Thursday, June 26, 2008

The Usability of Ratings

The Web 2.0 world we live in has been beneficial in creating a community of users that share their thoughts and opinions on the quality of just about anything ranging from websites, books and music, to services provided by vendors in their local.

The power is in the hands of the end-user and they now have a voice. Even more powerful and more prevalent on the web is the power of groups. A fundamental paradigm has adapted itself online in the world of Web 2.0: People tend to buy, use, do and think the way their peers do. In other words, there is a human need to fit in and be among people much like ourselves that we often follow the trend so as not to be left behind or seen as out of touch.

Ratings are often the means of expression of preference in the Web 2.0 world. It's easily understandable by all. Something rated "1-star" is often deemed as bad while something given an "A" or a "10" is deemed as good or excellent.

If you intend employ a ratings system in the design your software, website or web application be sure to choose a system that is not only understandable, but objective.

The more common system is referred to as a Likert Scale. It's a 5-point scale similar to the following:

  1. Strongly disagree
  2. Disagree
  3. Neither agree nor disagree
  4. Agree
  5. Strongly agree
Anything more than 5 points tend to muddy the waters of objectivity. The more choices, the less absolute the values will be. For example, in a 10-point scale where I say "1" is bad and "10" is the best, I may interpret "7" to be OK while another person may say "8" is also OK. Well, which is it?

The Likert Scale is the better choice when implementing a rating system, but it does have it's pitfalls. I find the 5-point Likert Scale to lack objectivity when you give the end-user a list of many things to rate. The trend usually produces a traditional bell curve where most of the choices fall under the rating of "3" - neither agree or disagree.

If you want your end-users to commit to a choice, do a 4-point scale such as the following:
  1. Strongly disagree
  2. Disagree
  3. Agree
  4. Strongly agree
By removing the middle choice you avoid neutrality and create greater objectivity. Greater objectivity means better usability to the community of users.

No comments: