The packet has to agree with itself
A review board reads a submission as a single document assembled from several files, and its first job is to check that those files describe the same study. Students build them at different times — the method in one month, the consent form in another, the instrument revised somewhere in between — and the contradictions accumulate quietly.
So the work here begins with a consistency pass across the whole packet: does the consent form promise what the method delivers, does the recruitment language match the population in the protocol, is the instrument in the appendix the one the analysis plan assumes. That pass is where most of the returns you would otherwise receive are removed.
- 01Protocol written on your board's own current template, not a generic one
- 02Consent and assent forms that match the method they describe
- 03Instruments, scripts and recruitment text consistent with the population
- 04The analysis plan the protocol implies, actually stated
- 05A point-by-point reply to every item in a revision letter
The route is decided by the study, not by preference
Which review category applies is a property of the study rather than a choice you make about it, and the determination belongs to the board. What you control is whether the submission you send matches the study you are actually proposing — because a packet that understates what it is doing does not get a lighter review, it gets returned, and then gets the heavier one anyway with weeks gone.
Exempt is a determination a board makes, not a conclusion you are entitled to reach on your own, and most programs want that letter on file before anything is collected. Deciding privately and gathering first is the expensive route: data obtained without an approval in place can be unusable, and there is no retrospective fix.
A revision letter is an invitation, answered properly
Most students reply to a revision request by editing the protocol and resubmitting it, which forces the reviewer to re-read everything to find what changed. A reply that walks the board's own items in the board's own order, quoting each one and stating exactly what was altered and where, can be approved in a single pass.
Which is why the object here is not a fast submission but one that does not come back. A returned protocol is rarely a small event: it moves everything downstream of it, and the things downstream of a protocol are data collection, analysis and a defense date that other people have already put in their calendars.
Three steps, then the record keeps itself.
Send what exists
The draft protocol, your board's template and handbook, and anything already returned to you. If nothing exists yet, send the research question.
Get the consistency pass
The whole packet read as one document, with every internal contradiction listed before it becomes a revision letter.
Answer in their order
If a letter does come back, the reply walks the board's items in their sequence, so the second read is quick and uncontentious.
Queries on record.
Can you guarantee approval?
No, and an offer of one is a reason to walk away. The committee is independent, it applies its own judgment to your particular study, and nothing outside that room touches the decision. What you can control is entirely upstream: a packet whose parts do not contradict each other, pitched at the level the study genuinely needs, in that board's current format.
How long should I budget for review?
Ask your own office what it is running at now, because the figure varies widely and last year's is not evidence. Then treat that period as a phase of the project with dates of its own rather than as dead time between phases. The thing worth planning for is not the first review but the chance of a second, since that is what moves a defense date.
My study involves my own students or staff. Is that a problem?
It needs designing around rather than avoiding. Where you hold authority over the people you are studying, a board will want to see that participation could genuinely be declined without consequence, and the usual answers are structural: somebody else recruits, or the design removes any route by which you could learn who took part. Settle it while the design is still movable, because occasionally the answer rules out the study as conceived.
Is a DNP project the same submission as a PhD study?
Often not, and it is the first thing to establish because it routes everything else. A project delivering a practice change measured locally may sit outside the definition of research entirely; the same activity written up for generalisable conclusions does not. Deciding privately in either direction is expensive — an unnecessary full review costs a term, and the wrong label on something that really is research can put the whole dataset beyond use.
Do you write the consent form?
It is drafted with you against your board's current template and the method as it actually stands, then reviewed alongside everything else in the packet so the promises in it match what the study does. You read it, you question it, and it goes in under your name — the same rule that applies to everything else here.
Elsewhere in the hall.
Send the assignment and the deadline.
A specialist credentialed in your own field reads what you actually have and replies within two hours with a plan and a price. The reading costs nothing, and if the honest answer is that we are not the right hall for it, you will get that instead.
