), X( 2 ), Y ) )
);
Why is this idea important?
- backward compatibility
- For many applications, it makes the most sense for subcategories to be sorted by order within the parent category.
