## Innumeracy at the Globe and Mail

In the June 23 print edition of the Globe and Mail (billed as “Canada’s National Newspaper”), there’s an article on data centres (“Hewers of wood, storers of data”), in which, on page B4, one can read the following:

Greenpeace recently released a report that said if the Internet were a country, it would be the fifth-largest consumer of energy, largely because of the massive data centres that run unseen in the background. The group estimated that the centres will use 1.9 billion kilowatt hours of electricity by 2020 — more than the amount currently used by Canada, France, Germany and Brazil combined. (The average US home uses 8,000 kilowatt hours a year.)

An exercise for the reader: How many logical fallacies, arithmetic errors, or contradictions of common knowledge can you find in this passage?

I haven’t tried to determine whether these fallacies originate in the (unidentified) Greenpeace report, or are original to the Globe and Mail.

Entry filed under: Society, Statistics, Statistics - Nontechnical.

• 1. Ken  |  2011-06-24 at 4:11 am

• 2. Radford Neal  |  2011-06-24 at 9:29 am

Thanks. From a quick glance, the relevant figures are on page 11, and are just taken from an earlier report. Their exact meaning is unclear without following that up, but one can see that the Globe and Mail article’s 1.9 billion figure was supposed to be 1,973 billion.

Not noticing that you’re off by a factor of 1000 is a pretty good indication that you shouldn’t be writing on a topic. Of course, maybe the error was introduced later by the copy editor.

• 3. jerzysblog  |  2011-06-27 at 12:15 pm

Could the factor-of-1000 issue be due to the fact that some countries use commas for decimal points and others use periods?

• 4. Radford Neal  |  2011-06-27 at 2:00 pm

I think that’s unlikely, since periods are the norm for decimal points in Canada (and the Greenpeace report uses a comma for 1,973 billion).

The real point though is not that the wrong figure was somehow obtained, but that it got by the reporter and editors without anyone realizing it was wrong by a factor of 1000. It’s common knowledge that electricity costs about \$0.10 per kilowatt-hour, so 1.9 billion is only \$190 million, which is not surprisingly high, be rather surprisingly low, being only about 0.0003% of world GDP. It’s also common knowledge that the total number of homes in Canada, France, Germany, and Brazil is well over 100 million, so at 8,000 kilowatt-hours a year per home (perhaps a bit less, if usage is less than in the US, but it’s the right order of magnitude), we get something more than 800 billion kilowatt-hours, which is considerably more than 1.9 billion.

Those are just the arithmetic errors. There’s also the fundamental stupidity (or intent to confuse?) of conveying the magnitude by saying that the Internet would rank fifth if it were a country. How does that help the reader understand the magnitude? You have to start by trying to figure out what the sixth-largest country is in energy consumption, which is impossible if you’re not sure whether the EU is counted as one country or many (and if many, how about the UK?). Wouldn’t just giving the percentage of world energy consumption used for the Internet be a lot more informative?

Then there’s the misleading device of comparing usage nine years from now (projected how?) with usage today.

Finally, the whole point of the article regarding the importance of energy consumption by data centres seems intuitively to be all wrong. I haven’t followed up the second Greenpeace report to see what they say, but I’m pretty sure that to the extent that the Internet is an important cause of energy usage, it’s the energy consumed by the personal computers used by the millions of people browsing the web that dominates.

• 5. Anonymous  |  2011-06-28 at 10:26 am

Might this “error” be simply explained by the dual usage of short and long scales in Canada: http://en.wikipedia.org/wiki/Long_and_short_scales#Canadian_usage

• 6. Radford Neal  |  2011-06-28 at 1:34 pm

I don’t think so. In Canadian English, “billion” means 1000 million. That’s certainly the convention at the Globe and Mail.
That’s certainly how the vast (really vast) majority of readers will interpret it

I think the most reasonable explanation for the original error (not the subsequent failure to see the error) is just that someone with poor eyesight mistook the comma in the small font used for a period.

• 7. Dan Freedman  |  2011-10-17 at 12:11 pm

I find a simliar innumeracy occurs in the US these days when discussing economic statistics. Here are two examples:

1) “GDP grew at 3.9% last quarter.” What is actually meant by these words is that GDP grew at an annualized rate of 3.9% last quarter, when compared with the same quarter from the previous year. But this is never mentioned. The difference between what is written and what is the meaning is stark.

2) “These cuts will result in FOUR HUNDRED BILLION dollars in savings over 10 years.” (I’ve emphasized the words that are usually stressed when expressing this sentence verbally). While this is indeed what is meant, one might ask the question: “How much is this over 1000 years?” (by which I really mean to say “why choose 10 years?” For decades, numbers were given as annualized, but there has been a collective decision to quote 10-year numbers in order to make costs or savings sound 10 times more impressive. If this were done universally and with open recognition, it would be fine. For example, the US GDP is about \$150 trillion dollars in size over 10 years. But that’s not what the politicians do. Instead, they mix and match without telling people. It is not uncommon to hear “The US budget deficit is \$1.8 Trillion in 2012, and the new health care system will shave \$980 Billion over 10 years.” It’s almost like mixing the imperial and metric systems together.

— Dan Freedman

• 8. Radford Neal  |  2011-10-17 at 1:52 pm

Hi Dan! Good to hear from you!

Your two examples are indeed regrettably common.

The GDP example is very frustrating, especially when the magnitude of the growth number is such that it’s not obvious whether or not they really meant the growth for that quarter alone (a number which is sometimes quoted). I think this practice became downright irresponsible during the last recession, when quarterly figures for contraction were multiplied by four, giving the impression that the recession was much more severe than it actually was. This hardly contributed to recovery. In that context the “annualized” figure is rather misleading even if correctly labelled, since annualizing the percentage change makes sense only if you think that change may continue at that rate, and it was pretty clear that economy wasn’t going to contract at the same rate for three more quarters.

Another perennial confusion is that between “inflation declined” and “prices declined”…

• 9. Mailing  |  2012-05-11 at 12:57 pm

It is true that the internet has grown to be one of the biggest consuming powers of energy. Nevertheless it helps to evolve the world a lot through instant messaging like mail, social media and other platforms

• 10. Nigel Goodwin  |  2012-05-19 at 10:25 am

A few comments. EU is not one country, this is not on anybody’s agenda (I hope).

UK is not a country, England is a country, but I admit that UK is the entity which should be used.

I dislike it when ‘annual’ usage is compared with ‘usage by 2020’. Do they mean the annual useage in 2020, or the total used between now and 2020? I assumed the latter, you assumed the former, it is not clear.