Junk science

It is difficult to see what good purpose was served by this study.

The Otago people (in conjunction with Auckland's public health group) put cameras on kids that would take snapshots every six seconds. Then they poured through the footage to see how often the cameras, and presumably the kids, saw things that Otago people have long wanted to have restricted, like ads for food they don't like or alcohol. They counted the number of times things were seen. And then published the numbers in (at least) two separate studies expressing horror at the number and calling for bans on the things that they counted.

Is there any number that would have been low enough? Almost certainly not.

Is there any context for the number that might assist in anyone telling whether a number is low or high? Heck no. The news story on it talks about kids being bombarded with 27 junk food ads per day. Would there be fewer than 27 ads for candy in any 80s kid's daily bundle of comic books? I'm not the only one who remembers being bombarded with ads for Life Savers, am I?

The news story also says they got $800,000 to do the study.

They also counted the number of times the cameras saw alcohol related stuff and used the number, I kid you not, to call for a ban on alcohol sales at supermarkets.

The abstract of their paper is almost parody. Here it is.

Background and aim

Exposure to alcohol marketing within alcohol retailers has been associated with higher rates of childhood drinking, brand recognition, and marketing recall. This study aimed to objectively measure children's everyday exposure to alcohol marketing within supermarkets.

Method

Children aged 11–13 (n = 167) each wore a wearable camera and GPS device for four consecutive days. Micro-spatial analyses were used to examine exposures within supermarkets.

Results

In alcohol retailing supermarkets (n = 30), children encountered alcohol marketing on 85% of their visits (n = 78). Alcohol marketing was frequently near everyday goods (bread and milk) or entrance/exit.

Conclusion

Alcohol sales in supermarkets should be banned in order to protect children from alcohol marketing.

I wonder what number would have had them saying "Ok, maybe we don't need to call for a ban." Would it be more than zero? Was there any point to the study? I don't think the 1989 legislation that allowed sales in supermarkets said anything like "Oh, and we totally expect that parents will cover their kids' eyes as they go past the wine aisle, so it's ok, but if anybody ever shows that kids might actually see what's down the aisle, then we totally need to re-think this."

Some context that didn't make it into any of the press reporting:

• The proportion of kids aged 15-17 who consumed alcohol in the past year dropped from 74.5% to 57.1% from 2006/7 to 2014/15.
• Binge drinking more than halved over the same period, dropping from 25% to 10.7% of kids aged 15-17.
• The number of hazardous drinkers among those aged 15-17 dropped from 19.5% to 10.8% from 2006/7 to 2014/15.

I don't know if this is the stupidest study in the world. Otago also had that one where they recruited 13 people, mostly from Facebook, interviewed them about their smoking, then called for a ban on smoking outside of bars on the basis of those conversations.

What would be sufficient basis for a call to ban alcohol sales at supermarkets? Strong evidence that the substantial inconvenience cost imposed on shoppers would be outweighed by reductions in external harm imposed by drinkers as result of the ban.

Being able to pick up a bottle of wine or beer with your normal shopping trip is a good thing that should count for more than nothing. It's a nice part of the Outside of the Asylum.