Create Your Very Own Biased Study

      25 Comments on Create Your Very Own Biased Study

A few weeks ago, the media were all abuzz about a new study published in the New England Journal of Medicine. The study’s conclusion: weight loss is all about the calories, period. High carb, low carb, high fat, low fat – it doesn’t matter.

Lead researcher Dr. Frank Sacks and his team put four groups of people on four different diets for two years. There was a high-fat/low-carbohydrate diet, a high-carbohydrate/low-fat diet, and two diets sort of in-between. At the end of the two years, the weight loss was about the same among all four groups.

Well, that settles it. Case closed. You can get off that silly, faddish, low-carb diet now. Drop the cheeseburger, pick up the whole-grain bun, and start counting your calories.

I was intrigued, because this study’s results don’t match my own. When I was on the Zone diet, I consumed about 1700 calories per day, which included about 175 carbs. I lost six pounds per month. On my fast-food diet, I consumed about 2000 calories per day, but limited myself to around 100 carbs. I lost 12 pounds in 28 days. That’s double the weight loss, despite an extra 300 calories per day. Exercise can’t explain the difference, because I walked the same number of miles each week on both diets.

At least that’s how I remember it. But according to Dr. Sacks, those numbers can’t be right. Either he’s wrong, or I’m a functional schizophrenic, walking around with memories of experiences that didn’t actually happen. I must have a beautiful mind. (My contact at the CIA, who has me decoding secret Al Qaeda messages in New England Journal of Medicine articles, does like the way I think.)

Or maybe, just maybe, there’s something else going on here. Ah, yes, that does seem to be the case.

I’ll spare you the details on the diets because Jimmy Moore and Mike Eades already did a full slice-and-dice on the numbers. But here’s the bottom line: Dr. Sacks and his fellow researchers got the result they wanted by coming up with a definition of “low-carb” that is, to put it charitably, rather creative: 35 percent of total calories.

If you consume 2000 calories per day, that’s 175 grams of carbohydrates. The Atkins diet recommends starting at 20 grams per day – barely one-tenth of the carbohydrate load Sacks calls a “low-carb diet.” Drs. Mike and Mary Dan Eades also advise dieters to start very low and gradually raise the carb count to perhaps 50 grams – less than a third of what Sacks calls a “low-carb diet,” and just 10 percent of the calories in a 2000-calorie diet.

In other words, Sacks didn’t bother to include a diet that would’ve actually restricted carbohydrates to the point where insulin levels would plummet and send the dieters into fat-burning mode.

After thinking long and hard about this little bit of scientific chicanery, I can only come to one conclusion: Dr. Sacks is brilliant! He’s given all the married men on earth what we need most – a clear demonstration of how to “prove” you are right even when you’re not. Speaking for married guys everywhere, I want to thank you, Dr. Sacks, from the bottom of our hearts.

Okay, gentlemen, wise up and follow this brilliant doctor’s example. From now on, when you need to prove a point, just change the definition of the terms so you can’t lose. Then conduct a quick study using the terms you created.

Suppose your wife accuses you of drinking too much beer. “What’s wrong with beer?” you ask. She then offers up some cockamamie theory about how beer makes you fat, stupid, and a dangerous driver. In the pre-Sacks era, you would’ve been roped into an argument that could last for days.

But thanks to the Sacks methodology, now you can disappear into the den for a couple of days, do some intense research, and re-emerge with a study like this:

Comparison of Weight, Cognitive Ability and Motor Skills In Moderate vs. Heavy Beer Drinkers.

No, the results were pretty straightforward.

We assigned one random husband to two groups of beer drinkers on random afternoons when the NCAA playoffs were on television. The first group, defined as the moderate-drinking group, consumed 1.0 bottles of Bud Light during the afternoon games. The second group, defined as the heavy-drinking group, consumed 2.0 bottles of Bud Light during the afternoon games. Both groups consumed the allotted bottles of beer well before half-time.

Both groups also consumed exactly 10 pepperoni Slim Jims during the games, and both groups participated in equal amounts of exercise, defined for the purposes of the study as 1) running to the bathroom during commercials and 2) occasionally jumping up from the couch and yelling, “What’s the matter, ref, are you BLIND?!!”

Weight was measured by bathroom scale. Additionally, abdominal fat was recorded by employing a carpenter’s tape measure, which was extended and bent carefully around the waist.

Cognitive ability was determined by the participant’s ability to read the final scores displayed on the television screen and calculate which team’s score was higher. (In layman’s terms: who “won” the game.) Each participant’s conclusion as to the highest score was later compared to ESPN news reports to assess accuracy.

Motor skills were determined by recording each participant’s average highest score in three rounds of “Grand Theft Auto,” played on an Xbox 360 immediately following the game.

Weight gain was remarkably consistent among both the moderate-drinking and heavy-drinking groups (0.5 pounds vs. 0.7 pounds respectively, P=0.05). Abdominal circumference was identical in both groups, although the participant in the heavy-drinking group suffered a minor laceration due to a belching incident that occurred as the tape measure was being cinched around his waist.

Cognitive ability was also remarkably similar. Both groups correctly identified which team scored the most points in every instance (1.0 vs. 1.0, P=0.01). The participant in the heavy-drinking group was noticeably unsatisfied with the scores, despite correctly calculating which was higher, and indicated verbally that the referee was afflicted with near-sightedness.

Motor skills were similar, but not identical. The average Grand Theft Auto score was slightly higher among the moderate-drinking group than the heavy-drinking group (523 vs. 489, P=0.11) but was not a desired outcome and therefore classified as statistically insignificant. The Grand Theft Auto score was also impacted by the heavy-drinking subject’s need to urinate more often (Pee=2.5/hour vs. Pee=3.5/hour, P=0.22) which could be considered a distraction.

The commonly-held belief among many researchers and most wives that heavy beer-drinking will lead to obesity, stupidity, and an increased risk of traffic accidents and DUI arrests is not supported, but may require further research in other research environments, preferably during baseball season.


That’s it. Hand your wife a copy of the study and explain that according to carefully controlled research, it doesn’t matter how much beer you drink – the only risk is a slightly higher P score, which is no big deal because you always remember to put the seat back down. Then turn triumphantly towards the fridge to grab a cold one.

Of course, if your wife happens to be a real scientist – or is blessed with a functioning brain – she may stop you with one of those looks and say, “Frankly, my dear, this study Sacks.”