How Common Is Parental Alienation? The 22-Million Prevalence Study

Love Over Exile is a plain-language research archive on parental alienation, written by Malcolm Smith — an alienated parent and author of the forthcoming book Love Over Exile — for non-specialist readers (other alienated parents, family members, therapists, lawyers) who want to understand the evidence base without a psychology degree or a journal subscription. This page is one entry in that archive.

What the Researchers Asked

How many people does this actually happen to?

It sounds like the first question any field should answer, but for parental alienation it was almost the last. For four decades, the research base was dominated by clinical samples — parents and children already in family court, already being assessed, already in crisis. Those samples told us what alienation looked like in its most visible forms, and they told us that a substantial minority of custody-disputing families had it. What they could not tell us was whether parental alienation was a rare feature of the most conflicted separations, or a widespread pattern of post-separation family harm that most people never see because it happens inside private households.

Jennifer Harman and her colleagues at Colorado State University had been arguing since at least 2016 that the prevalence question needed a different answer, and that it could only come from a different kind of sample. Harman, Leder-Elder and Biringen’s 2019 paper in Children and Youth Services Review is where that argument became the empirical answer the field has cited ever since.

The research question is formally narrow. How many adults in the United States, today, are being targeted by parental alienating behaviours from a current or former partner? How many of those adults have children who have been moderately or severely alienated as a result? What is the mental-health cost to those adults? Behind that narrow question sits a broader reframing: is parental alienation a marginal, edge-case phenomenon, or is it — as Harman’s earlier work had argued — a form of family violence on a scale that demands a public-health response?

What They Did — Methods in Plain English

The paper’s contribution is, first, methodological. Before this study, every prevalence estimate for parental alienation came from a sample that was already conflicted.

Fidler and Bala’s 2010 Family Court Review paper, the most cited earlier estimate, put the figure at roughly 11–15% of children in custody disputes. Bow, Gould and Flens (2009) surveyed custody evaluators and found about 60% had encountered cases they considered parental alienation in their practice.

Both numbers were useful. Neither was a population estimate. A 60% provider rate tells you how common the issue is on an evaluator’s desk, not how common it is in the general population of separated parents. An 11–15% rate in custody disputes tells you how common it is among families already litigating, not among separated families more broadly.

Harman and her colleagues took a different route. They commissioned three separate polls from YouGov, an online survey platform that maintains panels of respondents who are then weighted to Census-matched demographic targets using iterative proportional fitting. Weighting a panel against Census benchmarks — age, gender, race, education, region — does not eliminate selection bias, but it materially reduces it compared to convenience samples. The result is an estimate closer to what you would get from a genuine population survey.

Three polls were fielded:

Poll	Year	Country	Sample (parents subsample)	Measure
1	2014	United States	273	Single-item subjective self-report: “do you feel you are being alienated from your child(ren)?“
2	2015	Canada	397	Same single-item subjective measure
3	2016	United States	594	Behaviourally anchored measure — a battery of items identifying parents who were non-reciprocating targets of parental alienating behaviours (the other parent engages, the respondent does not)

Polls 1 and 2 asked the subjective question directly. That has a known limitation — some respondents may interpret “alienated” to mean any kind of estrangement, or any post-separation distance, rather than the specific pattern the researchers had in mind. Poll 3 deliberately fixed this by shifting to a behavioural measure. It asked parents not “do you feel alienated?” but “does the other parent do these things?” — drawing conceptually on the seventeen alienating behaviours catalogued by Baker and Chambers (2011) and identifying the subset of parents whose answers to those questions placed them on the receiving end of a pattern they were not themselves reciprocating.

The shift from Poll 2 to Poll 3 is a key piece of the paper’s methodological discipline. If the three polls had all used the same measure and produced convergent results, it would be a straightforward replication. If they had used different measures and produced wildly divergent results, the whole estimate would be unstable.

What the paper shows is that the two methods — subjective feeling of alienation, and behavioural exposure to the other parent’s alienating tactics — converge within a reasonable range: 35.5%, 32%, and 39.1%. Both lenses point to the same scale. That is the kind of triangulation prevalence studies rarely get.

Statistical analysis was straightforward: weighted prevalence estimates with 95% confidence intervals, chi-square tests for demographic subgroup comparisons, and regression of mental-health outcomes (depression, trauma symptoms, suicidality) on targeted-by-PABs status. The specific confidence intervals and exact subgroup test statistics live behind the Elsevier paywall; the core numbers and their direction are in the open abstract and the press summaries, which is how they became the field’s reference point.

What the paper did not do is worth naming too. It did not follow parents over time — it is cross-sectional, not longitudinal. It did not interview the other parent, the child, or any third party.

