How many PhD students can one supervisor take in Vietnam?

Direct answer: The number depends on the date a candidate enrols. Candidates enrolled before 1 January 2027 stay under Circular 18/2021, which allows a professor 7, an associate professor 5 and a PhD holder 3. Intake announced from 1 January 2027 follows Circular 53/2026, whose Article 17(4) keeps a ceiling but points to the ministry's intake rules.

Bộ Giáo dục và Đào tạo (2021) wrote the older limit into Article 6(2) of Circular 18/2021/TT-BGDĐT: "Tại cùng một thời điểm, người có chức danh giáo sư được hướng dẫn độc lập tối đa 07 nghiên cứu sinh" (at one time, a person holding the professor title may independently supervise at most 7 doctoral candidates). The same paragraph gives 5 for an associate professor or a doctor of science, 3 for a PhD holder, and counts co-supervision of one candidate as 0.5.

Circular 53/2026/TT-BGDĐT replaces that text. Signed on 30 June 2026 and published in Official Gazette No. 447 on 28 July 2026, it applies to master's and doctoral intake announced from 1 January 2027, and Article 34(1) leaves earlier enrolments under the 2021 rules (Bộ Giáo dục và Đào tạo, 2026b). Its Article 17(4) states: "Một giảng viên hướng dẫn số lượng nghiên cứu sinh đối với từng ngành đào tạo trình độ tiến sĩ không vượt quá số lượng tối đa quy định của Bộ Giáo dục và Đào tạo về xác định số lượng tuyển sinh trình độ tiến sĩ ở cùng một thời điểm" (a lecturer's doctoral candidates in each field may not exceed the maximum the ministry sets for determining doctoral intake, at the same time). The maximum is in Article 5(1)(c) of Circular 34/2026/TT-BGDĐT (Bộ Giáo dục và Đào tạo, 2026a): 8 candidates per professor, 6 per associate professor or doctor of science, 3 per PhD holder. That these ratios double as the personal ceiling is our reading of the cross-reference, since the circular does not spell it out, so confirm it with the department.

Source: Bộ Giáo dục và Đào tạo (2021, Arts. 5-6; 2026a, Art. 5; 2026b, Arts. 2, 17, 34), read 2026-10-31.

Direct answer: Appendix I of Circular 34/2026 defines doctoral training capacity as 3 per PhD holder, 6 per associate professor and 8 per professor. Maximum intake is that capacity minus three quarters of current doctoral enrolment. A department can therefore be full while one named professor still looks free.

The formula sits in Table 1D of Appendix I, which Bộ Giáo dục và Đào tạo (2026a) prints in the Official Gazette of 10 May 2026: capacity equals the number of PhD holders times 3, plus associate professors times 6, plus professors times 8, and maximum intake equals capacity minus enrolment times three quarters. The three-quarters factor implies a cycle of roughly four years, which is our inference and not a figure the appendix states.

A worked example shows the effect. Take a hypothetical field with 2 professors, 3 associate professors and 4 other PhD holders. Its capacity is 2 × 8 + 3 × 6 + 4 × 3 = 46. If 40 candidates are enrolled, three quarters of 40 is 30, so the field may admit at most 16. Add 6 more enrolled candidates and the maximum falls to 46 − 34.5 = 11.5. The example is ours, built from the circular's formula, and no real department is meant.

Two details limit who counts. Under Article 3(6), a lecturer counted for doctoral intake must meet the supervisor standard and hold a full-time contract of at least 3 years with the institution. Enrolment is counted on 31 December, institutions update their data by 30 January and report intake by 28 February under Article 6, and adjustments are republished before 31 May under Article 7 (Bộ Giáo dục và Đào tạo, 2026a). Room with a supervisor is a seasonal figure, not a fixed trait of the person. The 2026 round still uses the earlier counting method.

Direct answer: Vietnam fixes one national number by title, while the UK and France leave the number to institutions. Taylor (2021) found that 59 of 70 UK institutions with central limits set them between six and eight full-time equivalents. French doctoral schools set their own maxima, for example 5 candidates at Grenoble and 4 at Perpignan.

Taylor (2021) searched the public documents of 143 UK institutions in January and February 2021. Of them, 45 mentioned no limit, 28 left limits to schools or departments, and 70 published central limits. Of 20 institutions that limited every supervisor by full-time equivalent, 18 chose six. Overall 59 of the 70, or 84%, fell between six and eight, while 7 set more than eight.

