tant mentioned that the diversity training is particularly effective if the individuals

participating are volunteers, rather than being $$forced$$ to attend the training program. So$,$ I

agreed to start the evaluation by taking the first $40$ volunteers for the program. $($Strangely$,$ it

turned out that over half of the first $40$ volunteers were either working in the Human Resources

Department of the company even though HR actually makes up about $10\%$ of our workforce$).$

This training program was conducted off$-$site for a period of a full work week. This was done at

a very nice hotel, in order to increase the level of volunteering. When they were considering

volunteering, the potential participants were told that, after the training was concluded on

Friday afternoon, they would be permitted to have family members or significant others join

them for the weekend for a little $$rest and relaxation$$ before returning to work on Monday,

with the company picking up the weekend tab for the hotel. Most of the $40$ volunteers involved

took advantage of this family offer.

When they came back, I was really interested in how this diversity training might impact our

staff, so once the trainees returned to work, I spent some time observing their interactions with

others. I think maybe the trainees and those who did not go to the training both noticed that I

happened to be around a bit more often than usual, but this was $$unobtrusive$$ observation.

After all, they should always be on their $$good behavior$$ when they are at work.

The consultant and I puzzled about the appropriate outcome measure for this study. Did the

training make a difference? What could we measure? We decided that the tone of interactions

of the volunteers with others in the workplace was as good an outcome measure as any. But we

wanted a comparison group, so we asked each training program participant to nominate a co$-$

worker in the company who was as similar as possible to himself or herself with respect to

demographic factors $($race$,$ age, gender, education level, hierarchical level, and type of position$)$

but who did NOT attend the training session. We used these $$non$-$trainees$$ as our control

group. We then tracked, for a full week, who the trainees were choosing to interact with and the

tones of those interactions. We sent forms to these coworkers of the trainees and asked the

coworkers to rate the general nature of their interactions with these trainees, including the level

of positivity and professionalism in the interaction. We also watched to see who the $$control

match,$$ non$-$trainees were interacting the most with at work, and we sent the same surveys to

the coworkers of these non$-$volunteers.