Scholarship Recommendation Letters: Two Real Letters for One Applicant, Annotated

Most advice about recommendation letters stops at "choose someone who knows you well." That is true and it is not enough. The applicants who use letters best do not think about one letter at a time. They think about the set — who covers what, which referee vouches for the classroom and which for the lab, and which one is best placed to answer the doubt sitting in the file.

This article shows that thinking in action. Below are two real recommendation letters, quoted in full, both written for the same applicant — a chemical-engineering graduate who won a funded PhD place. One is from the head of the research institute where he worked. The other is from his master's supervisor. Read together, they are a small masterclass in division of labour, and one of them does something most letters never attempt: it takes the applicant's single biggest weakness and defuses it in two sentences.

These are genuine letters from a published collection of admitted applications. The applicant applied from China to a US graduate programme around 2001, so the test-score context is of its time. What carries over completely is the structure — what a strong letter contains, and how two letters should fit together rather than repeat each other.

What a Committee Actually Uses Letters For

Your statement is you speaking about yourself. Your letters are the only part of the file where somebody else does. That is their entire value, and it defines what a committee reads them for.

A committee uses letters to do three things a self-statement cannot: corroborate the claims you made about yourself, triangulate your character from people who watched you work, and calibrate the risk in anything that looks weak. A letter that only repeats your CV in someone else's voice wastes all three. A letter earns its place when it tells the committee something they could not have known from your own documents.

That is why the strongest applicants engineer their letters as a set. If both referees say the same three things, you have one letter signed twice. If they divide the terrain, you have coverage.

The Principle: Two Letters, Two Jobs

The applicant below — call him Mr. Zhang, as the letters do — was applying for a PhD after a master's in chemical engineering and a stint as a research associate at a microbiology institute. He had two obvious referees available: the head of the institute where he now worked, and the professor who had supervised his master's.

He did not ask both to describe the same qualities. The letters split cleanly:

• Letter 1 (research head) covers his present self — his work in the institute's industrial R&D, his conduct in seminars, his English in a working setting. It speaks to who he is now, as a researcher and colleague.
• Letter 2 (master's supervisor) covers his formative self — his degree, his thesis, his three years of training. It speaks to how he was built.

Between them the committee gets a complete arc — trained here, performing there — with almost no overlap. Now the letters themselves.

Letter 1, Annotated: The Research Head

How the letter establishes standing. The first move is to explain how the referee knows the applicant — recruited him personally, works with him daily. A committee weights a letter by how well the writer actually knows the candidate, so a strong letter answers that question in its first three sentences instead of leaving it to be guessed.

The seminar anecdote. The letter does not say "he is intellectually bold." It says he stood up at a named symposium and challenged the speakers by pointing out hurdles in their work and offering his own strategies. That is boldness shown, not asserted, and it is the single most memorable thing in the letter. One concrete scene outweighs a paragraph of praise.

Named outputs. Two book chapters, a patent application, two named bioproducts. A committee can, in principle, verify any of this. Verifiable specifics are what separate a letter that vouches from a letter that merely flatters.

Letter 2, Annotated: The Academic Supervisor

Different window, same person. This referee knew Mr. Zhang for three years as a student, so the letter speaks to the things a supervisor sees and a department head cannot: how he was trained, how his thesis went, how he grew over time ("his communication skills and cooperation capacity improved"). Notice that growth is a claim only a long-term supervisor can credibly make — and it is more persuasive than a static superlative, because it implies observation over time.

One honest limit. "I can confirm that he is one of the excellent students I have directed" — one of, not the best I have ever taught. A measured superlative from a credible source reads as more honest than an absolute one, and committees have learned to discount letters that call every student the finest in a decade.

The deliberate overlap. Both letters mention the same trait — his habit of finding the weakness in an argument and proposing a fix (Letter 1: "pointing out their potential hurdles and offering his alternative strategies"; Letter 2: "point out potential hurdles, and offer unique solutions"). This is the one repetition the set allows, and it is intentional. When two independent referees who watched you in different settings name the same signature strength, that trait stops looking like self-promotion and starts looking like fact.

The Move Most Letters Never Make

Reread these two sentences from Letter 1:

This is the most valuable thing in either letter, and almost no recommendation letter does it.

The applicant had a real problem: standardized test scores that undersold his English. He could have addressed it himself in his statement, but self-defence about a weak score is weak by nature — of course the applicant thinks his English is fine. Coming from the head of a research institute who has watched him "fluently discuss our research program in English with visiting scholars from Japan, Europe and the US," the same claim is evidence. The letter names the weakness, outranks it with first-hand observation, and closes the door.

This is the letter-writing equivalent of a rule that runs through winning applications: a weakness is safest when someone credible neutralizes it directly. A referee can do this even more powerfully than you can, because they have no incentive to flatter and every incentive to protect their own credibility. If there is one number or gap in your file that worries you, the single best use of a letter is to have the right referee address it head-on. For how to handle the same weaknesses in your own voice, see How to Explain a Low GPA, a Gap Year, or a Change of Field.

