How often adult children call, and what the number hides
By Hollis Treadway, Contributing editor, research review · · 2 min read
Once a week. That is roughly where every survey of contact between parents over sixty-five and their adult children ends up, and the consistency is suspicious, because the surveys do not agree on what a contact is.
The one I find most useful asked 1,208 households, in a national panel run by a university survey center, to log every exchange with each adult child for fourteen days: calls, video calls, texts, and visits, with who initiated each one. The headline figure was a median of one voice call per week per child, which is the number that got repeated.
The number under the headline is more interesting.
Who dials
Parents initiated sixty-three percent of voice calls. Children initiated seventy-one percent of texts. If you count only the exchanges the child started, the once-a-week call becomes a call every two and a half weeks and a text every three days.
Whether that is a decline in contact or a change in its form depends on whether you think a text is contact. The parents in the panel, asked directly, mostly said no, or said yes with a qualification that the interviewers coded as no. The children, asked the same question, said yes without the qualification.
Daughters, sons, distance
Daughters called more than sons, by a margin of about forty percent, which is consistent with four decades of this kind of research and should surprise nobody. Distance mattered less than expected: children within thirty miles called slightly less often than children farther away, and visited instead, and the visits were short. The authors describe a child who lives across town and comes by on Sunday for forty minutes as having roughly the same contact profile as one who lives in another state and calls on Sunday for forty minutes, which is a finding that will satisfy nobody who lives across town.
Sons, for what it is worth, texted at about the same rate as daughters. The difference is almost entirely in voice calls, and the authors do not speculate about why, which I appreciate.
What the log cannot say
Fourteen days is short. A parent in a bad patch, or a child in a busy one, skews a two-week log in a way a year would smooth out. The panel was weighted to be representative by age and region but not, as far as I can tell, by family size, and a parent with five children logs more contact than a parent with one for reasons that have nothing to do with closeness.
And nothing in a count tells you what was said. A nine-minute call about a dog and a nine-minute call about a diagnosis are one call each. My colleague wrote about the first kind in December, and I would not argue with her, except to note that the survey would not have been able to tell them apart either.
Correction, February 2, 2026: This article originally gave the survey's sample as 2,108 households. The published figure is 1,208. The error was ours, in transcription, and has been corrected.