Search

The assumption of cause-effect relationships in ex post facto research is a widely known issue in the field of research methods in social sciences. To address this important limitation, the use of causal inference techniques has become widespread in recent years. Causal inference establishes a set of statistical procedures for drawing causal conclusions in non-experimental research. Despite its wide popularity and diffusion in the social and health sciences, its use in educational research is still marginal. Thus, this paper introduces the main causal inference techniques available to the educational researcher when observational panel data are available. After addressing the key features and potential of propensity score matching, instrumental variables, and regression discontinuity design, we present an example application of each of these techniques. We used the available databases from the PISA 2018 assessments. We included the mathematical competence as the dependent variable in all the three models implemented. Given the different characteristics of each of these techniques, the independent variable used is different in the three models applied: attendance to early childhood education in propensity score matching; student academic expectations in instrumental variables; and size of the community in which the school is located in regression discontinuity design. The article concludes by discussing the potential of this set of techniques, taking into account the needs and methodological procedures most commonly applied in educational research. [ABSTRACT FROM AUTHOR]

Copyright of Electronic Journal of Educational Research, Assessment & Evaluation / Revista Electrónica de Investigación y Evaluación Educativa is the property of RELIEVE - Revista Electonica de Investigacion y Evaluacion Educativa and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Professional development for teachers is fundamental in the configuration and functioning of smart schools. This paper aims to characterize professional development with the participation of compulsory secondary teachers in Spain and to detect key factors in the functioning and organization of schools associated with higher levels of ongoing teacher training. A cross-cutting non-experimental design was used to conduct a secondary analysis of data from PISA 2018 tests, including over 20,000 teachers and more than 1000 schools in Spain. Descriptive results show great variability in teachers' commitment to their professional development; this variability is not associated with the grouping of teachers by school. The decision tree model completed with data mining tools shows that intensive professional teacher development in schools is associated with a better school climate and higher levels of innovation, cooperation, taking on shared goals and responsibilities, and leadership distributed among the education community. The conclusions highlight the importance of ongoing teacher training and how this improves educational quality in schools. [ABSTRACT FROM AUTHOR]

A central objective of schooling processes at the international level is reading skills development. Unfortunately, many students in the European Union underperform at this, and these low performances can be more pronounced in countries with lower Human Development Index (HDI) values. This study analysed the contextual effect of school socioeconomic status (SES) on reading performance using PISA data from 27 countries. We found that school SES had a positive contextual effect on student reading performance, especially countries with lower HDI values, via reading self-competence. However, in middle and higher HDI countries, the contextual effect was smaller, and we did not observe an indirect effect via reading self-competence. We conclude that school SES, rather than student SES, matters the most. Therefore, the modification of school composition and resources might act as a buffer for low student SES, particularly in lower HDI countries. [ABSTRACT FROM AUTHOR]

International large-scale assessments, such as PISA, provide structured and static data. However, due to its extensive databases, several researchers place it as a reference in Big Data in Education. With the goal of exploring which factors at country, school and student level have a higher relevance in predicting student performance, this paper proposes an Educational Data Mining approach to detect and analyze factors linked to academic performance. To this end, we conducted a secondary data analysis and built decision trees (C4.5 algorithm) to obtain a predictive model of school performance. Specifically, we selected as predictor variables a set of socioeconomic, process and outcome variables from PISA 2018 and other sources (World Bank, 2020). Since the unit of analysis were schools from all the countries included in PISA 2018 (n = 21,903), student and teacher predictor variables were imputed to the school database. Based on the available student performance scores in Reading, Math, and Science, we applied k-means clustering to obtain a categorized (three categories) target variable of global school performance. Results show the existence of two main branches in the decision tree, split according to the schools' mean socioeconomic status (SES). While performance in high-SES schools is influenced by educational factors such as metacognitive strategies or achievement motivation, performance in low-SES schools is affected in greater measure by country-level socioeconomic indicators such as GDP, and individual educational indicators are relegated to a secondary level. Since these evidences are in line and delve into previous research, this work concludes by analyzing its potential contribution to support the decision making processes regarding educational policies. [ABSTRACT FROM AUTHOR]

• Decision trees enable the detection of factors that predict school effectiveness. • The predictive model is more accurate for low effectiveness schools. • A single negative factor at school escalates the risk of low effectiveness. With the main goal of identifying the process factors associated with school effectiveness in secondary education, this work presents an innovative methodological proposal. Based on secondary data from the Spanish sample of PISA 2015, high- and low-effectiveness schools were selected by analysing the residuals of the school level (level 2) in multilevel models. Subsequently, decision trees were used to identify the process variables with a greater predictive power for the identification of high- and low-effectiveness schools. While the hierarchical linear models obtained show inter-school variance scores greater than 10%, decision trees achieve a precision greater than 90%. We conclude by analysing the suitability of using decision trees in data originating from large-scale assessments, and examining the obtained factors associated with school effectiveness. [ABSTRACT FROM AUTHOR]

• The studied variables affect achievement, with some gender and country differences. • Male achievement tends to be more affected by the variables in Finland and the UK. • All correlations yielded significant results except the feeling of belonging in Spain. • Parent support, belonging and motivation reduce gender differences in achievement. The aim of this study is to define the moderating effect of gender on the relationship between academic achievement and some explanatory factors, according to the PISA 2015 results in the United Kingdom, Finland and Spain. Based on an exploratory correlational study, multivariate statistical techniques were applied with moderate regression analysis in these countries, which were considered relevant examples for understanding the current situation. The results obtained suggest that gender is a variable that, at a general level, moderates the relationship between the explanatory factors studied, with the exception of anxiety, and academic achievement, mainly in the Finnish student population. Ultimately, the relevance of gender as a moderating variable for the improvement of student achievement and educational quality is verified. [ABSTRACT FROM AUTHOR]