Specific Behaviour Beats Every Adjective

The clearest line between a strong letter and a forgettable one is specificity. Watch the difference inside a single sentence from a third letter in the same collection, written for a medical applicant by his hospital supervisor:

A weak letter would have written "he is compassionate and patient-centred." This one shows the compassion as behaviour — chatted, cheered up, relieved anxiety — and then draws the larger point. The reader believes it because they can picture it.

Every adjective in a recommendation letter is a claim the reader has no reason to accept. Every specific behaviour is evidence they can weigh. The best letters are almost entirely the second kind. When you brief a recommender, the most useful thing you can give them is not a list of your virtues but a list of things you actually did that demonstrate them.

How to Brief a Recommender

In much of the world, and especially for Vietnamese applicants, the practical reality is that a busy referee will ask you to draft the letter, or to give them detailed notes. Treat that not as an awkwardness but as your chance to engineer the set.

• Assign each referee a job. Decide before you ask: this professor proves my research ability, this employer proves my professional judgement, this teacher proves how I work with others. Tell them the facet you are hoping they can speak to.
• Give them raw material, not adjectives. Send each recommender two or three concrete episodes — a project, a problem you solved, a moment in class — with dates, names, and outcomes. Let them choose the words. Your job is to supply the evidence.
• Point one referee at your weakness. If a score or gap worries you, ask the referee best placed to observe the real skill whether they would be comfortable addressing it. Do not script the sentence; give them the truth and let them vouch in their own words.
• Avoid identical letters. If you draft more than one, make sure they do not share sentences or cover the same three qualities. Two letters that read alike are read as one.
• Brief them on the target. A referee who knows the scholarship values leadership, or research, or development impact can angle honestly toward it. For what different funders weigh, see What Scholarship Committees Actually Look For.

What a Weak Letter Looks Like

By contrast, the letters that add nothing share a few traits:

• They only restate the CV. If everything in the letter is already on your transcript, the referee has vouched for nothing.
• They are all adjectives. "Hardworking, intelligent, responsible, and dedicated" with no scene attached is invisible; committees read it a thousand times a season.
• They never explain how the writer knows you. A glowing letter from someone whose relationship to you is unclear carries little weight.
• They over-claim. "The best student I have taught in twenty years" from a referee who taught you one course reads as inflation and taints the rest.
• They duplicate each other. Three letters making the same three points are weaker than two that divide the ground.

Checklist: Your Recommendation Set

• Does each referee cover a different facet of you — research, professional, academic, character — with minimal overlap?
• Does every letter explain, early, how the writer knows you and for how long?
• Does each letter contain at least one specific, verifiable episode rather than a list of adjectives?
• Is there exactly one deliberate overlap — a signature strength two referees independently confirm?
• Have you pointed the best-placed referee at your single biggest weakness?
• Have you given each recommender concrete raw material — projects, dates, outcomes — rather than a list of your virtues?
• Do the letters avoid sharing sentences or reading as if one hand wrote them all?
• Does at least one letter speak to what this particular scholarship values?
• Would a sceptical reader finish the set knowing something about you that your own statement could not have told them?

For the essay these letters sit alongside, read How to Write a Scholarship SOP That Gets You Shortlisted. To see a full statement annotated, read Anatomy of a Winning SOP: A 500-Word Statement That Won an MIT Fellowship.

Frequently Asked Questions

How many recommendation letters do I need for a scholarship?

Most scholarships and graduate programmes ask for two or three. The number matters less than the coverage: each letter should speak to a different facet of you, so that together they corroborate your research ability, your character, and your professional or academic performance without repeating one another. Two well-divided letters beat three that make the same points.

Who should write my scholarship recommendation letter?

Choose referees for how well they know you and what facet they can prove, not for their seniority. A professor who supervised your thesis and can describe your actual work is worth more than a famous name who barely knows you. The strongest sets pair someone who can speak to your academic training with someone who can speak to your research or professional performance.

Can a recommendation letter address a weakness like a low test score?

Yes, and it is one of the most powerful things a letter can do. In the real letter analysed above, a research head wrote that the applicant's true English ability exceeded his standardized scores, neutralizing the one number that could have sunk his file. A credible referee vouching against a weakness carries more weight than the applicant defending it, because the referee has no incentive to flatter.

Is it normal for the applicant to draft their own recommendation letter?

In many countries, including Vietnam, busy referees often ask the applicant to draft the letter or supply detailed notes. If this happens, use it deliberately: assign each referee a distinct facet, give them concrete episodes rather than adjectives, and make sure multiple drafts do not read alike. The referee must review, edit, and genuinely endorse the final letter.

What makes a recommendation letter strong?

Specific, verifiable behaviour over adjectives. A strong letter explains how the writer knows you, then shows your qualities through concrete episodes — a project, a question you raised, a problem you solved — that the reader can picture and, in principle, verify. Weak letters restate the CV and stack adjectives; strong letters tell the committee something your own documents could not.

Your Next Step

List your possible referees and, beside each, write the one facet of you they are best placed to prove. If two of them would prove the same thing, you have a coverage gap — find a third who covers different ground, or brief one of the two to speak to something else. Engineer the set before you send a single request.

ScholarshipHub is free to use and built specifically for students like you: track your applications, store recommender details and deadlines, and never miss a date. Start your free account →