🚀 go-pugleaf

LeRoy Blue <leroyblue@billon.net> writes:

> On Sun, 21 Dec 2008 22:04:27 -0800, nobody@nospam.pacbell.net (Bill
> Z.) wrote:
>
>>LeRoy Blue <leroyblue@billon.net> writes:
>>
>>> On Sat, 20 Dec 2008 23:17:32 -0800, nobody@nospam.pacbell.net (Bill
>>> Z.) wrote:
>>>
>>>>LeRoy Blue <leroyblue@billon.net> writes:
>>>>
>>>>No, I think you lie - you made up that 70 percent thing and the
>>>>CDC data simply is not consistent with it - the actual numbers
>>>>are lower.
>
>>No it isn't - you get that sort of error when you have around 1000
>>cases as it follows a Poisson distribution, and the standard deviation
>>(the statistical error) is the square root of that number. For a mean
>>of 500,000 cases, the statistical error (which measures the random
>>change from year to year) is 0.14 percent.
>
>
> You lesson in homosexual DOGMA mathematics is as flawed as your
> fundamental beliefs about how good and how pleasant homosexuality is.
> The 70% figure stands.

Nope - you are lying.  First, what you call "homosexual DOGMA mathematics"
is simple statistics: A Poisson distribution occurs when events occur
with a fixed average rate and independently of the time of previous
events. See <http://en.wikipedia.org/wiki/Poisson_distribution>. It
describes radioactive decay, the arrival of telephone calls at a central
office from subscribers using that office, and in many other contexts.

Your 70 percent figure does not "stand" - the citation you gave gives
a number of about 68 percent and the error is too small to round up to
70%.  Further, tht 68 percent is an overestimate - it assumes that all
individuals in the "male-to-male sexual contact & IV drug use"
catagory were infected via a sex act.  Obviously that is not the case
because the rate for IV drug use alone is not zero.

>>You can't find a citation but "know" anyway?  And you expect us to
>>believe you?
>
> Believe this Gump:
> Men
>
> http://www.cdcnpin.org/scripts/population/men.asp
>
> In the United States, HIV-related illness and death historically have
> had a tremendous impact on men who have sex with men (MSM). MSM
> accounted for 70% of all estimated HIV infections (including MSM who
> are also injection drug users) among male adults and adolescents in
> 2004 <snip>

Try again - we were talking about the cumulative number of cases and
that is a different number.  You quoted a particular URL and the data
in that URL gave a figure of 68 percent.  You were criticized simply
because the data did not match what you claimed.

The value per year can go up and down for a variety of reasons besides
statistical fluctuations. In the time frame you are now using (2004
and in a later post, 2005) a new HIV test that provides very quick
results was being introduced as part of a demonstration project,
<http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5647a2.htm?s_cid=mm5647a2_e>.
In particular, the CDC introduced the tests "venues where persons at
high risk congregated, resided, or sought medical care."  They
introduce a test that catches infections earlier than those infections
would otherwise be caught, and the rate of diagnoses goes up
temporarily as a result.