Tuesday, August 6, 2013

Gallup drinking preference poll is misleading: Beer still far outsells wine

Gallup released its annual poll of American drinking habits this week, and it's more good news for wine: Americans are about as likely to say they drink wine as beer.

Americans are full of shit. I'll show you.

First, here's the Gallup preference chart

It's interesting data: Gallup has been asking people some of the same questions since World War II. As you can see, 47% of Americans said beer was their favorite alcoholic beverage in 1992, and only 27% chose wine. Now it's nearly tied, 36-35.

Except it's not. The above is what Americans say when pollsters ask them what they like. Below is what they actually buy.

From the Wine & Spirits Wholesalers Convention, Apr. 2013
As you can see, this is not close at all. Americans drink more than 10 times as much beer, by volume, as wine. Wine costs more than beer, but even considering that, Americans spend more than 3 times as much money on beer as on wine. Heck, they spend more than twice as much money on spirits as on wine.

This is not to say the Gallup poll doesn't matter. It captures a snapshot of people's opinion of wine, which is generally positive. But as a country, we lie to pollsters, especially when they ask about what we eat and drink.

Example: We tell McDonald's that we want salads, and then we don't buy them. We say we're eating healthier as a nation, but in fact we're eating bacon-filled tater tots.

I could get more into the Gallup data, particularly the fact that women report themselves as much bigger fans of wine than men. Maybe that's true, and it's worth knowing for people trying to sell wine.

But for this post, let's stick to the facts. Americans may say we prefer wine, but we're still buying 10 times as much beer.

Follow me on Twitter: @wblakegray and like The Gray Report on Facebook.


Unknown said...

It seems that Gallup is counting people while you are counting beverages. Isn't possible that beer drinkers drink a lot more beer than wine drinkers drink wine, which would mean that no one necessarily lied?

W. Blake Gray said...

Kim: It's possible. But 10 times as much? It doesn't seem likely. Polls about what people eat have been well-established to be different from what they actually eat. I could have listed a bunch of examples; they're out there.

Robert said...

The poll is a great example of hubris. You can also notice that the american consumer drinks more liquor than wine also, which is why larger wine companies are diversifying their portfolios to include more spirits to offer to consumers. Wine-er is finer but liquor is quicker, and more profitable.

Unknown said...

You're right, 10 times is a stretch.

Dan Fishman said...

There is no reasons anyone needed to lie to create these data.

For one thing, beer has about 1/2 to 1/3 the alcohol of wine per serving (especially the most popular brands like bud/coors light, which are 1/3 or even less). Right there that cuts it down to about 4x, not 10x, by volume.

Then you could have a group of people that drink beer and only beer (approaching 36%). Then you have two other groups of people who prefer spirits or wine, but who also drink beer, contributing to the overall volume of beer sold as well.

Plus, I would guess that the vast majority of underage drinkers are buying cheap beer by the case or keg, which goes into volume sold but not the survey (assuming they only call people over 21). Judging by what I have seen on many college campuses, that adds up to a lot of beer.

I don't think it takes a stretch to see how a combination of beer drinkers drinking more (and only beer), with other drinkers drinking less and occasionally having beer could result in those preferences found by Gallup, while still seeing the sales by volume favoring beer.


Unknown said...

I have long complained about the press treatments of this poll. Back in 2005, when the wine preference exceeded that of beer, I remember seeing headlines which read something like, “Americans drink more beer than wine.” I recall the question having been worded differently in the past, as well. Instead of, “Do you most often drink liquor, wine or beer?”, which incorporates an element of quantity, it was something like, “What is your alcoholic beverage of choice?”, which does not. One can prefer wine without actually drinking much of it due to cost, peer pressure, etc.

I don’t doubt that some respondents where less than honest, thinking wine is the more sophisticated answer. However, I can believe that beer drinkers drink 10 more beer than wine drinkers drink wine. For many wine drinkers, one or two glasses on the weekend is all they drink. Most of the beer drinkers I know drink beer every day and at least 10 times more beer on the weekend than the quantity of wine I drink.

Unknown said...

Oops! That headline should have read, "Americans drink more wine than beer."

W. Blake Gray said...

Dan: I'd be very surprised if, for all the political fuss sometimes made about it, underaged drinking represents even 1% of the alcohol consumed in America.

It would be interesting to see a survey. But take a serious look around the next time you're in a place that sells lots of beer, wine and spirits. Stand there for an hour and watch purchasers. Calculate how many of them might either be underage or planning to turn that booze over to underage drinkers. It's the proverbial drop in a bucket.

Dan Fishman said...


You didn't really address any of my main points, the most important of which is that it is misleading to look at consumption in terms of volume, as opposed to in terms of beverages.

But if you want to look at college students, it has been a while, but I am not that far away from being 20 that I have forgotten that the vast majority of the alcohol we drank was on private property and purchased by someone else.

According to this study, 31% of College students meet the criteria for alcohol abuse. With 21.6 million college students, that's about 7 million alcohol abusers. Certainly those students, most aged 17-20, account for much more than 1% of drinks consumed.

Knight JR, Wechsler H, Kuo M, Seibring M, Weitzman ER, Schuckit M. Alcohol abuse and dependence among U.S. college students. Journal of Studies on Alcohol, 2002


W. Blake Gray said...

Dan: As interesting as it is, the student alcohol abuse survey is another survey. Sales figures are sales figures.

I dunno man, if you want to believe that the opinion survey is an accurate representation of people's drinking habits, be my guest.

Dan Fishman said...

Obviously it is true that more beer is consumed in the US than wine. What is not necessarily true is that "Americans are full of shit." or that "we lie to pollsters".

The survey data and the consumption data are not mutually exclusive. The claim that Americans drink 10x more beer than wine is misleading. All in all, it's an interesting issue, but the combination of inflammatory language and careless data assessment in this post does more to muddle the picture than advance the discussion.


Erika Szymanski said...

I actually have to weigh in on the side of Johannsen and Benson and Anon here. Living in a town dominated by a university that routinely makes Playboy's top ten party school list, and being in a position to observe both folks who sip wine and folks who chug beer, the volume-preference breakdown seems plausible. Of course, Pullman isn't a representative sample, and I don't have numbers, but let's just say that I wouldn't be surprised. Interestingly, the grocery store closest to campus devotes nearly as much retail space to wine as it does to beer, but unscientific checkout observations show occasional purchases of single bottles of wine (yes, often by older non-undergrads) and frequent purchases of entire shopping carts of cheap beer by what can most politely be described as frat boys. In other words, customer-by-customer, just as many people might prefer wine to beer, and the store obviously values wine-buying customers, but beer easily wins on volume (and profit, I'd imagine).

Jon Bjork said...

To me, this is still good news. In the same way that Americans talk dry, but drink sweet, perhaps they are lying, thinking that it is cool to say they're drinking wine instead of beer.