This research aims to develop a valid and consistent measure for socioeconomic status at the household level using census microdata available from the Integrated Public Use Microdata Series- International (IPUMS-I), the world's largest census database. First, we use principal component analysis to compute a wealth index based on housing characteristics and asset ownership. The validation strategies include comparing our proposed index with the widely used Demographic and Health Survey (DHS) wealth indices and then verifying the predictive power of our index on education enrollment and primary school completion. Moreover, we attempt to identify general conditions necessary to produce an internally consistent asset index based on census microdata. Our results show a consistently positive effect of the wealth index on education outcomes across four census samples (Peru 1993, South Africa 1996, Brazil
2000, and Colombia 2005). Furthermore, graphical analysis of kernel distributions suggests our measure is comparable to that of the DHS. Finally, through a stepwise elimination procedure, we find evidence supporting the internal consistency of the census asset index. As an important practical implication of our results, we are able to propose a methodology to determine which assets are more important in determining household socioeconomic status.