Decisions that shape education research: Investigating the implications of different operationalisations of academic resilience

Year: 2021

Author: Rudd, Georgia

Type of paper: Individual Paper

Academic resilience captures academic success despite adversity and thus is an important concept for promoting equity in education. However, despite relative consensus in conceptualisation, there is no agreed upon measure of academic resilience, resulting in different approaches to its measurement. This has complicated attempts to reconcile disparate findings in the field. Typically, a resilient sample is based on criteria that defines exposure to high-risk conditions and demonstrations of high achievement. However, currently each study nominates its own criteria. For example, high risk, often operationalised as low socio-economic status (SES), has been defined as being in the lowest 25% of the SES distribution, as well as being eligible for free school lunch. Therefore, this study explored how measures of high risk and high achievement intersect to influence the prevalence rates and strength of protective factors related to academic resilience.The sample comprised the Aotearoa New Zealand data from three international large-scale assessments: PISA (n = 53,000), PIRLS (n = 56,742), and TIMSS (n = 118,136). Two high-risk criteria (i.e., low-SES) and three high achievement criteria were applied to achievement data in reading, mathematics, and science, resulting in six operationalisations of academic resilience. These operationalisations were used to calculate prevalence rates– the proportion of high-achieving students within the high-risk sample. Logistic regression models were fitted to estimate the likelihood of being resilient. The model included students’ subject confidence, parental attitudes, and sense of school belonging, as protective factors. The prevalence rates of academic resilience varied depending on the operationalisation applied, ranging from 2.3% to 61.9%. This variation reflects the choice of measurement, as well as the actual performance of the education sector in question. Protective factors were much more consistent across operationalisations. Subject confidence was significantly predictive of resilience status, whereas parental attitudes were only predictive of resilience some of the time. This study provides support for different operationalisations based on the study purpose. Measures of academic resilience used to compare prevalence rates across contexts must be consistent in their measurement of high risk and high achievement; and identifying protective factors should be done in a culturally sensitive manner. Decisions about the operationalisation of academic resilience determine who is identified as resilient. Findings should be interpreted in accordance with how academic resilience has been measured to ensure that resilient students are accurately represented, providing greater visibility to a student group whose life experiences can offer important insights for reducing the achievement gap.