International large scale assessments of student achievement employ different indicators to operationalize the socio-economic status (SES) of students. As all of these indicators correlate significantly with school achievement it is common to use aggregates of these variables to measure SES. The question arrises whether the sociological construct ‘poverty’ provides additional explanatory power when describing social disparities in students’ achievement. For our analyses we use data from three studies, which measure families’ poverty explicitly: (1) a follow-up study to PIRLS 2006 that tests 4th grade students in the German-speaking Community of Belgium in the domain of reading; (2) TIMSS 2007 Germany testing 4th grade students in mathematics and science and (3) the first measurement of the Panel Study at the Research School Education and Capabilities in North Rhine-Westphalia (PARS) testing 5th and 9th grade students in different domains. Our analyses show a significant relationship between families’ risk of poverty and students’ achievement even if other indicators of students’ SES are controlled for. We conclude that for in-depth analyses it is beneficial to use a combination of different indicators in order to consider various aspects of SES when analyzing social disparities in educational context.