Abstract:

Evaluation of educational systems and practices and in particular teachers’ instructional practices has been an age-long practice in education. An educational system is evaluated to obtain feedback on the functionality of the system in order to improve the system and for the purpose of accountability to the stakeholders. Surveys are widely used to evaluate educational systems and practices. At tertiary education level, student surveys are commonly used in evaluating teaching effectiveness and the results of the students’ evaluations are often employed in faculty appraisals and promotions. However, at secondary school level, student evaluation is not often used to appraise the effectiveness of educational systems and teachers’ instructional practices. The contention is around the reliability and validity of student evaluation of teaching at secondary school level. There seems to be a general, though undocumented, believe that secondary school students may not be knowledgeable enough to give reliable and valid evaluations of educational systems and teachers’ instructional practices. This study investigated the reliability of students’ evaluations of mathematics teaching. The study is anchored on the positivist research paradigm and followed survey research design. The participants were 100 randomly selected students from 10 secondary schools in one District in a Province in South Africa. A student survey questionnaire consisting of 27 Likert scale items was developed and validated for the purpose of the study. The questionnaire consisted of items relating to the teachers’ knowledge of the subject, lesson preparation, lesson organisation, lesson presentation, student assessment, and student motivation. The validation of the instrument involved content and construct validity established through expert judgement of the suitability of each item in the instrument, calculation of content validity ratio of each item and conducting of exploratory factor analysis. The instrument was found to be valid for the purpose of the study. The reliability of the questionnaire was assessed by calculating the Cronbach’s alpha coefficient for internal consistency reliability. An alpha coefficient of .95 was obtained from data collected on a pilot study conducted in another secondary school to test the instrument. This Cronbach’s alpha coefficient implied that the instrument was very reliable. The instrument was administered to the students between 2 days to 5 days after the teachers finished teaching the topic (trigonometry) which was used for this study. Data were analysed by computing Intraclass Correlation Coefficient (ICC) in SPSS® statistical package version 24 using one-way random effects, absolute agreement, multiple raters/measurements model at 95% confidence interval. A high degree of reliability was found between the 100 measurements; the average measure ICC was .877 with a 95% confidence interval from .746 to .967 (F(9,90) = 8.880, p<.001). This ICC value implies both high degree of correlation and agreement between the students’ evaluations of the teachers teachings. Hence, the students’ evaluations of the mathematics teachings were reliable.

It was therefore concluded that secondary school students’ can give a reliable evaluations of mathematics teachers’ instructional practices and that the students’ evaluations of mathematics teachings can be used to make decisions on the teaching effectiveness of the mathematics teachers.

Keywords:

Intraclass Correlation Coefficient, Mathematics teaching, mathematics learning, Reliability, students’ evaluation, teaching effectiveness, teaching evaluation.