Solved
Change order of multiple options (likert) based on country/embedded data ?
+4I guess, this is a classic question... :-)
In Germany and some other parts of the world, my "customer effort score" multiple options needs to have the opposite order
80% of all countries it's:
Strongly Disagree
Disagree
Neutral
Agree
Strongly Agree
But in Germany it would be
Strongly Agree
Agree
Neutral
Disagree
Strongly disagree
ANyway I can achieve this in a smart and clever way without having to create two different questions and then do all kind of clumbsy stuff to merge the scores/results into 1 column again? :neutral:
Thanks!
Best answer by mcoverdale
I suggest using the display logic for each answer.
First you would need to add a question that asks which country they are from, then add display logic to the responses. To add display logic click on the response (e.g. strongly agree) then click on the blue button with a down arrow that is next to it and select "add display logic" from the drop down menu. With display logic you will be able to decide which responses are shown based on the question about which country they are from. You can have all 10 of the responses that you list above as possible answers (strongly disagree, disagree, etc.) but only display the first five if they indicate that they are not from Germany and only show the other 5 if they indicate that they are in Germany.
+3I suggest using the display logic for each answer.
First you would need to add a question that asks which country they are from, then add display logic to the responses. To add display logic click on the response (e.g. strongly agree) then click on the blue button with a down arrow that is next to it and select "add display logic" from the drop down menu. With display logic you will be able to decide which responses are shown based on the question about which country they are from. You can have all 10 of the responses that you list above as possible answers (strongly disagree, disagree, etc.) but only display the first five if they indicate that they are not from Germany and only show the other 5 if they indicate that they are in Germany.
+4Thanks mcoverdale!
Since I use the recode values to give it a score (strongly disagree = 0, disagree = 25 etc). I assume it wont conflict having the same option twice ?
Basically my answer options would be
Strongly disagree (RV=0)
Disagree (RV=25)
Neutral (RV=50)
Agree (RV=75)
Strongly agree (RV=100)
Strongly agree (RV=100)
Agree (RV=75)
Neutral (RV=50)
Disagree (RV=25)
Strongly disagree (RV=0)
?
Btw I have their location/languaue in the embedded data as I will use Site Intercepts for this survey.
+3I've assigned the same recode value to different answers in the same question before and it hasn't been an issue (although it probably wouldn't hurt to pilot it with a couple people first). When you download the data their answer should appear as the value (0-100) that you assigned.
+6@AbdulDezkam - Admittedly this is exactly what you said you didn't want to do, but it is really your best/easiest/cleanest option. It is also something that comes in very handy with trickier data sets.
I would make 2 separate questions that display based on logic, then use the survey flow to combine the scores into 1 field. This way it will always have the correct recode value and the logic is easy to maintain (nothing clumsy about it).
The 2 main important things are 1) make sure your embedded data names are exactly the same in both ED blocks and 2) make sure to choose the "selected choices recode" option.
This gets a little confusing in text, so I included pictures to explain, using your example.
!
!
And this is what the response data would look like. Obviously you'd want to key off of the "Score" field.
!
!
And this is what the response data would look like. Obviously you'd want to key off of the "Score" field.
!
+7@mcoverdale that is a genius solution to keep everything in one question and getting all recodes to work, thanks!
I wonder why Germans use the opposite scale?
+4Thanks @mcoverdale ! That actually works pretty smoothly. Just need to trust my embedded data then :-)
@lillianc There are countries like Italy, Germany and some Asian countries where it is pretty no-go to have the likert scale from low to high. I know it is slowly changing to "global standards" in Germany, but still not happening in several other countries.
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.