How to retain childcare staff: what actually lowers turnover

Start with the measurement problem: there is no single national turnover rate for childcare staff.

Published estimates run from under 10 percent, in school-sponsored centers, to as high as 46 percent a year, depending on the study, the state and the method — so a single headline turnover figure you read is only as good as the scope and year behind it.

Check what a study measured before you benchmark against it.

The spread is the finding.

A 2022 Minneapolis Fed analysis of the 2019 National Survey of Early Care and Education found center teacher turnover of 21.1 percent in "all other" centers versus 7.7 percent in school-sponsored centers, and 25.3 percent in centers serving subsidized children versus 15.8 percent in centers that do not.

State-level studies push higher: research on Louisiana centers finds annual teacher turnover as high as 46 percent, and a Virginia study puts annual center teacher turnover at 36 to 38 percent — both figures carried in a March 2026 working paper on early childhood teacher turnover.

Read the numbers as an operator, not a judge.

The same job title carries very different odds depending on the setting it sits in and the families the program serves — school-sponsored versus "all other" centers is one comparison, centers serving subsidized children versus those that do not is another, and each gap is read within its own comparison.

Why any individual teacher quits is program-specific — our research did not verify a ranked list of departure reasons for the industry — so the practical move is to find out in your own building: stay interviews with the staff you want to keep, exit interviews with the people who go, and a look for the pattern in your own data.

Retention is also the back end of the pipeline: hiring for your childcare center covers the stages before this one — where to find candidates, how to screen and onboard them — and the moves on this page land on whatever picture of the job a hire was given at those stages.

Direct retention payments have the strongest causal evidence behind them. Treasury reported in 2021 that in a University of Virginia experiment offering child care teachers $1,500 to stay, departures were cut nearly in half, to 13 percent — versus nearly a quarter of the teachers who received nothing leaving within eight months.

The bonuses were modest; the effect was not.

A payment tied to staying — rather than to signing on — changed enough stay-or-go decisions to move a program's whole staffing math.

The design matters more than the dollar amount.

The Virginia payment went to teachers already doing the work and was tied to remaining employed — which is what separates a retention payment from a recruiting bonus aimed at people who have not started yet.

If you structure one, define the window, put the terms in writing, and pay it on schedule; a retention program that gets renegotiated at the deadline keeps nobody, and the next one you offer buys nothing.

A bonus patches a pay problem; it does not replace one.

The pay scale you publish — base rates, and what each credential or added responsibility adds to them — is one of the first documents a restless teacher studies, and if it has not moved while your market has, the bonus is a delay, not a fix.

The schedule is the other half of the compensation you control, and it costs planning rather than payroll.

Make it survivable: consistent shifts people can plan around beat a schedule that changes every week.

Requests for time off get answered on a predictable timeline, not the week before the holiday.

And the coverage plan is built around the rooms' ratio needs first — so that when a teacher calls out, the floater absorbs the room and the remaining teachers keep their breaks, instead of breaks being the first thing that disappears on a hard day.

None of that shows up on a pay stub, and all of it shows up in whether people are still on your roster when the next enrollment push starts.

The practices below are the parts of the working environment a director controls outright, and they are the difference between a job people endure and one they build around.

Make the first months deliberate. A new teacher's first weeks are where retention is cheapest: pair every new hire with an experienced teacher, check in at the end of the first week and again after the first month, and separate a rough start from a bad fit honestly.

Early check-ins catch the fixable problems before they harden into a resignation — and a struggling new teacher's first need is a named person to ask, not another policy document.

Protect breaks and planning time in writing. A break that only happens when coverage allows is not a break; it is a lottery.

Build the weekly coverage plan before the week starts, name who absorbs a callout in each room, and keep the plan visible — teachers should not have to negotiate for a mid-morning break.

Recognize specifically, and let it come from you. "Good job" in the hallway is noise.

Naming what someone did — "the way you rerouted the block area before the collision happened" — tells a teacher their judgment is visible, and that visibility is what turns a job into a professional identity.

Appreciation that is specific, timely and tied to practice you want repeated costs nothing and compounds.

Give teachers real decisions. Room arrangement, curriculum choices, how the classroom budget is spent — the people who live with a decision should have a hand in it.

