Where each branch of the WGU tree ends, and what sends your file down one of them
Four landings, and only one of them is what most people mean by "IRB review". Which one your capstone reaches is decided by a small set of facts — what you record, who is in the population, what exposure would cost them, whether anything untested is being introduced — and by almost nothing about your topic. The level is read off the design, and it is read by the office, not by you.
Four landings: not human subjects, exempt, expedited, convened board. Identifiability, consequence of disclosure, the population involved and the risk of what you introduce decide which. You argue for one; WGU's IRB office confirms or re-reads it.
Which routes does WGU actually recognise?
WGU's Policy Handbook sets the frame in a few lines. Proposals to conduct research using human subjects go to the IRB so the IRB can determine whether full review and approval are required. Proposals that involve no human subject data, carry no research purpose, use no atypical methodology, touch no vulnerable population or sensitive topic and place nobody at risk may be deemed exempt from IRB oversight. Where risk is minimal, designated members of the board can review by an expedited procedure. Where it is not, the full board reads.
Alongside those sits the lightest landing, and the most common one for a WGU capstone. Published WGU capstone guidance describes a letter of determination obtained in writing, before collection starts, for a project that is not human-subjects research at all. That route is still a filing — just the shortest one in the folder.
One caution on vocabulary. WGU's handbook uses "exempt" loosely enough to cover projects a federal reader would call not-human-subjects, and the current route names, categories and forms are whatever WGU's current handbook or IRB portal says they are — material that sits behind university sign-on. The machinery underneath does not shift: it is 45 CFR 46, which every reviewer is trained on.
| Landing | What sends a file here | What the office issues | What the folder has to carry |
|---|---|---|---|
| Not human subjects | No living people, or nothing recorded that points back at one, and no aim beyond the site | A written determination | A short, exact statement of intent, data and site |
| Exempt category | Human-subjects research that falls wholly inside a listed exemption | An exempt determination naming the category | The full application, instruments, and consent information where the category needs it |
| Expedited | No more than minimal risk, and procedures on the federal expedited list | Approval from the chair or a designated reviewer | Consent file, data-security plan and site authorisation, all read closely |
| Convened board | More than minimal risk, or a protected population at the centre of the design | Approval, conditions, deferral or disapproval | Everything above, plus the justification for including that population |
What makes a capstone "not human subjects research"?
Two definitions have to fail, and only one of them needs to. Research at 45 CFR 46.102(l) is a systematic investigation whose design reaches for generalizable knowledge; a human subject at 46.102(e) is a living person an investigator interacts or intervenes with, or whose identifiable private information the project obtains, uses, studies, analyses or generates.
Published WGU capstone guidance puts the second failure in operational terms that are easy to check against your own plan. A project reads as not human subjects where no living participants are involved, where only de-identified information is used and the candidate holds no access to the identifier key or to participant identities, or where the data is not about human subjects in the first place. Read the middle condition slowly: it is not about how carefully the key is stored. It is about whether you can reach it.
The first failure — no generalizable aim — is the one most WGU capstones rely on, and it is argued in the intent paragraph. Improvement work versus research takes that argument apart properly.
What does "exempt" actually require?
More than the word suggests, and the burden sits with the applicant. Exempt is a category the office confirms after reading a complete application; it is not a status you enter by asserting it. Published WGU capstone guidance has historically opened only a subset of the federal exemption categories to capstone projects, which is a useful reminder that a category existing in the regulation does not mean it is available to your programme.
The categories that matter for capstone work sit at 45 CFR 46.104(d). Category 1 covers normal educational practices in established educational settings. Category 2 reaches educational testing, survey and interview procedures, and observation of behaviour in public — but only where responses are recorded without readily ascertainable identities, or where disclosure could not reasonably damage someone's standing, liability, employability, finances or reputation. Category 3 covers benign behavioural interventions with adults who agreed in advance, and category 4 the secondary use of information already collected.
