SIGN() formula returns unexpected results - excel

Question

In continuation of my previous question: Sumproduct with multiple criteria on one range
Jeeped provided me with an very helpful formula to achieve a sumproduct() which takes multiple criteria. My current case is however a bit broader:
Take these example tables:
First column is the ID number, second column a respondent group(A,B). Column headers are question types (X,Y,Z).
Table Q1
| | | X | Y | Y | Z | Y |
|----|---|---|---|---|---|---|
| 1 | A | 2 | 2 | 1 | | 1 |
| 2 | A | 1 | 1 | | | 2 |
| 3 | A | 1 | 1 | | | 1 |
| 4 | A | 2 | 1 | | | 1 |
| 5 | A | 1 | 2 | 1 | | 1 |
| 6 | A | 1 | 1 | | | 1 |
| 7 | A | | | | | |
| 8 | A | | | | | |
| 9 | A | 1 | 1 | | | 1 |
| 10 | A | 2 | 2 | 2 | | 2 |
| 11 | A | | | | | |
| 12 | A | 1 | 2 | 1 | | 2 |
| 13 | B | | | | | |
| 14 | B | 1 | 1 | | | 1 |
| 15 | B | 2 | 2 | 1 | | 1 |
Table Q2
| | | X | Y | Y | Z | Y |
|----|---|---|---|---|---|---|
| 1 | A | 1 | 2 | 1 | | 1 |
| 2 | A | 1 | 1 | | | 1 |
| 3 | A | 1 | 1 | | | 1 |
| 4 | A | 1 | 1 | | | 1 |
| 5 | A | 1 | 1 | | | 1 |
| 6 | A | 1 | 1 | | | 1 |
| 7 | A | | | | | |
| 8 | A | | | | | |
| 9 | A | 1 | 1 | | | 1 |
| 10 | A | 1 | 1 | | | 1 |
| 11 | A | | | | | |
| 12 | A | 1 | 2 | 1 | | 1 |
| 13 | B | | | | | |
| 14 | B | 1 | 1 | | | 1 |
| 15 | B | 1 | 2 | 1 | | 1 |
Now I want to know the amount of times a respondent answered 1 (yes) on Q2 for each question type (X,Y,Z). The catch is that if someone answered 1 (yes) on Q1 it should "override" the answer on Q2, as we assume that when someone answers yes on Q1 (implementation of a measure), their answer on Q2 (knowledge of said measure) has to be yes as well.
The second catch is that for the first two occurrences of Y there can only be yes in one of both columns, so in fact there can only be two yes answers for question type Y for each respondent.
I used the following formula (on sheet 3): =SUMPRODUCT(SIGN(('Q1'!$C$2:$G$16=1)+('Q2'!$C$2:$G$16=1))*('Q2'!$B$2:$B$16=Blad3!$D5)*('Q2'!$C$1:$G$1=Blad3!E$4)) to obtain the following results.
| | X | Y | Z |
|---|---|----|---|
| A | 9 | 19 | 0 |
| B | 2 | 4 | 0 |
For X these results are correct, as there are 9 1's in table Q2.
For Y the results for B are correct, for A however they are not, as there are only 9 respondents, answering max 2 questions would result in a max of 18, we have 19 however.

Answer 1

It turns out there is nothing wrong with the formula, just that it isn't suited for the way this data is organised. If you look at row 5:
Q1
| | | X | Y | Y | Z | Y |
|----|---|---|---|---|---|---|
| 5 | A | 1 | 2 | 1 | | 1 |
Q2
| | | X | Y | Y | Z | Y |
|----|---|---|---|---|---|---|
| 5 | A | 1 | 1 | | | 1 |
If we condense that to everywhere there is a 1 in any of the Y column we get this table:
| | | X | Y | Y | Z | Y |
|----|---|---|---|---|---|---|
| 5 | A | | 1 | 1 | | 1 |
When I ask for the sumproduct() for this combined table the result will be 3.
To prevent this I added a helper column (between the two Y and the Z column) to my tables, with the following formula: IF(OR(D1=1,E1=1),1,""). Removed the headers from the double Y columns, and re-running the query produced the correct results.
New table Q1 looks like this then:
| | | X | | | Y | Z | Y |
|----|---|---|---|---|---|---|---|
| 1 | A | 2 | 2 | 1 | 1 | | 1 |
| 2 | A | 1 | 1 | | 1 | | 2 |
| 3 | A | 1 | 1 | | 1 | | 1 |
| 4 | A | 2 | 1 | | 1 | | 1 |
| 5 | A | 1 | 2 | 1 | 1 | | 1 |
| 6 | A | 1 | 1 | | 1 | | 1 |
| 7 | A | | | | | | |
| 8 | A | | | | | | |
| 9 | A | 1 | 1 | | 1 | | 1 |
| 10 | A | 2 | 2 | 2 | | | 2 |
| 11 | A | | | | | | |
| 12 | A | 1 | 2 | 1 | 1 | | 2 |
| 13 | B | | | | | | |
| 14 | B | 1 | 1 | | 1 | | 1 |
| 15 | B | 2 | 2 | 1 | 1 | | 1 |