A teacher executing someone else's plan has a job; a teacher building one has a practice, and practices are harder to walk away from.

Fix problems while the person is still angry, not after they are gone. When a teacher raises the same issue twice, it goes on your list with a date attached — or it goes on the exit interview's transcript instead.

Stay interviews, short and direct, are how you hear the second raise of the same issue while you can still do something about it.

A teacher who can see the next rung — and the raise attached to it — has a reason to stay through the hard semester.

A ladder holds people when three things are true: the steps are named, the credential each step requires is funded, and the raise arrives when the credential does.

Miss any one of the three and the ladder is a poster on a break-room wall.

T.E.A.C.H. is the model to know. T.E.A.C.H.

Early Childhood scholarships use a cost-shared model that pays most of tuition, books and travel, often requires paid release time from the sponsoring employer, awards a bonus or raise on completing credit hours, and requires the recipient to stay with the sponsoring program for a set period.

That last clause is the retention mechanic: the scholarship is structured so the credential a teacher earns is anchored to your program for the length of the commitment.

Sponsor obligations and cost shares vary by state program, and the T.E.A.C.H.

National Center's directory (as of October 2026) lists programs in 21 states — AL, AR, CO, FL, IA, IN, ME, MI, MN, MO, NE, NV, NH, NC, OH, PA, RI, SC, UT, VT and WI, with Tennessee listed for WAGE$ only — so check the directory and your state's sponsor for what participation would ask of your budget and your schedule.

Registered apprenticeship builds the raises into the program. Federal standards for registered apprenticeship (29 CFR 29.5) require a term of at least 2,000 hours of on-the-job learning for time-based programs (or a demonstrated-competency or hybrid design), a recommended minimum of 144 hours of related instruction per year, and a progressively increasing wage schedule that starts at no less than the FLSA minimum wage.

The wage progression is the point: the raise is not renegotiated each year, it is scheduled.

Apprenticeship.gov lists "Early Childhood Educator" and "Early Childhood/PRE-K Teacher" among the education-sector apprenticeship occupations — check Apprenticeship.gov for a registered program in your state before you build one from scratch.

Or draw the ladder yourself. Name the steps — assistant, lead teacher, mentor or floater lead, assistant director — attach the credential each step expects (the CDA is the nationally recognized credential BLS says states and employers most often require; your state's lead-teacher rule names its own), and attach a raise to each completed step.

Fund the credential through T.E.A.C.H. where the directory lists your state.

Where it does not list one, check with your state's early childhood professional-development office before funding it yourself — our research found states that run their own scholarship programs the national directory does not list.

It also helps to see the market from the candidate's side: browse childcare teacher jobs and read how postings handle pay growth and step-ups — the teachers comparing offers are reading them.

The ladder does not have to be tall.

It has to be visible, funded and kept — the year a promised raise quietly does not arrive is the year the ladder stops working on anyone.

Retention work without a number is mood management.

Define turnover once and never change the definition: the count of departures over a period, divided by your average staff count over the same period, annualized.

A rate you have tracked the same way for three years is worth more than a more precise one you started last quarter.

Segment it before you average it.

A center-wide rate hides everything useful: track departures by classroom, by tenure band, and by role — lead versus assistant — and watch the first-year window separately.

The plan that keeps a second-year lead teacher is not the plan that gets a new assistant through their first winter.

Ask before people go, and after.

A stay interview with the staff you most want to keep — what keeps them here, what would pull them away — surfaces fixable problems while they are still fixable.

An exit interview catches the pattern for the next hire.

The pairing matters more than either conversation alone, and both are worthless if the answers are not written down and revisited.

Benchmark against yourself, not against a national figure.

Published turnover estimates run from under 10 percent to 46 percent depending on method, setting and state, so a "good" number borrowed from a study of somewhere else's centers tells you nothing about your building.

The trend is the benchmark: quarter over quarter and year over year, is the rate falling in the classrooms where you spent the effort?

This page is employer information, not licensing or legal advice. Scholarship terms, apprenticeship standards and workplace policies vary by state and by program — confirm current specifics with the program sponsor, your state child care licensing agency, or employment counsel before you act on them.