11a. Universes & units of measurement
This is a short thing I want to squeeze in that will help you visualize data accurately and responsibly. It’s very easy to misunderstand the “universe” of a number and its units of measurement, and therefore to make false comparisons or inaccurate descriptions. You can think of these as two parts that together create a measurement:
- What is each thing that is being measured? (unit)
- What are all the things that could be included? (universe)
You want to make sure you’re always comparing apples to apples, or making it clear that two things you’re presenting have different universes and therefore can’t be compared.
We’ve been working a lot with ACS data, and luckily the Census Bureau spells out its universes on each data table. Sometimes they’re pretty straightforward (total population, households, housing units). Sometimes they require a bit more understanding (population for whom poverty status is determined, population under 18 years in households), and sometimes they’re messy, especially when dealing with family types (own children under 18 years in families and subfamilies for whom poverty status is determined).
Other examples I encounter regularly:
- Birth outcomes: for some measures, unit might be babies; for others, parent giving birth
- Proxies: when I analyzed data on police stops, I tried to figure out a population to compare to. I didn’t have data on how many people in each census tract had a driver’s license, decennial data wasn’t out yet so I didn’t have reliable local counts of population 16 and up by race, so I just used population. It wasn’t ideal.
- Relationships: is a question being asked of parents, or of adults with a child in their household? These aren’t necessarily the same.
- Survey respondents: is a survey given to the population at large, or just adults?
- Households: is this rate the share of households, or the share of heads of households, or the share of households by some characteristic of the heads of households?
I usually use chart subtitles to summarize what is being measured, of whom, where, and when. Then, if necessary, I’ll go into more detail in a figure note or something similar. e.g. a chart with the subtitle
“Share of households without vehicle access by number of workers and race/ethnicity of head of household, 2020”
then had a note spelling out the universe as
“Households included here are those with at least one member between ages 25 and 64. When discussing race and ethnicity of households, values are based on the race/ethnicity of the person designated as head of household when filling out the census, which may differ from other members of the household.”
Because that universe is based on households, I wouldn’t want to directly compare it to another measure based on population.
Back to top