Raised to Resist · The Receipts

Is spanking actually harmful?

UPDATED MAY 13, 2026

Consensus

Confidence: strong

Spanking harms kids, and the research isn't mixed on it. A meta-analysis of 111 effect sizes covering more than 160,000 children found that 13 of 17 outcomes showed spanking predicting harm: more aggression, more antisocial behavior, more mental health problems. The most uncomfortable finding: those effect sizes look about the same whether researchers are measuring spanking or physical abuse. The line parents draw between "discipline" and "abuse" doesn't show up cleanly in the data. This isn't a correlation problem. Using propensity-score matching, which mimics random assignment by controlling for the obvious confounds, researchers found that spanking at age 5 still predicted more behavior problems at ages 6 and 8, even after adjusting for baseline behavior and family circumstances. Spanking doesn't fix behavior. It worsens it. In the ACE (Adverse Childhood Experiences) framework, spanking sits in the same cluster as physical and emotional abuse. Adults who were spanked as children have higher rates of suicide attempts, substance use, and heavy drinking, independent of other childhood adversity. And it doesn't even work as discipline. No study has consistently shown spanking effective at changing behavior long-term. What does predict kids growing into socially competent adults: warmth.

Spanking and child outcomes: Old controversies and new ... (2016)Strengthening Causal Estimates for Links Between Spanki... (2017)Spanking and adult mental health impairment: The case f... (2017)Hugs, Not Hits: Warmth and Spanking as Predictors of Ch... (2016)

Contested

The research consensus is real, but the debate around it isn't fake either. Three things are still actively contested. One: whether "ordinary spanking" (open hand, buttocks, with warmth, no objects, in a non-abusive home) should be statistically separated from harsher physical punishment. A minority camp led by Robert Larzelere argues the meta-analyses lump together cases that look very different in practice. The dominant view, articulated in Gershoff and Grogan-Kaylor's 2016 meta-analysis, is that even after isolating "ordinary spanking," the harm pattern holds. Two: whether residual confounding inflates the effect sizes. Propensity-score matching controls for what researchers can measure. Critics argue that what they can't measure (household chaos, genetic predispositions, parental mental health that predates the spanking) might still be doing some of the work. Most longitudinal studies addressing this find the effect holds. The methodological floor is open but narrowing. Three: whether spanking has different effects in different cultural contexts. Earlier work suggested possible moderation in Black families specifically; later replications largely did not support this. That literature has its own dedicated read on The Receipts. None of these debates change the headline. They sharpen the claim. They don't break it.

What is debated: One area of ongoing discussion involves whether race or cultural context moderates the association between spanking and child outcomes. Some studies report smaller or attenuated effect sizes in certain racial and ethnic groups, though the overall direction of findings remains negative across groups.

Race as a Moderator of Associations Between Spanking an... (2016)

What This Means

If you were spanked as a child, you're working against a deep groove. Many parents who don't want to spank still find themselves spanking when overwhelmed. That's an impulse-control problem in the parent, not a discipline problem in the kid. Naming it that way changes what you do about it. Here's what actually works: Warmth as the baseline. Not in a hallmark-card way. Specific warmth: eye contact, physical affection that isn't contingent on good behavior, narrating your kid's emotions back to them. This is the single variable that predicts social competence most reliably across studies. Clear expectations stated before the moment. Kids do better when they know what's coming. "We're going to the store. In the store, you'll stay in the cart. If you want something, you can ask once." The work happens in the parking lot, not in aisle six. Empathic responses to misbehavior, not soft ones. "I see you're frustrated. We don't hit. Let's figure out what you needed." Not "Oh sweetie, it's okay." It's not okay; the behavior was a problem. Empathy names the feeling. Discipline addresses the behavior. These are not the same. Parent training programs work. Triple P, Parent-Child Interaction Therapy (PCIT), and The Incredible Years are evidence-based programs with measurable effects. None of them ask you to be a perfect parent. They give you a different default move when your patience runs out. Here's the script for the moment you'd usually spank: "I need a minute. We're going to talk about this in five." Walk away. Drink water. Come back. Address the behavior. The kid learns that you got angry and still chose not to hit them. That's the lesson that ends the cycle.

Receipts