Then the detail almost everyone misses. Where identifiable responses are recorded, categories 2 and 3 remain open only if a board performs a limited IRB review to make the privacy and confidentiality finding at 46.111(a)(7). "Exempt" in that situation still means a reviewer reads and rules. It is lighter than full review; it is not an absence of review.
What opens the expedited route?
Under 46.110, two conditions have to hold together. Risk must be no more than minimal — which 46.102(j) pegs to the harm or discomfort of ordinary daily life, or of a routine examination, physical or psychological. And every procedure used must come from the list of expedited categories the Secretary of HHS publishes and revisits periodically.
Two features of the route repay knowing before you file. First, the reading is done by the chairperson or by one or more experienced reviewers the chair designates from the board — not by the full membership. Second, and more usefully: those reviewers may exercise every authority the board holds except the authority to disapprove. A project can only be disapproved after review by the convened board.
That changes how a returned expedited file should feel. What arrives is a request to modify, or a decision to send the file upward — not a verdict, and rarely the end of anything.
When does the convened board read it?
When risk exceeds the everyday, or when a protected population is central to the design. The approval criteria at 46.111 are worth reading once in full, because they are the questions every reviewer is working through: risks minimised and reasonable against anticipated benefit, equitable selection of participants, consent sought and documented properly or properly waived, monitoring where it is warranted, and adequate provision for privacy and confidentiality. Where participants could be pressured or unduly swayed, 46.111(b) wants extra safeguards written into the design.
WGU's own published application form makes the population question unusually concrete, and its list is longer than most people expect. It asks whether the population includes people with physical, mental, developmental or emotional disabilities, pregnant women, children, prisoners, residents of facilities such as nursing or group homes, people who do not speak English, people aged 65 or over, those who are traumatised, economically disadvantaged or terminally ill — or the researcher's own employees or learners. For nursing capstones run on a unit or in long-term care, several of those lines can be live at once, and the form asks for justification and context rather than a yes.
What moves a file up a level?
Five things, in roughly this order of force.
- Identifiability. Whether anything you keep leads back to a person — by name, or by a key you can reach. It decides more files than the other four combined, which is why it has its own piece.
- What disclosure would cost. If a leak could touch employment, licensure, standing or legal exposure, the light routes close regardless of how small the project is.
- Who is in the population. Children, prisoners, adults with impaired decision-making capacity, people you supervise or care for, and economically or educationally disadvantaged groups all pull additional safeguards into the file.
- Whether anything untested is introduced. Applying an accepted practice locally reads very differently from trialling something to see whether it works.
- Recording. Audio, video and photographs are identifiers in their own right, and they move files more often than the candidates making the recordings expect.
Can the office put you somewhere other than where you applied?
Yes, and it is normal. You file under the reading your design supports; the office reads the materials and confirms that reading or replaces it. Nothing we do changes that, and no honest consultant will tell you what the board will decide or when. What a well-built file does is remove every reason for the reader to reach for a heavier route out of doubt — by making the population, the data, the recording and the intent unambiguous on the page. The folder, document by document shows what that looks like assembled.
What to do next
Answer the five questions above in one written line each. If every answer points to the same landing, you have your branch and its folder. If two disagree, that disagreement is the real problem — and it is far cheaper to settle now than in a revision response.
Send those five lines to us if you want them read by someone whose whole practice is these files. The walk sets out what we carry, the questions page covers the rest, and the free application review names your branch in writing.
Sources
- WGU Policy Handbook — Institutional Review Board (IRB), cm.wgu.edu
- Western Governors University Institutional Review Board — Request for Approval to Conduct Research (published copy of the WGU form, PDF)
- 45 CFR part 46 (the Common Rule), full regulatory text — see 46.102, 46.104, 46.110 and 46.111, govinfo.gov
- OHRP — Quality Improvement Activities FAQs, hhs.gov
Find out which landing your capstone reaches, in writing.
Send the design as you actually plan to run it. One of the three consultants reads it against the forks and replies with the branch, the folder for it, and whether you can reasonably file alone.
Request the free application review- the landing your design honestly reaches
- what would move it up a level, and what would not
- the documents that landing's folder needs