CV Tips for Social Work PhD Students

Doctoral students have asked me about CVs often enough that it seemed easier to write it down once. I am not sure my own CV is a model of anything, and most of this is what I settled on for myself, so please take the parts that fit your program and leave the rest.

Research interests

Most social work doctoral students seem to have two things going at once: a population they came to the field for, and a problem they want to solve. Mine were immigrants with language barriers and access to mental health care. In practice the work rarely stays in one pairing. The same topic moves to a different population (e.g., language access in mental health care, first for Spanish and Asian language speakers, later for South Asian and Arabic speakers), and the same population gets a different topic (e.g., the same immigrant communities, studied for AI and technology adoption instead of service access). A single research interests sentence covers only one of those combinations.

What I propose instead is a TPM approach: topic, population, and methodological approach, each on its own labeled line.

• Topic is the substantive problem, for example access to mental health services, housing instability, or child welfare system involvement.
• Population is the group, for example immigrants with limited English proficiency, older adults aging in place, or youth in foster care.
• Method is the methodological approach, for example geospatial analysis, latent class analysis, or community-based participatory research.

Written out as a sentence, in the order topic, population, method, the three parts read as one claim: I study access to mental health services, among immigrants with limited English proficiency, using geospatial analysis and natural language processing.

On the CV the same sentence becomes three lines. Mine reads:

Line	Content
Topic	Minority Mental Health, Mental Health Services, Culturally and Linguistically Appropriate Services, Digital Mental Health
Population	Immigrants, Refugees, and Racial/Ethnic Minorities
Method	Multilevel Modeling, Geospatial Analysis, Computational Social Science, Natural Language Processing, Machine Learning

Three lines because different readers want different parts of it. A school hiring into a child welfare cluster is reading for the population, and a search that opened when the quantitative methods instructor retired is reading for the method. Keeping the three apart makes each one easy to find.

The version of this that belongs in prose belongs on a website. My research page carries the paragraph, the three levels I work across, and the current projects, while the CV carries only the three lines.

I keep each line to four or five items, since a longer list stops being specific.

TPM during PhD training

Naming the three lines early did something for me beyond the CV: it gave me a way to decide which opportunities to take during the doctorate. When someone offers you a co-authorship on a paper outside your area, an RA line on a project you did not design, an invitation to run a workshop, or a dataset nobody has analyzed yet, you can sort the offer by how many of the three lines it touches.

In the first year or two, I would treat one match as enough. If a project touches your topic, or your population, or a method you want to learn, it is probably worth taking even when the other two are wrong. The method line in particular should stay open at this stage. Most students arrive with a topic and a population and no clear sense of which methodological approaches those two will support, and the way to find out is to work on projects that use different ones. A paper on the wrong population with the right method still teaches you the method, the collaborators you meet doing it are the ones who will still be around in five years, and none of the three lines has to be settled yet.

From the third year I would raise the bar to two. By then there is a dissertation to protect and a CV that has started to have a shape, and a project matching only one line costs more time than it returns.

For the dissertation topic itself, I recommend checking whether it touches all three lines before you commit to it. It is the project a committee, and later a search committee, will take as your research agenda.

After the doctorate, the work belongs at the intersection of all three. Positions and grants at that point are awarded on the strength of a research agenda, and an agenda is what the intersection of the three lines looks like once it has been narrowed for a few years.

The effect across five years is a record that widens first and narrows later. The early entries look scattered because they were, and by the end the recent ones cluster.

The method line

The method line is the one I see filled in most vaguely. “Quantitative and qualitative methods” covers almost everyone, so it tells a reader very little, and I would name the designs you can actually run instead.

The words that do some work look like this.

• Qualitative: community-based participatory research, constructivist grounded theory, interpretative phenomenological analysis, ethnography, photovoice, rapid qualitative analysis, qualitative comparative analysis.
• Quantitative: structural equation modeling, latent class and latent profile analysis, multilevel modeling, difference-in-differences, regression discontinuity, propensity score methods, survival analysis, decomposition, and the specific causal inference designs you have implemented.
• Computational: topic modeling, network analysis, geospatial accessibility measures, natural language processing.

A useful way to sort the words is to ask whether one of them could put you on a shortlist to teach a methods seminar, and to cut the ones that could not.