It is a parent-report study, and parents reporting they are targeted is what it measures. Other designs would answer different questions. This one answered the scale question.

The two figures, drawn to scale

The single most-misread number in the 2019 paper is the gap between targeted and fully alienated. The diagram below puts the two figures alongside the rest of the US adult population to scale.

What They Found — Six Key Findings

1. 39.1% of US parents are non-reciprocating targets of parental alienating behaviours

The paper’s single most conservative, most behaviourally anchored finding comes from Poll 3: 39.1% of the 594 US parents in the 2016 YouGov panel met the threshold for being non-reciprocating targets of parental alienating behaviours. The word “non-reciprocating” matters. The measure does not count parents who say the other parent engages in alienating behaviours while also engaging in them themselves — that would capture high-conflict, bidirectional cases where no single alienator can be identified. It counts only those on the receiving end of a pattern they are not returning.

The two earlier polls using the subjective single-item measure — “do you feel you are being alienated from your child(ren)?” — produced 35.5% in the 2014 US sample and 32% in the 2015 Canadian sample. That three measurements using two different methods land within seven percentage points of each other is the triangulation the paper depends on. Whichever lens you use, around a third of separated parents in the Global North say they are on the receiving end of this pattern.

2. Extrapolated to the US parent population — approximately 22 million targeted parents

Applying the Poll 3 rate to the Census-weighted US parent population at the time of the 2016 poll produces the figure the paper is best known for: 22,141,650 — most commonly cited as over 22 million parents in the United States.

That extrapolation is only as good as the underlying rate and the weighting. The headline number is an estimate, not a census. But it is the single best estimate the field has, and it remains the most-cited prevalence figure for parental alienation in US advocacy and policy work six years on from publication. When a journalist or a court filing or a safeguarding report says “22 million parents”, this paper is where the number comes from.

Two qualifications matter.

First, the figure is US only. The Canadian Poll 2 was not extrapolated to a national total in the paper. For UK prevalence, the reference is Hine, Harman, Leder-Elder and colleagues (2025) in Journal of Family Violence, which studied separated parents specifically and found rates of 39–60% depending on the behavioural threshold.

Second, the poll was fielded in 2016. Population growth and divorce-rate changes since then mean the 22 million figure is now somewhat conservative; no updated prevalence poll has been published as of April 2026.

3. 6.7% of targeted parents have moderately-to-severely alienated children

This is the finding that keeps the 22 million figure honest.

Being targeted by alienating behaviours is not the same as having a child who is fully alienated. Many parents who are targeted continue to have a working relationship with their child; the alienating behaviours are present but have not succeeded, or have not fully succeeded, in fracturing the parent-child bond. A smaller subset have children who are actually refusing a relationship — what Bernet and Greenhill’s Five-Factor Model calls Factor 1, contact resistance or refusal.

Harman and colleagues stratified their targeted subsample by severity of the child’s alienation. 6.7% of targeted parents — about one in fifteen — had children who were moderately or severely alienated. Applied to the US adult population, that works out to approximately 1.3% of all US adults, or around 3.8 million children.

The distance between the two numbers is worth sitting with.

Alienating behaviours are relatively common: roughly 1 in 3 separated parents is on the receiving end. Full parental alienation — where the child has actually rejected the parent — is much rarer: about 1 in 75 of the total US adult population is in that position. The ratio is roughly 30 to 1.

A responsible reader of this paper holds both figures together. Understating the wider pattern misses the scale of the behaviour. Overstating the narrow pattern conflates exposure with outcome and invites reasonable methodological pushback.

4. Alienation occurs across all demographic groups

The paper reports that parental alienation targeting cuts across socio-economic and demographic categories. Both men and women are targeted at substantial rates. It is not a phenomenon of any one income band, education level, or race. The earlier 2016 North Carolina companion paper found higher rates among Black and Native American respondents and among those with high-school-only education, but the 2019 broader sample points to a pattern that is present everywhere, not concentrated in any one demographic.

The gender point deserves particular care because it is the most commonly misreported single finding of this literature. The 2019 prevalence paper does not support the claim that “mothers alienate more” or “fathers alienate more.” It reports that both sexes are targeted at substantial rates and that alienation occurs across demographic groups. A separate 2019 Harman, Matthewson and Baker paper in Journal of Family Violence looked specifically at gender differences in the type of alienating tactics used — finding mothers somewhat more indirect aggression and fathers somewhat more direct — but that is a finding about style, not prevalence. The single strongest predictor of who alienates is not gender; it is who holds the custodial advantage, which varies with jurisdiction and with era.

