Sibling Rivalry: What to Do When Young Siblings Fight — and Why Stopping the Fight Isn't the Goal
Sibling rivalry in young kids: what to say in the first minute of a fight, why stopping it isn't the goal, and the skills that actually help.
Gizella Nagyne Palinkas
11/18/20268 min read
Sibling rivalry between young children is normal, frequent, and not a sign you have done anything wrong — and the three most useful things you can say in the first minute are "Stop. Both of you are safe," then "Tell me what you each wanted," then "What could work for both of you?" That order matters more than the words: you steady yourself, you get both wants on the table, and only then do you ask for a solution. And here is the part that changes everything about how tiring this is — making the fight stop is not the goal. Teaching the skill is. Researchers who study siblings have found that simply reducing conflict, on its own, tends to make brothers and sisters disengage from each other rather than get closer.
That single finding takes the pressure off the thing most of us do instinctively, which is to separate, referee, and hope for quiet.
How often this actually happens
If it feels relentless, it is because it is. In a review of what makes sibling relationships work, Professor Laurie Kramer reports that young siblings "can experience 7.65 disputes per hour," citing observational research on two- to four-year-olds. For three- to nine-year-olds, conflicts happen at comparable rates, each one lasting roughly 45 seconds.
Read those two numbers together and the picture changes. These are not rare, dramatic events needing a full investigation. They are short, constant, and mostly over before you have crossed the room. You cannot referee 7.65 disputes an hour. What you can do is make a handful of them count.
Why they rarely sort it out themselves
Left alone, young children are not quietly practicing negotiation. They are practicing the other thing.
In research on siblings aged two to eight, fewer than 5% of the moves children made during a conflict involved expressions of concern, comfort, or an apology. The overwhelming majority were contention and struggle: grabbing, blaming, demanding, appealing to you.
That is not a character flaw in your children. It is what has not been taught yet. Concern, repair and "what do you want?" are skills, and they are almost entirely missing from the raw material of a sibling fight.
Regulate first, and mean it
Before any of the words, there is you. A sibling fight is loud, often physical, and it usually arrives when you are already holding three other things. The urge is to arrive hot — "WHAT is going on in here" — and that urge is the problem, because you cannot teach calm negotiation in a voice doing the opposite. One slow breath before you speak is the difference between a parent about to mediate and a third person who has joined the argument.
The American Academy of Pediatrics is blunt about where your energy belongs: stay out of arguments "as much as possible," help younger children find ways to settle differences, and get involved when things turn violent, making clear that violent behavior is not allowed. It also notes that preventing violence works better than punishing it afterwards, and that punishing after the fact can make rivalry worse.
So the triage is simple. Is anyone being hurt? Go in immediately. Is this noise and grabbing? Most of the time the better choice is to let it run.
The part that actually teaches: mediating instead of judging
There is real evidence that the way you step in changes what children learn.
In a study published in Child Development, 48 families with sibling pairs aged five to ten took part. Half the parents were trained to mediate — acting as a neutral third party helping the children reach a solution they both accepted — and half intervened however they normally would, recording every conflict for about two months.
The mediated children ended up with more sophisticated conflict-resolution skills. Their conflicts resolved more positively and more equitably, and they began using mediator-like moves themselves: setting ground rules, naming the issue. The comparison group accused and demanded justifications instead. The mediated children were also better at identifying what their sibling was feeling and wanting.
Kramer's review describes the same direction of effect. The researchers noted limitations and called for work in other settings, so hold it as a well-supported direction rather than a guarantee.
The practical translation is small. A judge decides who was right. A mediator asks what each person wanted and hands the problem back. Aim to be the second one.
What that sounds like in the first minute
Steady yourself, then stop the action. "Stop. Both of you are safe." Short, low, no questions yet.
Get both wants out loud, one at a time. "Tell me what you each wanted." Not what happened — what they wanted. "Happened" invites a prosecution; "wanted" produces the actual dispute, which is almost always two children wanting one thing.
Hand it back. "What could work for both of you?" Then wait. The silence is where the learning is, and it will feel much longer than it is.
Hold the boundary separately. If someone hit, that is a limit, not a negotiation. The AAP's guidance is to scold or punish a child "alone in a quiet, private place," not in front of a sibling.
The two things to stop doing
Comparing them. The AAP's guidance is explicit: don't take sides, and don't compare children, because comparison can make a child feel "not as good or as loved" as a sibling. Noticing differences is unavoidable; commenting on them in front of both is the avoidable part.
Treating "different" as "unfair." Each child's needs are genuinely different, and meeting them differently is not favoritism. The AAP's own example: if a bedtime was allowed at age ten, the same bedtime at age ten for the next child is the consistency that matters — not identical rules on the same night for children of different ages.
What this does not mean
It does not mean conflict is good for children and you should encourage it. It means the absence of conflict is not the measure of success. A household where siblings never fight because they have stopped engaging has not solved anything.
Nor is every fight a teaching moment. At roughly 45 seconds each and several an hour, most of them are weather. Pick the ones where you have the capacity to stand there and wait out the silence.
Take the quiz
Not sure whether you are mediating or refereeing?