The same three lines look like this for projects other than mine.

Topic	Population	Method
Reunification and system involvement	Families in rural child welfare systems	Survival analysis, administrative data linkage, multilevel modeling
Housing instability and health	Older adults aging in place	Longitudinal structural equation modeling, geospatial analysis
Substance use treatment access	Native American communities	Community-based participatory research, mixed methods, qualitative comparative analysis
School-based mental health	Immigrant-origin adolescents	Latent class analysis, difference-in-differences, survey psychometrics
Intimate partner violence services	Deaf and hard of hearing survivors	Constructivist grounded theory, participatory design, cognitive interviewing

Section order

There is no fixed order below the header, and the order itself is an argument. My rule is to put the material you most want counted where the eye lands first.

University career centers say much the same thing, though not in the same order. Washington tells doctoral students to organize the document so the most relevant material comes near the beginning, and Penn says “whatever is most important merits the most space.” The guides also disagree with each other about the order. Colorado’s graduate school puts service and conference activity ahead of publications. Toronto Mississauga puts community involvement there. A CV guide from a medical school’s faculty affairs office states that “teaching and mentoring information does not go in the CV,” which no social work source would say. The advice is local, so I would read several before adopting any of them.

The social work version, from Jennifer Luna’s CV guide for social workers, lists heading, education, research interest, teaching interest, research experience, teaching experience, publications, presentations, grants, scholarships and awards, clinical and practice experience, university service, service to the profession, community service, and professional associations. That is a reasonable default and it puts the three blocks in the order used here.

The header is fixed by convention: name, contact, education, research interests. After that, the document falls into three blocks in this order: research, teaching, then service and everything else.

Inside each block the order is yours, and the research block is where the choice bites. It holds awards, grants, publications, and presentations, and which of the four goes first depends on which one you want the reader to see. A student with four first-author papers might lead with publications, while a student with a funded F31 and one publication might lead with grants. Neither order is wrong, and each one is a claim about where your strength is.

That logic reverses for a teaching-focused position, where teaching moves up ahead of the research block. Michigan’s career center calls this normal practice: “you can have multiple CVs for multiple purposes,” and if you are applying to research postdocs and teaching-focused lecturer positions in the same season, “it probably makes sense to have both a research focused and teaching focused CV. The order and content would differ with each document.” Reordering takes about ten minutes, and it is the cheapest tailoring I know of.

Institution type

Order matters because the two kinds of social work faculty job weight the same CV differently, and there is survey evidence about how much.

The CSWE and GADE report on doctoral education, built from the 2019 CSWE Annual Survey and the 2020 GADE Director Survey, asked program directors how important various factors are in the academic job search, on a five-point scale. Research doctorate directors rated research productivity 4.39 and practice doctorate directors rated it 2.00. A focused research agenda was 4.19 against 1.86. Practice experience reversed: 3.56 for research doctorate students and 4.80 for practice doctorate students, the highest rating any factor received in the survey. Teaching experience came in at 3.94 and 3.80, and the difference was not statistically significant.

Teaching is the one factor both kinds of program weight about equally. Everything else moves, so the research block and the service block trade places depending on where you are applying.

GADE, the association of social work doctoral program directors, publishes what a PhD student should have accumulated by graduation, and the list amounts to a table of contents for the CV. Its 2023 quality guidelines say that before graduating, students will have “presented at least 2-3 juried presentations at national or international conferences,” “published at least one sole or first authored, peer-reviewed journal article,” published co-authored articles, “independently taught 1-2 BSW and/or MSW courses as the instructor of record,” and “substantively contributed to a research proposal for internal or external funding.” The document also reports that among surveyed programs, 57 percent said their students had one to three co-authored peer-reviewed articles and 32 percent said four to six. These are aspirational targets set by program directors, not a hiring bar, but they tell you which five sections of your CV those directors expect to be non-empty.

The research side also carries a published expectation about volume. Lightfoot, Franklin, and Beltran, writing in the Journal of Social Work Education, report that search committees at research-extensive institutions “expect new PhD social work graduates to have at least four to five peer-reviewed publications listed in a candidate’s curriculum vitae.” They immediately qualify it by quoting Khinduka, who argued that a candidate’s publications need to reflect “a thematically integrated body of work” with consistent and coherent themes, and warned that “a glut of inconsequential publications needs to not be confused with serious intellectual work.” So the count is a floor, and a reader still has to be able to see one line of inquiry across the whole list, one more argument for the three lines at the top of the document.

