Shouting at children when they misbehave, rather than reasoning with them, is more harmful to their overall behaviour, according to new research from LSE.
Dr Laure de Preux, a research officer from LSE’s Centre for Economic Performance, presented her findings at the Royal Economic Society’s 2014 annual conference this week.
Her study applies quantitative analysis to a question that has previously largely been the preserve of qualitative studies: the impact of parenting styles and parenting practices on children’s wellbeing. Among the findings:
Excessive shouting, punishing or ignoring naughty children increases their behavioural problems;
Reasoning with children does not have an adverse impact on their behaviour;
In general, there is surprisingly no parenting style that appears to be particularly worse or better for the child;
Participation in physical activity encouraged by the mother improves physical health but harms mental health;
Higher socio-economic classes clearly create environments that benefit their children, and in which it is difficult for researchers to pinpoint exactly what makes a difference.
Dr de Preux says:
"Economists have generally focused on the socio-economic determinants of children’s wellbeing. For example, children from educated mothers and/or from wealthier families are generally better off.
"But from a public policy point of view, these results are not helpful to improve the worse off; the government cannot send mothers back to school, and it has no guarantee that monetary transfers will be spent in a way that improves the child’s wellbeing.
"It is therefore important to get a better understanding of the impact of specific mothers’ practices."
Psychologists have focused for decades on the impact of parents’ involvement and strictness on their children’s wellbeing and behaviour. Adopting an economics approach, Dr Laure de Preux shows that the mother’s overall involvement and strictness are affected by many other factors that bias the results.
She corrects for these unobservable factors, and considers the impact of specific parenting style measures that could affect children’s physical and mental health. Her analysis reveals that punishing a child a lot when she/he is naughty, apart from reasoning with the child, is very harmful. But there is not a specific parenting practice that appears to be better than another.
Starting with Baumrind in 1966, psychologists have advocated the benefit of the parents’ overall involvement and strictness. Baumrind defines authoritative parents as the ones who are more demanding, while at the same time being attentive to their children’s needs and showing affection. In this literature, children from authoritative parents are more independent and self-confident, and show pro-social behaviour among others.
But what are the dangers of relying on small and observational samples? There are a number of issues when trying to estimate the causal effect of maternal practices on children’s wellbeing.
First, there is the problem of ‘reverse causality’: a mother is more likely to shout if her child is naughty, but the child might be more naughty if her mother shouts a lot. Simply looking at the correlations between the two does not make it possible to disentangle the one-way effect.
The second issue is the ‘omitted variable’ problem: for example, mothers who are more educated may be more likely to reason with their naughty child, but might also be more likely to do many other things that benefit their children, such as providing healthier meals and more frequent physical activities. Therefore, when in this case education is not observed, one has the impression that reasoning with a naughty child may be better.
Much of the evidence from psychology relies on small samples and a limited set of socio-economic variables. In her research, Laure de Preux applies various econometrics methods to address these problems.
She shows that once the two identification problems are taken into account, overall involvement is beneficial to the child, contrarily to strictness. Particularly excessive shouting, punishing or ignoring a naughty child increase her behavioural problems. Only reasoning does not impact the child’s behaviour.
In general, there is surprisingly no parenting style that appears to be particularly worse or better for the child. The only specific qualitative measure of parenting style that becomes relevant is the participation in physical activity encouraged by the mother, which improves physical health but harms mental health.
Despite the fact that overall involvement and strictness are important determinants of parenting style, the role of socio-economic factors remains almost unchanged. This result suggests that higher socio-economic classes are able to create a whole environment that is not observed by researchers, but that clearly benefits the child. Further research should aim to identify the possible pathways explaining these relationships.
Notes for editors:
For further information or for a copy of the paper, contact Romesh Vaitilingam from the Centre for Economic Performance at firstname.lastname@example.org, +44 7768 661095
9 April 2014