Using IPUMS datasets as as pseudo panel/multilevel repeated cross-sectional design


I would like to use the IPUMS datasets to investigate the impact of climate change. Can the IPUMS datasets from countries and survey years be pooled to form pseudo panel/repeated cross-section for econometric analysis?

I would like to use a methodology similar to this paper -

I would be grateful for your help and suggestions. Thank you.




Using IPUMS International data, what you are proposing to do sounds reasonable. Similar to the design of the paper you shared, individual samples should be nationally representative and comparable across countries at GEOLEV1. In addition to understanding the comparability of the variables you are interested in (which can be found on the comparability tab of each), you may also want to explore the detailed sample information provided here as some samples have notes about specific geographies within the country (i.e. excluded regions/states).


Dear Michelle,

Thank you so much for your reply. I appreciate it. I will go through the information regarding regions/states.

Are the dates of interview available for the IPUMS-International datasets? I am happy to provide more information for this request if necessary. Thank you again!




You can find information on the census/survey day for each sample here. For example, if you click on Argentina 1970, you will see under census/survey characteristics that the census/survey day was September 30, 1970.

If I’ve misunderstood your question, please feel free to clarify here or you can email us directly at