Little-noticed amid the social upheaval of the past half-century has been one change that is almost universally considered positive: the increasing amount of time residential fathers spend with their children.

In 1965, dads spent an average of 2.5 hours a week caring for their children; by 2011, that figure had almost tripled to reach seven hours a week. According to Pew Research Center surveys, forty-six percent of fathers worry that they spend too little time with their children, and 50 percent of them say it’s very or somewhat difficult to balance their job and family responsibilities.

As that last figure implies, many fathers sense a conflict between providing for their families and being active, involved fathers (though roughly as many seem to feel they’re doing OK). One would expect men’s work hours  to have a dramatic effect on the amount of time they spend with their kids, but as researcher Brittany S. McGill found in a recent study in the Journal of Family Issues, that’s not necessarily the case.

Using data from the Child Development Supplement to the Panel Study of Income Dynamics, McGill analyzed the association between dads’ work hours and attitudes and their involvement with their children. Fathering attitudes were measured by respondents’ affirmation or denial of statements like “fathers play a central role in the child’s personality development” and “it is essential for the child’s well-being that fathers spend time interacting and playing with their children.” (Deeming fathers’ involvement essential, and as important as mothers’ involvement, was classified as a “nontraditional” or “new father” attitude.) Involvement was measured by the time dads spent with kids both in general and on specific activities, as well as their responsibility for certain aspects of child care such as bathing, disciplining, and playing with kids.

To my surprise, McGill found that fathers’ work hours generally were not associated with the amount of time they spent with kids. More work hours meant less time spent on basic physical care of children, but on other activities, even dads working more than 50 hours a week spent roughly the same amount of time with their kids as other dads. The very small number of dads who were not employed at all did spend more time on child care, however, and when McGill examined individual fathers’ changes in work and child care hours over time,  she found a stronger relationship between the two.

Fathering attitudes, however, were “significantly positively associated” with father involvement in the bivariate results:

Fathers with nontraditional attitudes toward fathering, for example, spend an average of 17.3 hours per week with the focal child [the child analyzed in the panel study], compared to 13.9 hours among fathers with the most traditional fathering attitudes.

Note that these numbers are higher than the aforementioned number of hours dads spend specifically on child care; these numbers include all time spent with kids.

Cross-sectional analyses of whether dads’ attitudes shape the association between their work hours and the time they spend with kids show that having “new father” attitudes reduces the association between the two. That is, traditional fathers who work long hours spend less time with their kids than traditional fathers with shorter hours, whereas nontraditional fathers with long hours spend roughly as much time with their kids as their counterparts working only part-time. McGill speculates, based on studies of mothers’ time allocation, that these highly involved fathers may juggle long work hours and time with their kids by reducing the hours they spend on personal leisure.

That dads are making an effort to spend time with their sons and daughters is good news for kids—but not all children are benefitting from this new norm. Close to three in ten children live apart from their fathers, and few absent dads are regularly involved in kids’ day-to-day lives. Unfortunately, trends in divorce rates and nonmarital births suggest that this father involvement gap will only continue to grow.