Take the 2-minute tour ×
User Experience Stack Exchange is a question and answer site for user experience researchers and experts. It's 100% free, no registration required.

Say we have a list of items which are 'action items', in the sense that each item represents a task for the user.
The list is compiled automatically in a way resembling a news-feed - there is some heuristic algorithm that tries to guess which items will be most relevant to the user.

For each item we want the user to have the following options:

  • A way to designate that the task is 'done' (e.g. a button that says "dismiss").
  • A way to give us feedback about that item, specifically by choosing between two options that generally mean "I want to see more items like this one" and "I want to see less items like this one".

As you might have guessed, the feedback is used to improve the heuristic algorithm. Therefore the feedback is both important to us as the developers, and will help improve the user experience over the course of time.

Should we force the user into giving us feedback?

By force I mean having only two buttons on each item, where both of them perform the action of 'dismissing' the item, but one of them represents positive feedback and the other one negative. This way the user has no way of dismissing the item without providing feedback for it.
If we take this route: What could we use as the text for these buttons?, i.e. is there a short and clear way to express the content of "Dismiss this item and show me more like it in the future"? (and the same with "less").

If we don't force the user into giving us feedback, meaning we have a 'dismiss' button which is neutral in terms of feedback, what would be a good way the encourage the user to give us feedback?
For example: should we put the feedback buttons on each item, in addition to the 'dismiss' button, or should we ask for feedback only after the item is dismissed?

One last thing: can anyone point us to existing interfaces that deal with similar problems?

share|improve this question
2  
No no no no. Please No. Forcing users to provide feedback when they don’t want to won’t give the data quality you need. –  Benny Skogberg Sep 9 '13 at 8:25

2 Answers 2

up vote 4 down vote accepted

Two buttons with the same functionality (Dismiss) for a user and some different meaning for you (Feedback) could lead to some issues:

  • user has no "neutral" choice, so you do force him to take some desicion (like vs dislike).
  • mixing two function in each button (Like+Dismiss and Dislike+Dismiss) could lead to false results. The only meaningful operation for a user is Dismiss, so he could ignore Like vs Dislike and press the nearest button to Dismiss. And this leads to wrong algorithm work.
  • having forced to choice (Like vs Dislike) some users could afraid of not they assess the item but the system assesses them based on their choice. That could lead to unexpected behavior.
  • two buttons make users to think more. "Don't make me think", remember?

So, the conventional arbitrary choice could work better:
enter image description here
or
enter image description here

share|improve this answer
    
+1 You pretty much convinced me that the 'dismiss' and 'like' actions should have separate buttons. However, I don't think the terms 'like' and 'dislike' are what we should use here. Some of the action items might have a negative meaning, such as an item alerting the user to some problem. What we're looking for is something that means "show me more/less items like this one". Any idea how to convey this message in one or two words? –  Joe Sep 9 '13 at 11:06
    
@Joe – I've updated the answer, please, take a look. For me, as non-native English speaker, such kind of message would be clear enough. Also it doesn't contain emotional assessment of an item content, like "Like vs Dislike" does. –  Alexey Kolchenko Sep 9 '13 at 11:39
    
additionally, you could still display the thumb up/down button after the user clicked dismiss by only minifying the action for the time beeing (eg: only show title, gray out content, etc) that way he can still provide feedback which might entirely clear the item. on the other hand if he don't provide feedback the item should not reapper the next time the list is shown –  Mathieu Sep 13 '13 at 21:00

If a user need to do some work with task, maybe it is better for him to evaluate task after (s)he's done. Does "I haven't tried this yet, but give me more of these tasks!" way of doing things OK in your situation?

Maybe it is better to provide three buttons like: [Done Up] [Done] [Done Down]

Benefits are: just one click for user, and you don't force a user to make a decision when he's unsure.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.