How do UK study visa refusal rates differ by nationality?
In the year ending June 2026, the Home Office refused 4.9% of decisions on UK study visas. For applicants from some countries the share was above 10%. For others it was close to zero. Here are the published numbers, with the source beside them.
Source: UK Home Office, Immigration system statistics, year ending June 2026 (table Vis_04). Checked against the table on 16 October 2026.
of all study visa decisions were refusals: 19,608 of 403,063.
of decisions for applicants from Pakistan were refusals: 4,778 of 26,105.
of decisions for applicants from China were refusals: 102 of 86,716.
Share of study visa decisions that were refusals
Refusals as a share of decisions (grants plus refusals), sponsored study visas, year ending June 2026. The chart shows the five nationalities with the largest shares among the 20 with the most applications, the overall figure, and three others for comparison. All 20 are in the table below and in the data file.
Source: UK Home Office, Immigration system statistics, year ending June 2026, table Vis_04, published 27 August 2026. Refusals as a share of decisions, calculated by B4A from the Home Office counts. Includes main applicants and dependants. Provisional from the first quarter of 2026. Open Government Licence v3.0.
The 20 nationalities with the most applications
Show or hide the table
| Rank | Nationality | Decisions | Granted | Refused | Refused share |
|---|---|---|---|---|---|
| 1 | India | 87,804 | 83,958 | 3,846 | 4.4% |
| 2 | China | 86,716 | 86,614 | 102 | 0.1% |
| 3 | Nigeria | 37,424 | 33,080 | 4,344 | 11.6% |
| 4 | Pakistan | 26,105 | 21,327 | 4,778 | 18.3% |
| 5 | Nepal | 17,977 | 17,150 | 827 | 4.6% |
| 6 | United States | 16,920 | 16,850 | 70 | 0.4% |
| 7 | Bangladesh | 11,656 | 10,021 | 1,635 | 14.0% |
| 8 | Turkey | 5,911 | 5,834 | 77 | 1.3% |
| 9 | Saudi Arabia | 5,740 | 5,626 | 114 | 2.0% |
| 10 | Ghana | 5,040 | 4,155 | 885 | 17.6% |
| 11 | Kuwait | 5,080 | 5,051 | 29 | 0.6% |
| 12 | Malaysia | 4,871 | 4,859 | 12 | 0.2% |
| 13 | Sri Lanka | 4,140 | 3,706 | 434 | 10.5% |
| 14 | France | 4,133 | 4,117 | 16 | 0.4% |
| 15 | Canada | 4,005 | 3,982 | 23 | 0.6% |
| 16 | Germany | 4,118 | 4,102 | 16 | 0.4% |
| 17 | Hong Kong | 4,076 | 4,068 | 8 | 0.2% |
| 18 | South Korea | 3,780 | 3,765 | 15 | 0.4% |
| 19 | Spain | 3,308 | 3,292 | 16 | 0.5% |
| 20 | Thailand | 3,155 | 3,152 | 3 | 0.1% |
| Other nationalities | 61,104 | 58,746 | 2,358 | 3.9% | |
| Total | 403,063 | 383,455 | 19,608 | 4.9% |
Rank is by number of applications. A decision is a grant or a refusal. Refused share is B4A’s calculation: refused divided by decisions, rounded to one decimal place. Other nationalities are all those outside the top 20.
A new limit for universities, from 1 June 2026
Since 1 June 2026, a UK university or college that sponsors international students must keep its own visa refusal rate below 5% to pass its annual Basic Compliance Assessment. The limit was below 10% before that date.
The rule applies to each institution’s own record, not to any nationality. Because refusal rates differ by nationality, the countries an institution recruits from can affect how close it sits to the limit. The overall figure in this table, 4.9%, sits just under 5%.
What these numbers do not show
- Why applications were refused. The table gives counts, not reasons.
- Withdrawn and lapsed applications. They are not counted as decisions, so they sit outside every rate here.
- Who is in the group. Main applicants and dependants are counted together.
- The chance for one person. A nationality is a large group. Each application is decided on its own evidence, so a group rate says nothing certain about one applicant.
- Timing. Outcomes are counted by the date of the decision, so a decision can relate to an application made earlier. Figures from the first quarter of 2026 onwards are provisional.
- Countries outside the top 20. They are grouped as other nationalities: 2,358 refused of 61,104 decisions, or 3.9%.
From the Home Office table to this page
- We took the counts for each nationality from table Vis_04 in the Home Office summary tables.
- We recalculated every rate as refused divided by decisions. In every row, grants plus refusals equal decisions, and the rows add up to the total.
- AI helped us find and sort the figures. A person checks and approves the key figures and charts before a story is published.
Contains public sector information licensed under the Open Government Licence v3.0. Refusal rates are B4A calculations. The Home Office does not endorse this page and is not responsible for our calculations. This page is information, not immigration advice. Data sources and disclaimer.
Planning to study in Europe?
New Life Europe is our sister practice. It helps students and families compare study routes in Europe and prepare for the move. The numbers on this page are public statistics, not advice on your own case.
Questions people ask
What does refusal rate mean here?
It is the number of refused applications divided by the number of decisions. A decision is either a grant or a refusal. Applications that were withdrawn or lapsed are not counted as decisions. The Home Office table gives the counts, and we calculated each rate from them.
Does this mean my own application will be refused?
No. These are totals for large groups of applicants. Each application is decided on its own evidence. The table cannot tell you the chance for one person, and it gives no reasons for refusals. For advice on your case, speak to a licensed immigration adviser.
Which visas and applicants are included?
Sponsored study visa applications made from outside the UK, for main applicants and their dependants together. Applications to extend a stay from inside the UK are in separate Home Office tables.
How often do the figures change?
The Home Office publishes new figures every quarter. This page shows the year ending June 2026, published on 27 August 2026. Figures from the first quarter of 2026 onwards are provisional and can be revised.
Can I reuse the data?
Yes. The Home Office publishes it under the Open Government Licence v3.0. Our file adds one column, the refused share, which is our own calculation. Please name the Home Office as the source.
More data stories
Public data on migration, trade, finance, education and tourism, checked against the primary source.
New Life Europe
Our sister practice helps students and families compare study routes in Europe and prepare for the move.
Have data like this? We will turn it into a decision.
Bring a report, a spreadsheet or a question. We will tell you what the numbers can and cannot say.