Does Spanking Work? What Research Shows

Spanking does not work, in the sense most parents mean it: research spanning five decades finds it associated with worse behaviour over time, not better, once studies account for how a child was already behaving beforehand. It does produce faster compliance in the moment it happens, and that gap between what it looks like it is doing and what it is actually doing is the whole story.

This guide walks through what the evidence actually shows, why the short-term effect fools even attentive parents, and what to do with a child who is not responding to any of it. Our parenting and mental health pillar covers the wider evidence on what actually predicts good outcomes for a child.

Does spanking work in the short term

Spanking does typically stop the behaviour it is aimed at, in the moment it is used. A child who is hit is frightened or in pain, and stopping the behaviour is how a frightened child escapes an immediate threat, which is a different mechanism from learning that the behaviour was wrong or figuring out a better alternative.

That distinction matters because “it worked” usually means “the behaviour stopped right then,” and by that narrow measure, spanking often does exactly what a parent wanted. The problem is that this is the only timeframe in which it looks effective, and it is the timeframe a stressed parent in the middle of a difficult moment is least equipped to check against anything longer.

What does the research actually show

The most comprehensive answer to this question comes from a meta-analysis of over five decades of research covering more than 160,000 children. Gershoff and Grogan-Kaylor combined effect sizes across seventeen different child outcomes and found spanking significantly associated with worse results on thirteen of them, including more aggression, more antisocial behaviour, more mental health problems, and a worse parent-child relationship. [gershoff-2016-spanking-metaanalysis] Not one of the seventeen outcomes showed spanking associated with a better result.

Crucially, this held even when the analysis isolated spanking from more severe physical punishment and compared it against the single outcome parents care about most: immediate compliance. Spanking was associated with more immediate compliance in the moment, the short-term effect described above, but it was also associated with decreased long-term compliance, meaning children spanked more often tended to behave worse over time, not better. [gershoff-2016-spanking-metaanalysis] A discipline method that trades long-term behaviour for a moment of obedience is not a method that is working; it is a method that is borrowing against the future.

The American Academy of Pediatrics reached the same conclusion from a different direction. Its 2018 policy statement recommends against spanking, striking, or shaking a child of any age, on the grounds that it does not improve behaviour and carries a documented risk of harm, and it recommends non-physical, consistent discipline in its place. [sege-2018-aap-discipline]

Why does it feel like it works

It feels like it works because the feedback a parent gets is immediate and the cost is not. The child stops crying, stops hitting a sibling, stops the tantrum, right there, in front of you. That is a powerful, instant reward for the parent, which is exactly the kind of feedback loop that keeps a behaviour in a parent’s own repertoire even when it is not achieving what they actually want.

What is missing from that moment is any signal about what happens the next time the same situation comes up, or what the child concluded about why it happened. A child who stops a behaviour out of fear has not necessarily learned what to do instead, only that getting caught is dangerous, which is a lesson about hiding behaviour from a parent rather than about changing it. Longitudinal research is what catches this gap, because it follows the same children over months and years rather than judging a single moment in isolation, and that is exactly where the worse outcomes show up.

What to do instead

Structured time-out and consistent, calmly delivered consequences are what the evidence actually points toward, and the specifics matter more than the phrase suggests. A time-out works by removing attention and stimulation, not by causing distress, so the technique that gets results looks different from how it is often pictured:

• Keep it brief and age-scaled. Roughly one minute per year of age is the commonly used starting point (two minutes for a two-year-old, five for a five-year-old). Longer is not more effective; it just makes the technique harder to follow through on consistently.
• Deliver it calmly, every time. The consequence should look and sound the same whether you are having a good day or a terrible one. A time-out delivered with obvious anger teaches the child to watch your mood rather than their own behaviour.
• State the reason briefly beforehand, not during. One short sentence naming the behaviour (“we don’t hit, so it’s time-out”) before it starts. Do not narrate, negotiate, or re-explain once the time-out has begun; a child who is upset is not absorbing a lecture.
• Return to normal afterward, without a long debrief. A brief, warm re-connection once it is over does more than an extended conversation about what happened. The consequence already delivered the information.
• Use it every time the behaviour happens, not just when you have the energy. Consistency is doing more of the work here than severity. A consequence that sometimes happens and sometimes does not teaches a child to gamble on whether this is one of the times it will be enforced.

Consistency between two parents or carers matters as much as consistency within one. A rule enforced by one adult and waved off by another teaches a child to sort people into who to test, which undercuts the whole approach regardless of which non-physical method is used.

Is there a screener for this

No. Nothing on this site covers parenting or discipline as a self-assessment topic; the screeners here measure things like anxiety, depression, sleep, burnout and loneliness in the adult answering them, not a child’s behaviour or a parenting approach. Our self-assessment hub lists everything actually available, and it is worth a look if you are also carrying your own stress through a hard stretch of parenting, since that stress is itself something the research above treats as relevant to a child’s outcomes, not a separate problem.

When to seek help

Raise it with a doctor if you are finding it hard to control your own anger during discipline, particularly if you notice yourself escalating toward hitting harder or more often than you intend to. That is a common and treatable pattern, not a character flaw, and naming it early is what keeps it from becoming the discipline method by default.

It is also worth raising if a child’s behaviour is not responding to consistent, calmly delivered, non-physical discipline over several weeks. That can mean the approach needs adjusting, or it can point to something else going on that a brief description of the pattern, what triggers it, how long it has lasted and what you have already tried, will help a professional assess properly.

How MyFreud can help

Discipline that works is discipline you can actually deliver the same way under pressure, and pressure is easiest to see in hindsight rather than in the moment you are in it. Daily tracking of your own stress and mood gives you the pattern across weeks rather than the memory of one bad evening, which is exactly the detail that helps you notice when you are drifting toward reacting rather than following the plan you set for yourself, and it is the same detail worth bringing to a doctor if your own anger is part of the picture.

Download MyFreud and start today: App Store or Google Play.