5. Targeted parents show elevated depression, trauma, and suicidality

Prevalence without impact is just counting. The 2019 paper’s second contribution is the mental-health layer. Parents targeted by alienating behaviours reported significantly higher rates of depression, trauma symptoms, and suicidal ideation than non-targeted parents in the same sample.

The single most striking figure is in the suicidality domain. Across the combined samples, approximately 23% of targeted parents reported severe suicidal ideation or suicide attempts. Put in context, that is roughly 288 times the general-population rate for attempted suicide. Read the figure carefully: it is ideation plus attempts, combined. It is still, by any reasonable reading, a public-health signal on a scale that should provoke a policy response.

The depression and trauma-symptom findings are directionally clear but the exact prevalence rates live in tables behind the Elsevier paywall. Follow-on Harman work synthesising the 2019 data with later Canadian samples elaborates on these figures; for this article we cite the directional finding and the specific 23% suicidality figure that has passed into the downstream literature. The broader point the paper wants to make is the one we should take: the mental-health cost to targeted parents is large, concrete, and measurable. It is what makes prevalence a public-health question, not a private grievance.

6. The methodological step-change — from clinical samples to population panels

The sixth finding is not a number but a methodological move, and it is arguably the paper’s longest-lasting contribution.

Every prior prevalence estimate for parental alienation had come from a sample that was already in crisis. Fidler and Bala’s 2010 estimate of 11–15% came from custody-disputing families. Bow, Gould and Flens’ 60% came from evaluators reporting on their own caseloads.

Baker’s earlier work drew on adults who volunteered to recount childhood experiences. All useful. None a population estimate.

Harman’s move to Census-weighted YouGov panels was the methodological step the field needed.

The sample includes parents who have never litigated, parents whose children are on good terms with both parents, parents whose separations were amicable. Some of them — by the measure used in Poll 3 — still report being targeted.

That the prevalence figure stays high in a sample that is not pre-selected for conflict is the real argument the paper makes. It is not that alienation is common in already-conflicted families (which we knew). It is that alienation is a feature of the general population of separated parents, not only of the ones visible to family courts.

This shift is why the 2019 paper has become the reference point for downstream work on scale. It is also why subsequent prevalence research — Hine et al.’s 2025 UK paper, the ongoing international replication effort — has adopted the population-panel framework. The methodological upgrade, not the specific numbers, is what the paper bequeaths to the field.

Why This Matters

A prevalence paper does two kinds of work. It tells you how common a thing is. And, if it is the first credible population-level estimate of its kind, it changes what kind of thing the thing is allowed to be.

Before 2019, parental alienation could be plausibly framed as an edge-case pathology of the most conflicted separations — rare, unfortunate, and a private family tragedy rather than a public issue. After 2019, that framing does not survive contact with the numbers.

Twenty-two million US parents is not an edge case. It is a feature of the US adult population on the scale of major depression, or of adults living with a chronic pain condition.

Even if the narrower figure — 1.3% of US adults with a moderately-to-severely alienated child — is the more operationally useful one for clinical work, that still represents a population roughly the size of the entire US deaf or hard-of-hearing community. The phenomenon is not small.

The paper’s influence has played out in three overlapping ways.

First, it reshaped advocacy and policy framing. The “22 million parents” figure is the single most-cited statistic in US parental alienation advocacy, in Parental Alienation Awareness Day campaigns, in submissions to family-court reform enquiries, and in press coverage of the phenomenon. When a journalist writes that parental alienation “affects 22 million American parents”, they are citing this paper, whether they know it or not. The figure turned what had been a scattered collection of clinical and advocacy voices into a coherent public-health claim.

Second, it enabled the classification argument to scale. Harman, Kruk and Hines’ 2018 Psychological Bulletin paper argued that parental alienating behaviours meet the definitions of family violence. A classification argument is only politically significant if the classification applies to a meaningful population. This paper is the empirical floor under that argument: 22 million US parents targeted means a form of family violence happening at scale, not a fringe claim.

Third, it sharpened the diagnostic conversation. Bernet and Greenhill’s 2022 Five-Factor Model commentary explicitly cites this paper’s 30%+ targeted versus roughly 1.3% fully-alienated distinction as the reason clinicians need a threshold gate (Factor 1, contact refusal). The prevalence paper did not just establish how common the phenomenon is; it made the exposure-versus-outcome distinction unavoidable. Any responsible diagnostic framework has to respect the 30-to-1 gap.

For readers of this site, the reason the paper matters is simpler and harder. If you are a targeted parent, you are not in a rare category. You are in the company of millions. If you are a professional who has never taken parental alienation seriously as a clinical or legal issue, the numbers are the reason to start.

What This Means for You

The practical weight of a prevalence number depends on what you do with it.