Header and education

The header is short and almost every part of it is settled convention.

Name, email, a website if you have one, and an institutional address. I leave out the photograph, the date of birth, the marital status, and the citizenship status. Drexel’s guide says “do not include any personal information such as date of birth, sex, marital status, race, or citizenship status.” International students are the readers most often tempted to explain a visa situation here, and that explanation belongs in a conversation with the search chair after an offer.

Education is in reverse chronological order and carries more than the degree name. Mine lists the advisor, the dissertation title, and the full committee, because a committee is a list of people the reader may know and because it is the fastest available evidence about training. Luna’s social work guide gives the same instruction: “if you are a doctoral student or seeking a faculty position, include your dissertation title and committee chair.”

For education I include degrees and substantial training, and nothing smaller. A one-day webinar, a completion certificate from an online course, and a departmental workshop look out of place next to a doctorate, and they read better in a training section further down. My CV keeps a skills and training section for that material, which leaves the education block to three degrees and their committees. Minor credentials get the same treatment, since a list padded with small items makes the substantial ones harder to see.

A degree in progress gets the word expected and a date. Penn’s example is “PhD Immunology, expected May 2025.” Washington adds that ABD candidates should “include the word ‘expected’ with your anticipated completion date and consider including a brief statement regarding the status of your dissertation,” which in practice means a defense date if one is scheduled and a chapter count if one is not.

Publication status

The status label is the part of a CV I see stretched most often, and it is also the part a reader can check most easily.

I keep published work and unpublished work in separate sections. My CV has Publications, and then a separate Work-in-Progress section that is broken into Revised and Resubmitted and Under Review, with the target journal named. Nothing in the publications section is unpublished, so a reader counting peer-reviewed articles can count them without reading the labels. Vanderbilt’s psychology guide says to list under peer-reviewed publications “only papers that are published, in press, or accepted pending revision,” and “don’t co-mingle ‘in preparation’, ‘under review’, etc. with things that are published, in press, or in press pending revision.”

Each label has a specific meaning, and APA draws the boundary at acceptance: “when a work has been submitted for publication and accepted by the publisher but not yet published, we consider it an in-press work,” and everything short of acceptance stays in the unpublished category.

• In press or accepted: the editor has said yes and the article has not appeared yet.
• Revised and resubmitted: the journal invited revisions and the revision is back with them.
• Under review: it is with a journal and no decision has come.
• In preparation: it is not with anyone.

A manuscript that is under review has not been accepted, and using the stronger label because the reviews looked positive is easy for a reader to check.

The list is in reverse chronological order, and I do not alphabetize it. Vanderbilt’s guide says that alphabetizing “makes it very difficult to gauge current productivity, which is something people pay close attention to.”

Whether to name the target journal for work that has not been accepted is contested, and both positions are worth knowing before you decide.

The strict position says no. Victoria Reyes, writing in Inside Higher Ed, tells applicants “don’t list where a paper is submitted or under review unless it received a revise and resubmit letter.” Justin Weinberg at Daily Nous rules out the journal name at any pre-acceptance stage: “do not do that thing where you list on your cv, next to the title of your as-of-yet-unaccepted submission, the name of the journal to which you submitted it,” because “trying to impress people by telling them how highly you rate yourself is rarely a winning move.” Vanderbilt calls naming the journal “generally not good practice,” while noting that “opinions differ on that though.” APA’s citation format assumes the same answer, since the reference for a submitted manuscript ends with the author’s department and university, never the journal it was sent to.

My own CV names the journals. A doctoral student’s under-review list is often the only visible record of a year of work, and I would rather show it than leave a gap. It holds up because those entries stay outside the publications section, each one carries its actual status, and the count of peer-reviewed publications does not depend on any of them. If you cannot keep that separation clean, I would take the journal names off.

What I would not do is attach a journal name to a manuscript in preparation, since nobody has read it yet, including the journal.

