thiagi.com Freebies Email Games POLL & PREDICT
There were 20 items, and 118 people voted on them. Each participant gave their first, second, and third choices.
Each item received 3 points for a first-place vote, 2 points for a second-place vote, and 1 point for a third-place vote.
Since 118 people voted, there were (3 + 2 + 1) * 118 = 708 total points available (though the rules prevented any single item from receiving more than 3 * 118 = 354 points). The items are listed below in order from most popular to least popular. Note the tie for first place.
In spite of the fact that we wanted everyone to participate in the poll and the predictions, we only received 106 predictions. Perhaps our web form was too complicated.
We scored the predictions as follows: Each person received 3 times the score of their first-place choice plus 2 times the score of their second-place choice, plus the score of their third-place choice. A "perfect" score would have been 602 points (that is, 3*102 + 2*102 + 92). Wendy Danner missed a perfect score by 18 points, or about 3%! (She correctly pegged the top two items, but pegged the fourth-place item as the third.) Leanne M. Hoagland-Smith did very nearly as well by correctly naming the top three, but with first- and third-place items reversed.
Runners-up included Paul Kovac, Renee Whited, Mary Downs, and Don Horncastle, all of whom scored 544 points (90%) or better.
From around the world, 118 participants took part in this game. This healthy number ensures that the results are generalizable -- to the group of people who play e-mail games :-).
Interestingly, 12 people participated in the polling phase but not in the predicting phase. Obviously our instructions were not clear enough. We did fix the PREDICT page to make sure that players don't assume that it is the same page. This improved the return rate.
Three people were kind enough to write me notes (two of them fairly lengthy ones) to explain why didn't want to waste their time participating in this e-mail game. I agree with their impeccable logic.
The 20 terms identifying facilitator characteristics in the list are undefined. For example, "Listen empathically" means different things to different people. We are not sure about what meanings different people attributed to each of these items. It is not defensible to pretend everyone was thinking of the same thing when they selected the top three. If we had defined each term with behavioral indicators, we would have received more reliable results.
The context was not defined. Effective facilitation requires flexible adjustments to suit the situation. For example, an effective facilitator working in Peoria may fail miserably in Pretoria. A group of engineers may not like the same facilitation styles as a group of social workers. People working through a tightly-structured technical approach may have different preferences from people working through a loosely-structured participatory approach. And so on... If we had specified the context in which the facilitator is operating, we would have received more reliable results.
It's difficult for the human brain to compare and select among 20 items. Other approaches would have helped us filter down the list to the magical seven plus or minus two and made the processing easier (and more reliable).
Primacy and recency effects (the tendency to pay more attention to the first few and the last few items) may have impacted people's choices. We could have reduced this effect by giving people different, randomized lists.
All of the facilitator characteristics in this list have a potential for both positive and negative impact on participants. For example, too little assertiveness on the part of the facilitator will result in chaos in the facilitated session. On the other hand, an extremely assertive facilitator may not produce any useful results. All items are polarities and it is important for the facilitator to identify and maintain an optimum state along each characteristic that fits the situation. (For a practical discussion of this phenomenon, please see my article, The Secrets of Successful Facilitators.)
With all of the limitations, we still have some interesting and impressive data. I'll ask my friend the statistician to comment on the validity and generalizability to the choices.
We have a couple of follow up games planned to explore facilitator characteristics. One of them, called DEFINE, pits different definitions against each other to identify the critical features of each of the facilitator characteristics. The second one called GOOD AND BAD helps us discover the positive and negative aspects of each of these dimensions. Please stay tuned for these games.
Copyright © 2000. Workshops
by Thiagi, Inc. All rights reserved
Revised: June 30, 2000