The resilience concept contains two critical conditions, namely, exposure to significant threat or severe adversity and achievement of positive adaptation despite major assaults on the developmental process (Kiswarday, 2012). In Educational Equity Research, children who are from socioeconomically disadvantaged families yet manage to turn around their academic performance by defeating the severe challenges and setbacks in their education are academically resilient. The assets and mechanisms that these children rely upon when facing the adversities in academic contexts is important to study as it has strong policy implications. However, academic resilience is understudied, especially in the aspects of variation and changes across different countries and over time. Limited research however showed, countries with a low proportion of disadvantaged children tend to have a high proportion of resilient children and vice versa. High performing countries usually also have a high proportion of resilient children, and academic aspiration is then an important factor predicting the probability of being resilient (Sandoval-Hernandez & Cortes, 2012; OECD, 2011).

Using Australian high school students’ self-reported questionnaire information, Martin & Marsh (2006) was able to identify a 5-C model of academic resilience, namely, confidence (self-efficacy), control,  coordination (planning), commitment (persistence) and composure (low anxiety). They conducted a path analysis and found that the 5-C dimensions highly predict student’s psychological and educational outcomes such as enjoyment of school, class participation, and general self-esteem. However, resilience is not only affected by student-level factors, classroom and school level factors reported by teachers and school principals, such as teacher’s confidence and expectation in student’s academic success, and school emphasis on school success, and parental support may also predict the likelihood of being academic resilient and in turn to predict cognitive and non-cognitive outcome measures (e.g., Borman & Overman, 2004; OECD, 2011). Martin and Marsh (2006) also pointed out that in the future studies of academic resilience, multiple sources of data from home, classroom and schools, as well as the community and system level should be relied on and information from different level of observations should be taken into account simultaneously in predicting academic resilience.

In the recent decades, Swedish education system has undergone series of reforms characterized by decentralization, deregulation, privatization and school choice. These reforms have transformed Swedish school system into a market-like enterprise (Lundahl et al., 2013, SOU, 2014:5).  Meanwhile the welfare system in Sweden was under restructuring and deliberate school choice and residential segregation in Sweden lead to increasing school segregation with respect to school composition of student SES and ethnicity. These have brought negative impacts on educational equity (Levin, 1998). Schools that attract more high-SES and native children also tend to have better teachers and more resources, which in turn strengthen between-school differences in learning environment (Yang Hansen & Gustafsson, 2016, Han, 2018). These changes put children from vulnerable family circumstances in double jeopardies. We thus have the reason to believe that the compensatory abilities in supporting academic resilience of these disadvantage children vary across schools, education systems, and over time.  It is thus of great importance to investigate changes in mechanisms of academic resilience both within-country and across different educational systems.

The current study is to investigate the changes in the amount of resilient students during the last 15 year in Swedish compulsory schools, and to identify school related factors that function as compensation to their disadvantaged family background.

Using Swedish data from PISA studies between 2000 to 2015, academic resilient students can be identified. In this study, and we define academic resilience as students being 25% lowest in family socioeconomic status and yet achieve at or above the median proficiency level (Agasisti, et al., 2018). We study different educational resources and supports from individual and school levels and try to predict the probability of being resilient in a multilevel analysis for categorical outcome. This analysis will be replicated with all recurring data of PISA cycles. The factors identified to predict academic resilience in each study will be compared and the changes in the set of factors will be examined in relation to policy reforms in terms of educational resource and support allocation. we used gender, age, mothers and father’s education, the highest value of the International Socio-Economic Index among parents, number of books at home, and attitude and motivation as individual-level factors and school social and ethnical composition, teacher education and certification, school instructional resources, teacher and students behavior, learning environment will be used as school level predictors. Two-level logistic regression analysis will be used in Mplus.

The analysis is on-going. The preliminary results so far showed that showed that the percentage of Swedish 15-year-olds being resilient among all the disadvantaged children in PISA samples has reduced from 2000 to 2011, and then raised somewhat in 2015. The factors that may account for the difference in the likelihood for a student being resilient with multilevel logistic regression analysis is related to student disciplinary factor and school socio-economic composition of the student body.

Agasisti, T. et al. (2018), “Academic resilience: What schools and countries do to help disadvantaged students succeed in PISA”, OECD Education Working Papers, No. 167, OECD Publishing, Paris. http://dx.doi.org/10.1787/e22490ac-en Borman, G. D., & Overman, L. T. (2004). Academic resilience in mathematics among poor and minority students. The Elementary School Journal, 104(3), 177-195. Retrieved from http://www.jstor.org/stable/3202948 Han, S.W. (2018). School-based teacher hiring and achievement inequality: a comparative perspective. International Journal of Educational Development, 61, 82-91. Kiswarday, V. (2012). Empowering Resilience within the School Context. Methodological Horizons, 7(14). Levin, H. M. (1998). Educational vouchers: Effectiveness, choice, and costs. Journal of Policy Analysis and management, 373-392. Lisbeth Lundahl, Inger Erixon Arreman, Ann-Sofie Holm & Ulf Lundström (2013) Educational marketization the Swedish way, Education Inquiry, 4:3, DOI: 10.3402/edui.v4i3.22620 Martin, A. J., & Marsh, H. W. (2006). Academic resilience and its psychological and educational correlates: A construct validity approach. Psychology in the Schools, 43(3), 267-281. OECD. (2011). Against the odds: Disadvantaged students who succeed in school. Retrieved from http://dx.doi.org/10.1787/9789264090873-en. Rosenbaum, P.R. and Rubin, D.B.(1983). The central role of the propensity score in observational studies for causal effect. Biometrika, 70, 41-55. Sandoval-Hernandez, A. & Cortes, D. (2012), Factors and conditions that promote academic resilience: A cross-country perspective. Presented at the 25th International Congress for School Effectiveness Improvement (ICSEI), Malmo, Sweden. Retrieved: http://iea.academia.edu/AndresSandovalHernandez/Talks/65278/Factors_and_conditions_that_promote_academic_resilience_A_cross-country_perspective SOU (2014:5). Staten får inte abdikera – om kommunaliseringen av den svenska skolan Betänkande av Utredningen om skolans kommunalisering, Statens offentliga utredningar [The state must not abdicate -- on the municipalisation of the Swedish school system. Report on the inquiry into the municipalisation of the school system]. SOU, 2014:5, Stockholm. Yang Hansen, K., & Gustafsson, J.-E. (2016). Causes of educational segregation in Sweden – school choice or residential segregation. Educational Research and Evaluation, http://dx.doi.org/10.1080/13803611.2016.1178589