Does science say smacking causes harm?

‘Smacking children makes their behaviour worse, scientists find.’ (The Times, 29 June 2021)

‘Just one smack is enough to harm a child’s mental health.’ (The Daily Telegraph, 13 January 2021)

If you’ve seen newspaper headlines on smacking, you’ll know how ‘certain’ the research findings apparently are.

But as with many other academic studies described in the press, there is an unhelpful willingness to skip over some important details.

Those arguing for a smacking ban often point to the scientific research to claim that smacking is both ineffective and bad for children’s development.

But ‘what the science says’ isn’t always so clear.

Let’s look at some of the limitations of the ‘science’ on smacking:

Correlation and causation

‘Correlation does not equal causation’ is a cliché for researchers, but it’s a really important rule when it comes to studies on smacking and childhood outcomes.

For example, does smacking actually cause aggressive behaviour and mental health problems and other negative outcomes? While there is good evidence of associations between these things, there are several reasons why it does not prove causation.

  1. Omitted variables

Child development is complex involving a huge array of factors. Even when trying to be unbiased and accurate, it’s impossible to account for every characteristic that is relevant to parents’ use of reasonable chastisement and children’s development. These are called ‘omitted variables’ or ‘confounders’.

We cannot automatically assume that a poor child outcome was caused by smacking. Some important factor is often absent from the analysis, such as maternal depression, family stability, household income or child temperament.

Without recognising these other variables and accounting for them, the relationship between smacking and a negative outcome may appear stronger than it is. Even studies which control for some of these variables inevitably omit countless other aspects of parent-child relationships which are not captured in the data.

Science cannot isolate the effect of one parenting strategy on child development and rule out all the possible confounders.

  1. Reverse causality

Where studies show links between smacking and a negative outcome, such as bad behaviour, it could be that the relationship is actually the other way around – reverse causality. In at least some cases, bad behaviour may be prompting the smacking. But this is a commonly overlooked complexity in interpreting smacking data.

  1. Measurement challenges

Another reason not to confuse correlation with causation is that it is difficult to actually quantify the negative outcomes being observed. How do you get a proper grasp of a child’s mental health from a few questions in a survey?

As well as that, it is also difficult to get unbiased data when asking people to report about controversial or sensitive topics, such as disciplining children.


Despite hundreds of studies on the subject, none have established a causal link between smacking and poor child outcomes.

While there is much to be gained from researching child development, to pretend that we can measure all the factors that shape a child’s trajectory in life is an obvious mistake. Correlations are interesting and worth giving thought to, but they should not be used as the basis for changing the criminal law and interfering in family life.