targets and decide on an appropriat

targets and decide on an appropriate action (ignore, monitor, warn, ready,
or defend). The participant’s score is based on the distance between the
participant’s decision and the correct defense decision, resulting in five
possible outcomes for each trial: hit (2 points), near miss (1 point), miss (0
points), incident (–1 point), and disaster (–2 points). After each of the 40
trials, the participants received a score and accuracy feedback (including
the correct decision).
Procedure
Participants were recruited in groups of two to arrive at staggered times.
Participants were led to believe that the study concerned distance learning.
Participants learned that they would be assigned to the role of trainer or
trainee and that they would engage in a computer-mediated training interaction
with a partner who was in a different state (in reality, both participants
were in the same lab space). Thus, trainers and trainees never
actually saw one another. To lend credence to the photo manipulation,
experimenters took a digital photograph of each participant.
The first participant to arrive was assigned to the role of trainer. During
the hour before the second participant arrived, trainers read the task
instruction manual and completed a practice and scored version of the task.
Next, the experimenter opened a digital photograph of an obese or averageweight
woman on the trainer’s computer that purportedly depicted the
trainee. Before the training interaction began, the trainer completed a
measure of training and trainee expectancy.
Roughly an hour after the trainer arrived, a second participant arrived
and was assigned to the role of the trainee. The trainee had no knowledge
of the photograph manipulation and had no reason to believe the trainer
might hold negative expectations regarding the trainee or training. The
experimenters instructed both the trainer and the trainee to put on a
combination headphone–microphone headset, and the experimenters enabled
the communication software.
Because we were interested in how expectations might mold the quality
and the content of a training interaction, the content of the training was not
regulated—trainers were given total control of the interaction. During the
training interaction, the trainers had access to (but were not required to use)
the training manual and a three-item practice version of the task that
remained open on the trainee’s computer. Trainers were instructed to
terminate the training sessions when they felt the trainee was instructed to
“the best of his or her ability.” Following the interaction, contact between
the trainer and trainee was terminated. Trainees completed a scored version
of the trained task. Trainers were not given any information regarding the
trainee’s performance. Both participants completed an evaluation of the
training and of their interaction partner.
Measures
Unless otherwise noted, the factor structure for each set of items was
determined using Cattell’s (1966) scree test method. Factor loadings were
determined using a principal axis factor analysis (with Promax rotation if
more than one factor emerged).
Trainer expectancy variables. After viewing the image of the trainee and
prior to the training interaction, the trainer completed a 12-item expectancy
questionnaire. Two items assessed expected training success (“I expect to
be satisfied with the interaction” and “I expect the training to be successful”).
Ten questions assessed the trainer’s expectations of the trainee.
Because these sets of questions were expected to be theoretically distinct,
the 2 items regarding the training were combined to form Expected
Training Success (
.78). A two-factor structure emerged for the 10
trainee expectancy items (Expected Trainee Work Ethic and Expected
Trainee Success; see Table 1).
Trainer postinteraction variables. Immediately following the training
interaction, the trainer evaluated the training interaction on an 11-item
questionnaire. Two items were included to bolster the cover story and were
not included in the analyses (“How flexible was instruction using a computer?”
and “How easy was it to communicate using a computer?”). A
one-factor structure was used on the remaining 9 items (Trainer Evaluation
of Training; see Table 2). In addition, the trainer evaluated the trainee on
the 10 qualities originally presented in the expectancy questionnaire,
generating a one-factor structure (Trainer Evaluation of the Trainee; see
Table 3).
Trainee postinteraction variables. Following the completion of the task,
the trainee responded to an 11-item evaluation of the training. The first 2
items were the same cover-story enhancers given to the trainer and were
not included in the analyses. A one-factor structure was used on the
remaining 9 items (Trainee Evaluation of Training; see Table 4). In
addition, the trainee evaluated the trainer on 10 items, resulting in one
factor (Trainee Evaluation of Trainer; see Table 5).
Trainee performance. Trainees performed 40 trials of the trained task.
To avoid a practice effect that could confound the information taught
during training (the task provided feedback after each trial), we averaged
the trainee’s first 10 trials to represent her score (Trainee Performance).
Results
Unless otherwise noted, analyses are univariate analyses of
variance on a 2 (trainee weight: obese, average)  2 (trainer
gender: male, female) between-participants design. Participant
race did not interact with any dependent variable (all Fs  1), thus
analyses collapse across race.
Hypothesis 1: Pretraining Expectations
Trainers completed a practice and scored version of the experimental
task immediately before reporting their expectations of the
training and the trainee. Thus, because the trainer’s performance
likely influenced subsequent training and trainer expectations, we
used the trainer’s performance as a covariate in the analyses of all
trainer expectations.
Expected training success. Supporting Hypothesis 1a, a main
effect of trainee weight emerged on expected training success such
Table 1
Factor Loadings for the Trainer Expectancy Questionnaire
Item
Loading
Factor 1 Factor 2
Expected trainee work ethica
I expect the trainee . . .
will be unmotivated. (R) .98 .27
will put forth effort. .83 .02
will not pay attention. (R) .71 .06
will be a disciplined learner. .67 .19
will be highly motivated. .64 .11
will be intelligent. .57 .26
will listen well. .55 .20
Expected trainee successb
will have success in the task. .02 .90
will learn quickly. .17 .76
will not succeed in the task. (R) .13 .75
Note. All questions were assessed on a 7-point Likert-type scale ranging
from 1 (strongly disagree) to 7 (strongly agree). R
item was reverse
scored.
a Eigenvalue
5.37, % variance
53.74,
.89. b Eigenvalue
1.38,
% variance
13.76, 
.84.