If you are a targeted parent. The 2019 paper gives you a number that cannot be dismissed. You are not one of a few unlucky outliers.

Your experience sits within a pattern affecting more than 22 million parents in the United States, and — per Hine et al. 2025 — 39% to 60% of separated parents in the United Kingdom. When a family member, friend, lawyer or therapist suggests your situation is unusual, or overstated, or a private matter you should handle privately, the paper is the reference that answers back.

The isolation that targeted parents report is partly the structural result of the experience itself; it is also a function of believing you are alone in it. You are not.

If you are a lawyer or therapist reading this. The prevalence data means the population of parental alienation cases is large enough that competence on the topic is not optional specialist work — it is general practice.

A family-law practice that handles post-separation disputes will encounter these cases. A therapy practice that sees separated parents will encounter these cases. Treating the phenomenon as marginal means misclassifying a sizeable share of the presenting population.

The paper’s methodological shift to population panels is also the evidence you can point to when a judge, a colleague, or a safeguarding officer asks whether you are overstating. You are not. The figures are from a peer-reviewed study in an indexed journal using Census-weighted samples.

If you are a family member, friend, or colleague of an alienated parent. The numbers give you a way to take what you are hearing seriously. The experience your friend or relative is describing — the loss of contact, the erosion of the relationship with the child, the sense of invisibility — is being reported by millions of parents simultaneously. The appropriate posture is the one you would take with a friend living through another kind of large-scale, under-recognised harm: listen, take the harm at face value, do not offer premature advice, do not ask the targeted parent to carry the burden of proof that their experience is real.

If you are the alienating parent reading this with discomfort. Prevalence is the part of the research that tells you the other parent’s experience is a pattern, not a personal failing. It is also the part of the research that tells you the behaviours that targeted parents describe are what you are being described as doing. You may have a different account of events. The question this paper asks you to sit with is not whose account is correct but whether the pattern you recognise in your own conduct — over the many small decisions that make up post-separation life — is one of the patterns that 22 million other parents are experiencing from their end. If you recognise yourself in the behavioural description, the statistic is no longer an abstraction.

What the Paper Doesn’t Tell Us — Limitations

Four honest limitations, named plainly.

First, everything is self-report. Parents report that they are targeted. Parents report on their children’s behaviour. Parents report on their own mental-health outcomes. No corroboration from the other parent, the child, or any third party. That is a genuine limitation — some of the parents in the sample will be reporting the experience of alienation while themselves engaging in alienating behaviours; some will be in high-conflict dynamics where both parents feel targeted and no single alienator can be identified. The paper’s use of the “non-reciprocating targets” measure in Poll 3 mitigates this concern but does not eliminate it. Future work using dyadic measurement — the same question asked of both parents — would sharpen the estimate.

Second, the measure of “being alienated” in Polls 1 and 2 is a single-item subjective question. “Do you feel you are being alienated from your child(ren)?” invites respondents to bring their own interpretation of “alienated”. Some may read it as “estranged”, which is a different construct, or as “generally distanced”, which is a vaguer one. Poll 3 addresses this by using a behaviourally anchored measure, and the convergence between 35.5% (subjective) and 39.1% (behavioural) is modest evidence that the two methods are tracking the same thing — but it is not conclusive. A cleaner prevalence study would use the behavioural measure across all three polls.

Third, the data is US and Canadian from 2014 to 2016. Generalising to other legal systems, cultural contexts, or more recent years is inappropriate. The paper does not cover the UK, Europe, Asia, Australia, or Latin America. Hine, Harman, Leder-Elder and colleagues’ 2025 Journal of Family Violence paper is the UK analogue and finds higher rates in UK separated-parents samples, but on a different sampling frame. Readers outside the US should treat the 22 million figure as a reference point for how a population-level estimate looks, not as a number that applies to their jurisdiction.

Fourth, critics of the broader parental-alienation prevalence literature have fair points to make. Janet Johnston (2020, Family Court Review) and Varavei (2025, International Journal of Social Welfare) have raised concerns that population-level self-report may inflate the true prevalence by capturing cases where both parents feel alienated from the child without a clear primary alienator, or by capturing the fallout of separation rather than a specific alienation pattern. No formal published methodological critique of this specific paper has appeared in the literature I can locate, but the broader critiques apply. The paper is the best available prevalence estimate. It is not the final word.

None of these limitations is a reason to discard the paper. They are reasons to cite it carefully — with the specific numbers attributed to the specific polls, with the targeted-versus-alienated distinction preserved, and without the casual conflations that have accumulated around the 22 million figure in the six years since publication. The aim of a good article about a paper is to leave the reader better able to cite it well. That is the aim of this one.