Authorship

I bold my own name in every citation. It is a small formatting move that removes work from the reader, who otherwise has to scan each author list to find you and to see your position in the order. Washington’s career center says to “consider bolding your name on each publication, particularly in fields where multiple authors are common.”

Co-first authorship needs a manual note, because the equal-contribution footnote a journal prints under the author list does not travel. As CASRAI’s guide puts it, “because the upstream databases mostly don’t preserve the designation, it is the author’s own responsibility to manually annotate a co-first-authored publication on a CV.” Reference managers, Scopus, Web of Science, and Google Scholar all drop it. So I would add “(co-first author)” after the citation myself, every time.

The other move worth copying is a legend. My presentations and work-in-progress sections carry a line saying that an asterisk marks a student collaborator and a dagger marks a community collaborator, and the symbols then do the rest of the work. Mentoring and community partnership otherwise go unrecorded, since there is no section for either one, and a symbol in the author string is something a reader can count.

The same device works on training, where my skills and training section marks competitively selected or funded programs with an asterisk. That separates a summer institute you were admitted to from a webinar you registered for, without editorializing about either.

Datasets, tools, and reports

Peer-reviewed articles are not the only citable output, and I think social work doctoral students produce others and then leave them off more often than they need to.

Deposited datasets get a DOI and a citation, and two of mine are archived at OpenICPSR and appear in the CV under Other Publications with their DOIs, alongside the book chapter. A deposited dataset is a finished product that a reader can open, unlike an unpublished manuscript. ICPSR’s data citation guidance gives the fields: author, title, distributor, date, version, and a persistent identifier. I cite my own data the way I would cite anyone else’s.

The hiring literature has begun arguing for giving these outputs their own space. Puebla and colleagues, writing on recognizing data and software contributions in hiring, promotion, and tenure, argue that application forms and CV templates “should also have dedicated sections where applicants can list their most relevant data and software outputs.”

Public tools, dashboards, and code releases belong in the same neighborhood. So do book chapters, technical reports, policy briefs, and a book proposal under contract or review. Princeton’s CV guide puts all of these in as subheadings under Publications: peer-reviewed articles, book reviews, conference proceedings, technical reports, web-based publications, and other publications. I prefer that nesting to a section called Public Scholarship, which is not a heading I have seen used consistently. Media coverage of your work gets its own short section near the end, with the outlet, the date, and the link.

Presentations

I group presentations by what they cost to get, since a reader who knows the field is already doing that mentally. Penn tells applicants with long lists to “subdivide by topics (peer reviewed papers, reviews or posters, invited talks),” and a medical school CV guide states the rule for the other category: “list only presentations (poster or oral) that were the result of an accepted abstract submission” in that section, because “invited lectures and other presentations go elsewhere on your CV.”

That gives four lists.

1. Peer-reviewed conference presentations, from accepted abstracts.
2. Invited talks, colloquia, and methods workshops, since an invitation is a different signal from an accepted abstract.
3. Roundtables and panels, where it is worth marking whether you organized the session or spoke on it.
4. Community presentations, trainings, and webinars, which carry weight in social work in a way they do not in some neighboring fields.

Presentations that are accepted but not yet delivered belong on the CV with the wording that matches their actual status. Mine carry either “submitted to” or “invited to” and the scheduled month, so a January conference appears in the fall CV without pretending to have happened.

Grants

Most doctoral students hold few grants as principal investigator, and I would still list everything else.

A grant entry needs the project title, the funder, the award period, and your role. The Arizona guide asks for “title, percent effort, role (e.g., PI, Co-PI, etc.), all co-PIs, source of funding or agency, full funding amount,” and the role is the field I see left blank most often. Co-investigator, graduate research assistant, project coordinator, and analyst are all real roles on funded work, and naming yours is more informative than omitting the line. GADE’s own outcome list for PhD students asks that they have “substantively contributed to a research proposal for internal or external funding” before graduating, a contribution that carries no PI line at all.

Submitted proposals get their own subsection with their status attached. The Arizona guide keeps a “Submitted Grants and Contracts” heading separate from awarded work and says to “indicate if ‘pending’ or ‘unawarded.'” For a doctoral student this section carries real information, because a scored but unfunded F31 shows a reader that you have written and submitted one.

