What are “integrated variables” in IPUMS-International?
Integration – or “harmonization” – is the process of making data from different censuses and countries comparable. For example, most censuses ask about marital status; however, they differ both in their classification schemes (one census might recognize only a general category of “married,” while another might distinguish between civil and religious marriages) and in the numeric codes assigned to each category (“divorced” might be coded as a “4” in one census and as a “2” in another). To create an integrated variable for marital status we recode the marital status variable from each census into a unified coding scheme that we design. Most of this work is carried out using correspondence tables like the one here.
Because some censuses provide more detail than others, a coding scheme that reduced variables down to the lowest common denominator across all samples would inevitably lose important information. As a result, many IPUMS integrated variables use composite coding schemes. The first one or two digits of the code provides information available across all samples; the next one or two digits provide additional information available in a broad subset of samples. Finally, trailing digits provide detail only rarely available. All meaningful detail in the original enumerations is therefore available to researchers if they need it, but they can confine their attention to the less-detailed digits if they wish.
For some variables, the composite coding structure is recognized in our system by formally distinguishing separate “general” and “detailed” versions, as described here.
The other component of integration is the variable documentation. The documentation aims to highlight important comparability issues that are not self-evident from the coding structure for the variable. A general comparability discussion emphasizes issues for international comparisons, and country-specific discussions note comparability concerns when making intra-national comparisons over time. IPUMS staff must exercise their judgment in composing this documentation – there is no formula for it. But users need not depend totally on us: the variable documentation provides links to both English-language and original-language census questionnaires and instructions. This material is readily available on every variable description page through the link to “enumeration text.”