# TRANSFORM ROW WISE DATA INTO ONE COLUMN CONSECUTIVELY based on the number of values in each row

I've learned how to use filters to organize row-wise data into columns and am taking that to heart in this spreadsheet! But I've hit a major snag.

If you look at the example I've attached here, you'll see that I've successfully built out columns S through X - and everything is working with either filtering or SQL queries to data located elsewhere in this workbook - some from this sheet, some from other sheets . and it all aligns perfectly!

But here's the problem. As you can see in columns Z through AG, the answers to each survey columns are arrayed row-wise. When I try to filter columnns AA through AG and place the results in column V, I run into two problems: 1) it stacks everything from AA, then AB, then AC etc. That's exactly wrong! I need it to stack Row 3's values from AA through AD as I have done in V3 through V6. But the problem is that I haven't figured out how to do this using an array formula - and the numbers of answer options for each question differ - anywhere from 2 to 7. I've got a helper column in Z which shows how many data points there are per question - and I was hoping to get that to operate as a control to determine how many datapoints I would need to write into column V. Couldn't figure that one out. So - I wrote a ridiculous formula to process the control in column Z in order to determine how many cells (from 1 to 7, AA to AG) to write out, transposed. But then, I can't do that in a single column using ARRAYFORMULA - which defeats the entire effort.

Everywhere you see a light green cell, there's a formula in there.

So - here are the formulas:

Column S (easy enough): `=query(\$O3:\$Q,"select O where O > 0 order by O asc", 0)`

Column T (also easy enough): `=arrayformula(if(isblank(\$S3:S),"",if(\$S3:\$S<>\$S2:\$S,vlookup(\$S3:\$S,ExitPollSETUP!\$C28:\$E,{2,3},false),"")))` ---this is a simple breakpoint tool to only label the first row in each group so the labels don't repeat (and it also pulls in column U).

I'll jump over Column V for a second - that's the "problem area".

Column W and X are simple queries retrieving already-filtered data which matches up correctly with Column S. (I've validated all of this and it's working perfectly). Here's that formula: `=query(\$O3:\$Q,"select P,Q where O > 0 order by O asc", 0)`

Now - for Column V - you can see that I want to fit in exactly the columns AA through AG, transposed into a single column, but only based on the number of answers, which is the control in Column Z. I want to do this recursively so that it all falls in as shown in the document. But I couldn't figure out how to do that in an array formula - which is the big problem I'm having - so instead I've had to paste in the same dumb formula for each new question and that's exaclty what I don't want to do!

Here's the formula currently being used for each answer in Column V: `=if(\$T3<>"",IF(\$Z3=7,transpose(\$AA3:\$AG3),if(\$Z3=6,transpose(\$AA3:\$AF6),if(\$Z3=5,transpose(\$AA3:\$AE3),if(\$Z3=4,transpose(\$AA3:\$AD3),if(\$Z3=3,transpose(\$AA3:\$AC3),if(\$Z3=2,transpose(\$AA3:\$AB3),"")))))),"")`

See what I'm doing here? Selecting the number of cells to transform using the helper column.

So - bottom line - how can I go a row at a time through the entire set of AA to AG and then transpose all of it - left to right per row, and continue moving through the data left to right per row, and place it correctly in the first row of each answer, in Column V ????

I'm bewildered by this - and would LOVE an answer. Thanks for reading this.

Gary K.

(BTW-I'm running this with an API query to a highly secure data set and can't share the sheet because the data is being generated with a KEY. That's why I've cut out a sample of the sheet and pasted it here and provided the relevant formulas above)

This formula takes each question and its standard answers in a single row, and re-formats the data to the desired output.

`=query({query({A:I},"select Col1,Col2,Col3,Col4");query({A:I},"select Col1,' ',' ', Col5 where Col5 is not null LABEL ' ' '', ' ' ''",0);query({A:I},"select Col1,' ',' ',Col6 where Col6 is not null LABEL ' ' '', ' ' ''",0);query({A:I},"select Col1,' ',' ',Col7 where Col7 is not null LABEL ' ' '', ' ' ''",0);query({A:I},"select Col1,' ',' ',Col8 where Col8 is not null LABEL ' ' '', ' ' ''",0)},"where Col1 is not NULL order by Col1")`

• The query is nested, each additional response separated by an array separator (`;`)

• For the second and subsequent responses, there is a "where" to test for the existence of the response.

• The formula (as written) assumes that the maximum number of responses = 6. But if the maximum is (or becomes) greater than 6, then it is a simple matter to add an extra query into the nested array.

SAMPLE

• Hi. Thanks so much for sending me your very helpful and logical code. Fascinating, really. Problem: I'm getting an array literal error. Here's a link to the spreadsheet. docs.google.com/spreadsheets/d/… Please, if you can, take a look at the last sheet, Filter2. The code you provided is in L2. The four red columns are intended to be where the output of the nexted query is written. Thanks so much! -Gary Dec 3, 2022 at 7:58
• Perhaps you mean "FILTER:T2" BUT... just had to refresh json on "ExitPollSETUP". Note: your formula in "ExitPollSETUP!C27" is a bit convoluted. Refer my simplified formula `=if(\$B\$6=true,transpose(importJSON(\$B\$4,if(\$B\$7=true,B5,""))),"Don't load JSON")` (using checkboxes) on "tedinozJSON". Dec 4, 2022 at 1:55
• Note:, Q#2074 has 13 responses so you need to expand the formula by adding Col13, Col14, Col15 & Col16. Refer sheet="tedinozFilter:T2" for an example. Dec 4, 2022 at 1:57
• This is all incredibly helpful. Wow! And thanks for catching my error on filtering an insufficient quantity of columns. I actually thought it was my DATA filter that was wrong producing the mismatch - and was digging around - not realizing that it was actually the questions and answers where I had come up short. Dec 4, 2022 at 3:42
• You have saved me with this - can't tell you how much I appreciate your guidance! Dec 4, 2022 at 3:42