10
Wallhead and Ntoumanis
Fox, & Underwood, 1995) on teaching styles and student motivation has utilized
the CBAS in order to examine the consistency of teacher behavior across differing
curricular programs.
The transcripts from the videotapes were coded by the researcher and one
other person who was blind to the purposes of the study. The second observer was
trained prior to the intervention to identify the categories of teacher behavior rel-
evant to the adapted CBAS. Due to the number of categories of teacher behavior
included in the CBAS, an extensive training was undertaken. Sample 10-min vid-
eotape segments of teaching episodes that were not part of the intervention were
filmed. During observations of the first two teaching episodes, exemplars of each
defined category of teacher behavior were identified. Subsequent segments of teach-
ing were then observed and coded independently until a criterion level of 80%
interobserver agreement was reached for each segment. After this criterion was
met, each intervention sample lesson was coded independently. Observations were
recorded by writing down every coded behavior the teacher exhibited during the
two sample lessons. Interrater reliability was found to be 0.88 for the Sport Educa-
tion lesson and 0.82 for the traditional lesson observation.
Design and Procedure
Due to the use of intact classes, it was not possible to make a random assign-
ment of participants to the two levels of the independent variable (Sport Education
and traditional programs). Therefore the study utilized a nonequivalent control
group design (Campbell & Stanley, 1963) which signifies that the groups may be
nonequivalent prior to intervention due to some systematic difference between the
two classes. The dependent variables were the student motivational indices of en-
joyment, perceived effort and perceived competence, student perceived motiva-
tional climate, achievement goal orientation, and perceived autonomy. All variables
were assessed before and after the intervention.
In order to reduce investigator selection bias, prior to the start of the inter-
vention the Sport Education curriculum model (n = 26) and a traditional approach
model (n = 25) were randomly assigned by an assistant to the two intact groups.
One week prior to the basketball program, all students completed a series of baseline
questionnaires in a quiet classroom setting. They were informed that they would
be taking part in a study that “would look for new ways to teach PE,” but they were
not informed of the exact purposes of the study. The questionnaires took approxi-
mately 20 minutes to complete and were administered to each class separately.
The students were encouraged to be as honest as possible and were assured that
their responses would be confidential. At the end of the 8-week intervention all
students again completed the same questionnaires.
Sport Education Intervention. In the experimental condition the teacher
implemented the Sport Education model. The intervention model followed a three-
phase format: a teacher-directed skill development phase, a preseason scrimmage
phase, and finally a formal competition phase. The teacher-directed skill develop-
ment phase involved 3 lessons, during which students led warm-ups but were given
teacher instructions on the generic skills of scoring, passing, and dribbling. The
preseason phase also involved 3 lessons and was designed primarily for students
to work in their teams with practices led by the student-coach and facilitated by the
teacher. In this phase the students took responsibility for refereeing and the choice