I spend some of my time (technically half my time, but #gradstudentlife) working as a Graduate Research Assistant at the Minnesota Population Center primarily with the Integrated Public Use Microdata (IPUMS) Project. Most of my work is with the User Support Team helping users figure out how to make use of IPUMS data and fixing strange quirks in the data. After over a year of this job, I began to notice some frequently asked questions. In a new blog series on the Minnesota Population Center blog “Use It For Good“, I answer some of these FAQs.
Here are links to the first three posts in this series:
Most IPUMS data is a sample of some larger population. Therefore when users want to make some sort of statistical statement about the larger population, frequencies and proportions need to be weighted properly to reflect the complex sample selection design of most nationally representative surveys. In this post I briefly introduce the reasoning for using sample weights and link to existing resources.
A lot of people want to perform U.S. analysis at the county level, however, for reasons relating to confidentiality this is challenging when using modern U.S. census samples. In this post I explain these “missing” U.S. counties and introduce common work-arounds.
Before anyone can actually download and use IPUMS data, they must register for an IPUMS account. This strikes some as strange, because IPUMS provides public use data. In this post I discuss the reasoning behind requiring user registration and explain the expectations users should have when using data from various IPUMS projects.