# statistics INSANITY

**URL:** <https://boards.straightdope.com/t/statistics-insanity/31492>\
**Category:** Factual Questions\
**Created:** [August 31, 2000, 2:39pm UTC](https://boards.straightdope.com/t/statistics-insanity/31492 "2000-08-31T14:39:58Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![zwaldd](https://avatars.discourse-cdn.com/v4/letter/z/a9adbd/32.png) [@zwaldd](https://boards.straightdope.com/u/zwaldd)\
**Post date:** [August 31, 2000, 2:39pm UTC](https://boards.straightdope.com/t/statistics-insanity/31492/1 "2000-08-31T14:39:58Z")

</div>

in this article: [http://www.motherjones.com/mother\_jones/MJ98/wellbeing.html](http://www.motherjones.com/mother_jones/MJ98/wellbeing.html), the statement is made that “melanoma rates are rising about 6 percent each year, and sunscreen sales are continuing to climb.” in another thread and forum, i made a comment about the relevance of a similar statement in a cecil column. i believe that without knowing how this compares to annual population growth, not to mention the location where the statistic applies (usa? world?), that statement is no more meaningful than an advertisement. sales of everything should be expected to climb annually, so the sunscreen comment is absolutely meaningless. the melanoma percent increase figure would only be meaningful if it was per capita, which it does not appear to be. my comments in the other thread were refuted, so can statisticians or the learned confirm or negate my theory?

---

<div class="post-metadata">

**Author:** ![LazarusLong42](https://avatars.discourse-cdn.com/v4/letter/l/87869e/32.png) [@LazarusLong42](https://boards.straightdope.com/u/LazarusLong42)\
**Post date:** [August 31, 2000, 3:06pm UTC](https://boards.straightdope.com/t/statistics-insanity/31492/2 "2000-08-31T15:06:52Z")

</div>

Disease rates, when pulbished, are almost _always_ given as _per capita_. In fact, I cannot think of an instance I’ve seen where they weren’t.

As to sunscreen sales, I’d bet that’s based on total dollars, and as you said is totally useless.

I’d lay money that both statistics are US-based.

LL

---

<div class="post-metadata">

**Author:** ![zwaldd](https://avatars.discourse-cdn.com/v4/letter/z/a9adbd/32.png) [@zwaldd](https://boards.straightdope.com/u/zwaldd)\
**Post date:** [August 31, 2000, 3:30pm UTC](https://boards.straightdope.com/t/statistics-insanity/31492/3 "2000-08-31T15:30:28Z")

</div>

> [@](#):
>
> Disease rates, when pulbished, are almost always given as per capita.

not being a statistician, i would take that to mean a 6% annual increase per capita means that if 1% of the population gets melanoma this year, 7% will get it next year, and depending on when the study began, we’ll all have melanoma in the near future. that doesn’t sound right, but i don’t think i’m doing the math correctly.  
a little help here?

---

<div class="post-metadata">

**Author:** ![Robot\_Arm](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/robot_arm/32/18280_2.png) [@Robot\_Arm](https://boards.straightdope.com/u/Robot_Arm)\
**Post date:** [August 31, 2000, 4:00pm UTC](https://boards.straightdope.com/t/statistics-insanity/31492/4 "2000-08-31T16:00:40Z")

</div>

I believe a “six percent increase per capita” goes like this.

In one year, 100 people per 100,000 get melanoma.  
The next year, 106 people per 100,000 get melanoma

(I’ve no idea if those numbers are even in the ballpark.)

zwaldd, that “six percent” increase is taken from the number of cases, not from the entire population.

---

<div class="post-metadata">

**Author:** ![zwaldd](https://avatars.discourse-cdn.com/v4/letter/z/a9adbd/32.png) [@zwaldd](https://boards.straightdope.com/u/zwaldd)\
**Post date:** [August 31, 2000, 4:16pm UTC](https://boards.straightdope.com/t/statistics-insanity/31492/5 "2000-08-31T16:16:23Z")

</div>

ok, got it. thanks for the info.
