The author describes that the boys of a father who worked 55 hours or more in a week showed more externalizing behaviors. They recruited thousands cases and observed them for 10 years at maximum. The behaviors of the children were measured with the Child Behavior Checklist, which was used to evaluate problematic behaviors of children, by their mothers. They also gathered some factors of the condition of the families such as working time ,financial status, level of education, and so on from the parents.
The result was that the children whose father worked more than 55 hours in a week showed higher scores of problematic behaviors.
The authors seemed to claim that time spent with their children was so important for them. However, this survey has failed to show the evidence about it. No association was found between the time spent with children and their behaviors. The author guessed that the quality of time was more critical rather than the quantity of the time they spent with their children. But I think it is not persuasive.
My hypothesis is more simple: aggressive children are born from hyperactive fathers. Hyperactivity is a characteristic which is difficult to interpret. In childhood it may be problematic. They may tend to externalize one's failure. In adulthood, aggressive lifestyle is preferable to some extent. Some of the fathers can have bipolar spectrums or ADHD trait. These concepts are still vague, and so likely to be underestimated. The author declared that mental problem of the parents did not significantly influence the result. But potentially hypomanic persons are seldom assessed as mentally disordered.
In total, we should be cautious to evaluate the result of this research.
By the way, I was noticed about the research through a certain Japanese news site. The site introduced this research as an evidence of the harm of bad companies in which the employees were forced to work for a long time. The website did not show the link to the article itself. It is an example of warped interpretation and insincere citation, I think.