Joffm wrote: But what do you expect as answer?
Do you really think respondents will enter something and then rate "Don't like it".
Normally they will only enter things they like.
True, they would normally only write things they like at least a little. However, they could still compare their other-option to prefilled options; "looking reindeers" could be given as even better than the midnight sun, or then as additional "it was nice too" -thing.
In this case I am not creating a survey; I just try to implement what the research group wants.
Metaquestion: Should I open a new question about removing the header block from one question?
