Abstract
Examinees who skip questions or have prior knowledge of the test content will
generate an inaccurate and invalid test score. Assurance that test scores reflect the
examinee's abilities is challenging for test administrators. Thus, this study examined
students' response aberrance in the unified mathematics test based on their test
anxiety and mathematics ability. The research design used was a cross-sectional
descriptive survey. Through a multi-stage sampling procedure, 550 subjects were
sampled. Data were collected using the Mathematics Achievement Test (MAT) and
Test Anxiety Inventory, which had reliability coefficients of 0.84 and 0.82. Using
the person fit package implemented in R programming language version 4.0.1
software, response aberrance (W* and B*) indices for each student were computed.
A 2 x 2 x 2 factorial multivariate analysis of variance (MANOVA) was used to
analyse the data descriptively and inferentially. Results showed that high school
students have a response aberrance in mathematics with a mean of 0.96 and 4.83,
which are greater than 0.50 for both aberrance indices. The study found a significant
interaction between student ability and test anxiety, with a significant main effect on
student ability and test anxiety. The results indicated a significant MANOVA, and a
follow-up analysis of the variance test showed a statistically significant difference in
the response aberrance as measured by W* and B* on students’ ability, with low
ability having the highest mean scores. The study also indicated a significant
difference in the response aberrance as measured by B* on test anxiety, with high
anxiety having higher mean scores than low anxiety. The study concluded that high
school students' ability and test anxiety affect the severity of response aberration in
Mathematics.