This study investigates the relationship between school-level accountability practices and science achievement of 15-year-olds, across four counties: Australia, Korea, Portugal, and the United States. We used PISA 2006 data, since 2006 is the only administration that has focused on science. School-level accountability practices are here defined as activities that: (a) provide school achievement data to external stakeholders, or (b) establish consequences according to the achievement results. Using linear regression analysis, we found that school-level accountability practices varied across these four countries, albeit not all pairs of countries were significantly different from each other in this regard. Using hierarchical linear modelling, we found that school-level accountability practices had a small effect on science achievement. Importantly, this effect was not independent of schools' and students' socio-economic status.