- Posts: 4
- Thank you received: 0
Welcome to the LimeSurvey Community Forum
Ask the community, share ideas, and connect with other LimeSurvey users!
Analysis of Randomized Question Position
- JohannRoeben
- Topic Author
- Offline
- New Member
Less
More
3 years 4 months ago #209232
by JohannRoeben
Analysis of Randomized Question Position was created by JohannRoeben
Hello there,
I conducted a survey in which I wanted to randomize the question order within a specific group, so in "Randomization Group Name" i wrote "1" for that group. Now I want to analyse, if the order of questions has had an effect on the answer. Where do I find information regarding the positioning of questions per respondent? (eg. Respondent 1 had question order 2-3-1) I do not see it in the data, is it possible?
Thanks already,
Johann
I conducted a survey in which I wanted to randomize the question order within a specific group, so in "Randomization Group Name" i wrote "1" for that group. Now I want to analyse, if the order of questions has had an effect on the answer. Where do I find information regarding the positioning of questions per respondent? (eg. Respondent 1 had question order 2-3-1) I do not see it in the data, is it possible?
Thanks already,
Johann
The topic has been locked.
- Joffm
- Online
- LimeSurvey Community Team
Less
More
- Posts: 12916
- Thank you received: 3976
3 years 4 months ago #209234
by Joffm
Volunteers are not paid.
Not because they are worthless, but because they are priceless
Replied by Joffm on topic Analysis of Randomized Question Position
Hi, Johann,
nowhere.
Because you assume that there is an effect, you do the randomization to avoid a bias.
Nothing else.
If you wanted to analyze the effect of the order of the questions/groups you should have included some more things, like the use of ".qseq" or ".gseq"
E.g. questions of type equation that store this, {Q1.gseq}
As you see here
manual.limesurvey.org/ExpressionScript_-...#Access_to_variables these properties return the sequential number of a question/group.
Or you should have displayed the groups in predefined orders to be able to analyze this effect.
Sorry, but you did not test your survey. Then you'd have seen that there are no informations about the random order in the dataset.
Joffm
nowhere.
Because you assume that there is an effect, you do the randomization to avoid a bias.
Nothing else.
If you wanted to analyze the effect of the order of the questions/groups you should have included some more things, like the use of ".qseq" or ".gseq"
E.g. questions of type equation that store this, {Q1.gseq}
As you see here
manual.limesurvey.org/ExpressionScript_-...#Access_to_variables these properties return the sequential number of a question/group.
Or you should have displayed the groups in predefined orders to be able to analyze this effect.
Sorry, but you did not test your survey. Then you'd have seen that there are no informations about the random order in the dataset.
Joffm
Volunteers are not paid.
Not because they are worthless, but because they are priceless
The topic has been locked.
- JohannRoeben
- Topic Author
- Offline
- New Member
Less
More
- Posts: 4
- Thank you received: 0
3 years 4 months ago #209237
by JohannRoeben
Replied by JohannRoeben on topic Analysis of Randomized Question Position
Thanks for the quick answer! You are right, I should have tested the survey for this. It is the first time I am using it so I guess this is what we call "learning money" in Germany...
But as you said, since it is randomized, the effect is neutralized anyways. I just wanted to make sure it worked. Next time I will equations.
Thanks, Johann
But as you said, since it is randomized, the effect is neutralized anyways. I just wanted to make sure it worked. Next time I will equations.
Thanks, Johann
The topic has been locked.
- holch
- Offline
- LimeSurvey Community Team
Less
More
- Posts: 11655
- Thank you received: 2742
3 years 4 months ago #209247
by holch
I answer at the LimeSurvey forum in my spare time, I'm not a LimeSurvey GmbH employee.
No support via private message.
Replied by holch on topic Analysis of Randomized Question Position
War da nicht mal was mit dem "seed" Wert und dass man über diesen die Reihenfolge rausfinden kann?
Habe da leider nur was im Hinterkopf.
Hier wurde das diskutiert: forums.limesurvey.org/forum/german-forum/118416-variable-seed
Habe da leider nur was im Hinterkopf.
Hier wurde das diskutiert: forums.limesurvey.org/forum/german-forum/118416-variable-seed
I answer at the LimeSurvey forum in my spare time, I'm not a LimeSurvey GmbH employee.
No support via private message.
The topic has been locked.