My somewhat tongue-in-cheek piece on Labor’s threat to banish licensed characters like Shrek from the supermarket has provoked predictable reactions suggesting that I am an apologist for an evil empire whose aim is to “trick” young children into eating unhealthy food (comments, yesterday).

As a parent, I’m very concerned about childhood obesity. In fact, as an obese adult with a family history of premature cardiovascular disease, I’ve got more reason to be concerned than many. But when it comes to the role of Shrek in childhood obesity, perhaps it’s time we sorted the ogres from the donkeys.

Notwithstanding Dr Rosemary Stanton’s eminence in the field of nutrition, she seems to have a very narrow view of how statistics should be applied to the issue of cause and effect.

Yes, it’s true that the average Australian supermarket now stocks 30,000 items (in the industry known as “stock-keeping units” or SKUs). To begin with, thousands of these are accounted for by numerous different types and flavours of toothbrushes, shampoo, hair colour, toilet paper, and dog and cat food.

And yes, there are many more varieties of chips, chocolate and fizzy drinks in the supermarket than when I was growing up. But there are also perhaps 20 or 30 different SKUs containing different types of lettuce or other salad greens when, in the 1970s, there was just “lettuce” (one kind), and “rocket” was what took Apollo astronauts to the Moon.

Likewise, you couldn’t get a “snack pack” of diced mango in natural juice or kiwifruit puree or countless other variants in the 1960s (I didn’t eat a mango until I was in my twenties). Yoghurt was something exotic you could only buy at the local “Continental delicatessen” – today, Coles offers more than 150 different yoghurt SKUs, the majority of which are “lite” or “98% fat free”.

The point is that it is simplistic to observe the larger number (ie variety) of snack food items on supermarket shelves and conclude that this either proves or explains higher consumption of junk food and obesity. If the increase in SKUs in the supermarket is the cause of childhood obesity, then it would be just as reasonable to conclude that our children are fat because they are eating pet food or panty liners!

What about advertising? In 2004, the UK Office of Communications (Ofcom), the independent regulator of the communications industries, published an extremely comprehensive and measured report entitled “Childhood Obesity – Food Advertising in Context”.

A literature review undertaken for Ofcom by the London School of Economics concluded that there was evidence for a “modest” direct effect of food promotion on children’s food choices, but that there was “insufficient evidence to show that TV advertising has a larger, indirect effect on children’s food choices”.

Quantitative research commissioned especially for Ofcom demonstrated that, while TV played an important role, a number of other factors – including the child’s own taste preference, price, familiarity, peer pressure, healthiness and convenience – were actually more important than advertising in shaping children’s food choices and attitudes.

The Ofcom report found that obese children may have somewhat different responses to food advertising, and that parents of obese children may have different shopping behaviours, although they were actually less likely to report that their child pesters them while shopping. But, again, none of this research allowed cause and effect to be disentangled.

Of course it would be silly to ignore the potential power of advertising – like all forms of TV programming – to indirectly affect the views of children and parents about issues such as diet, body shape and exercise. But as a fictional, non-human, cartoon character, Shrek would almost certainly be a whole lot less influential in this regard than characters on Neighbours or contestants on Australian Idol and Big Brother.

The use of licensed characters and celebrity endorsers is undoubtedly effective in many promotions, but the sales data show that short-run licensed products (like Arnott’s Shrek cheese snacks) almost inevitably take share away from other established snack products (like Twisties and Cheezels), rather than promoting additional consumption in the category. In other words, if cheese snacks are a problem then it would be a whole lot more effective to ban the category rather than the character.

Dr Stanton flatters me beyond belief if she truly thinks my provocative “outburst” in Crikey provides all the proof we need that banning the use of licensed characters would remedy childhood obesity. I think we all deserve more evidence than that.