The Group Selection of Variables that Effected to Science Scores of Indonesia^s PISA using Group LASSO Vera Maya Santi (a*), Rizkha Hayati (a), Bagus Sumargo(a)
a) Program Studi Statistika, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Negeri Jakarta, Jl. Rawamangun Muka, Kota Jakarta Timur, DKI Jakarta, 13220, Indonesia
*vmsanti[at]unj.ac.id
Abstract
Scientific literacy is a person^s ability to apply knowledge and develop a reflective mindset so that they can participate in overcoming issues and ideas related to science. The quality of Indonesian scientific literacy based on the results of the Program for International Students Assessment (PISA) which was followed from 2000 to 2018 is still relatively low. In 2018, Indonesia^s PISA science score average is still well below the Organization for Economic Cooperation and Development (OECD) average. PISA data has a high data complexity and has multicollinearity, so it requires appropriate statistical methods in conducting analysis. Unfortunately, it is still very rare to conduct research quantitatively. The group LASSO method is one of methods that can be used to select groups of variables that affect science literacy in Indonesian PISA while overcoming multicolliearity. The results of the analysis showed that there were 11 groups of explanatory variables that affect the Indonesia^s PISA science score with an RMSE value of 25.00 and an R2 value of 0.31. This means that only 31% of the variance of the average science score can be explained by the explanatory variables, while the rest is explained by other factors outside the study.
Keywords: Group lasso- Multicollinearity- Scientific literacy