How to Create a Fit Model with Different Sized Data Tables

[#1]

I have 12 serialized assemblies. Each assembly has 8 parts, each part has a set of data of the same column structure. 

Each assembly goes through 1 qualification test.

Each assembly then goes through 10 rounds of life testing (each round has the same column structure).

That leaves me with 3 tables 

Table 1 has 8 rows per assembly

Table 2 has 1 row per assembly

Table 3 has 10 rows per assembly

I am interested in only one output, and I want to do a leverage analysis to determine the impact of each column in each table on that output. The output is a column in table 3. 

How do I combine the tables in such a way that I do not have duplicated data in any column or row that will skew the weight of the leverage analysis?

I have attached much smaller versions of the tables for reference.

[#2]

Welcome to the community.  I know someone will check in and help you soon (Jim might have a script).  I will say, you need to have all of the data in one table.  There is a Join Table option (Tables>Join Tables), have you tried it?

https://www.jmp.com/support/help/en/16.2/?os=mac&source=application&utm_source=helpmenu&utm_medium=a...

[#3]

Like @statman said, Looks that you need to join Table 1 with Table 3 via TEST SN and TEST TYPE,

and then join the result with Table 2 via UNITSN.

The result is then 120 rows with all columns.

In the result table you can filter and evaluate according to your needs.

I attach the two Scripts here:

Data Table( "Table 1" ) << Join(
	With( Data Table( "Table 3" ) ),
	By Matching Columns( :TEST SN = :TEST SN, :Test Type = :Test Type ),
	Drop multiples( 0, 0 ),
	Include Nonmatches( 0, 0 ),
	Preserve main table order( 1 )

Data Table( "Untitled 33" ) << Join(
	With( Data Table( "Table 2" ) ),
	By Matching Columns( :UnitSN = :UnitSN ),
	Drop multiples( 0, 0 ),
	Include Nonmatches( 0, 0 ),
	Preserve main table order( 1 )
)