Topics of grade retention in Portugal through the PISA: quality and equity

Abstract

In the framework of measures required for the Europe 2020 strategy – and with regard to the prevention of early school leaving based on frequentist definition of probability and multilevel logistic regression model, we identify the risk groups of grade retention. PISA 2012 data analysis suggests that early retention is a strong predictor of late retention and that the probability of early retention is 35 times higher in the first tenth of the distribution of the students’ socioeconomic status in comparison with the top tenth. The odds ratio calculated using the multilevel logistic model indicates that male students are on average 1.7 times more likely to be retained than female students; for each additional unit of standard deviation in the socioeconomic status of the student, the odds ratio of non-retention increases 1.5 times, and four times per unit of school socio-economic composition. With regard to mathematics self-concept, for each additional unit of standard deviation, the odds ratio of non-retention increases 1.5 times. Bearing in mind the international debate on the long-term effect of grade retention on individual careers, and knowing that Portugal is among the OECD countries with the highest degree of inequality in income distribution, the interpretation of the results recommends the design, planning and implementation of intervention programs aimed at improving the learning in risk groups, with the 1st and 2nd cycles of basic education as priorities.