I__. - International Military Testing Association

-__- ..___ -..-
A FIRST LOOK AT THE EFFECT OF INSTRUCTOR BEHAVIOR
,IN A COMPUTER-BASED TRAINING ENVIROrJMENT
STANLEY D. STEPHENSON
SOUTHWEST TEXAS STATE UNIVERSITY
Computer-based Training (CBT) research has typically focused
on comparing a CBT course with a corresponding traditional
instruction (TI) course. Compared to a similar TI course, CBT
generally, but not always, produces increases in learning and
retention while concurrently requiring less time than TI
(Fletcher & Rockway, 1986; Goodwin et al., 1986; Kuibte;eulik,
1986, 1987; McCombs, et al., 1984; O'Neil, 1986).
, CBT
results have not always been positive; there are many instances
in which CBT did not produce increases in performance or
decreases in learning time (Goodwin et al., 1986; McCombs et al.,
1984). In general there has been very little research on maximizing
performance within a CBT system (Gillingham & Guthrie, 1987).
Conversely, there is a long history of research on variables
which influence achievement in TI systems. One of the most
researched variables is instructor behavior. TI research has
produced a relatively high degree of consensus as to what an
effective instructor does versus what a not-so-effective instructor
does, with effective being defined in terms of academic
achievement (Brophy, 1986; Brophy & Good, 1986; Rosenshine,
'1983). Yet, CBT research has neglected the role of the instructor
(Moore, 1988). Therefore, little is known about whether or
not TI instructor variables transfer to CBT.
In one of the few studies which did examine the role of the
CBT instructor, Moore (1988) found that students who had positive
teachers scored significantly higher than those in classes with
negative teacher. McCombs et al (1984) reviewed various early
CBT courses and found that two factors were critical to the success
of the CBT courses. These were: (a) adequate opportunities
for student-instructor interactions, and (b) the incorporation of
group activities with individualized training. McCombs (1985)
reviewed the role of the instructor in CBT from a theoretical
perspective and developed several practical suggestions for
instructor use. One of her suggestions was that "...instructors
must have meaningful roles in the management and facilitation of
active student learning, if the CBT system is to be maximally
effective" (p. 164).
As noted from McCombs et al (1984), student-instructor interaction
was a critical factor with regard to success of a CBT system.
This is a significant finding since one of the most consistently
reported positive TI instructor behaviors is frequent but
short student-instructor interactions; i.e., an increase in student-instructor
interactions produces an increase in achievement
(Brophy, 1986; Brophy & Good, 1986; Rosenshine, 1983). Therefore,
a TI instructor behavior which may transfer to CBT is student-instructor
interaction.
The purposes of this study were two-fold. First, this study
was an attempt to begin to explore the effect of instructor
behavior on achievement in CBT. Second, this study specifically
examined the effect of student-instructor interaction in CBT.
Based on the TI instructor literature, it was hypothesized that
144
.