Hello ACS Data Community,
I recently stumbled upon a perplexing issue in the 2022 ACS 5-year estimates that I believe warrants discussion. According to the official enrollment statistics provided by BYU-Idaho (byui.edu/executive-strategy-and-planning/institutional-data/official-enrollment-statistics), the ratio of male to female students is roughly balanced, hovering around a 1:1 ratio.
https://www.byui.edu/executive-strategy-and-planning/institutional-data/official-enrollment-statistics
However, this seems to be in stark contrast with the ACS 2022 5-year estimates for the Rexburg, ID Micro Area, specifically in the S12002 table.
https://data.census.gov/table/ACSDT5Y2022.B12002?q=B12002:%20Sex%20by%20Marital%20Status%20by%20Age%20for%20the%20Population%2015%20Years%20and%20Over&g=310XX00US39940
The ACS data indicates a surplus of 5643 never married women in the area, which seems implausible given the university's gender ratio. This discrepancy raises serious concerns about the accuracy of the ACS estimates. Could this be a methodological error or a misinterpretation of the data?
I reached out to the ACS team regarding this issue, but they firmly denied any error in their data. This response is a bit disconcerting as it challenges our understanding of the demographic dynamics in university-centric areas like Rexburg. I'm curious to hear your thoughts and experiences regarding similar discrepancies in ACS data, especially in areas with significant young adult populations.
How do we reconcile such stark differences in data from reliable sources? Are there any methodological nuances of the ACS estimates that might explain this discrepancy? I look forward to a vibrant discussion on this issue.
On the one hand, your experience shouldn't count for nothing, you're on the ground and that gives you insight that can't be gained through a survey. On the other hand, you only know what you know, and…
Aefcmj, I need to further affirm your suspicions. The discrepancy seems to be centered around the BYUI campus, which in the 2018-2021 data was 5% male.Even stranger, looking at the 2013-2017 data, it was…
I pulled up the College enrollment in the Rexberg Micropolitan area and I include it here. It does show the typical gap between men and women enrolled.
From a quick look at data on the area, there was…
Interesting. The Rexburg, ID Micropolitan Area seems to be all of Fremont and Madison counties (https://www2.census.gov/programs-surveys/economic-census/2007/reference-maps/id/metro/2007-us39940m-rexburg-id-micropolitan-statistical-area.pdf), so I'm guessing that there are other residents there who are not part of the university.
It might also be worth looking at some of the school enrollment tables (any that start with B14), for example, B14002 actually lists people enrolled in college undergraduate years by sex. One thing to note though is that ACS only captures yes/no on school enrollment, not on any particular institution. So this would also include people who are enrolled in any other institutions who live in the micropolitan area (community colleges, other universities, and could even be online programs).
Right, BYUI's report supports this. Note the higher counts of women enrolled online
Online data make no sense, because online school can have no building but just a website and be registered in some city.
That makes no difference to the ACS. You seem to take issue with the ACS reporting more women than men because it's inconsistent with BYUI enrollment counts. But the ACS is not counting enrollment, either online or in person.
This is all based on my personal experience, I observed that when I lived in Rexburg, there were more men than women, so there were some men who couldn't find wives
On the one hand, your experience shouldn't count for nothing, you're on the ground and that gives you insight that can't be gained through a survey. On the other hand, you only know what you know, and unlike you, the ACS data is based on a random sample, which reaches people likely outside of your orbit.But there's a third hand- The difference between the male and female populations is most stark at 18 and 19 years old, never married. 2,481 male, 8,510 female. That is a very large difference, much larger than for any other age group. That's weird!
Aefcmj, I need to further affirm your suspicions. The discrepancy seems to be centered around the BYUI campus, which in the 2018-2021 data was 5% male.Even stranger, looking at the 2013-2017 data, it was much more balanced, at 44% maleAccording to the data, the male population dropped from 541 to 80 over the course of these 5ish years. Of course, I'm looking at block groups and weird things happen in block groups, but clearly it's affecting the data for the whole area.
Male vs. female residence halls?
NO, check this: brightspotcdn.byui.edu/.../approved-housing-map-june-2023.pdf
This map doesn't prove or disprove anything. In fact, it's not relevant at all. Your question was why the ACS counts of never-married women vs. men are so disparate in Rexburg. College enrollment -- on campus or online -- doesn't matter. There are plenty of unmarried/never-married women (and men) that are not in college.
No, they wouldn't come here if they didn't go to college, very few local residents here
Really? So there are 64,349 people in the Rexburg Micro Area, and around 22,000 are BYUI on-campus enrollees. Around 4,000 people work for the university. What do the other 38,000 people do? You've observed them all and inquired as to their marital status?
Glenn, I had the same response you did, but when I looked at the granular data, it seems like most of the strange data is in the city of Roxburg (about half the population of the MuSA), and looking at census tracts, it's in the area of the city where the university is.Question: Doesn't the ACS use administrative data submitted by institutions for group quarters? Could this be an issue with data submitted by the university?
According to the IPEDS Feedback report 57% of the Students are female.
The report is available here and relevant graph is below: nces.ed.gov/.../ReportPDF.aspx
Maybe LDS women are more similar to other women than you think. Note the so-called 35 Peer Schools have 58% women. These numbers are from 2019 per IPEDS. Or maybe LDS guys drop out real fast :).
That could be, sure. The ACS methodology mentions a two phase GQ survey -- facility level and person level.
https://www2.census.gov/programs-surveys/acs/methodology/design_and_methodology/2022/acs_design_methodology_ch08_2022.pdf
Seems weird that mis-reporting at the facility level would not be corrected in the later phases of personal contacts and QC, but I'm sure it's possible.
Glenn--
I don't think there was misreporting. The ratio reported by BYUI to IPEDS is an exact match to the ratio in the ACS.
IPEDS 55% women ACS 52.99% Women 10,150 or 17,803 Private College Undergrads.
This based upon two official sources: the ACS, the IPEDS data as reported by BYUI.
Andy
I did some misreporting of my own-- IPEDS 57% female. ACS 57.01 Female 10150 of 17803 undergrads and private colleges or universities.
This based is upon two official sources: the ACS and, the IPEDS data as reported by BYUI.