Getting in

Two in five undergraduates went to a Japanese language school first. One in seven doctoral students did.

"Do I need to go to a language school before I can apply?" is one of the first questions anyone asks, and the honest answer is that it depends on what you are applying to. JASSO asks all 7,867 students in its survey whether they entered their current school straight from abroad — and the split changes dramatically with the level you are aiming at.

There is a question in the survey that almost nobody quotes, and it answers something people ask constantly. Students are asked whether they entered their current school directly after arriving in Japan, or whether they were somewhere else in Japan first. Across everyone, 63.1% came straight in and 36.9% did not.

That average is useless on its own, because the two groups are not distributed evenly. Here is the same question sorted by what the student is enrolled in now — the share who reached that programme by way of a Japanese language school.

An undergraduate applicant is roughly three times as likely to have come through a language school as a doctoral applicant. If you have been told "everyone goes to a language school first" or "nobody does", both statements are describing a real group of people — just not necessarily yours.

Where the other route actually goes

2,905 students did not enter their current school directly. The survey asks where they were immediately before, and the answer is overwhelmingly one thing.

So the indirect route is, in practice, the language school route. Everything else put together accounts for about a quarter of it.

The full picture, by level

Two numbers matter and they are not the same. The first is how many people at your level came in directly. The second is how many of your future classmates arrived by way of a language school.

The last row is worth a second look. Even at language schools themselves, 19.8% were somewhere else in Japan first — and most of those, 15.7% of all language school students, were at a different Japanese language school. Arriving in Japan and arriving at your final school are not the same event, and for a lot of people they are years apart.

At master's level the picture diversifies rather than simply thinning out. Of the 358 master's students who did not enter directly, 209 came from a language school and 59 came from a Japanese undergraduate degree — that is 7.2% of all master's students who did their bachelor's here and stayed on.

What this does not tell you

This is a description of what students did, not of what any school requires. The survey does not ask whether a language school was necessary — only where people were. Some of those 41.5% of undergraduates could have applied directly and chose not to. Some could not have.

Admission requirements are set school by school. A Japanese-taught undergraduate degree generally expects Japanese language ability that most applicants do not have on arrival; a doctorate supervised in English may expect none. That difference is very likely what the gradient in the first chart is made of — but the survey does not say so, so treat that as a reading, not a finding.

What to take from this

• The answer changes by level, not by country. If you are aiming at an undergraduate degree, the language school route is the majority experience. If you are aiming at a doctorate, it is the exception.
• The indirect route is essentially one route. 75.2% of it is language schools. Preparatory courses and transfers between Japanese institutions are small by comparison.
• A language school is a real destination in its own right, not only a corridor — 140,174 people were enrolled at one in May 2025, and that number is growing faster than the universities'.
• Budget for the whole path, not the destination. If your route runs through a language school, that is a period with a different scholarship regime and a different working week — both of which we have measured separately.
• Ask the school, not the internet, whether you can apply directly. This data tells you what is common. It cannot tell you what you are eligible for.

Sources. Japan Student Services Organization (JASSO), FY2023 Survey on Living Conditions of Privately Financed International Students. Table 3-3 (direct entry versus indirect, by level of enrolment, with 2019 and 2021 comparisons), Table 3-4 (where indirect entrants were immediately before). Survey PDF (Japanese). The 140,174 language school enrolment figure is from JASSO's Survey of International Students in Japan, May 2025. On the calculation: the "via a language school" column is ours, not JASSO's. Table 3-3 gives the number of indirect entrants at each level; Table 3-4 gives what share of indirect entrants at each level came from a language school. We multiplied through to the base of all students at that level — for undergraduates, 816 of 1,967, or 41.5%. Both tables use the same 7,867 respondents. On what we left out: short-term university, professional graduate school, research students and preparatory courses are in the survey but have between 36 and 117 respondents each, which is too few to report a percentage from. The doctorate row is doctoral candidates in the later stage of the programme.