ESTIMATING MEASUREMENT ERROR OF SENIOR SECONDARY SCHOOL ECONOMICS EXAMINATIONS USING GENERALIZABILITY THEORY IN IMO STATE
Abstract
Scores obtained by the students (object of measurement) in examinations are affected by multiple sources of error (facets) and these scores, are used in taking relative and absolute decisions about students. With poor performance of students in Economics, there is need to estimate measurement error and dependability of the scores to find the extent of the contributions of these sources of error in examination scores. This study estimated measurement error of the 2020 SSCE Economics essay examinations using generalizability theory in Okigwe Education Zone of Imo State, Nigeria. To guide the study, two research questions were answered and one hypothesis tested at 0.05 level of significance. A random effect two-facet fully crossed s x i x r design for generalizability (G) and decision (D) studies was adopted for the study. The population of the study comprised 6,799 Senior Secondary Three (SS3) students from the 137 public secondary schools in the Okigwe Education Zone for the 2023/2024 academic session. The sample for this study comprised 680 students representing 10% of the total number of SS3 students and 68 schools representing 50% of the total population of schools in the zone. Paper 2 sections A and B of the 2020 WAEC Economics essay paper and 2020 and 2021 WAEC Economics Marking Guide were adopted for this study. The instruments were already validated and standardized by WAEC. Data collected and analyzed for research questions using the software EduG version 6.0-e based on analysis of variance (ANOVA) and generalizability. Hypotheses were tested using 95% confidence interval using standard error of measurement (SEM) and the variance components at 0.05 level of significance. The findings of the study revealed among others that, the students’ facet (?2S) made the highest contribution to measurement error in examination scores followed by the residual (?2SIR), that is, the interactions of students x items x raters; and the interaction between students x items (?2SI) in the 2020 SSCE economics examination. Based on the findings, it was recommended among others that, Generalizability analysis should be carried out by test developers, psychometricians, policymakers in assessment in the estimation of reliability to identify multiple sources of error and to reduce or eliminate measurement error, hence maximize reliability of candidates’ scores.
Keywords
Generalizability theory, instrumentation facets, students