Show me the data

I will never forget a lesson my 7th grade science teacher taught on reading the fine print. She presented research linking artificial sweeteners and cancer in lab rats.

There’s something about the words ‘research’, ‘lab’, and ‘study’ that makes things sound so believable, isn’t there? The same goes for statistics. Nearly 87% of the lab rats contracted terminal cancer within 2 years of first ingesting the artificial sweetener aspartame on a daily basis.**

Our science teacher then went on to tell us how the lab rats were fed their body weight in artificial sweetener daily. Well, I don’t know about you, but I’ve never met anyone who drank 168,000 cans of Diet Coke per day. That’s right. I did the calculation.

My engineer of a husband likes to quote a college professor of his who also taught a valuable lesson: correlation does not imply cause and effect. I can’t tell you how many times I’ve had to remind a co-worker of this basic principle. It can be a tough one to remember.

One of my favorite examples was put forth by Steven D. Levitt, the economist, in Freakonomics. He found a strong correlation between the number of books in a child’s household and the child’s standardized test scores. Isn’t it so tempting to assume causation? The data, however, held no correlation between standardized test scores and actual books read.

All of this is to say – be wary of what you read. Scrutinize. If something sounds alarming, look for the fine print. Or as I like to say, “Show me the data.”

A few weeks ago I read an article referencing studies that, “show a decline in the well-being of American children over the past 50 years.” I cringe. Not for the decline in the well-being of American children, but for the absurdity of that statement.

Pray tell me who facilitated this 50 year long experiment. I’d love to hear how “well-being” was quantified and measured. Let’s pretend for a minute that there was actually a “study”, that this was anything more than someone’s opinion. Now tell me how anyone could possibly go about determining why.

Let’s all help each other out and refrain from backing up our parenting decisions with ridiculous assertions.

**I totally made this statistic up while writing this post.

Advertisements

11 thoughts on “Show me the data

    • I thought you would appreciate that!

      As for the article – hahahaha! My post was in no way supposed to be about artificial sweeteners. I totally just threw a dart at the wall (figuratively) when I picked aspartame. So I find it pretty awesome that it wasn’t far from the truth.

  1. People love to grab on to the tiniest sliver of info and run with it. A great example is the idea that kids are much more likely to be abducted by strangers these days, when in reality, it’s actually less likely to happen today than in the 60s and 70s. Cable news effect!

    • That’s a good one! In reality the bigger issue is probably childhood obesity, which could be due to less play time outside (says the girl that always preferred indoor activities like coloring…).

  2. That’s why I don’t believe the “Back to Sleep” campaign. If you look at the numbers, the amount of total infant deaths didn’t reduce, just the number of data points that were catalugued as SIDS-related, meaning some cases that were being classified as SIDS were just reclassified. Yay statistics!

    … But don’t tell my pediatrician!

    • I won’t! I know a number of people who put their babies to sleep on their stomach, and I’m not opposed to it myself. Our biggest issue is laying Ruby down without her waking up. All of my practice has been laying her down on her back so I can’t imagine how I would flip her over without waking. Otherwise I think I would try the stomach – just to see if she slept better that way.

      • We gently roll Addilyn from our arms to her stomach in the crib and then bounce her like a basketball (very gently) and she loves it.

  3. my favorite studies! and then you see that the group on which the study was conducted was something like 10 specimens. In which case, even if it did imply causality, which it doesn’t, it would certainly not be a statistically relevant sample.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s