In Pursuit of the Perfect Letter of Reference

Campbell, R.W., M. Boersma, J. Dower, G. Muller-Parker, C.S. Weiler

Reference letters are an important part of a young scientist's life - recieving a good recommendation can tip the scales when applying for a grant, fellowship, or job.  As well,  as one becomes involved with the supervision of students, etc., one will eventually begin to recieve requests for letters of reference,  and there are not many resources available on the topic of what exactly consitutes a "good" letter of reference.   If one has not had the opportunity to see many reference letters,  writing those first few letters can be a difficult task.  Below are some resources for both writing, and receiving, a good reference letter,  put together by Rob Campbell from discussions at the DIALOG VI symposium and suggestions by John Dower, Sue Weiler, Gisèle Muller-Parker, and Maarten Boersma.



Links:

Books:




Tips for writing a good reference letter:

1.  As you begin,  remember that your job is to honestly represent the candidate's strengths,  and focus on them (not their weaknesses). Some referees seem to feel obligated to include something about a candidate’s weaknesses as well as their strengths (this seems to be particularly common among government researchers). If you do not think the person is appropriate for the opportunity, whatever it is, tell them and give them the option of then selecting another reference.

2. Don't equivocate.  Make it clear, right from the start of the letter, what you think about this person's appropriateness/strengths for the job in question. For instance, whereas many (most?) reference letters start with something banal like "I have known Dr. Blogs for three years" (a particularly bad example, from the NACE site). Try to open with something more like "Dr. Blogs is without doubt one of the top young prospects in the field of Y and will be an outstanding addition to your department." Key point: If the person writing the reference letter can't get excited about the candidate, neither will the Search Committee!

3. Be concise. Don't waste undue space prattling on about the candidate's academic record. You can safely assume that the members of the Search Committees can read, and that they will no doubt have noticed the CV in the application package. Spend a sentence or so summarizing the highlights: "In terms of academic excellence, Blogs' CV speaks for itself, as does the fact that he was awarded the prestigious Whatchamacallit Fellowship for his doctoral studies." Be sure to mention anything in the CV that make the candidate particularly suited to the specific opportunity.

4. It is sometimes useful to give some kind of 'rating'. For example, "of the 50 students I have supervised or otherwise interacted with over the past 10 years, Dr. Blogs is among the top x". If  you are a relatively new faculty member, it might be appropriate to mention that, but say something like: "of the x graduate students and post-docs I have interacted with, Dr. Blogs is among the top x". Use this technique sparingly! If you grade every candidate you write a letter for as in the top 5%, people will notice, and your opinion will become less valuable.

5. Give tangible examples of things that make the candidate suitable for the job in question. Anecdotes are particularly useful, because they give the person reading the reference letter a sense of what the candidate is really like. For instance, if the position specifically seeks a candidate with sea-going experience then instead of writing that "Dr. Blogs has considerable sea-going experience", write something like "Blogs is as much at home at sea she is in the lab. In fact, her dogged determination to get the job done (coupled with an iron-clad stomach!) have earned her considerable admiration within the sea-going oceanographic community."

6. Put the candidate's key scientific contributions in perspective: Keep in mind that, in most cases, only one or two members of the Search Committee will be bona fide experts in the particular field in question. Therefore, it is important to make it clear why the candidate's accomplishments are noteworthy. So, rather than saying that "Dr. Blogs has published several papers in reputable journals and has presented his results at international meetings", try something like "Dr. Blogs' papers have appeared in the top two journals in her field, and are already being regularly cited by her peers. Her work has generated considerable international interest as well, and this past year she gave an invited talk at the ASLO Ocean Sciences Meeting (the most important annual meeting in her field)".  Remember that you, the referee, can 'brag' about the work a bit more than might be seemly by the candidate.

7. Hit all the key points (and hit them hard) but know when to stop. Opinions on the “ideal length” of a good reference letter vary from between 3/4 of a page to 2 pages. John Dower suggests: “If at all possible, I keep reference letters to a single page - if one well-crafted page hasn't convinced the Committee that you can walk on water then a 2nd page is just going to annoy them even further”. One way to keep things succinct is to avoid repeating information that the candidate has already presented elsewhere in their application (e.g. in their CV, their statement of research interests, etc.).

