What the Study Found
- Childfree adults range from 22% of California’s population to 9.4% of Louisiana’s, with no clustering into multi-state regions.
- State context shifts one person’s predicted odds of being childfree by 13.1 points, from 14.1% to 27.2% across the extremes.
- Individual traits matter far more: predicted odds run 3.3% to 43.8% between two different adults in an average state.
- Higher per-pupil education spending accompanied fewer childfree adults, a link that appeared only once other factors were held constant.
A map of the United States shaded by the share of adults who neither have children nor want them comes out looking oddly scattered. California tops it at 22 percent, Michigan sits just behind at 19.9, and at the far end Louisiana manages 9.4% and South Dakota an even 10. Low-prevalence states do bunch a little across the Mountain West and the Great Plains, but the high ones turn up all over the place: Illinois, New York, Michigan, California. Individual traits, being single, older, male, white or politically liberal, track with being childfree far more closely than anything about the state a person happens to live in.
Childfree is not childless, and the distinction does a lot of work here. It means having no children and wanting none, as against wanting them and being unable, or simply not having got round to it yet.
Zachary Neal and Jennifer Watling Neal, both psychologists at Michigan State University, went looking for that group inside the Civic Health and Institutions Project, a cross-sectional survey that samples all 50 states and weights each state’s results to resemble its own population. Their analytic sample came to 20,118 adults, surveyed in 2024 (the fieldwork ran across two months of the autumn). Three questions did the sorting: have you had children, do you plan to, and do you wish you could. Only a no to all three counted, which is a stricter screen than most national surveys bother with.
No Childfree Belt on the Map
The obvious hypothesis was regional. Nothing doing. The authors tested whether neighboring states resemble one another and found no evidence of clustering, so being childfree is not a coastal phenomenon, nor a blue-state one, nor a Mountain West one in reverse.
Four state-level features went into the analysis: total population, cost of living, abortion protections and per-pupil education spending. Taken one at a time, the first three tracked with prevalence and the spending figure did not. Bigger, costlier, more permissive states held more childfree residents. Roughly what you’ woul’d expect, if the story were about states at all.
Then the model put individual and state characteristics in the same room. Watling Neal, quoted in a university statement, was clear about which won: “Those state factors matter much less than individual characteristics like marital status and age.”
The Person Beats the Place
In a mixed-effects model (which asks what predicts an individual’s childfree status while accounting for the state they sit in) nearly all the explanatory weight landed on the person. Standardized effects for individual characteristics ran several times larger than anything the state contributed. The authors turned that into scenarios, which helps. A 30-year-old conservative partnered non-white woman living in a state at the national average came out at 3.3 percent likely to be childfree, while a 50-year-old liberal single white man in that same average state came out at 43.8 percent, a spread of 40.5 percentage points. Send one person on a tour of every state instead, from the least populated and cheapest to the most populated and costliest, and the swing was 13.1 points, which is real but hardly the same animal.
By the authors’ reckoning, being single was associated with being childfree about 11 times more strongly than a state’s population was. All of it cross-sectional, though, so associations rather than causes, and the sample is a weighted online panel rather than a random draw.
One line in the press release deserves a footnote. Michigan State’s announcement has Watling Neal saying that a state’s cost of living and its protection of reproductive rights are both related to how many of its residents are childfree, and the reproductive-rights part of that holds only in the simpler analysis: once individual characteristics entered the model, abortion protections fell away, while per-pupil education spending, which had looked irrelevant on its own, surfaced as a state factor pointing the other way, with more spending accompanying fewer childfree adults (a reversal the release does not mention).
Which leaves the policy reading mostly negative. Neal noted in the same statement that “these findings suggest that reducing the cost of living, spending more on education or changing restrictions on abortion are unlikely to lead childfree residents to have children.” Affordability is certainly on people’s minds, as a Pew Research Center survey published in 2024 found, though naming a cost is not the same as saying a subsidy would change the answer. The UN Population Fund reached a comparable verdict in 2025, reporting in its State of World Population that measures which sideline reproductive choice have a long record of failing. Weak associations can be ruled out as levers. Strong ones cannot simply be ruled in, because the direction of causation is unknown, and the paper is careful to add that its own biggest correlate, being single, is not therefore a lever either.
What the map does say is that childfree adults live everywhere, in numbers no state can sensibly treat as marginal. “Although some states have large childfree populations, they remain a hidden group because their decision not to have children violates a strong norm,” said Watling Neal, whose team is now turning to how those adults’ social networks differ from those of parents.
- Study type: Cross-sectional survey with generalized linear mixed-effects modelling (random intercepts by state); peer-reviewed, published in PLOS One
- Sample size: 20,118 adults across all 50 states, from a cleaned pool of 20,901; Washington DC residents excluded
- Instrument: Three-item sequence on having, planning and wishing for children, classifying childfree status; state measures of population, regional price parity, abortion-policy protection and per-pupil spending
- Sampling frame: Civic Health and Institutions Project wave 33; online quota panels post-stratified by race, ethnicity, age, gender, education and geographic region for state and national representativeness
- Period covered: Single wave fielded 30 August to 8 October 2024; no follow-up
- Funding / conflicts of interest: Authors report no specific funding for this work and declare no competing interests; the underlying survey is a multi-university collaboration supported by National Science Foundation awards
- Data availability: Data and analysis code deposited on the Open Science Framework; source survey data distributed by the Civic Health and Institutions Project
- Preregistration: Not reported
- Main limitation: Author-stated: the cross-sectional design identifies associations only, not causal effects, and the state level is a coarse spatial unit that could hide clustering at finer scales. Weighting covers a limited set of characteristics, so representativeness on others is unverified
Reference
Neal, Z. P., & Neal, J. W. (2026). The spatial distribution of the U.S. childfree population: Does state context matter? PLOS One, 21(8), e0352872. https://doi.org/10.1371/journal.pone.0352872
Frequently Asked Questions
Is childfree just another word for childless?
Childfree is not another word for childless. It describes people who have no children and do not want any, which sets them apart from people who wanted children and could not have them, and from people who still plan to have them later. The study counted someone as childfree only after a no to all three of those questions, which is why its estimates are lower than a simple count of non-parents.
Does moving to a state like California make someone less likely to want children?
Moving to a state like California would not, on this evidence, make someone less likely to want children. The survey was cross-sectional, so it captures associations at a single moment rather than causes, and the direction of any relationship between state context and being childfree is unknown. What the model does show is that a person’s own characteristics account for far more of the difference than their state does.
Could a state raise its birth rate by making itself cheaper to live in?
A state is unlikely to raise its birth rate by making itself cheaper to live in, at least not by way of its childfree residents. Cost of living was associated with prevalence only weakly once individual characteristics were taken into account, and one of the authors said the findings suggest that cheaper living, more education spending or looser abortion restrictions are unlikely to lead childfree residents to have children.
What is stopping researchers from saying what causes someone to be childfree?
What is stopping researchers from saying what causes someone to be childfree is the shape of the data. A single cross-sectional survey wave shows who is childfree at one point in time, not what made them so, and following the same people over years would be needed to identify causes. The state level is also a coarse unit, so any clustering at the scale of cities or neighborhoods would be invisible here.
Cite This Page