Team Testing for Individual Success and the Development of Interpersonal Skills Among Aspiring School Leaders 133Participants
One hundred eighty graduate-level school leadership students participated in the study.
These 180 students were selected based on their enrollment in 1 of 10 sections of the course
Educational Research. Section sizes ranged from 8 to 27 students.
Data Collection
As part of this study, 180 students’ scores on a criterion-referenced final exam and quali-
tative survey data from a subgroup of these students were collected and examined. Although
the exam is thought to have content validity, its statistical reliability was not determined.
The treatment under investigation included two-person instructor-assigned and self-
selected pairs. The control group (n = 133) for this study included students in sections of Edu-
cational Research taught by the same instructor and given the same exam taken individually.
The three treatment groups used in this study were also taught by the same instructor and
were given the same exam. Treatment Group 1 (n = 20) included students assigned to two-
person teams by the instructor. Treatment Groups 2 (n = 5) and 3 (n = 22) included students
who chose to work individually or in pairs on the exam, respectively (see Table 1).
Table 1. Test Format.Group Test format Number of students
Control individual test 133
Treatment Group 1 paired test 20
Treatment Group 2 individual test by choice 5
Treatment Group 3 paired test by choice 22
An open-ended qualitative survey was given to all students in Treatment Group 1 (n = 20).
This survey consisted of four questions soliciting feedback on the participants’ experience in
the team-testing format. These questions were as follows:
- Would you want to do a team test again in the future? Why or why not?
- What do you see as the benefits of a team test?
- What do you see as the drawbacks of a team test?
- Do you have any suggestions for ways to make the team test better? If so, explain.
This survey instrument was given after the close of the course and data were not collected
or read until after student grades were posted.
Data Analysis
Student scores on the exam were entered into SPSS. Scores on the exam could range from
0 to 22. Mean scores and standard deviations were tabulated by test format (individual, paired,
individual by choice, and paired by choice). A series of t tests was then run to determine if the
scores of the treatment and control groups were significantly different. Effect sizes using
Cohen’s d were calculated for any significant findings.