8. Give examples that indicate personal characteristics/strengths. Key point: In addition to looking at your professional credentials, Search Committees look at job candidates as future potential colleagues (i.e. who they will have to interact with on a daily basis for the next 20 years). Couching the candidate's skills in a way that makes them seem like a real person, and someone that you like to work with, with can help to get the Search Committee on their side. John Dower suggests:  "After the initial introductory sentence, think about referring to the candidate by their first name in the rest of the letter - it’s a bit like the strategy used by hostage negotiators to get the kidnappers to see their hostage as a real person".  Not everyone does this, and Sue Weiler notes: "In recent DIALOG letters, reviewers are increasingly using the last instead of first name, and sometimes even just the initials, in the body of the letter after the first mention (e.g., "Mary Jane Blogs" or Matthew Jason Blogs" become "MJB". Referring to "MJB" or "Blogs" instead of "Mary Jane" or "Matthew Jason" tends to make the entire letter seem more gender neutral. Maybe I'm too old-fashioned here--I started publishing in the era when many women would use initials instead of names for published papers so the reader would not be able to identify the gender. Hopefully this new generation carries less baggage.... See for yourself how it works on you -- For example, compare "Blogs", "Mary Jane", "Matthew Jason" and "MJB" in the sentences from section 6. Of course, as John indicates, it is important to show the candidate as a person. Use of first or last name is situation-specific, and something to keep in mind at all times." In addition to the professional credentials, remember to include perspectives about personal attributes. For example, "Having worked extensively with MJB, I can predict that you will enjoy having Blogs in the lab. MJB is bright, energetic, works well in a group or alone, fits in well in any situation and is absolutely dependable and delightful to work with. MJB is one of those rare people that has the personality, maturity and combination of curiousity and intellect to make almost any group better." Specific examples are useful as well. For example, "MJB is an extremely intelligent, well educated, and extraordinarily flexible scientist who is willing to take on any challenge. Blogs recently taught two weeks of an introductory course I am teaching with XX students and by all accounts did exceptionally well."

9.  Remember that cultural differences are enormous between countries. If you are a non-US person writing a letter for the US system, keep in mind that letters in the US are normally much more glowing than letters from other countries. In many countries a letter of reference is simply a statement that the person worked in the organization, and of course this will not be received well if sent to a US institutions.  This does not need to be highlighted (e.g. avoid statements like "I am a European so I write honest letters", which is less uncommon than you might think), but  make sure that the cultural difference if present is explained well.  Be aware that there are sometimes "secret codes" that are used in some countries. For example, something like "Blogs was well liked by her colleagues", could in some circles be interpreted as "Blogs spends too much time socializiing". If in doubt ask the person applying for the position, or better still a person familiar with the specific institution.  Try to avoid sentences which can be interpreted in more than one way. A few words of explanation can avoid much ambiguity.

10. In summary, be honest, clear and concise, and focus on the candidate's strengths.



Things to do to receive a good reference letter:

1. Pick your referees carefully. Remember that a letter from a leader or leaders in your field will carry much more weight that someone less well-known. Select your referees so they can address different aspects of your background experience and abilities (i.e., at least one or more of your 3 referees should be able to talk about more than just your research). If at all possible, include your Ph.D. advisor as one of the referees. If you do not, readers often assume the worst. If there is a good reason for not including your advisor, explain yourself or ask one of the other references to explain why the advisor is not being used (away at sea? maternity leave? deceased? jealous of your fabulous work? stole your ideas...)

2. Give your referees all of the background information about you they will need to write a complete and thoughtful letter of recommendation (CV, statements of research and teaching interests, list of courses you can teach, etc.). Update the information on a regular basis. Mention to them any accomplishments/expertise/etc. that you think make you particularly well suited for the opportunity.

3. Don't just send the referee links to the university or position announcement. Give your referees a short synopsis of the announcement, and include the following information in a short bulleted list of job specifics, including:
• Size of university/college and terminal degree offered (BS, MS, PhD);
• Mission of the university (or mission of the department to which you are applying);
• The job expectations (the focus on research/teaching, undergrad/grad, disciplinary emphases);
• What you would like them to write about or emphasize (things in your CV that are most noteworthy, or relevant);
• Why the job is particularly well-suited to your skills and interests;
• Why you think you are the right person to teach the courses required;
• Optional (should definitely be in the candidate's cover letter, but sometimes worth including in reference's letter as well):

• How your research fits in with that of others on the faculty;
• Your ability to get external funding within the context of the university environment.

4. Be nice to your referee! Give your referees plenty of lead time and thank them for their efforts. Keep them posted on developments in your job search.

Acknowledgments: DIALOG is supported by the U.S. National Science Foundation (NSF), National Oceanographic and Atmospheric Administration (NOAA), National Aeronautics and Space Administration (NASA), and Office of Naval Research through NSF/OCE-0217056 and ONR-N00014-98-1-0590 to Whitman College, C.S. Weiler PI.