Hi Chris - thanks for your post. Appreciate everything you do. I would be cautious about using this data to conclude that international student volumes went up. As Clay mentioned, big R1 universities are reporting declines, other universities reporting declines, and when using other data sources we are seeing declines - for example, our team downloaded and analyzed the i94 data of foreigners entering the U.S., and when looking at student visa holders in the month of August, it was down -19% vs previous August.
Now, this data is counting all entries in the U.S. (including those who visited home for Summer vacation and came back) but how could it be possible that this is so down, yet the SEVIS data shows an increase? It's not adding up.
I agree that there seems to be a disconnect in the data that's really hard to make sense of, and we need to be cautious of over-interpreting any one data point. The people I've talked with are *really* curious about OPT and whether it bolstered otherwise sagging new enrollment numbers. It's too bad SEVIS data doesn't disaggregate "enrolled students" vs "OPT", so we won't know until IIE data comes in (or institution-level impacts).
It does seem that the data picture is mixed, and the full picture will come into view as we have more sources. It does seem a 16% difference in projections and SEVIS is hard to account for by a single factor.
Thanks for sharing the link to the i94 data who are part of the Distributed Progress community -- I'm adding it as an appendix to the end of the article, so others can benefit from your insights and this important data source. You and the cracker jack data team at Shorelight do outstanding work.
If anyone reading this wants to add, feel free to jump in.
Thanks Chris! Appreciate it. And appreciate the shoutout in the article and on LinkedIn. The person who built the dashboard is Hilary O'Haire - she is our head of Analytics and basically downloaded the source data and built the dashboard within a couple of hours!
Looking forward to more insights from you and the industry as we work to make this a more welcoming place for international students.
Thanks for the question, Clay. Bloomberg reported that "Student arrivals to the US dropped 19% in August year-on-year to just over 313,000 – a fifth consecutive month of declines, according to data from the International Trade Administration that tracks arrivals to the country." My analysis for this article is based on the SEVIS data just posted by the Department of Homeland Security through September. The need to "triangulate" various data sources will be important, especially as the USDOS Monthly Nonimmigrant Visa Issuance Statistics become avaialble (they are only available through May at the moment) and IIE does its annual Fall Snapshot survey. It will take time for us all to make sense of these multiple data points, especially if they send conflicting signals. SEVIS is one data point -- an important one for sure -- but there will be more to come.
I would add that the Bloomberg data is 'arrivals' - and this year, anecdotally there is a lot of talk about how many students decided to stay in the US this summer. As such there is potential that a lot of the 'arrivals' are really a lot of first year students, while the second or third or fourth year year students didn't 'arrive' as normal - as they stayed! I am seeing a lot of enrollment reports from larger colleges which are quite stable on average, some are up and some are down a bit, some smaller colleges in the middle of nowhere seem down a lot, but overall I was expecting down 5-8% region after looking at the data college by college so far.
Great insight -- not as many "arrivals" -- perhaps because fewer summer trips home (or abroad) by second, third, or fourth year students -- thanks for sharing. The link Clay shared is great -- and aligns what Ryan Allen calls an "Age of Conquest" (see link in article to his analysis of US higher sector change over time) -- with market concentration "at the top" with declines at regional publics, small privates, particularly ones that took "big bets" on countries where we see declines (or turbulence) in enrollment, e.g., China, India, etc. Open Doors will be the go to source for this institution-level data (and good reporting between now and November, like Ashley Mowreader did for Inside Higher Ed!)
A couple of points I would make (as a non professional in this field) is
1. From what I can see from searching through daily reports from various colleges which randomly report their enrollment figures (mostly including intl students) I would say that even if total F1 visas are down say 15% in 2025, the dollar weighted average fees paid will be down a lot less. This is due to mix - the more expensive colleges 'sell out' every year, the constraint is not demand it's supply. So if for example you hear that "Indian students are shunning the US" it doesn't matter for the top collages as they will always fill up and they charge 3-4x what some regional colleges charge that may be bearing the brunt of any demand destruction.
2. Aligned with this is a possibly controversial point in that the students attending the top schools are just better able to plan and navigate their way through adversity in terms of attaining visas etc - yes they are just a bit smarter! Also, these top colleges will have better infrastructure to help these students navigate a path to the US, plus these students will have higher confidence that the college will have backup plans such as online or temporary overseas campuses should the need arise. You can see why Trump made the comments he made at the time he talked up the 600,000 Chinese students when he referenced some colleges being in trouble - these are often red state often Christian colleges that are being hurt most by the visa issues we see so he is likely getting an earful from certain interested parties.