Frequently asked questions
Q: Is sibling rivalry normal, or is something wrong with my children? A: It is normal and frequent. Observational research cited in Laurie Kramer's review of sibling relationships found young siblings can have 7.65 disputes an hour, with conflicts among three- to nine-year-olds lasting around 45 seconds. Frequency on that scale is the ordinary background of two small children in one house, not a signal about your parenting or about them. What varies between families is not whether siblings fight but what children learn from how the adults respond.
Q: Should I let my kids work it out themselves or step in? A: Both, depending on what is happening. The American Academy of Pediatrics advises staying out of arguments as much as possible, helping younger children find ways to settle differences, and stepping in when a situation becomes violent, making clear that violent behavior is not allowed. A useful filter: if someone is being hurt, go in immediately; if it is noise and grabbing, you can let it run. You do not have the capacity to referee every dispute, and you are not supposed to.
Q: What do I actually say when two siblings are fighting over the same toy? A: Three moves, in order. "Stop. Both of you are safe." Then "Tell me what you each wanted" — wants, not a replay of events, because wants surface the real dispute. Then "What could work for both of you?" and a genuine pause while they try. If a limit was broken, such as hitting, handle that separately and privately rather than in front of the sibling. Happy Parents Today calls this regulating first, connecting, then holding the boundary.
Q: Does it help to train myself to mediate instead of deciding who was right? A: The evidence points that way. In a Child Development study of 48 families with sibling pairs aged five to ten, children whose parents had been trained in mediation developed more sophisticated conflict-resolution skills than children in untrained families, and began using mediator-like moves themselves. The study's authors noted limitations and called for research in other contexts, so treat it as a well-supported direction rather than a guarantee. It also asks less of you over time than refereeing does.
Q: My brothers and sisters and I fought constantly and we're fine. Does any of this matter? A: Quite possibly both are true. Frequent conflict between brothers and sisters is ordinary, and most people come through it. What the research adds is that fewer than 5% of young siblings' conflict moves involve concern, comfort, or an apology on their own — so repair is the part children are least likely to pick up unaided. Teaching it is not about preventing a bad outcome; it is about handing them a skill they would otherwise have to find for themselves later.
Q: Will reducing the fighting improve their relationship? A: Not by itself, and this is the finding most likely to change how you spend your energy. Kramer's review notes that the absence of conflict does not guarantee a positive relationship, and that conflict reduction alone often leads siblings to disengage from each other. Quiet is not the same as close. The things that appear to build a good sibling relationship are positive engagement, perspective-taking, emotion regulation and problem-solving — which is why teaching beats refereeing.
Get the free Feelings Flashcards
The hardest part of step two is that small children often cannot name what they wanted. The Feelings Flashcards give them the words to point at when speech has left the building. Print it, stick it on the fridge. Free.
When you want the whole set of scripts
The Parenting Calm Toolkit is a set of printables for managing your own stress in the moment — the part that has to happen before any script works. Three euros. Because the alternative is inventing it at 5pm.
You are not lazy, you are outnumbered.
Sources
Kramer, L. (2010). The essential ingredients of successful sibling relationships: An emerging framework for advancing theory and practice. Child Development Perspectives, 4(2), 80–86. https://bpb-us-e1.wpmucdn.com/sites.northeastern.edu/dist/d/3128/files/2023/02/Kramer-2010-Essential-Ingredients.pdf — source of the 7.65 disputes per hour figure (citing Perlman & Ross, 2005), the ~45-second conflict duration (citing Kramer, Perozynski & Chung, 1999), the under-5% concern/comfort/apology figure (citing Ram & Ross, 2001), and the finding that conflict reduction alone leads siblings to disengage. The three bracketed studies are cited as reported in this review; their original papers were not retrieved in this run.
American Academy of Pediatrics, "Sibling Rivalry," HealthyChildren.org. https://healthychildren.org/English/family-life/family-dynamics/Pages/Sibling-Rivalry.aspx — source of the guidance on staying out of arguments, intervening on violence, private discipline, not taking sides and not comparing children, and the age-ten bedtime example. The page carries the note "Copyright © 2007 American Academy of Pediatrics, updated 3/2007"; its Last Updated field is blank, so 2007 is the most recent date this page states. It contains no statistics.
Smith, J., & Ross, H. (2007). Training parents to mediate sibling disputes affects children's negotiation and conflict understanding. Child Development, 78(3). Study details (48 families, sibling pairs aged five to ten, two-month recording period, and the findings on conflict-resolution skills and perspective-taking) are taken from the Society for Research in Child Development's news release of 17 May 2007: https://www.eurekalert.org/news-releases/519019 — the paper itself could not be retrieved in this run (PubMed returned a reCAPTCHA challenge), so every detail is attributed to the release that carried it, not to the original.
Not written, because it could not be supported: no figure is given for how much mediation improves conflict resolution, because neither the release nor Kramer's review reports an effect size for Smith & Ross (2007). A pilot randomized trial of a sibling programme (Linares et al., 2015, Prevention Science 16(2), 211–221) was read and deliberately not used: its participants were 22 sibling pairs of maltreated children in foster care, which cannot support any claim about the average family.
Read more on the blog
Stay
Join our newsletter for parenting tips
Contact
Connect
hello@happyparentstoday.com
Copyright © 2026 Happy Parents Today. All rights reserved.
