This study compares the Scientific Literacy of Brazilian and Japanese students, based on the results of PISA and seeks to answer the following research questions: a) Are there differences of cognitive competence in Science among Brazilian students and those from other countries, especially Japan, in the PISA ?; B) Are there items of PISA 2006 that present differential functioning, with Brazil as a reference ?; and c) Is it possible, based on the PISA data and the complementary adoption of a qualitative approach, to identify different curricular emphases and / or pedagogical practices in Science Education in Brazil and Japan that contribute to the understanding of differences in performance among their students? To answer the first question, we compared the results of Brazil and Japan in the 2006 and 2015 editions of PISA, where the focus was Science. Brazil is disadvantaged compared to almost all the countries that participated in the Program, which is partly explained by the repetition. In the performance scale, Brazil remains at level 1 and Japan at level 3 in 2006 has moved to level 4 in 2015. In order to answer the second question, we used the Differential Item Functioning (DIF) analysis in the 2006 test s Science items and we conclude that there is a large presence of DIF in these items comparatively between Brazil and Japan. Although they are not able to compromise the evaluation process by favoring one group over the other, these items suggest different curricular emphases in Science. Taking into account this hypothesis and to answer the third question we adopted a qualitative approach, with observation of the use of class time; record of curricular emphases and the occurrence of activities related to interaction, investigation, experimentation and application from the perspective of students, teachers and observation; and interviews with experts and PISA managers. The results showed that more than 20 percent of official classroom time observed in Brazil is wasted with questions other than effective teaching of science; 10 times more than in Japan. In Brazil, there is a more pronounced curricular emphasis in Natural and Biological Sciences. The curriculum is distributed more homogeneously in Japan and is sectioned in Brazil. According to Japanese students, activities of interaction, investigation, experimentation and application are not frequent. The most recurrent activities observed and perceived by Japanese teachers are those of experimentation and interaction; In Brazil, those of interaction and application. Interviews with experts in Science Education and PISA managers in Brazil and Japan have shown that Japan s success in this assessment is associated with the existence of a common national curriculum and the teachers ongoing in-service training as well as the educational system reforms raised by the PISA results. The low performance of Brazilian students in PISA would, in turn, be related to the lack of preparation of the students, their lack of familiarity with the test, poor teacher training and the limited use of the evidence produced by the large scale evaluations.