Showing posts with label Numbers. Show all posts
Showing posts with label Numbers. Show all posts

Does “Repeat 3 Times” Mean to Do It 4 Times Total, or 3?

Technical writer Paul Stregevsky clarifies an ambiguity you almost certainly never considered:

“Repeat three times.”

So, do it four times, right? Literally, yes. But as often as not, the writer meant, “Perform three times” — in other words, “Repeat twice.”

percentage. percentage points.

A percentage increase is not the same as a percentage point increase.

If a number increases from 20% to 22%, that’s a difference of two percentage points. It’s not two percent.

Use percentage to describe the difference between two whole numbers. For example:

“Sales jumped from $90K to $150K, or 66%.”

Use percentage points to describe the difference between two percentages:

“Trump’s approval rating has dropped 10 percentage points, from 40% to 30%.”

Derek Thompson Shows How to Contextualize a Statistic

One of my favorite writers, Derek Thompson, of the Atlantic, does a superb job of bringing clarity to a statistic that most readers would otherwise skip right over. Here’s Derek:

“In June, researchers from N.Y.U., Stanford, and Microsoft published a paper with a title that made their position on the matter unambiguous: ‘Digital Addiction.’ In closing, they reported that ‘self-control problems cause 31% of social media use.’ Think about that: About one in three minutes spent on social media is time we neither hoped to use beforehand nor feel good about in retrospect.”

There’s a lot to like here:

1. The transition, “Think about that.” (Most of us would use a cliché such as “in others words.”)

2. The conversion of a percentage (31%) into a fraction (one in three). (This is one of the tips I teach in my workshop on humanizing big numbers.)

3. The vivid and concrete translation from “self-control problems” to “time we neither hoped to use beforehand nor feel good about in retrospect.”

Finally, Derek packed all these tricks into a single sentence. Well-done!

When Being Inexact Is Perfectly Correct

Two years ago, the New York Times reported that on his 1996 income-tax returns, Donald Trump declared a $916 million loss.

When discussing this fact, most of us round “$916 million” up to “$1 billion.”

That’s both appropriate and advisable. In short, “one billion” is more memorable than “916 million.” Indeed, “a billion bucks” is not only easier to recall; it’s also easier to articulate.

Note: You could also round down, to $900 million. But most people, as a matter of tradition, tend to round up.

Postscript: One of my students points out an important nuance: While the headline should cite “$1B,” the article itself should use the actual number.

Which Sentence Is Better?

1. A DHS report estimates that over 40% of illegal immigrants entering the U.S. since 2004 have traveled through the Mona Pass.

2. Between 2004 and 2010, more than 4 in 10 illegal immigrants traveled through the Mona Pass.

Leave your answer and explanation in the comments section.

How Do You Concretize 37 Grams of Fat?


A memorable story from Chip and Dan Heath’s book, Made to Stick: Why Some Ideas Survive and Others Die (2007):

Art Silverman stared at a bag of movie popcorn. It looked out of place sitting on his desk. His office had long since filled up with fake-butter fumes. Silverman knew, because of his organization’s research, that the popcorn on his desk was unhealthy. Shockingly unhealthy, in fact. His job was to figure out a way to communicate this message to the unsuspecting moviegoers of America.

Silverman worked for the Center for Science in the Public Interest (CSPI), a nonprofit group that educates the public about nutrition. The CSPI sent bags of movie popcorn from a dozen theaters in three major cities to a lab for nutritional analysis. The results surprised everyone.

The United States Department of Agriculture (USDA) recommends that a normal diet contain no more than 20 grams of saturated fat each day. According to the lab results, the typical bag of popcorn had 37 grams.

The culprit was coconut oil, which theaters used to pop their popcorn. Coconut oil had some big advantages over other oils. It gave the popcorn a nice, silky texture, and released a more pleasant and natural aroma than the alternative oils. Unfortunately, as the lab results showed, coconut oil was also brimming with saturated fat.

The single serving of popcorn on Silverman’s desk—a snack someone might scarf down between meals—had nearly two days’ worth of saturated fat. And those 37 grams of saturated fat were packed into a medium-sized serving of popcorn. No doubt a decent-sized bucket could have cleared triple digits.

The challenge, Silverman realized, was that few people know what “37 grams of saturated fat” means. Most of us don’t memorize the USDA’s daily nutrition recommendations. Is 37 grams good or bad? And even if we have an intuition that it’s bad, we’d wonder if it was “bad bad” (like cigarettes) or “normal bad” (like a cookie or a milk shake).

Even the phrase “37 grams of saturated fat” by itself was enough to cause most people’s eyes to glaze over. “Saturated fat has zero appeal,” Silverman says. “It’s dry, it’s academic, who cares?”

Silverman could have created some kind of visual comparison— perhaps an advertisement comparing the amount of saturated fat in the popcorn with the USDA’s recommended daily allowance. Think of a bar graph, with one of the bars stretching twice as high as the other.