Whether to print dollar amounts is field-specific, and the guides disagree. Medical school guides want the total, the direct costs, and the indirects. Humanities and social science guides often carry no figure at all. My own CV names the amount where the amount is the fact, as with a multi-year national scholarship, and leaves it off otherwise. You might ask your advisor what your subfield does before deciding, since this is one of the places where following the wrong guide makes you look like you came from another discipline.

Fellowships and scholarships go in their own block, separate from grants, since they are awarded on different grounds and a reader is counting them separately anyway.

Teaching

If you are on the job market, this block opens with teaching interests, a short list of the courses you are prepared to teach. I write mine against real curricula, pulling the course list from two or three schools I would apply to and naming courses that exist there, including at least one required MSW course and one methods or statistics course if you can teach them.

Then the teaching experience, where the thing to get right is naming your role. Duke’s graduate school defines instructor of record as “the formally designated instructor (or professor) of a credit-bearing, college-level class who is responsible for instruction, evaluation and final grade submission,” and adds that “for many faculty positions, this type of experience may be critical in advancing in a search.” That is a different job from teaching assistant, graduate student instructor, lab instructor, co-instructor, or guest lecturer, and a reader will assume the weakest reading of an ambiguous entry.

My teaching table names the role first in every row, then the course, then the level and modality in parentheses, then the institution. Repeated offerings are collapsed into one row with both terms in the date column, which keeps the section short without hiding anything. Chicago’s graduate school tells students to “use subcategories to bring order and focus attention,” and gives “Instructor / Teaching Assistant / Tutor” as the example, which does the same work a different way.

Guest lectures are contested. The Professor Is In rules them out entirely: “under no circumstances may guest lectures in courses be listed here or anywhere on the CV. That is padding.” Chicago’s sample CV folds one in as a bullet under a lecturer position instead. My own CV keeps them as a short labeled subsection under teaching, with the course and the school, on the grounds that a guest lecture in a required MSW course is evidence about teaching and belongs nowhere else. None of the three files a guest lecture with invited talks.

Service

Service is the block where social work CVs differ most from the CVs of neighboring disciplines, and in my experience it is where doctoral students under-report the most.

The block has three parts and they go in this order.

1. Practice experience: your MSW field placements, your post-MSW practice positions, your licensure.
2. Community service: board membership, research committees at community organizations, consulting for a community agency, volunteer roles that have lasted years.
3. Academic service, which has its own internal order of department and school committees, then professional association roles, then peer review.

Peer review deserves a line of its own once you have done any, with the journals named. If you reviewed under a faculty member’s supervision, I would say so.

Community-engaged work is worth more space than doctoral students usually give it, and the promotion documents back that up. The University of Minnesota School of Social Work’s promotion and tenure guidelines carry a section on publicly engaged scholarship, defined as work that “combines research, teaching, and service in projects that involve community members including families or other stakeholders outside of academia as co-creators and collaborators.” The products it counts are named: program development, “practice manuals, treatment manuals, and training materials,” “practice guidelines, assessment tools, and evaluation tools,” curriculum development, public education materials, and policy initiatives. Its service section separately counts “interviews with the media (e.g., radio, newspaper, television), blogs, podcasts, YouTube or other social media videos, and webinars” and program evaluations.

If you built a training curriculum for a community agency, wrote the evaluation report, or gave a radio interview about your findings, there is a school of social work that counts each of those toward tenure. It helps to put them on the CV now, with the partner named and the year, while you still remember the dates.

Licensure and practice

I give licensure its own short section with the state, the credential, and the license number, though the number is optional and only if you are comfortable publishing it. Mine lists a New Jersey LSW and a Korean credential, because both are issued by a licensing body that can be checked.

I list practice experience in full even when it predates the doctorate and even when it feels far from my research. Field placements, direct practice positions, and case management jobs are qualifications in this field in a way they are not in psychology or sociology.

The job postings say so directly. In one count, Lightfoot and Zheng monitored 150 announcements covering 198 tenure-track social work positions advertised between May 2017 and April 2018 and found that “a large majority (87.4%) of the positions specifically stated that having a master’s degree in social work was required or preferred” and that “nearly four-fifths of the positions (79.8%) required applicants to have at least 2 years of post MSW experience.” Research 1 universities accounted for about a third of the openings, so this is not a pattern confined to teaching institutions.

