We've been evaluating speaker performance, quality of sessions, satisfaction with events and almost every other measurable/metricable activity for the past few years on the One to Nine scale. And believe it or not, the Average bar is not 5.0 but 7.0 especially for Speaker ratings and overall satisfaction. 8.0 is good, 8.5 is Top 10. Obviously on a scale of 1-9, it's next to impossible to get a 9.0.

Recently i've been seeing a lot of other Microsoft events that started using the 1-5 scale. Those of you who attended MIX06 would know. I can see the merit of using the 1-5 scale especially when it is much simpler to understand. Although one might argue that a 1-9 scale provides more detail, I can safely say that is not true. How many people can't actually tell the difference between 3 and 4. or 6 and 7? usually its 1-2 for bad, 7-8 for acceptable, and 9 for impressive/excellent. With that in mind, wouldn't using a 1-5 scale be sufficient?

So here lies the rub,

How are we going to link the new 1-5 scale to the old 1-9 scale? you can't simply divide or multiply scores.

for example, a great speaker on the 1-5 scale might score 4.5. If you scale up by 2 and then multiply by 0.9, you'll get 8.1, where if you rate that same presentation on a 1-9 scale, the speaker might score 8.4 - 8.5.

so, what do you all think? 1-5 or 1-9?