Preventing duplicate responses

I ask participants to enter 10 attributes, one per screen. Is it possible to prevent them from entering the same attribute twice? Thanks!

Not super-easily. But..

Do you know how to use a [I]responses.xls[/I] file (from the [I]Advanced Features[/I])? If so, you could probably create a conditional statement in Excel to test whether the given response is equal to a previous response. And that if such proves true then you could set the skipto value to the current variable name. Basically, that means you wouldn't be able to proceed until you entered something new.

You could also create a custom item via HTML.

Thanks for your reply -- I do know the basics of how to use responses.xls, but I'm not familiar enough with Excel to know how to create a working conditional statement for this.

Also, in the control condition, they still list 10 attributes, but in later questions, they are given two different attributes (in the experimental condition, they are given 2 of their 10 attributes). The two different attributes are randomly chosen from a list of the 25 most common attributes, which means that it's possible there is overlap between the control list and the 10 attributes they listed. I need to make sure that the two control attributes they receive are different from the 10 attributes that they listed. I've figured out how to write functions to select two random attributes from the control list, but that's as far as I've gotten.

Would you be able to give me a sample conditional statement that I could modify for either of those two problems? Thank you!!

Sure. Can you start us off? Create a very simplified version of the que you are trying to make. Say where they are instructed to list 4 attributes only. Include a responses.xls file with the appropriate variables and then zip and post those two files here as an attachment. I'll see what I can do with the responses.xls file so that they can not repeat a response. Then, hopefully, you will be able to take that logic and extend it to your larger scenario with 10 attributes and the control group also. At least it will give us a place to start and the simplified sample files will let me get a better sense of what you are trying to do.

Thank you very much! The attached questionnaire file is for an experimental condition. In a control condition, something like <randctrlword1> and <randctrlword2> would replace <maxideal1> and <maxideal2> for the essay questions.

Please let me know as soon as possible -- the grad student I am doing this for needs to run her experiment next week, and she just recently changed her experiment, otherwise I would have posted my question much earlier. Thanks again!

Good news and bad news. I've attached a responses.xls file revision that shows how you can determine a skipto value depending on whether a response is a repeat of a previous one. Check out the code in the cells from C2:C5. Then enter different values in B2:B5. See how the skipto values are affected by repeating an attribute. If the code looks complicated, there is a nice thorough explanation of it under "IF" in the Excel Help guide.

The bad news is that MediaLab will not accept the current item's variable name as a skipto value if it's generated in the responses.xls file. It only works if you do it directlyin the QUE file which isn't an option here. I was unaware of this and will make a note of it for the next version. In the meantime, a workaround might be to create a repetition of each item, e.g., ideal2b, ought2b, etc. that starts off with "You already used that attribute, please try again". You could then adapt the skipto logic in the attached responses.xls file to SKIP those items if there is in fact NO repetition. Certainly a little more work, but it has the advantage of being able to give the subject feedback about why they are having to try again. If you wanted to be extra safe, you could include 2 or 3 repetitions.

Anyways, this wasn't the super-fun response I was hoping to provide but hopefully the conditional logic alone will provide something useful for a solution of some kind.

Thank you!!!! That was very helpful.