That two-year figure is set by the accreditor, which is why it recurs across postings. CSWE’s 2022 Educational Policy and Accreditation Standards.pdf) require, in standard 4.2.2, that “faculty who teach social work practice courses have a master’s degree in social work from a CSWE-accredited program and at least two years of post-master’s social work degree practice experience in social work,” with the same threshold for anyone supervising field placements. A program that hires you for a practice course has to document your practice hours to its accreditor, and your CV is the first place it looks.

Which position you apply for decides whether the requirement reaches you at all. A current opening at the University of Kentucky College of Social Work lists required education as a “PhD in Social Work, Sociology, Criminal Justice, or related field” and required license or certification as “N/A.” A tenure-track line at UW-Whitewater with a 4/4 teaching load required “an MSW with two years of post-MSW practice experience.” Both lines are in social work, and the requirement differs because of what the person hired will teach. So I would not bury practice experience at the end of the CV, where a reader looking for it has to hunt.

ORCID and biosketches

I recommend registering for an ORCID iD early and putting it under your name. ORCID’s own guidance is to “include your ORCID iD in your CV or resume, web page, email signature, business cards and anywhere else you have a public profile.” It costs nothing, takes ten minutes, and keeps your record attached to you across two name spellings, three institutions, and a common surname.

There is now a federal reason to do it early. NSF’s biographical sketch policy, updated April 2026, says “Proposers must use SciENcv to prepare their biographical sketches for proposals to NSF,” and the same requirement applies to current and pending support. NIH moved to the same Common Forms, and NOT-OD-26-079, released April 21, 2026, says that “On May 8, 2026, system warnings will change to errors that will stop any submission not using compliant Common Forms,” and that “All individuals submitting Common Forms to NIH must use SciENcv to produce digitally certified PDF(s).” Applicants have to obtain an ORCID iD and link it to eRA Commons. NIH has moved this deadline more than once, so check the current Guide Notice before you rely on any date, including that one.

For fellowship applicants the requirement is older than that. NIH has required an ORCID iD from individuals on training, fellowship, research education, and career development awards since NOT-OD-19-109, with individual F and K applications covered from January 25, 2020. An F31 applicant needs the iD in place before the application goes in.

A biosketch is a different document from a CV. It is short and fixed-format, with a personal statement and a selected list, produced from SciENcv, which pulls from your ORCID record and My Bibliography. NIH’s forms page says: “SciENcv must be used to prepare the form. There is not a blank template that can be downloaded and completed outside of SciENcv.” The current Common Form has no page limit and uses per-field character caps instead, so any advice you find describing five pages and Sections A through D is describing the old format. The practical consequence for a doctoral student applying for an F31 or an NSF fellowship is that the ORCID record comes before the application, so it should be populated before you need it.

Metrics

Citation counts and the h-index turn up on doctoral CVs, and I would leave them off.

The San Francisco Declaration on Research Assessment asks readers not to “use journal-based metrics, such as Journal Impact Factors, as a surrogate measure of the quality of individual research articles, to assess an individual scientist’s contributions, or in hiring, promotion, or funding decisions.” The declaration does not ban all numbers: its seventeenth recommendation tells researchers to “use a range of article metrics and indicators on personal/supporting statements, as evidence of the impact of individual published articles and other research outputs.” The line it draws is between a journal’s average and a paper’s own record.

The h-index attracts a sharper objection than that. University College London’s research metrics guide, from a DORA signatory institution, says “we recommend avoiding the journal impact factor or the h-index; these are two commonly seen metrics that often give misleading results.”

Funders have started moving in the same direction, and Chris Woolston, writing in Nature, reported that in December 2021 UK Research and Innovation, the largest public funder of British science, abandoned the conventional CV in funding applications in favor of a format meant to “enable people to better demonstrate their contributions to research, teams, and wider society.” The narrative CV is not a US norm yet, and it may never fully arrive, but the direction of travel is away from counting.

For a doctoral student the practical problem is that a citation count invites a reader to compare it with someone else’s, and at this career stage the comparison is usually between a two-year record and a five-year record. What I do instead is keep a Google Scholar profile public so anyone who wants the number can find it, under a personal Google account, since institutional accounts close when you leave.

