The relationship between school educational resources and students’ literacy has long been focused in educational and psychological research. With restrictions of statistical analysis techniques, most of the studies overlooked the hierarchical characteristics of data and thus conclusions obtained from different levels of data analysis varied. Besides, previous researches were mainly concentrated on certain nations or regions, lacking of an international comparative perspective. The program for international student assessment (PISA) with national representative samples, strict evaluation standards and comparable observed variables provides highly reliable data. Based on the PISA 2006 data, the present study investigated the effect of school educational resources on the mathematical literacy from a perspective of international comparison. The result of this research can offer some valuable implications including establishing educational quality evaluation system and paying close attention to the validity of educational investment for our country. This research focused on data of Hong Kong–China, Japan, Finland and the United States from PISA 2006 database. The school educational resources consist of 11 variables, i.e. school size, class size and school type, the index of availability of computers, the overall ratio of computers available for instruction to school size, the proportion of computers connected to the internet, the student-teacher ratio, the proportion of fully certified teachers, the proportion of teachers with an International Standard Classification of Education( ISCED) 5A qualification, the index on teacher shortage, the quality of school’s educational resources. The hierarchical linear model realized by SAS 9.0 was employed to analyze the data. The results indicated that mathematical literacy scores of students in Finland, Hong Kong and Japan are higher than the average of OECD countries (498), while that of American is the lowest in four countries (regions) and also lower than the OECD average. School size, student-teacher ratio had demonstrated significantly positive effects on the students’ mathematical literacy in Hong Kong–China: literacy score will improve by 12.59 points with each additional unit of student-teacher ratio. In Japan, school size, class size, school type and the proportion of teachers with ISCED 5A had significantly positive effects on mathematical literacy, while the ratio of computers available for instructi