Junk Science: Baby Video a No-No?

"Is Baby Einstein doing your child more harm than good?"

The answer to that question, posed by the cover story of Time magazine (Aug. 27), may depend on how you feel about drive-by product disparagement committed by anti-TV fanatics.

The Time article was spurred by a recent study published in the Journal of Pediatrics by University of Washington researchers who reported that each hour per day of viewing baby DVDs and videos (such as the Baby Einstein and Brainy Baby products) by 8- to 16-month-olds reduced their vocabulary by seven words as compared to those who did not watch.

The university aggressively marketed the UW study with an unusual media release that targeted specific brands that parents were urged to "limit the amount of time they expose their children to DVDs and videos such as Baby Einstein and Brainy Baby."

Researchers Dimitri Christakis and Frederick Zimmerman fueled the fire with colorful but rather non-scientific sound bites, including: "It’s not the first blow to baby videos and likely won’t be the last," and "I would rather babies watch 'American Idol' than these videos."

Ensuing media headlines followed a predictable course, ranging from "Baby Einstein Videos Ineffective, Study Finds" (National Public Radio) to "Study Shows 'Smart Baby' DVDs Slow Language Acquisition" (NBC News) to "Videos as a Baby Brain Drain" (Los Angeles Times).

Web sites ran even more sensational headlines: "Baby Einstein Sucks the Vocabulary Out of Your Kid's Brain" (Seattlest.com) and "The Parent Crap: Babies Raised by Videos Approximately as Dumb as Expected" (Gawker.com).

While I don’t know whether infants actually benefit from baby DVDs and videos, that's an extremely complex issue that would take a great deal of intensive and intrusive study to resolve, and the inadequate UW study certainly fails to show they cause harm.

First, the study is small. Although the researchers touted the inclusion of 1,008 children in the study, only 384 children were between the ages of 8 months and 16 months. Only 215 of those engaged in any TV, DVD or video viewing. It’s not clear how many children watched baby DVDs, but the answer is likely fewer than 215.

The number watching for one hour or more per day is likely fewer still. Next, the validity of the raw data is questionable. Data on viewing habits were collected by a telephone poll of parents. The researchers didn’t observe or validate any of the data collected and parents may easily have over- or underestimated their children’s actual viewing habits.

It’s also quite possible the data are biased, potentially skewing study results, since they weren’t collected from a representative sample of the general population. Study subjects were drawn from limited geographic area on the basis of telephone number availability.

Telephoned parents could decline to participate. Moving past the study’s questionable data, the researchers' technique for measuring child language development also is problematic. While the so-called MacArthur-Bates Communicative Development Inventory, or CDI, seems to be a reliable assessment tool for older toddlers, its application to 8- to 16-month-olds is not so reliable, according to a study published in the journal Child Development (March-April 2000).

The researchers nevertheless compared CDI scores to the claimed time spent per day watching the following types of media: baby DVDs and videos; children’s educational shows; movies and children’s non-educational TV; and grownup TV.

The only activity negatively correlated in a statistically significant way with CDI score was watching baby DVDs and videos. That correlation reportedly was about 10 times greater than that for grownup TV.

So the researchers jumped to the conclusion at least for the purposes of media consumption that the baby DVDs and videos reduced vocabulary development.

But not only is it highly inappropriate to leap from mere statistical correlation to cause-and-effect conclusions particularly based on a small and novel study with dubious data, the researchers would have us believe that watching baby DVDs and videos is much worse than watching grownup TV in terms of child vocabulary development.

Why would this be so? Absent a plausible explanation, the University of Washington didn’t return my call or e-mail, so none seems forthcoming that the result may be chalked up to unreliable data and analysis.

Baby DVDs and videos weren’t associated with reduced vocabulary development among the study's 17- to 24-month-olds. For the older toddlers, watching baby DVDs and videos correlated with a similar positive effect on vocabulary development as story-telling and music-listening.

Did the alleged adverse effect of baby DVDs and videos disappear with age or was it entirely bogus to start with?

The researchers admitted in their study's fine print that they didn’t directly test whether baby DVDs and videos had an actual positive or negative effect on vocabulary acquisition. They also quietly acknowledged that the study's correlative nature "precluded" drawing causal inferences and that their results could have been affected by biased and incomplete data.

While they remembered or were compelled by the Journal of Pediatrics' editors to note these "major limitations" in their write-up, Drs. Zimmerman and Christakis seemed to suffer mental lapses when it came to statements they made in media interviews.

And why let a few facts including the researchers' history of alarming parents about children watching TV, DVD and videos (more than 10 publications since 2004) get in the way of their scare?

This sort of junk science often goes unchallenged by its victims in the vain hope that the fuss will blow over. In this case, however, the Walt Disney Co. (owner of Baby Einstein) demanded that UW retract its media release, according to the Denver Post.

Disney refused to say whether it might sue, but the egregious conduct surrounding this study would seem to make for a pretty good test case of whether agenda-driven researchers can blithely smear commercial products despite knowing that their attack is groundless.

Steven Milloy publishes JunkScience.com and DemandDebate.com. He is a junk science expert, an advocate of free enterprise and an adjunct scholar at the Competitive Enterprise Institute.

Respond to the Writer