First Pass Yield

Corona, Dr. S is correct. Anything that comes off the line and is not ready “box & ship” is a defect.
Clearly your first pass yield “number” will suffer for this but isn’t that the point. You are seeking quantification and identification of the defects so you can fix them.
Don’t worry about how bad things “are” – worry about them staying that way. Look at it this way – you have plenty of opportunties to improve performance and get a huge raise and bonus. 

Brandon, I wish more people understood that uncovering and studying defects is a good thing! Early in my career, I had a Japanese professor who said, “When it comes to improving quality, you can’t be afraid to open your kimono!”

HBGB B^2
 
Usually your statement is correct, but not always.  We (used to) have a product that was installed through the use of large amounts of heat (think similar to welding).  During the install the heat changes the characteristics of the item being installed.  Tuning was done as second and separate step in the process.  The process performed exactly as designed – no defect.  The process was (I readily admit) not a good process and it has since been changed such that the heat from install is reduced, the item is shielded and no adjustment is needed.  If you re-read my post you will notice that I called it an adjustment and said the original post called it a test.  I made this distinction because in our original process the adjustment step was also called a “test”.  I am only trying to give an example of how one size does not fit all.  Not make a specific recommendation to this situation.  I would hate for good advice (such as Dr Scott’s) to be applied regardless of ALL of the facts.  Again I do not know if this situation in any way mirrors the one I described, but I stand by my 1st post for the reasons stated.  BTW I agree if it is a test then it is a defect – no questions.

The fact that your process allowed heat to change the product characteristics  is symptomatic of a defect.  Your subsequent Process improvement (shielding) serves to mitigate/eliminate the defect.
the pre-change process FPY would be lower than post-change, no?
When a widget, or transactional vehicle passes through a process and any “tweaking” is needed to influence the meeting of specs, that is a defect, or source of variation and thus impacts FPY.
I think we are aprox on the same train of thought, just differing in level of in-process/out of process focus.