Ranking questions - changes to data structure in responses table

Hi @Ante Thanks for stopping by the community. Could you explain a bit more what is making it impossible to view the data? Any screenshots you can provide would be helpful. 

In the meantime, I’d also suggest taking a peek at the workaround below, which might help you see the data in whatever format you choose!

• Create semi-automated reports with Typeform and Google Data Studio Tutorial

Hi Liz, it’s not impossible to see data, but to analyze it.

Downloadable response table reports answers as a string. Like Option 2, Option 3, Option 1, Option 4. This is not structure that’s ready to be summed up, averaged, or the frequency of top 2 ranks found.

Summary of responses shows a nice and neat chart and average ranks, but I want to slice my data per subsamples.

The tutorial you sent doesn’t cut it, because still I’d need to find some way to parse and structure the data so that it can be easily statistically manipulated.

(Which is beyond my skills. I was hoping to simply type in “=average() in a cell and be done :) )

The solution would be to change the structure of reported data in the table. Instead of a string, researcher would get one column per option in the ranking questions. And in the row, we’d get respective ranks given to those options by a respondent. Then it’s much simpler to make a pivot table.

Any thoughts on that, @john.desborough? 

Hi @john.desborough thank you for your response.

My problems start exactly where you’ve finished your answer: “once that is done you can do the analysis that you want to do”.

Could you please walk me through the steps you’d take after splitting strings into nine columns? (In my case, there were nine options to rank).

____

It still seems to me that this problem would be solved if Typeform reported the results in a ready-to-analyze form instead of a string. @Gabi Amaral do you have any info if this is in the product roadmap?

@Ante - it all depends on what you are trying to analyze, your thesis

the ‘values’ will be split out into columns 1 through 9 (in your case) with the highest ranked item in col 1 and down the line. note that even if it is numeric ie the answer is “3”, it will be  a text field that is created when the split is done. 

then you can do ‘normal spreadsheet things’ like frequency/counts of the items that people ranked first.. ranked last.. 

what do YOU want to achieve with the analysis??  You will need to figure out how you want to use the data.. 

des

I’m interested in average ranks of different options. To see if they’re different for different subgroups.

I’ve been able to get those averages out of the sheet. But it’s manual work that needs to be redone if some additional answers come by.

Had I had each option as a column with their rank by respective respondent, I’d be able to the analysis in a few minutes.

I was hoping you might have some neat trick up your sleeve @john.desborough :) 

All in all, this whole ordeal with prepping ranking questions for the analysis feels like Typeform is offloading the hard work onto its users.

@Ante - if your ranked choices are B,D,J,F,E,A,C,H,G,I  then those choices would appear in the columns left to right in that order using the split function. 

Unfortunately a lot of the functions work on columns and not on rows, from an analysis point of view. 

you would have to do something like this in the spreadsheet: 

• create a separate tab
• set up some sort of macro that would take each new row of input on the main tab and then create a column with the header of the email address or some other unique identifier in the data row
• then grab the “split” data elements from the row and use the arraytranspose() function to then write this set of data into the column with the email address that matches the email address in the header row, starting at row 2
• then you could do the analysis.. 

but all of that is outside of typeform and not really what this community is really designed to discuss ie Google Sheet analytics lol.. there are plenty of communities and fora that discuss those topics.. 

ideally you need to start with the outcome you want ie what decisions do you need to make and what information is required in ‘the report’ that you will use to make those decisions. Then you need to do the walk back in what data is required to be manipulated to give you the output you want - and can you make the tool you are using to do the analysis (maybe it’s Tableau and not Google Data Studio or Sheets)  do what you need it to do. Then you need to design the questions that will give you the data you need to do that work. If you can get it all in one tool ie Typeform then great. If you can’t get all the data in the format you need from Typeform, how much of it CAN you get out of typeform and what do you need to use to get the rest into the analysis tool. 

but you know all that.. Typeform is simply one tool in the ecosystem that supports business processes in this app-based society in which we live and work. it won’t solve everything but if it moves you 80 of the way to where you want to go and you have yet to find another tool that does that, then start there and refine the rest of the ecosystem around it.. 

des