3. The trend of concentration of wealth is something that is happening in most countries. Why is that a factor? Because it ensures that there is enough demand from wealthy families from many countries to seek a US education, which remains the gold standard alongside certain UK colleges. Witness the rise in applications from Vietnam or Nepal.
One point related to the accuracy of the data - Chris if you look at your summary of the state by state stats, Indiana stands out to me in that this data shows an 8% increase, however if you look at the press reports from IU it's down 14% and Purdue is down 6% (didn't find anything on Notre Dame). That down 14% may end up being down a lot less as these students show up - it's not clear if they included the students who gained admission but didn't show up as enrollees yet,
due to visa delays. Anyways the best way to gauge what's going on right now is just reading through what the bigger schools have said about Fall enrollments as most break down their international intake - so far it's a modest decline and nothing like the 30%+ some (NAFSA) were talking about as worst case (due to visa delays).
That's really useful to know about Indiana and the disparity between the reported data from the institutions with the largest international enrollments historically and what the SEVIS data indicate. It might be another "case of OPT" where enrollments are down but OPT participation is buffering (or counterbalancing ) enrollment decline (since SEVIS combines and does not disaggregate the two, for anyone who may not be familiar with the SEVIS dataset).
It will be important to watch as more data comes in and a "fuller picture" comes into view. I also appreciate the candid assessment of your three points. You've come to the right place -- Distributed Progress is a community of people thinking together about these issues -- variation in perspectives is one of the best parts, and you raise a number of issues I had not considered and will now consider more seriously, as more data comes in.
Agree with the concern expressed that the SEVIS data may not be reflecting what many universities are reporting. While we expected regional publics & smaller tuition-driven privates to struggle, their share of all international undergraduates within the U.S. higher education ecosystem is relatively modest. But if R1s (especially flagships & land-grants) are seeing substantial declines, that’s a different story, on a very different scale. https://www.jsonline.com/story/news/education/2025/09/29/international-student-enrollment-down-at-uw-madison/86355163007/
Hi Chris - thanks for your post. Appreciate everything you do. I would be cautious about using this data to conclude that international student volumes went up. As Clay mentioned, big R1 universities are reporting declines, other universities reporting declines, and when using other data sources we are seeing declines - for example, our team downloaded and analyzed the i94 data of foreigners entering the U.S., and when looking at student visa holders in the month of August, it was down -19% vs previous August.
https://public.tableau.com/app/profile/lusiad/viz/I-94StudentVisasDashboard/Dashboard
Now, this data is counting all entries in the U.S. (including those who visited home for Summer vacation and came back) but how could it be possible that this is so down, yet the SEVIS data shows an increase? It's not adding up.
I agree that there seems to be a disconnect in the data that's really hard to make sense of, and we need to be cautious of over-interpreting any one data point. The people I've talked with are *really* curious about OPT and whether it bolstered otherwise sagging new enrollment numbers. It's too bad SEVIS data doesn't disaggregate "enrolled students" vs "OPT", so we won't know until IIE data comes in (or institution-level impacts).
It does seem that the data picture is mixed, and the full picture will come into view as we have more sources. It does seem a 16% difference in projections and SEVIS is hard to account for by a single factor.
Thanks for sharing the link to the i94 data who are part of the Distributed Progress community -- I'm adding it as an appendix to the end of the article, so others can benefit from your insights and this important data source. You and the cracker jack data team at Shorelight do outstanding work.
If anyone reading this wants to add, feel free to jump in.
Thanks Chris! Appreciate it. And appreciate the shoutout in the article and on LinkedIn. The person who built the dashboard is Hilary O'Haire - she is our head of Analytics and basically downloaded the source data and built the dashboard within a couple of hours!
Looking forward to more insights from you and the industry as we work to make this a more welcoming place for international students.
Ah, Bloomberg is posting through August…
Remind me, what is the difference between the SEVIS numbers you are citing and the ones that Bloomberg references here - https://www.bloomberg.com/news/articles/2025-09-22/us-student-visa-arrivals-fall-to-four-year-low-led-by-asia-drop?embedded-checkout=true
Are Bloomberg citing earlier/ lagging totals?
Thanks for the question, Clay. Bloomberg reported that "Student arrivals to the US dropped 19% in August year-on-year to just over 313,000 – a fifth consecutive month of declines, according to data from the International Trade Administration that tracks arrivals to the country." My analysis for this article is based on the SEVIS data just posted by the Department of Homeland Security through September. The need to "triangulate" various data sources will be important, especially as the USDOS Monthly Nonimmigrant Visa Issuance Statistics become avaialble (they are only available through May at the moment) and IIE does its annual Fall Snapshot survey. It will take time for us all to make sense of these multiple data points, especially if they send conflicting signals. SEVIS is one data point -- an important one for sure -- but there will be more to come.
