Abstract:
The study‟s main purpose was to investigate the fair testing practices of a test
developer and teachers in a district-mandated testing programme in the Ashanti
Region of Ghana. The research design adopted for this study was the multilevel
mixed methods triangulation design. Critical case sampling technique was utilized
in selecting 3 key informants and 9 test questions. Also, the two-stage cluster
sampling technique was adopted in selecting 251 teachers from 162 public JHSs.
The main instruments used for this study included interview protocol and
document review for the test developer‟s strand and a 72-item questionnaire for
the teachers‟ strand. The reliability coefficient of the questionnaire was 0.82. The
data was analysed using qualitative content analysis, frequency and percentages,
one-sample t-test, and independent-samples t-test. The study‟s results showed that
the practices of the test developer, to a large extent, were fair to all test takers.
The results of the study further showed that in terms of test preparation,
administration, interpretation, reporting and uses made of test results, the
practices of teachers, to a large extent, were fair to all test takers. However, in
terms of grading students‟ test performance, the study‟s results showed that
student‟ grades were influenced by factors such as students‟ comportment, which
are irrelevant to the constructs as measured by the district-mandated test and thus,
are unfair. It was recommended that the test developer develops a comprehensive
test manual and share with all participating schools. The test developer should
also engage the services of experienced test administrators as chief proctors who
could serve as test monitoring teams.
