Suppose you wanted to know the population of the United States for the years 1920-1970. Where would you go to get that information?
The Census Bureau, of course. Counting people is their job.
And yet, if you hop over to their website you won't find this. Nor will you find much of anything else related to simple population measurements. If you want to know how many 18-year-olds owned cars in 2016, you can find out. Not easily, but you can do it. There are also reports galore on specific demographic subjects. But if you just want a straightforward count of people, or of the Hispanic share of the population, or the male-female ratio, or anything like that, forget it.
Why is this? Just to give you an idea of what I'm talking about, here's the kind of thing I'd like to see:
Don't bother critiquing this. I'm not actually proposing it as an interface, and it has some obvious drawbacks. It's just to provide a sense of what I think the Census Bureau should provide in a fairly simple-to-use manner.
This is the kind of thing the Census Bureau should make available, right? So join me in my jihad to get them to do it. Their job is to count people, and they should make their historical population figures easily accessible to anyone who wants them. Who's with me?
POSTSCRIPT: It's possible, of course, that the Census Bureau already has something like this and I'm too dim to have ever located it. That would be OK! I'm perfectly willing to accept the ridicule I'd deserve in return for finding out where this is.
For the kind of basic info you're looking for, I used to buy The World Almanac every year. They seemed to have a good sense of what us common folk want to know.
I have no idea whether that estimable publication is still around. But if it is, I think your post has motivated me to find it and buy one.
An BTW, Google is pretty good at finding that kind of information in less than a second.
I found the appropriate census department page using google. Most government websites feature very poor navigation, and the census seems to be no exception. So google works better than any native search function. Unfortunately, as I recall, this comments section does not allow posting of links.
This comments section does allow posting of links. Just not "too many" which is a number greater than one but less than or equal to four based on my experience. I've not bothered to try to narrow it down further.
I found a couple promising links but every one said 'Sorry, page removed'.
For any kind of research, IPUMS out of the University of Minnesota is a million times better than even trying to use the Census website.
For casual users, yeah, the Census website is hopelessly bad. The data explorer at data.census.gov is *supposed* to do what Kevin is talking about.. but good luck getting simple answers or having any confidence that you're using the right dataset as a casual user.
There are a bunch of products out there that curate and format Census data since it's such a time consuming mess. ESRI/ArcGIS for example now makes certain datasets available in their products in order to cut down that frustration.
IPUMS is great -- I'm a mega user. But, it's not a place to get population counts across time. It's super at providing information on characteristics
The UN Population Division seems to have the sort of website you are looking for: https://population.un.org/wpp/ , It covers all countries, but only goes back to 1950, and does not break down by race. Good for age bands, women, and children.
How funny that I looked this up a couple days ago for some flash fiction I was writing. I found this useful page: https://worldpopulationreview.com/
Historical census data is kept at the National Archives-- moved there 72 years after the census, or some period like that-- but seems oriented mostly toward genealogists and based on my experience is hard or impossible to maneuver through for this kind of information.
The Census does produce a bunch of pre-selected charts under the American Community Survey (age, ethnicity, native-born, commuting time, etc). For long time-series data there's the classic Historical Statistics of the US, taken over by Cambridge UP because the census bureau didn't have the budget to keep it going. It's best available through research libraries if yours has it.
Admittedly I haven't looked very deeply, but for the kind of aggregate statistics Kevin is looking for, I'm not seeing any way to do dynamic queries like you can do on Fred, say. The Census is really about enumerating people in certain mandated categories and collecting economically-useful information. Having to abandon HSUS is an indicator of that direction.
The values for the non-census years would be artificially "accurate" estimations, and give a false sense of "smoothness" to the resultant data. It would be better just to request the census-year data, because it's from onsite enumerations, at least until 2000.
quickfasts gives you the snapshot you are looking for
https://www.census.gov/quickfacts/fact/table/US/PST045219
BTW the US has added 28 million people since 2000. And people who look at that thru the lens of 'percentage growth' are missing 3/4 of the story. Any ecologist will tell you absolute numbers count.
https://www.census.gov/quickfacts/fact/dashboard/US/PST040219 is a better link. But still, information is limited to three points in time (2010, 2019, 2020), so not exactly a database that can be queried.
You want a simplified database search query interface? Impossible! This is the Census Bureau. It might be easier to search for the data within Fred.
For recent years, try CPS: https://www.census.gov/cps/data/cpstablecreator.html
For historical data that's free (far too much of it has been monetized behind closed doors, IMHO), try IPUMS NHGIS: https://data2.nhgis.org/main (requires login, but is free).
Both require a certain amount of experimentation to get the hang of their table builders. As your mock-up makes clear, it's not trivial to make interfaces that both accommodate multiple needs, and are intuitive.
The census bureau tries hard to provide some guidance, eg here -- https://www.census.gov/acs/www/guidance/which-data-tool/ -- but they are quite lacking in historical data tools.
CPS does not provide the kind of population counts you're looking for -- it does provide information on social/economic/demographic charaacteristics.
But, IPUMS-CPS is much better than the Census Bureau's website.
The Census Bureau's website is not easy to use and, with the last several redisigns, it's gotten worse IMO
The Population estimates Program actually has the data you're looking for, going back to 1900. But, you have to know where to look...
The Program "homepage" is here -- https://www.census.gov/programs-surveys/popest.html
The historic data would be what they call "intercensal" estimates. There are links on that page for 1900-1990 for the US and states.
The others are "buried" but hey are there.
There's an FTP page with all the Pop Estimates tables (but no links or easy explanatory links, although the documentation is also in the FTP tree)--
https://www2.census.gov/programs-surveys/popest/datasets/
Wonder if development has been outsourced to Tata.
I know this doesn’t help you much today, but... The Census Bureau used to have a fantastic tool called the American Fact Finder which could pull and even visualize any kind of data imaginable, but it was retired (without explanation and possibly for nefarious reasons) during the Trump administration.