Guides for Parents · 14 min read

When Parents Separate: What Children Need

By Muhammad Salman Afzal, Owner & Director · 23 August 2026

Written by the Inclusive Developmental and Therapy Center therapy team — our Speech & Language Therapist, psychology and ABA staff. Clinical reviewer for this site: Dr Muhammad Suffyan, MB BS (GMC 8023727) · Updated August 2026; this page has changed since our last clinical review.

Either you are deciding, or it has already happened, and in both cases the same question is running underneath everything: what is this doing to the children.

You will get certainty from everyone. Relatives who say a broken home ruins a child. Friends who say children are resilient and adapt in months. Both are talking past the evidence, which is more specific and more useful than either.

Illustration of a family going through a change

Start with the honest baseline

We are not going to open by reassuring you, because the first finding does not.

A meta-analysis of 67 studies comparing children of divorced parents with children whose parents remained continuously married found the divorced group scored significantly lower on academic achievement, conduct, psychological adjustment, self-concept, and social relations. That gap narrowed during the 1980s and widened again in the 1990s.

So the difference is real and it is measured across five domains. Anyone who tells you it makes no difference at all is being kind rather than accurate.

But a group difference is not a prediction about your child, and — more importantly — it does not tell you what is causing the difference. That is the question the rest of the research answers, and the answer changes what you should do.

The conflict, not the separation

This is the central finding and it reframes the whole decision.

A study of 1,076 children aged 10 to 13 found that perceived conflict between parents significantly and negatively predicted academic self-concept — how capable the child believed they were. It also found something schools should know: high well-being at school buffered the effect when conflict at home was low, but when conflict was high, that buffering weakened and even reversed. The authors call it an anti-buffering effect. A good school does not cancel out a war at home, and past a certain intensity it may stop helping at all.

The decisive study, though, is about what happens inside households that have not separated.

Research on 346 families with primary-school children identified three patterns using both parents' reports:

Positive spillover — good interparental relationship, few hostile or disengaged conflict strategies, more positive parenting.
Overt negative spillover — poor interparental relationship, open hostility, and more corporal punishment of the children.
Covert negative spillover — fewer hostile strategies but more disengaged ones, and more corporal punishment than average. The cold house. The parents who do not fight in front of the children because they have stopped speaking.

Children in the positive group had the most prosocial behaviour and the fewest problems, as you would expect. Children in the openly hostile group had the worst outcomes on parent-reported measures.

And then the finding that should stop anyone considering staying together in silence: although parents in the covert group reported their children had only moderate difficulties, the children themselves reported more anxiety than children from either the openly hostile homes or the positive ones.

Read that twice. In the cold, disengaged household, the parents thought the children were doing reasonably. The children reported the highest anxiety of any group. Children detect the atmosphere whether or not you raise your voice — and when nothing is ever said out loud, there is nothing for them to make sense of, so they carry it privately.

"We never argued in front of them" is not the protection it is believed to be.

Conflict handled well is not harmful — it may be the opposite

Here is the finding almost nobody is told.

Two longitudinal studies with 243 and 238 families of preschool children, using multiple methods and multiple informants, examined exposure to constructive interparental conflict — disagreement that is expressed, addressed and resolved with warmth intact. The result: constructive conflict predicted greater social competence and lower externalising symptoms, and the mechanism ran through the children's effortful control — their capacity to regulate themselves. The more rigorous three-wave design produced a comparable pattern.

Children who watch two adults disagree, stay respectful, and repair afterwards are learning something they cannot learn any other way. The goal is not a household without conflict. It is a household where conflict is conducted in a way a child can survive watching — and, ideally, learn from.

Which means the real choice is rarely "stay or separate". It is:

Conflict that is open, respectful and repaired — associated with better outcomes
Conflict that is hostile — associated with the worst parent-reported outcomes and more corporal punishment
Conflict that is silent and disengaged — associated with the highest child-reported anxiety

What happens to children after a separation

The living arrangement research is more encouraging than the headline meta-analysis, with one important caveat we will come to.

A Swedish national survey compared children in joint physical custody (n = 391), sole parental care (n = 654), and nuclear families (n = 3,639), aged 10 to 18. Children in joint physical custody did not report higher psychological complaints than children in nuclear families, while children in sole parental care reported elevated complaints compared with joint custody. Adding socioeconomic variables and parental ill health only marginally changed the result.

A second national study of 4,823 children aged 10 to 18 found no significant difference in self-esteem between children living equally with both parents, mostly with one parent, and those in nuclear families — while children in single care showed lower self-esteem. Again, not explained by socioeconomic factors.