Published limits and real loads can still differ. In the UK Research Supervision Survey 2024, which gathered 5,174 usable responses from supervisors in 141 institutions, the average supervisor held 5.5 candidates in total and 3.1 as main supervisor (n = 1,314) (UK Council for Graduate Education, 2024). Some 30% reported no known limit and 34% were unsure. Of those who knew of a policy, only 29% said it was always followed. The report states: "While 74% of respondents felt that one to four candidates were optimal, only 20% of policies had a maximum this low" (UK Council for Graduate Education, 2024, p. 8). Where hours were allocated, a main supervision slot carried an average of 42 hours a year, against 69 hours actually reported.

In France, Ministère de l'Enseignement supérieur et de la Recherche (2016) leaves it to each doctoral school council to set the maximum number of candidates a thesis director may supervise, in its Article 16. Université Grenoble Alpes (n.d.) states a maximum of 5 at once for its ED STEP school, and Université de Perpignan (n.d.) a maximum of 4 for ED305. Our guide to co-direction in France shows how that works for a joint supervision.

Source: the documents named in the last column and the school pages listed in References, read 2026-10-31.

Direct answer: Nobody has measured supervisor headcount against completion in Vietnam. Studies of other systems link finishing to the support a supervisor gives, not to the title or the cap. In a survey of 985 Russian graduates, significant supervisor support more than doubled the chance of a successful defence.

Zhuchkova and Bekova (2023) surveyed 985 doctoral graduates from 119 Russian universities; 74% reported significant support from their supervisors and 45% from their department. Research experience related to the dissertation topic predicted a successful defence until support entered the model, after which it lost significance. The authors report: "The strongest effect is observed for support from the supervisor, receiving which increases the chances for successful defense by more than two times" (Zhuchkova & Bekova, 2023, p. 9). Their odds ratio was 2.50 in one model and 2.11 in the next, and department support added 55% on top. The finding comes from one country's graduates, so it shows an association, not a rule.

Mosanya et al. (2022) interviewed 8 recent graduates of clinical pharmacy doctorates and surveyed 47 current candidates across eight accredited schools in Nigeria. Having more than one supervisor, supervisor commitment and departmental support appeared as facilitators of timely completion, and the authors write that "students who had more than one supervisor experienced a reduced risk of experiencing burnout" (Mosanya et al., 2022, p. 9). The sample is small.

On the supervisor's side, the UK survey found that supervisors formally supporting three to four candidates were more likely to say worry about their candidates kept them awake (35%) than those with one to two (30%) (UK Council for Graduate Education, 2024, p. 8). Read together, these studies make load a reasonable thing to ask about, not a reason to rule someone out.

Direct answer: Surveys in Denmark, Spain and North America describe supervision as a set of behaviours and not a headcount, so when you ask how many PhD students can one supervisor take, also ask what the supervisor actually does for each of them.

None of the studies below counts students per supervisor, which is why they sit beside the legal cap and not in place of it.

Wichmann-Hansen and Nielsen (2024) matter most for the load question, because a supervisor with many students may drift toward controlling or toward being absent, and the Danish pattern shows those are different problems.

Direct answer: Reviews of co-supervision, department money and completion suggest that a second supervisor and departmental support matter alongside the cap, so how many PhD students can one supervisor take is only one of several questions you should put to a department.

Kálmán et al. (2022) are useful because Vietnamese rules already allow a second supervisor, so the review tells you what such an arrangement is usually meant to achieve.

Direct answer: Ask the department, not only the professor: whether the field expects to announce doctoral places this cycle, which rulebook the cohort falls under, how many candidates the supervisor already carries, and whether a second supervisor is possible. A professor cannot read the capacity formula from a title.

Because the formula sits at department level, "Are you accepting students?" is a question the professor may not be able to answer alone. Four questions are answerable. First, will the field announce doctoral places in the coming cycle, and for which cohort year, since the cohort year decides between the 2021 and 2026 rules? Second, how many candidates does the supervisor already carry, counting co-supervision at 0.5? Third, does the supervisor meet the publication requirement in Article 17(1) of Circular 53/2026, namely a Scopus or Web of Science paper, book, patent or utility solution in the past 5 years (Bộ Giáo dục và Đào tạo, 2026b)? Fourth, would a second supervisor be possible, given that each candidate may have 1 or 2?

Practical guides for this stage include how to find a PhD supervisor and build a shortlist, the first email to a professor and questions to ask at a PhD interview. A doctoral advisor is also the person who confirms your topic within 18 months of admission under Article 17(5)(c), so the conversation about load belongs before you accept an offer, not after. The qualification at stake is the Doctor of Philosophy, and the body that sets these rules is Vietnam's Ministry of Education and Training.

Choosing which supervisor to approach, and when, is part of a PhD application strategy built with an expert in your field. MAAS support here is advisory: you write and submit your own application, no admission or grade target is promised, and no refund is tied to an outcome. If the expert is not a good fit, you can ask for a different one.