What Children Actually Need During Divorce: Evidence Over Instinct

The question every parent asks first is the wrong one. 'Will this damage my children?' has a researched answer, but it's the wrong question because it treats divorce as a single event that either harms or doesn't. Four decades of developmental research says that's not how it works.

The actual question — the one Paul Amato spent his career trying to answer across two meta-analyses synthesizing hundreds of studies — is closer to this: under what conditions do which children struggle with which aspects of adjustment, and for how long? That question has answers. Some of them are clear enough that they should change what you worry about.

The finding that replaced the old consensus

Until the mid-1990s, the dominant narrative was that divorce damaged children, full stop. The research that supported that narrative had a specific flaw: it compared children of divorce to children of intact marriages generally, without distinguishing between high-conflict and low-conflict intact marriages. Once researchers started making that distinction, the finding changed.

Amato's 2001 meta-analysis of 67 studies found that children from high-conflict marriages that ended in divorce showed better long-term adjustment than children from high-conflict marriages that stayed intact. The inverse was also true: children from low-conflict marriages that ended in divorce did worse than if those marriages had continued. The variable that mattered most was conflict, not structure.

That 75–80% figure is the one people don't hear, because it's not news. The minority who struggle longer — approximately 20–25% — are the children the research community focuses on, because they're where intervention matters. But the aggregate picture is that most children adjust.

Why conflict is the variable that matters

Parental conflict harms children through several pathways that developmental psychology has mapped in detail. The most direct is modeling: children learn relationship patterns by watching. Chronic hostile conflict teaches them that this is what intimate partnership looks like, and it shapes their expectations and their own relational scripts.

The second pathway is chronic stress. Exposure to ongoing parental conflict activates children's stress-response systems repeatedly and unpredictably. Over time, this contributes to the developmental outcomes the research flags — sleep disruption, academic difficulty, behavioral dysregulation, physical health complaints, anxiety and depression.

The third pathway is what researchers call 'triangulation' — being drawn into the conflict as a messenger, a confidante, or a surrogate partner. Kelly & Emery (2003) identified triangulation as one of the most reliably harmful parenting patterns during and after divorce. Children who are asked to carry adult emotional content about the other parent fare worse on nearly every measured outcome.

What actually helps: the protective factors list

Kelly & Emery's (2003) synthesis identified the protective factors that reliably predicted better child adjustment across studies. The list is unglamorous. These are not dramatic interventions — they're the ordinary work of parenting, preserved through an extraordinary period.

• At least one warm, authoritative parent-child relationship maintained through the transition. (This is the single most predictive factor in most studies.)
• Stable routines — bedtimes, meals, school schedules, extracurriculars — preserved as much as possible across both households.
• Conflict exposure minimized: no arguing, criticism, or interrogation at handoffs or in earshot.
• Age-appropriate, honest communication about what is happening — not every detail, but enough that children aren't making up worse versions in their heads.
• Preservation of relationships with both parents when both parents are safe, and with extended family on both sides.
• Explicit, repeated reassurance that the divorce was not caused by the child and is not their responsibility to fix.
• Continuity of school, friendships, and neighborhood when possible. Geographic stability is protective.

What's notable about this list is how much of it is within parents' control. The research does not say children need their parents to stay together. It says children need the things in that list, and those things can be preserved whether or not the marriage does.

Age matters — but for what they struggle with, not whether they do

Every age has its characteristic struggle. The struggles are different, but the research does not find an age at which divorce is 'safe' or an age at which it's catastrophic. What changes is the type of support children need.

Infants and toddlers (0–3)

At this age, children can't process divorce conceptually. What they're sensitive to is disruption of attachment relationships and caregiving routines. The protective priority is stable caregiving from both parents when both are available, predictable routines, and low-stress transitions. The overnights debate (covered in a separate article) applies most intensely here and is still contested in the research community.