The caveat, stated by the researchers themselves. A review of 40 studies on joint physical custody found evidence of positive effects, but noted the samples are dominated by self-selected, highly educated, high socioeconomic status parents with low conflict levels and children aged 6 to 15. The risks and benefits are therefore not settled, and are actively debated.

Which is the honest translation: shared arrangements look good in families able to cooperate. That is not proof they work where cooperation is impossible — and it is another reason the conflict, not the arrangement, is the thing to work on.

Does anything help the parents?

A comparative study evaluated psychoeducational programmes for divorced and separated parents — 62 participants across 11 weekly sessions with six-month follow-up, delivered in person or online.

Across both formats, participants showed reduced overall symptoms, reduced covert conflict, and increased forgiveness. The in-person format produced greater improvement than online for total symptoms, depression and forgiveness; online improved accessibility.

It is a small study without a no-treatment control, so treat it as encouraging rather than conclusive. But note what it moved: covert conflict — the same disengaged, silent pattern associated with the highest child-reported anxiety.

What to do, whichever way this goes

If you are still deciding

Ask the better question. Not "should we stay together for the children" but "can we conduct this in a way our children can survive watching". Both staying and separating can be done well or badly.
Do not assume silence is protection. The cold household produced the highest child-reported anxiety in the study above.
If there is violence, the calculation is different. Children living with violence between adults are not being protected by an intact household. See discipline without hitting or shouting for the related material, and get help locally.

Telling the children

Tell them together, if it is safe to. One version of events, from both of you.
Say it is not their fault, explicitly. Children construct guilt privately, from an argument they overheard or a wish they once had.
Say what will not change before what will. School, room, grandparents, who loves them.
Give concrete detail. Where they will sleep, on which days, who takes them to school. Young children cannot hold an abstract reassurance; they can hold a timetable.
Do not give the reasons. The adult narrative is not theirs to carry, and telling it recruits them into taking sides.
Expect to repeat it. The same questions will come back for months.

Afterwards — the things that matter most

Never use a child as a messenger. Adults communicate with adults, directly.
Never criticise the other parent to the child. A child hearing that half of them is bad has nowhere to put it.
Never ask them to report on the other household.
Keep both parents involved where it is safe. The living-arrangement studies consistently show children in sole care faring worse than those with contact with both.
Protect the routine. Same school, same bedtime, same rules in both houses where possible.
Tell the school this week, so a teacher does not misread a sudden change in behaviour as defiance.
Get your own support. Your functioning is the mechanism through which most of this reaches your child — see the break you are not taking.

What it looks like in a child

By age, roughly:

Under five: clinginess, sleep disruption, regression — bedwetting, baby talk, needing help with things they had mastered. See bedwetting.
Five to eight: open sadness, longing for reunion, fantasies of getting the parents back together, worry about the absent parent.
Nine to twelve: anger, often at the parent perceived as responsible; loyalty conflicts; somatic complaints; falling marks.
Teenagers: withdrawal, risk-taking, early independence, or a hyper-competent child who parents everyone else and is assumed to be fine.

The last one deserves emphasis. The child causing no trouble is not automatically the child doing well — and the study of covert conflict is the evidence for that.

When to get help

Changes that have not begun to settle after several months
School refusal or a marked fall in performance — see school refusal
Regression lasting more than a few weeks
A child expressing responsibility or guilt for the separation
A child who has become the confidant or carer of a parent
Refusing contact with one parent, which needs careful and neutral handling rather than pressure from either side
Aggression at home or school — see when your own child hits you
Any self-harm, or talk of not wanting to be here — a same-week conversation with a doctor

Where we fit

We are not a legal service, we do not assess custody, and we will not take a side between two parents.

What we do is the child — how they are actually doing rather than how each household reports them, which the covert-conflict research shows can be two different things. That includes the anxiety, the school refusal, the regression, and giving a child one adult in the arrangement whose only job is them. Where both parents are willing, work on how conflict is conducted is the intervention with the clearest link to the outcomes on this page.

Our pages on helping an anxious child, helping children manage big feelings and childhood anxiety cover what that work involves.

A consultation is Rs 1,500 and lasts up to 50 minutes, we work in Urdu or English, and you do not need a referral.

