David uses what he calls the 10–80–10 rule: 10 percent of any audience will like anything you do, and 10 percent will dislike anything you do. It is the 80 percent in the middle to whom you are really speaking. If you focus on either of the 10 percent groups at the extremes, you will likely not connect with the majority of your audience.
If you give too much emphasis to those in the 10 percent segments, the feedback you receive may be skewed. For example, at the end of the spectrum in which the "negative" 10 percent reside, though their suggestions for improvement may be valid, it is possible that the motives behind their criticism are mixed or mean spirited.
We have found that the best way to deal with the latter type of criticism and still maintain or enhance your credibility is to summarize the evaluations and ask the meeting planner to send out a copy of the evaluations to all of the participants. This allows the dissenters to see how their criticism stacks up with the majority of the participants.
One excellent technique to mitigate against the effects of extreme scores is to use Olympic scoring. In Olympic scoring, you throw out the highest and lowest scores and then average the remaining scores. Olympic scoring eliminates the fact that one unusually high or low score can bias the average, which gives a result that is closer to the true mean score (see Figure 7-8). For example, if there are fewer than 30 participants, and if one score is either much higher or lower than the average, then that score can radically skew the average ratings for any particular question. It is for this reason that we recommend Olympic scoring with groups of 30 or fewer. To use Olympic scoring, you simply list all of the scores for any particular question and then cross out the highest and lowest scores for that question. By using this process, you get a truer approximation of what the real rating should be and it is not subject to extreme scores as demonstrated below.