As a good Catholic girl at Mount Mary Immaculate Academy, I studied Latin for five years in school, and still have a particular affection for this dead language. So does science. Take the medical field, for example, and its common Latin phrase “post hoc ergo properter hoc”, which means “after this, therefore because of this”.
This belief is applied by researchers when they conclude that because a result happened after something else happened, the ‘something else’ must have caused the result. But in medicine, as in life, we know that correlation does not equal causation.
For example, the weekend before my first heart attack symptoms, I attended a birthday party (my own) where we celebrated the occasion by enjoying lots of very good wine and my friend Lynnie’s gorgeous homemade birthday cake. If correlation did in fact equal causation, we might reasonably conclude that wine and birthday cake are in fact the causes of heart attack.
Here’s another example, courtesty of Photon In The Darkness:
“An illustration of the perils of concluding that correlation equals causation is the strong correlation in elementary school children between their reading ability and shoe size. Obviously, bigger shoes (or bigger feet) do not help children read better, nor does better reading ability cause feet to grow, so the correct explanation must be something else. In fact, this is an illustration of the third reason for correlation: C (increasing age) causes both A (bigger feet) and B (better reading ability).”
Here’s an even better example from award-winning journalist and blogger Sandy Szwarc,who once made a pretty frightening announcement in Ideas in Action from PBS.
“You are about to learn of a beverage so dangerous that we must ban or restrict its sales, or at least enact tax penalties on it to deter consumption. Here’s what the research shows:
- Every person who drinks it dies.
- It has been linked to obesity. In fact, bigger people drink the most of it.
- It’s associated with type 2 diabetes, and all diabetics drink it in especially large amount.
- All heart attack victims drink it, and it’s a known factor in heart failure.
“There are been hundreds of studies finding these correlations — correlations so strong they make the evidence irrefutable. This is bad stuff. Everything you’ve just read is true. What is it?
Of course, says Sandy Szwarc, you could have filled in the blank with almost anything reported in health “news” that is frequently blamed for obesity, type 2 diabetes, heart disease or premature death such as:
- sodas/soft drinks
- high fructose corn syrup
- dietary fat
- high LDL (bad) cholesterol
- fast food
- trans fats
- watching television
Sandy gives this example:
“A study from the University of North Carolina at Chapel Hill, was released which claimed soda/soft drink consumption had increased 135% since 1977 and since rates of type 2 diabetes and obesity were rising, too, that was evidence that consuming these [drinks] increase weight gain in children and adults.”
This conclusion is known as “data dredge”. Based on that correlation alone, researchers then leapt in reverse to conclude:
“Reducing soft drink and fruit drink intake would seem to be one of the simpler ways to reduce obesity.”
Of the thousands of foods and beverages we consume, this study picked soda. But in typical data dredge fashion, says Sandy Swarc, the researchers could have mined that databank and pulled out anything. And they have!
For example, in a previous study, the same researchers found grains, legumes and low-fat milk intake up among adults since 1965, along with significant decreases in calories and percentages of dietary fat. Yet they didn’t tie these overall “healthful” eating trends to rising rates of obesity or type 2 diabetes.
According to the always enlightening Skeptic’s Dictionary, this is merely an example of what’s known as confirmation bias. Confirmation bias refers to a type of selective thinking whereby one tends to notice and to look for what confirms one’s beliefs. Not only that, but we then want to ignore, or undervalue, or not even look for the relevance of anything that contradicts those beliefs.
For example, if you believe that during a full moon there is an increase in admissions to the hospital Emergency Department where you work, you will take notice of admissions during a full moon, but then be inattentive to the moon when admissions occur during other nights of the month. A tendency to do this over time unjustifiably strengthens your belief in the relationship between the full moon and accidents.
Sandy Szwarc also reminds us that a major earmark of junk science and pseudoscience — or science being misused to sell us something or advance a special interest — is continuing to test and retest a belief long after it’s been disproven in well-designed studies.
Building a body of research can lead unsuspecting consumers to believe that there’s a body of evidence in support of a belief. It’s a good reminder to carefully consider the quality of research behind those miracle medical breakthrough health newsbites.
All studies are not created equal, and the weight of the evidence does not come by tallying up the number of studies on one side or another.
Find out more from Junkfood Science.
Do you know where Sandy Szwarc is now? Is she ill? Does she intend to return to publishing her web site?
I wish I did know – no new postings for a long time, and a cryptic ‘thanks’ message on her site.
Hello Carolyn. I have often suspected this to be true about what CAUSES certain outcomes. Too much room for subjective interpretations… Thanks for this.
Pingback: Nursing Station
Data dredge! This explains a lot! Love this post – thanks for this. Have just discovered your site, love your writing, smart, sassy, in your face – just the way I like it. Have just subscribed to your blog, too. Keep up the great work….
Thanks for your kind words, Natalie. Glad you found my site.