Cynics Unlimited

Critical analysis of media events is now a hot topic, thanks to several exposés of misleading information propagated through major news services. These criticisms are to be expected on highly emotional and divisive topics like the Middle East Conflict; however, news junkies and casual readers alike should be vigilant in dissecting articles on topics that at first may be seen as common knowledge. Consider a recent front-page article in the Globe and Mail concerning the perception of Canadian parents about the health of their children:

Click to Enlarge

Firstly, the terminology in the article is inconsistent. The title states that 9% of parents believe that their children are fat, whereas the article itself states that “9 per cent of Canadian parents believe their children are overweight or obese”. While these terms are sometimes used interchangeably in conversation they are not 100% congruent. “Fat” is a qualitative observation defined as having an unusual amount of fat. It is essentially a personal judgment call and what one person calls fat may not be fat to someone else. Conversely, “overweight” and “obese” have clinical definitions – having a body mass index (BMI) between 25 and 29.9 qualifies one as being overweight, while someone with a BMI of 30 or higher is defined as obese. This may seem like a small point, but one can easily be overweight without looking “fat”. Hence asking a parent if her child is fat is an indication of little more than her child’s appearance and her perception of what constitutes fatness.

More problematic is the interpretation of the numbers themselves. Let’s assume the inconsistency described in the previous paragraph is a non-issue. According to the article, 9% of parents think their children are fat, whereas 26% of children are actually obese or overweight. One could easily conclude that Canadian parents are clueless about the health of their children. However, the statistical comparison is invalid – a parent can have multiple children and make the same erroneous evaluation for each child. Consider the following example:

1. Assume a population of 100 parents and a population 200 children. Each parent has 2 children.
2. A survey of the 100 parents reveals that 9 of them believed their children were fat.
3. A survey of the children reveals that 40 are in fact overweight or obese

Based on this scenario one could conclude that 9% of parents think their children are fat whereas 20% are actually fat. However, because each parent has 2 children, a maximum of 18% of children might actually be labeled as fat by their parents (the 18% is a “maximum” because a parent could conclude that only 1 of her children is fat). This makes the percentage figures misleading because the parents are almost certain to be portrayed as less knowledgeable than in reality. The article did not mention a 1:1 parent to child relationship in the survey, and for that matter mentions “children” several times in relation to parental opinions.

The point is that critical thinking should always be employed when reading news reported in the mainstream media or elsewhere. Distortions and misleading information (sometimes intentional, sometimes accidental or due to space constraints) are a natural part of human communication and we should actively filter this information rather than passively accepting more favorable sentiment as fact.