"Children of confident fathers who embrace parenthood are less likely to show behavioural problems before their teenage years," The Guardian reports.
A study found a link between positive attitudes towards fatherhood and good behaviour at age 11. The UK study involved more than 6,000 children born in 1991 or 1992 as well as their parents.
Fathers were interviewed during the first year after their child's birth about their positive and negative reactions to becoming a father. Both parents were also asked about the amount of time the father was involved in childcare or domestic work.
After taking account of other factors, children of men scoring highly on confidence and emotional response to fatherhood were 13% and 14% less likely to have behaviour problems at age nine, and 11% less likely at age 11.
Factors such as a father's emotional response and confidence were found to be more important than the amount of time spent involved in the actual, sometimes messy, side of day-to-day childcare.
Attitudes to parenting have changed in the 25 years since the study began, so these results may no longer apply. Other factors linked to a reduced chance of children having behavioural problems included having older, better-educated parents.
And observational studies like this can't prove cause and effect. But perhaps it's not surprising that having positive, confident fathers at an early age is linked to better outcomes for children in later life.
For men concerned about upcoming "dadhood", there is training and advice available from a range of organisations, such as the National Childbirth Trust (NCT).
The study was carried out by researchers from the University of Oxford and was funded by the Department of Health, the UK Medical Research Council, the Wellcome Trust and the University of Bristol.
The UK media covered the study reasonably accurately. Different media sources picked different figures to illustrate the size of the effect, with some (including The Guardian) using figures that had been adjusted to take account of possible confounding factors, such the social status of the family.
Others (including the Daily Telegraph and the Daily Mail), used the unadjusted figures highlighted in the study's press release.
Unadjusted figures often sound more impressive but adjusted figures are usually more reliable.
This is a longitudinal cohort study, which recruited children and their parents while the mother was pregnant, and followed them for many years to assess how factors from their early childhood might affect outcomes in later life.
This type of study is good at spotting links between factors, but cannot prove that one factor causes another. For example, some children with behavioural problems might have been difficult babies who cried a lot, which might have affected their father's emotional adjustment to fatherhood, rather than the other way around.
Researchers used information from a long-running ongoing study, the Avon Longitudinal Study of Parents and Children, which recruited more than 14,000 pregnant women in the Bristol area in 1991 and 1992. Researchers used questionnaires filled in by parents at 8 weeks, 8 months, 9 years and 11 years after birth.
They only included children who'd been living with both parents at eight months, and for whom there was follow-up data at 9 or 11 years.
They used the questionnaires filled in by men to identify three factors – emotional response, time spent on childcare or domestic work, and confidence as a partner and father – which might affect children's behaviour.
They used the responses to the questionnaires to construct a statistical model to assess high or low scores to allocate to men for each of these factors. Behavioural scores for the children were assessed by questionnaires filled in by the mother.
Researchers took the following potential confounding factors into account in their calculations:
These were used to adjust the odds of children having behavioural problems, for fathers with high or low scores on emotional response, time spent on domestic work, and confidence in their role.
Children of men with a positive emotional response to fatherhood were:
Children of men who felt confident as fathers and partners were:
Researchers found no statistically significant link between children's behavioural problems and the amount of time their fathers had spent on domestic and childcare activities in early childhood.
However, parents who were older, had more education and higher social and economic status were less likely to have children with behavioural problems. Working more hours per week and worse mental health scores were linked to worse behavioural problems in children. Older children and boys were more likely to have behavioural problems than younger children and girls.
The researchers say: "We found that the children of fathers whom we characterised as having a positive emotional response to parenting and a sense of security in their role as a parent and partner early in the child's life … were less likely to exhibit behavioural problems at 9 and 11 years of age."
They say these factors may be "a marker of favourable parental characteristics and positive parenting in the longer term", while involvement in work such as shopping, cleaning and childcare "may simply reflect temporary circumstances" such as lack of other family support.
They conclude that their results suggest "psychological and emotional aspects" of paternal involvement in early years are "most powerful" in children's later behaviour.
It may seem obvious that children would benefit from having fathers who are happy and confident about their role. But there hasn't been much research on which aspects of a father's role are important for children, so this study adds some useful information.
It's important to remember that all the children in the study had both parents living with them in early childhood, so this isn't a comparison of children in single parent families with dual parent families.
The study only looked at the attitudes of fathers who were living with their children, asking questions including whether they had a strong bond with their child, regretted having the child, enjoyed spending time with the child and felt confident looking after them.
It's surprising that paternal time spent on childcare and domestic work did not seem to affect the results.
However, as the researchers say, this apparent anomaly might not reflect the father's long-term parenting, but might be a short-term factor. Some mothers were probably able to take a lengthy maternity leave and had help from other sources, but opportunities for paternal leave were far more limited during the 1990s.
The study has some strengths. It is a big study, carried out over many years, collecting a large amount of data.
However, there are many limitations. Observational studies can't prove that factors such as men's attitudes to fatherhood are the reason for the children's behavioural outcomes.
The researchers took account of some potential confounding factors when presenting their results (although not in the results they highlighted in their press release) but not all of them. For example, we know that education level of parents affected chances of behavioural problems, but these were not adjusted for in the results. In addition, we do not know what other major influences the children may have had, such as grandparents, other extended family, or their experience of nursery or primary school.
The analysis is based on questionnaires filled out by the mother and father which may not be entirely accurate and subject to recall bias.
Finally, the questionnaires regarding the children's behaviour and psychological well-being did not cover any mental health or behavioural condition, such as autism spectrum disorder, which could account for more challenging behaviour.
It's also true that attitudes to childcare and family make up have changed a lot over the 25 years since the study began. It's possible that we would see different results if the study was run again in today's society.
For those men struggling to cope or worried about the future after birth, there is help available from a wide range of different sources.