I would add that the Bloomberg data is 'arrivals' - and this year, anecdotally there is a lot of talk about how many students decided to stay in the US this summer. As such there is potential that a lot of the 'arrivals' are really a lot of first year students, while the second or third or fourth year year students didn't 'arrive' as normal - as they stayed! I am seeing a lot of enrollment reports from larger colleges which are quite stable on average, some are up and some are down a bit, some smaller colleges in the middle of nowhere seem down a lot, but overall I was expecting down 5-8% region after looking at the data college by college so far.
Great insight -- not as many "arrivals" -- perhaps because fewer summer trips home (or abroad) by second, third, or fourth year students -- thanks for sharing. The link Clay shared is great -- and aligns what Ryan Allen calls an "Age of Conquest" (see link in article to his analysis of US higher sector change over time) -- with market concentration "at the top" with declines at regional publics, small privates, particularly ones that took "big bets" on countries where we see declines (or turbulence) in enrollment, e.g., China, India, etc. Open Doors will be the go to source for this institution-level data (and good reporting between now and November, like Ashley Mowreader did for Inside Higher Ed!)
Thank you! Appreciate your thoughtful analysis and your detailed response. Cautiously hopeful that the negative impact of flawed policies will be delayed or blunted, if not avoided. Most reporting from institutions is that they are flat (especially among more prominent R1 universities & blue state/ coastal LACs) or down (among other Carnegie classes) - https://www.insidehighered.com/news/global/international-students-us/2025/09/23/international-enrollment-down-small-regional
A couple of points I would make (as a non professional in this field) is
1. From what I can see from searching through daily reports from various colleges which randomly report their enrollment figures (mostly including intl students) I would say that even if total F1 visas are down say 15% in 2025, the dollar weighted average fees paid will be down a lot less. This is due to mix - the more expensive colleges 'sell out' every year, the constraint is not demand it's supply. So if for example you hear that "Indian students are shunning the US" it doesn't matter for the top collages as they will always fill up and they charge 3-4x what some regional colleges charge that may be bearing the brunt of any demand destruction.
2. Aligned with this is a possibly controversial point in that the students attending the top schools are just better able to plan and navigate their way through adversity in terms of attaining visas etc - yes they are just a bit smarter! Also, these top colleges will have better infrastructure to help these students navigate a path to the US, plus these students will have higher confidence that the college will have backup plans such as online or temporary overseas campuses should the need arise. You can see why Trump made the comments he made at the time he talked up the 600,000 Chinese students when he referenced some colleges being in trouble - these are often red state often Christian colleges that are being hurt most by the visa issues we see so he is likely getting an earful from certain interested parties.
3. The trend of concentration of wealth is something that is happening in most countries. Why is that a factor? Because it ensures that there is enough demand from wealthy families from many countries to seek a US education, which remains the gold standard alongside certain UK colleges. Witness the rise in applications from Vietnam or Nepal.
One point related to the accuracy of the data - Chris if you look at your summary of the state by state stats, Indiana stands out to me in that this data shows an 8% increase, however if you look at the press reports from IU it's down 14% and Purdue is down 6% (didn't find anything on Notre Dame). That down 14% may end up being down a lot less as these students show up - it's not clear if they included the students who gained admission but didn't show up as enrollees yet,
due to visa delays. Anyways the best way to gauge what's going on right now is just reading through what the bigger schools have said about Fall enrollments as most break down their international intake - so far it's a modest decline and nothing like the 30%+ some (NAFSA) were talking about as worst case (due to visa delays).
That's really useful to know about Indiana and the disparity between the reported data from the institutions with the largest international enrollments historically and what the SEVIS data indicate. It might be another "case of OPT" where enrollments are down but OPT participation is buffering (or counterbalancing ) enrollment decline (since SEVIS combines and does not disaggregate the two, for anyone who may not be familiar with the SEVIS dataset).
It will be important to watch as more data comes in and a "fuller picture" comes into view. I also appreciate the candid assessment of your three points. You've come to the right place -- Distributed Progress is a community of people thinking together about these issues -- variation in perspectives is one of the best parts, and you raise a number of issues I had not considered and will now consider more seriously, as more data comes in.
Agree with the concern expressed that the SEVIS data may not be reflecting what many universities are reporting. While we expected regional publics & smaller tuition-driven privates to struggle, their share of all international undergraduates within the U.S. higher education ecosystem is relatively modest. But if R1s (especially flagships & land-grants) are seeing substantial declines, that’s a different story, on a very different scale. https://www.jsonline.com/story/news/education/2025/09/29/international-student-enrollment-down-at-uw-madison/86355163007/