The big-fish-little-pond effect on grade 9 learners in South Africa

Effective schooling remains a challenge in many countries, including South Africa. In addressing this challenge, a focal area is often neglected: student emotional wellbeing. Self-concept is defined as the perception of one’s ability, as well as motivation and academic enjoyment, and is a multi-dimensional concept used in studies of individual’s behaviour. One such dimension is academic self-concept (ASC), the process by which students perform social and academic comparison within and between classrooms. ASC is associated with the social context and comparative nature embedded in schools and classrooms. It is for this reason that ASC is often used to explain the Big-Fish-Little-Pond effect (BFLPE). The BFLPE hypothesizes that higher-achieving students’ academic self-concept can be negatively influenced when surrounded by similarly high-achieving peers, but positively influenced when surrounded by lower-achieving peers. This paper adds to the literature on the BFLPE in assessing the relationship between ASC and South African Grade 9 achievement in mathematics and science using the 2015 wave of the Trends in International Mathematics and Sciences Study (TIMSS). This dataset was chosen as it makes use of standardized assessments in mathematics and science, with the addition of a student questionnaire making use of Likert-type response questions related to student perceptions on relative standing and ability, subject enjoyment, as well as student motivation. These responses were captured using polychoric principal component analysis (PCA) to allow for the construction of three separate self-concept constructs: subject self-concept, extrinsic motivation, and subject-specific enjoyment. A multilevel modelling approach was adopted to capture the relationship between self-concept and achievement at a within- and across-classroom level. In addition to estimating random intercept and slope models, a cross-level interaction model was estimated to allow for the within-classroom relationship to differ by school socio-economic status. The results of the model indicate moderate-to-strong positive correlations between the three constructs, with all three were being positively related to achievement in mathematics and science. Concurrent with existing findings, this paper finds that each of the three constructs present varying degrees or relation to academic achievement in the wealthier and poorer subset of schools.