A 5 point scale provides a simple and effective mechanism for quantifying ratings. In our 31 years of collecting and analyzing consumer opinions, we've seen a wide variety of rating scales and have been able to form a pretty solid data-based position on the subject.
Here are the key reasons for this position:
- An odd numbered scale provides a mid-point for a neutral rating
- 2 points on either side of that neutral point provides the opportunity to differentiate between a strong positive (or negative) and somewhat positive (or negative) rating
For example, take a look at typical attribute ratings for a guest room below; we use a satisfaction scale
By only offering 2 points on either side of the neutral point, we provide a fair measure of standardization of responses. Some folks in our business like to use a larger point scale; 11 points is common (0 - 10). The main problem we see with this is that it's difficult to accurately discern differences between all the gradations on either side of the midpoint; what I may consider an "8", you may consider a "7". Tracking these metrics becomes problematic because the meaning of, for example, an average score of 7.3 is muddled.
Consider the difficulty in tracking an 11 point rating metric when you read the following real example of guest feedback- [we actually have some clients who we've designed custom feedback solutions for to their specifications, employing a 0-10 rating scale ]. This is a guest's verbatim explanation of their rating of an attribute; "No one is perfect but my 8 and 9 are good for me."
The blog of Database Sciences and its GuestInsight & ListenKeenly services.