A 2007 survey linked baby videos to a smaller vocabulary

The short answer

A 2007 telephone survey of over a thousand families found that infants aged eight to sixteen months who watched more baby DVDs and videos knew fewer words, on average, than infants who watched less, after researchers adjusted for other factors. It is a widely cited finding, and also a good example of why an association in survey data is not the same as a demonstrated effect of the video itself.

What the evidence says

Frederick Zimmerman, Dimitri Christakis and Andrew Meltzoff surveyed 1,008 parents of children aged two to twenty-four months by telephone in February 2006, identified through birth records, and published the results in October 2007 as Associations between media viewing and language development in children under age 2 years in the Journal of Pediatrics. Parents reported their child's viewing of different types of television and video content, and children's vocabulary was assessed using a standard parent-report checklist, the MacArthur-Bates Communicative Development Inventory. In a model adjusted for demographic and family-interaction factors, each additional hour a day of baby DVD or video viewing among infants aged eight to sixteen months was associated with a 16.99-point lower vocabulary score. No such association appeared among toddlers aged seventeen to twenty-four months, and a parent watching alongside the child made no measurable difference in either group. A 2005 review of the wider evidence base, Television and Very Young Children, had already noted that some studies in this literature found negative associations between viewing and language development of this kind.

For context

This was a survey, not an experiment: parents chose how much their own children watched, and no video was assigned to one group and withheld from another. The authors themselves say further research was needed to explain the association, leaving open whether heavier baby-video viewing displaces the talk and interaction that builds vocabulary, or whether families already managing a less verbal infant reach for videos more often for reasons the survey could not measure. The finding is also specific to the youngest infants studied and to video content marketed at babies; it does not describe toddlers in this sample, or television in general.

A practical next step

Rather than treating a single hours-per-day figure as a target to hit or avoid, it can help to notice what watching a baby video is replacing in a particular day, since the association here was with the youngest, least verbal children.

• What would this baby be doing with that time if the video were not on?
• Is a marketing claim for a baby video based on this kind of survey, or on any comparison at all?
• Has a pattern of more or less talk with this particular child changed alongside their viewing?

The 2007 finding does not show that a baby video directly causes a smaller vocabulary; it shows a statistical association, in one age band, that the authors themselves called unexplained and worth investigating further.