« Criminal chart | Main | Picking up the right file »


Bob Dobalina

I've read this blog regularly for a few months, and I've never thought you to be an ideologue, but your effusive praise for Rabinoff's book (where he uncritically perpetuates the questionable science behind some* of the antismoking movement's claims) has given me pause.

* Of course not everything the antismoking movement claims is suspect.


I couldn't agree more on the percentage issue. The one that confuses me at present is the Euopean decision to: "increase the use of renewable energy to 20% by 2020" (taken from: http://www.eu2007.de/en/News/Press_Releases/March/0309BKBruessel.html)

What does that mean? Does that mean 20% of all energy consumed in all of Europe? 20% of electricity consumption/production? Is nuclear power included? Given that electricity production accounts for only 1/3 or so of all energy consumption, the difference would be substantial.

Anyway, whatever the mystery 20% actually is, it sounds like a good start.


I took it as meaning that of the people in that age bracket today, when they are all eventually dead, 25% will have died from smoking-related conditions.

That's a bit contradicted by the "dying in your prime" part, but supported by "will die".

In the UK ~20% of all deaths are smoking-related so ~25% of deaths of people now middle aged sounds about right.

On this site

there's a similar description:

"Half of all teenagers who are currently smoking will die from diseases caused by tobacco if they continue to smoke. One quarter will die after 70 years of age and one quarter before, with those dying before 70 losing on average 21 years of life. [3]"

I don't understand what you mean by
"Looking at the legend, the red areas were regions in which deaths from tobacco use accounted for over 25% of "total deaths among men and women over 35". This explained some, as perhaps there were more reasons to die (warfare, other diseases, mine accidents, etc.) in developing nations than in developed nations, or that they had larger populations (so more deaths even at lower rates)."

Wouldn't the percentage of smoking-related deaths of all deaths in men and women over 35 correct for differences in population size?

Two particular things you didn't mention as reasons for the developed/developing discrepancy:

- AIDS-related deaths
- that many smoking-related illnesses like CHD are also related to other aspects of Western lifestyles like lack of exercise and diet

John S.

Another possible explanation for the north/south difference: people in developing countries have less disposable income to spend on cigarettes.

"One quarter will die after 70 years of age and one quarter before... [3]"

So does the other half live forever? I've gotta take up smoking.


Rose: putting smoking-related deaths over all deaths does correct some for population size; that's why I qualified by saying "more deaths even at lower rates"... I'd be less surprised if it were the case that overall more smokers died in developing nations but not as a percentage of total deaths.

Like the nameless commentator, I'm mystified by the "other half".

One other possibility is that the two statistics (number of smoking-related deaths over all deaths; number of predicted smoking-related deaths over all living) are unrelated, it just happened that 25% was a good cut-off, and the spatial correlation was accidental. This view is somewhat supported by the women's map which had no red (25%+) area.


Wouldn't the percentage of smoking-related deaths of all deaths in men and women over 35 correct for differences in population size?

The comments to this entry are closed.


Link to Principal Analytics Prep

See our curriculum, instructors. Apply.
Kaiser Fung. Business analytics and data visualization expert. Author and Speaker.
Visit my website. Follow my Twitter. See my articles at Daily Beast, 538, HBR.

See my Youtube and Flickr.

Book Blog

Link to junkcharts

Graphics design by Amanda Lee

The Read

Keep in Touch

follow me on Twitter