Thanks for all this. I'm not trying to over-engineer my teams but to assess empirically what factors might actually make a difference in group performance. The two key factors I use to make teams are grades and majors, trying to get diversity for each team. I consider other things like in-state versus out-of-state students, gender, and class level.
My thoughts on this were actually spurred by the following, which appeared in the Wall Street Journal earlier this week:
_____________________________________________
Organizational Behavior
Smart Groups & Women
Just because a group is full of intelligent individuals does not mean the group as a whole will necessarily be smart. But having more women does, according to a new study.
The researchers found that a group's collective intelligence is affected not by individual intelligence but by social sensitivity and the ability to make inferences-skills women are more likely to possess than men.
The study observed groups' intelligence over multiple tasks, such as completing puzzles or problem solving exercises, in order to come up with an overall measure to predict the groups' performance on new, novel tasks.
Adding smart people did not raise the group intelligence measure, nor did throwing dominating personalities into the mix. (In fact, loud-mouthed, strong-willed additions tended to set the group back.) Social sensitivity was the key to better performance.
The good news for employers: It's easier to make a group more intelligent than it is to make individual employees smarter.
"Evidence for a Collective Intelligence Factor in the Performance of Human Groups," Anita Williams Woolley, Christopher F. Chabris, Alexander Pentland, Nada Hashmi and Thomas W. Malone. Science (forthcoming)
_____________________________________________
Molly Espey, Professor
Applied Economics and Statistics
263 Barre Hall
Clemson University
Clemson, SC 29634
(864) 656-6401
-----Original Message-----
From: Team-Based Learning [mailto:[log in to unmask]] On Behalf Of Sweet, Michael S
Sent: Thursday, October 14, 2010 9:45 AM
To: [log in to unmask]
Subject: Re: team composition research
Dean's right--one or two (or maybe three) criteria is usually enough.
That said, a teacher here has begun using Blackboard and Excel to sort his 345 students into teams using the Big Five Inventory. (We don't have any data on it yet, but he likes how it's turning out.)
The Big Five Inventory is a personality meta-scale, derived from hundreds of personality inventories like the Meyers-Briggs, etc..
The Big Five personality factors are:
Extraversion
Agreeableness
Openness
Conscientiousness
Neuroticism
I have attached everything you need if you want to use Blackboard 9 and Excel 2007 or above to sort your students that way next semester.
1) Import the Blackboard Quiz file into your course on Blackboard (it's a Test, worth 0 points).
2) Have all your students take the Test.
3) Follow the directions on the Team Formation Handout (<- screenshot-by-screenshot).
FAIR WARNING: I cannot be available for tech support on this. If you are Blackboard and Excel savvy, it's pretty simple. But if you don't consider yourself an Intermediate level user of both, then it would be best to arrange a time with someone who is and they can walk you through it.
-M
-----Original Message-----
From: Team-Based Learning [mailto:[log in to unmask]] On Behalf Of Dean Parmelee
Sent: Thursday, October 14, 2010 6:43 AM
To: [log in to unmask]
Subject: Re: team composition research
It is hazardous to attempt to carefully engineer the team formation process, especially if you want to keep it transparent to all. Select one or two criteria that will help with wealth distribution and diversity - however defined. Then make it random - stratified random sort. Dean
Sent from my iPad
On Oct 13, 2010, at 1:52 PM, Larry Michaelsen <[log in to unmask]> wrote:
> A number of years ago, I did some analysis of the relationship between gender mix and RAT scores using data from several hundred teams. The differences were extremely small. The only observable difference (I think it was even statistically significant) was that all-male or all-female groups did ever so slightly better.
>
> Larry
>
>
> -----
> Larry K. Michaelsen
> Professor of Management
> University of Central Missouri
> Dockery 400G
> Warrensburg, MO 64093
>
> [log in to unmask]
> 660/429-9873 voice <---NEW ATT cell phone
> 660/543-8465 fax
>
>
>
>>>> Molly Espey <[log in to unmask]> 10/12/10 9:26 AM >>>
> Has anyone quantitatively analyzed how various components of a team contribute to team success? I'm interested in those components that we as teachers might have access to information about the students BEFORE forming teams: gpa, gender, class level, age, in-state versus out-of-state, major.
>
> I'm also curious: is it common or not for faculty to have access to this sort of information about students enrolled in their classes? We have something called the "Student Data Warehouse for Teaching Faculty" that provides lots of data but just about students enrolled in your classes for the current semester (available shortly before the semester starts in easy to use Excel format).
>
> Thanks in advance for any responses!!
>
> Molly Espey, Professor
> Applied Economics and Statistics
> 263 Barre Hall
> Clemson University
> Clemson, SC 29634
> (864) 656-6401
|
