5 Replies Latest reply on Aug 11, 2011 2:18 PM by Andrew Kratz

    Rating

      I received this question from a user - anyone know the anwer to this?  Thanks in advance!

       

      Some content types can be rated by the viewer, I know the action of rating content is captured, but is the rating itself captured ? What I'm wondering is: based on the rating information captured, can we determine if the user thought the content was useful?

        • Re: Rating
          crossman

          The rating is captured, though it may be difficult to pull that data out of the data warehouse (a whole separate topic).

           

          >based on the rating information captured, can we determine if the user thought the content was useful?

          Depends on your definition of useful. Is three stars or higher considered useful? Four and above?

            • Re: Rating
              jhoehne

              Kevin Crossman wrote:

               

              Depends on your definition of useful. Is three stars or higher considered useful? Four and above?

              That is one of the keys right there. The definition of useful, and then you have to hope the user is applying the same definition of useful. I think it could be challenging to measure usefulness in one dimension. It seems like it could be a formula of views, ratings, comments, reviews, links, etc.

               

              Also the variable nature of the content itself presents a challenge. Jive defines the stars with "My rating: Poor" through "Exceptional" Perhaps those could be customized more accurately suggest that user evaluate the "usefulness" or whatever dimension you want.

               

              I think Netflix is a good example of this, the star ratings on the movies ask how much you "Liked It", not if you thought was an artistic piece of cinema, or a travesty of film, or if the star was good or bad in it. You can imagine the dilemma I face when I see a movie with an interesting concept that was poorly done.

               

              Perhaps there's an opportunity to customize and expand the ratings like Wikipedia has now.

                • Re: Rating

                  Hi,

                  To add another wrinkle, with 5.0 we can also Like it.  Getting anyone to add a rating has been a  challenge.

                  Mike

                    • Re: Rating
                      John Schwiller

                      We had a huge debate within our internal community about content types where we have both like and ratings.

                       

                      I put the point that these were the same dimension or pretty close, and asked if it made sense that you could like and rate 1 or 2, or rate 4 or 5 and not like.

                       

                      Several users tried to convince me that the dimensions are different, ie you like a point someone is making but think they made the case badly - so rating it 1 or 2.

                       

                      In the absence of a lot of complex words on screen or a context sensitive '?' taking you to guidance, I think there's a high risk that most new users will not get the multi-dimensional nuance, will treat the like and rate as the same, and will use like for speed and rate when they have time to spare.

                       

                      No doubt this has been discussed many times before here?

                       

                      When I have time I'll probably scale back on the content types we let people rate.

                       

                      However re the issue of individuals using their own judgement about rating, I guess after many ratings have been done it all averages out so you are left with a lot of content and a useful (?) relative ranking of the set of content.

                • Re: Rating
                  Andrew Kratz

                  I recall when we were in the RFP process to select a vendor (Jive won of course), one vendor joked,  "Ratings are not used that much, but you need them to get through the RFP process."

                   

                  I always thought that minimally would use them to help differentiate my own content.  I hate it when I go to a one of my hard drive folders and don't have a way to flag, rate or put meta data around which files are important to me.  But as of now, I have not done that.  I think it might be because every time I create a piece of content I think it is awesome and can't take the long view to differentiate it....or maybe all my content is just awesome