Sources

Amato PR. Children of divorce in the 1990s: an update of the Amato and Keith (1991) meta-analysis. Journal of Family Psychology. PubMed 11584788
Huang L, et al. Typologies of family dynamics in Chinese families and implications for child adjustment: a person-centered approach. Journal of Family Psychology. PubMed 40244957
Davies PT, Cao VT, Baker AA, Sturge-Apple ML. Why is constructive interparental conflict beneficial for children's mental health? The role of effortful control and positive affect. Developmental Psychology. PubMed 40354259
Sun Y, et al. Effects of perceived interparental conflict on academic self-concept in late childhood: a moderated mediation study in China. Acta Psychologica. PubMed 41967267
Fransson E, et al. Psychological complaints among children in joint physical custody and other family types: considering parental factors. Scandinavian Journal of Public Health. PubMed 26553250
Turunen J, Fransson E, Bergström M. Self-esteem in children in joint physical custody and other living arrangements. Public Health. PubMed 28595062
Steinbach A. Children's and parents' well-being in joint physical custody: a literature review. Family Process. PubMed 29963700
Ordóñez-Carabaño Á, et al. Effectiveness of in-person versus online psychoeducational interventions with divorced/separated parents: a comparative analysis. Family Process. PubMed 41802725

FAQ

Frequently asked questions

Does divorce harm children?

There is a real measured difference, and we will not soften it. A meta-analysis of 67 studies found children of divorced parents scored significantly lower than children of continuously married parents on academic achievement, conduct, psychological adjustment, self-concept and social relations. But a group difference is not a prediction about your child, and it does not identify the cause — which is where the more useful research begins.

Should we stay together for the children?

That is the wrong question. The better one is whether the relationship can be conducted in a way your children can survive watching, because both staying and separating can be done well or badly. Research on 346 families found children in openly hostile households had the worst parent-reported outcomes — but children in cold, disengaged households reported more anxiety themselves than children from either the hostile or the positive homes.

We never argue in front of the children. Does that protect them?

Not as much as is believed. In that same study of 346 families, parents in the covert group — fewer hostile strategies, more disengaged ones — reported their children had only moderate difficulties, while the children themselves reported the highest anxiety of any group. Children detect the atmosphere whether or not you raise your voice, and when nothing is said out loud there is nothing for them to make sense of, so they carry it privately.

Is all conflict bad for children?

No — and this is the finding almost nobody is told. Two longitudinal studies of 243 and 238 families with preschool children found that exposure to constructive interparental conflict, meaning disagreement that is expressed, addressed and resolved with warmth intact, predicted greater social competence and lower externalising symptoms, working through the children’s capacity to regulate themselves. Children who watch two adults disagree, stay respectful and repair afterwards are learning something they cannot learn any other way.

What living arrangement is best after a separation?

The evidence favours keeping both parents involved where it is safe. A Swedish national survey found children in joint physical custody did not report higher psychological complaints than children in nuclear families, while children in sole parental care reported elevated complaints. A study of 4,823 children found no significant self-esteem difference between joint custody, mostly-one-parent and nuclear families, while children in single care scored lower. Neither difference was explained by socioeconomic factors.

Is that shared-custody evidence reliable?

With an important caveat the researchers state themselves. A review of 40 studies found evidence of positive effects but noted the samples are dominated by self-selected, highly educated, high-income parents with low conflict levels and children aged 6 to 15. Shared arrangements look good in families able to cooperate — which is not proof they work where cooperation is impossible, and is another reason the conflict rather than the arrangement is the thing to work on.

How should we tell the children?

Together, if it is safe, so there is one version of events. Say explicitly that it is not their fault, because children construct guilt privately from an overheard argument or a wish they once had. Say what will not change before what will — school, room, grandparents, who loves them. Give concrete detail: where they sleep, on which days, who takes them to school, because young children cannot hold an abstract reassurance but can hold a timetable. Do not give the adult reasons. And expect to repeat all of it for months.

What should we never do afterwards?

Never use a child as a messenger — adults communicate with adults. Never criticise the other parent to the child, because a child hearing that half of them is bad has nowhere to put it. Never ask them to report on the other household. And do not mistake a child causing no trouble for a child doing well.

What does distress look like at different ages?

Under five: clinginess, sleep disruption, regression such as bedwetting or baby talk. Five to eight: open sadness, longing for reunion, worry about the absent parent. Nine to twelve: anger often aimed at the parent seen as responsible, loyalty conflicts, stomach aches, falling marks. Teenagers: withdrawal, risk-taking, or a hyper-competent child who parents everyone else and is assumed to be fine.

When should we get help?

If changes have not begun to settle after several months; if there is school refusal or a marked fall in performance; if regression lasts more than a few weeks; if your child expresses guilt or responsibility for the separation; if they have become a parent’s confidant or carer; if they refuse contact with one parent, which needs neutral handling rather than pressure; or if there is aggression at home or school. Any self-harm or talk of not wanting to be here is a same-week conversation with a doctor.