Characterizing the Academic Achievement of Middle School Students in Science Subject based on Latent Profile Analysis

This study explored an approach to enhance the function of educational assessment results as educational feedback information so that educational policy makers and teachers have more accessibility with the assessment results based on the cognitive diagnosis model (CDM). Using secondary data of the National Assessment of Educational Achievement (NAEA), the latent profile analysis was applied based on the estimated probabilities of mastery in eight detailed learning areas of science subject, and as a result, five latent groups were derived. In detail, latent group 1 showed excellent achievement in all eight areas. Latent group 2 was named as ‘a group needed to cultivate scientific communication’ due to their lack of understanding in the detailed learning area 8 (Making argument based on evidence). Latent group 3 was named as ‘a group needed to cultivate scientific thinking’, because this group particularly needs to foster scientific thinking skills, such as ‘explaining phenomena related to scientific inquiry and problem solving.’ Latent group 4 showed relatively high capacity in ‘general data analysis and interpretation’ while had low ability in ‘scientific knowledge and understanding.’ The last latent group showed the low performance level on all eight detailed learning areas. Given the cross-tabulation with the latent group membership and achievemnet levels of NAEA, It was found that more detailed and useful achievement characteristics information could be grasped through the potential group classification, particularly for students in ‘intermediate’ or ‘basic’ achievement levels in NAEA.