In this study, it is aimed to establish a hybrid model that reveals the relationship between the Science Literacy level of the students in the Singapore and Turkey samples in PISA 2015 and the students' test anxiety, science interest, ambitions and beliefs in the accuracy of knowledge. In this context, the research is a relational survey model. Hybrid models were established using the data of students in the Singapore and Turkey samples. With the established models, the effect levels of the variables affecting the science literacy levels of the countries were tried to be compared with the hybrid model. In the research, hybrid model and measurement models were established with 5668 student data in the Singapore sample and 5084 student data in the Turkey sample. It is seen that the multivariate normality assumption cannot be achieved in the measurement models established with the sample of Turkey and Singapore and in the hybrid model. In the study, it is seen that the measurement models of all variables and subscales of science literacy were confirmed for both samples. At the same time, it is seen that the hybrid models established with the variables related to both models are confirmed. It was concluded that the fit index values of the hybrid model established with the data of Turkey were slightly higher, and all pathways were defined. In both samples, it was concluded that the variable that most affected the students' science literacy levels was their belief in the accuracy of science, and the variable that affected the least was their ambition. In both samples, the test anxiety variable affects the science literacy levels negatively. At the same time, it was concluded that students' ambitions do not have a direct effect on their science literacy in the Singapore sample.

Science literacy, interest, belief, ambition, test anxiety.