Preschool (3–5)

Preschoolers are developmentally egocentric — they're prone to magical thinking, and they often believe they caused the divorce. Regression (toileting, sleep, separation anxiety) is common and temporary. Explicit, repeated, age-appropriate reassurance that it's not their fault is the central intervention. Avoid long explanations; short, concrete, reassuring messages repeated often.

School-age (6–11)

School-age children understand divorce conceptually but often blame themselves or fantasize about reuniting their parents. They tend to be loyal to both and distressed by criticism of either. They may struggle with school performance and peer relationships during the acute adjustment period. The intervention is honest communication, explicit permission to love both parents, and protection from conflict.

Adolescents (12–18)

Adolescents often appear more 'fine' than they are because they externalize less to parents and more to peers. They may express distress through anger, withdrawal, academic disengagement, or risk-taking. They are also more aware of financial and logistical changes than younger children, and they are more vulnerable to being drawn into adult content about the divorce. Protection from triangulation and maintenance of the parent-teen relationship are the core interventions.

The 'stay together for the kids' question, handled carefully

This is the question where research findings are most often misused — in both directions. It's worth stating the evidence carefully, because reasonable parents can read the same studies and reach different conclusions about their own marriages.

What the research supports: children in low-conflict intact marriages generally fare as well or better than children whose parents divorce. Children in high-conflict intact marriages generally fare worse than children whose high-conflict parents divorce. The conflict level is the variable that flips the comparison.

What the research does not support: the simple claims 'staying is always best' or 'leaving is always best.' The evidence does not support either. What it supports is that the conflict children are exposed to is the mechanism, and interventions that reduce conflict (whether through repair or through low-conflict separation) are what helps.

How to tell them

The research on how children are told about divorce is clearer than the research on when. Consistent findings across clinical and empirical literature support a specific approach: tell them together if possible, tell them before physical changes happen, keep it simple, and expect to tell them more than once.

• Together and coordinated. Both parents present, with an agreed message. This signals that adult decisions are adult decisions and children are not being asked to choose sides.
• With enough lead time. Not months, but enough that they can process before moving day. Surprise and secrecy are what the research identifies as harmful, not information.
• Short and concrete. 'We are not going to be married anymore. We are going to live in two homes. We both love you. This is not because of anything you did. You will still see both of us.' Long emotional explanations overwhelm children.
• Expected repetition. Children do not fully process the news the first time. Expect questions to resurface for weeks and months. Answer them calmly each time, without burdening children with adult emotional content.
• Without blame. Naming the other parent as the cause places the child in an impossible loyalty position. The research on triangulation is clear that this increases harm.

When to get professional help for your child

Most children do not need therapy because of divorce. Routine adjustment distress — sadness, temporary regression, anger, sleep disruption — is normal and expected in the first months. The indicators that professional support is warranted are duration, severity, and functional impairment.

• Distress that has not gradually eased after 2–3 months.
• Significant and sustained changes in school functioning, sleep, eating, or peer relationships.
• Signs of depression or anxiety beyond what the child is developmentally expressing as grief.
• Expressions of self-harm, hopelessness, or wishes not to exist — regardless of duration, these warrant immediate evaluation.
• Being drawn into adult conflict to the point where they are trying to mediate or take sides.

Evidence-based child therapy for divorce adjustment is typically short-term and focused. Long-term therapy is not automatically better; sometimes it prolongs identification as a 'child of divorce' when the child would otherwise naturally move through the adjustment.

What to do with this

If you're deciding whether to divorce: the research does not answer that question for you. It does suggest that if your children are currently exposed to chronic parental conflict, the comparison between 'stay and keep exposing them' and 'separate and protect them from the conflict' is not obvious.

If you're already divorcing: the list of protective factors is mostly within your control. Warm parent-child relationships, stable routines, conflict minimized at handoffs, honest age-appropriate communication, preservation of relationships with both parents and extended family. None of this requires you to reconcile with your ex; all of it requires you to protect your children from the relationship's residue.