But that was too scientific somehow. Too rational. The amount of fat in this popcorn was, in some sense, not rational. It was ludicrous. The CSPI needed a way to shape the message in a way that fully communicated this ludicrousness.

Silverman came up with a solution.

CSPI called a press conference on September 27, 1992. Here’s the message it presented: “A medium-sized ‘butter’ popcorn at a typical neighborhood movie theater contains more artery-clogging fat than a bacon-and-eggs breakfast, a Big Mac and fries for lunch, and a steak dinner with all the trimmings—combined!”

The folks at CSPI didn’t neglect the visuals—they laid out the full buffet of greasy food for the television cameras. An entire day’s worth of unhealthy eating, displayed on a table. All that saturated fat—stuffed into a single bag of popcorn.

The story was an immediate sensation, featured on CBS, NBC, ABC, and CNN. It made the front pages of USA Today, the Los Angeles Times, and the Washington Post’s Style section. Leno and Letterman cracked jokes about fat-soaked popcorn, and headline writers trotted out some doozies: “Popcorn Gets an ‘R’ Rating,” “Lights, Action, Cholesterol!” “Theater Popcorn Is Double Feature of Fat.”

The idea stuck. Moviegoers, repulsed by these findings, avoided popcorn in droves. Sales plunged. The service staff at movie houses grew accustomed to fielding questions about whether the popcorn was popped in the “bad” oil. Soon after, most of the nation’s biggest theater chains—including United Artists, AMC, and Loews—announced that they would stop using coconut oil.

You Can’t Spell “Numbers” Without “Numb”

Mike Long is one of the best writers I have the pleasure of reading. His latest e-newsletter is worth reprinting in full.

Numbers Don’t Mean Anything Unless You Put Them in Human Terms

The cold precision of numbers is, to most readers, just that: cold. The vast majority of people don’t like numbers, don’t understand numbers, aren’t good with numbers, and wear their number-phobia with pride.

But as writers, the number problem we need to deal with most often isn’t the reader’s inability to divide the check three ways at dinner. The real problem is not knowing what numbers amount to in real-world terms. The United States federal budget is about $4 trillion. That sounds like a big number, but how big? The average reader has literally no idea.

How many thousands of dollars are in a million? How many millions are in a billion? How many billions in a trillion? Everybody’s lost. When we write or say $4 trillion, it impresses only because of the intimidating feeling we associate with big numbers. A good writer knows this and addresses it.

What’s the fix? Put big numbers in human terms.

For instance, a little division produces this nugget: $4 trillion in spending amounts to spending about $11 billion a day, every day. That’s a little easier to imagine, but few people really know how much a billion is, either. Dig deeper.

$4 trillion a year is about $500 million an hour. That’s more familiar, but not much. Keep going.

It’s also $7.5 million a minute — closer, but eh. I don’t really know what $7.5 million is like, do you?

But if you do the division one more time, we strike gold: $4 trillion comes down to $127,000 in spending every second. That’s a number I “get.”

If I tell you the government spends $4 trillion a year, well, you’re impressed (or worried) in a vague sort of way. But if I tell you that the government spends $127,000 a second every hour… of every day… of every year — now I have your attention.

And the real-world comparisons start writing themselves:

In one second, the federal government spends what two typical American families earn in an entire year.

Every two seconds, the federal government spends enough money to buy the average house—that’s your 30-year mortgage paid off in less time than it takes to breathe in.

In the time it takes you to read this sentence out loud, the federal government spent enough money to buy about half the homes on your block.

Every six weeks, the federal government spends enough money to buy Apple, the most valuable public company in America.

Five more weeks of spending and it could also buy ExxonMobil.

Another five and they — rather, you — have paid for Google, too.

Get the idea?

Cranking down numbers into familiar comparisons elicits something far more valuable than the typical “gee whiz” you’ve been settling for.

(By the way, it’s not hard to do the math. Just type it into your calculator, or even into Google: 4 trillion divided by 365 — that gives you dollars per day. Divide that by 24 to get dollars per hour. Divide that by 60 to get dollars per minute. See?)

Your writing has value only if you are providing something the reader does not know. So do some homework—put in effort that the reader has not or will not. Otherwise you’re just spouting uninformed opinion and if you wanted that, you could call the average “friend” on Facebook.

$127,000 a second — that’s one they’ll remember. Stop settling for vaguely scary numbers. Start comparing numbers to things everybody understands. There’s such a payoff.

Addendum (12/8/2013): Here's a good video from BuzzFeed that contextualizes 12 big numbers about Amazon:



Addendum (8/19/2014): Let’s keep this thread going. Here’s another example, from Derek Thompson:

1. According to a McKinsey Global Institute paper, email consumes an average of 13 hours per week.

2. Thus, email consumes 28% of the average workweek.

3. Thus, for every hour of work, we spend 17 minutes on email.

4. The typical “knowledge” worker — that is, somebody whose professional output is creative — earns $75,000 a year. Thus, the time spent on reading and answering email costs a company $20,990 per worker per year. Hence Thompson’s headline:

Typical White-Collar Worker Is Paid $21,000 a Year to Be on Email