Format and delivery

I number my publications. A numbered list lets you and anyone else refer to an entry by number, and it lets a reader see the count without counting.

I put a running header on every page with my name and the month the version was made, and a footer with the page number and the total. Mine says “Curriculum Vitae (Updated Jun 2026)” on all eighteen pages. Penn’s faculty job search guidance says to “include your name on every page, and a page number.” The date stamp tells a reader whether they have the current version, and the page total tells them whether they received all of it.

I send a PDF. Many searches go through Interfolio, which “automatically converts the document files to PDF format” whatever you upload, so a Word file will be converted by a machine you cannot see and with your formatting at risk. I would rather convert it myself and check the result. Interfolio’s size ceiling is generous at under 100 MB, but applications submitted through other systems can be far tighter, so a CV padded with high-resolution images can fail to upload at all.

I put my name in the file name, because it lands in a folder with eighty others, all of them called cv.pdf. One program’s application instructions specify the format: “save the file as LastName_FirstName_CV.pdf.”

There is no page limit and no expectation that a doctoral CV be short. Penn’s guidance says only “although there is no page limit, be concise.” Michigan sets the test on the entries instead of the page count: “CVs can be as long or as short as you want them to be” and “your decisions about what to include and not include should be dictated more by the quality and purpose of the experience than the length of the document.”

Other people’s CVs

The most useful hour you can spend on this is reading other people’s. Michigan’s career center gives the same instruction and then says to ask: “each discipline has its own nuances to CV design and content. Ask junior faculty, postdocs, or recent alums if you can take a look at their CVs.” It also suggests asking your advisor what they look for in a CV when they serve on a hiring committee, a question most students never think to put to someone who has done it.

Look at faculty pages at the schools you would apply to, at the CVs of people hired into the jobs you want in the last two or three years, and at the CVs of people one or two years ahead of you in your own program. Read for structure instead of content: where they put things, what they named the sections, what they chose to count.

It also helps to read at the level you are aiming at. A full professor’s CV will not tell you how to present four publications and a dissertation in progress.

This matters more in social work than in some fields. CSWE, GADE, and the schools themselves publish almost nothing on the mechanics: what a status label means, where a dataset goes, how to mark a role on someone else’s grant. Most of the guidance a student finds was written by career centers for psychology, engineering, and medicine, so reading actual social work CVs is how you correct for assumptions that came from somewhere else, like the medical school guide that says teaching does not belong on a CV at all.

Maintenance

Most of the difficulty in a CV is record-keeping. The version that is painful to produce is the one rebuilt from memory in the week an application is due, and the cost of that rebuild is things left off.

What has worked for me is updating it on the day something happens instead of the week something is due. An accepted abstract, a submitted manuscript, a rejection that sends a paper to a new journal, a committee assignment, a review completed for a journal: each of these is a two-minute edit on the day it happens and a thirty-minute archaeology project four months later.

This habit does not end with the doctorate. Faculty at most research universities file an annual activity report with a current CV attached, and several campuses have moved the whole thing into a system that keeps the record year-round. Boston University’s faculty annual report pulls from a standing “My CV” record, and Maryland requires that “all campus-level evaluations require faculty to submit their curriculum vitae (CV) in the official University CV format,” generated from the same system. A doctoral student’s version of that system is one plain file, kept current, for five years.

If you want the record structured instead of typed, the vitae package for R builds a CV from data frames and can pull entries from an ORCID record, and moderncv does the same job in LaTeX. Neither is necessary, and a Word file you actually update will serve you better than a pipeline you abandon in year two.

Keep one master file that holds everything and cut shorter versions from it for specific uses, deleting nothing from the master itself.

One of those cuts is a different document altogether. Research positions outside the academy, government analyst jobs, evaluation roles at nonprofits, and industry research want a resume, and the conversion is more than deletion. Yale’s career office describes the difference this way: a CV “uses titles and nouns to describe your tasks” while a resume “uses action verbs to highlight skills, experiences, and achievements,” and a resume for a PhD is two to three pages at most. SSWR keeps a page of employment resources that points at nonprofit, evaluation, and prevention research job boards alongside academic ones, a fair signal about where social work doctoral graduates go.