| Author |
Message |
Kubo
Joined: Aug 24 2005
Location: Mount Holly, NJ
Posts: 1062
|
Allow me to add a little bit of my own science into the conversation...
It seems like we want to know if there REALLY is a difference being made by the new vaccine. A t-test should be able to test that there is a difference or not.
t = (difference in means, on the basis that the value of a person's HIV status is 1 if they got it, 0 if they didn't) / (standard deviation of the mean score of getting HIV for the test group minus the mean score of getting HIV for the control group)
Mean for the test group: .006375
Mean for the control group: . .00925
Difference in variances: .00197
t = (.006375 - .00925) / .00197 = about -1.47
T-distribution: Critical value of -1.47 with 7999 (same as infinity) degrees of freedom = 0.075
So what this means, for those that don't dig statistics, is that the likelihood that the vaccine produced these results out of PURE CHANCE is less than 7.5%. Put another way, there is a 92.5% chance that the vaccine is the reason that people in the experimental condition (got the vaccine) contracted HIV at a smaller rate than those that received the placebo. This is provided the individuals in the study were given the vaccine or the placebo at random. This is probably the case, since the likelihood that people in either condition would have some inherent similarity across an 8,000 person sample is pretty slim.
Even with large sample sizes and small effects, that's pretty fucking good. Social science looks for 95% certainty, and medical science looks for 99% certainty, but depending on how much wiggle room you are comfortable with, it looks like the vaccine is definitely doing what it's supposed to be. It's at least 92.5% likely that that is the case anyway.[/u]
|
 Thou, because I am wroth, be not dismayed, for I shall win the strife, whoever circle round within for the defence. This their insolence is not new, for of old they used it at a less secret gate, which still is found without a bolt. Above it thou didst see the dead inscription; and already on this side of it
descends the steep, passing without escort through the circles,
One such that by him the city shall be opened to us. |
|
    |
|
Douche McCallister
Moderator
Title: DOO-SHAY
Joined: Jan 26 2007
Location: Private Areas
Posts: 5672
|
God I hated statistics...but atleast I completely understood all of your calculations. However I still feel it may be more likely that chance could have a larger factor then it actually looks. I mean hell the Lions won today! Chance is always present.
|
|
|
   |
|
FNJ
2010 SLF Tag Champ
Joined: Jun 07 2006
Posts: 12294
|
hey guys, remember when reed richards used statistics and algorithms to come to the conclusion that superhero registration was a good idea, and it led to the civil war?
|
|
|
  |
|
Optimist With Doubts
Title: Titlating
Joined: Dec 17 2007
Posts: 5042
|
| FNJ wrote: |
| hey guys, remember when reed richards used statistics and algorithms to come to the conclusion that superhero registration was a good idea, and it led to the civil war? |
Jew, I think we need to have a convo about real life vs comics.
|
|
|
    |
|
Dr. Jeebus
Moderator
Title: SLF Harbinger of Death
Joined: Sep 03 2005
Location: Wakefield, MA
Posts: 5228
|
| FNJ wrote: |
| hey guys, remember when reed richards used statistics and algorithms to come to the conclusion that superhero registration was a good idea, and it led to the civil war? |
Yeah, but Stark and SKRULL Pym were helping him.
|
dr.jeebus.sydlexia.com - Updated sometimes, but on hiatus!
| UsaSatsui wrote: |
| The three greatest heels in history...Andy Kaufman, Triple H, and Dr. Jeebus |
|
|
     |
|
Kubo
Joined: Aug 24 2005
Location: Mount Holly, NJ
Posts: 1062
|
| Douche McCallister wrote: |
| God I hated statistics...but atleast I completely understood all of your calculations. However I still feel it may be more likely that chance could have a larger factor then it actually looks. I mean hell the Lions won today! Chance is always present. |
I hate statistics too. I just have to learn them to do the work that I do, unfortunately. Chance is accounted for here, especially due to the large sample size. There's an issue of statistical power here too (the confidence you can have in your statistical findings), and based on a sample size of 16,000, its like 99.99%. Now what MIGHT be fudging the findings is some sort of covariate that wasn't accounted for. Like if one group DID actually have some inherent quality (less likely to go pick up hookers, for example) that differentiated them from the other group. But as far as random chance goes, the 7.5% is (supposedly) mathematical certainty if mediators or moderators are present.
|
 Thou, because I am wroth, be not dismayed, for I shall win the strife, whoever circle round within for the defence. This their insolence is not new, for of old they used it at a less secret gate, which still is found without a bolt. Above it thou didst see the dead inscription; and already on this side of it
descends the steep, passing without escort through the circles,
One such that by him the city shall be opened to us. |
|
    |
|
SoldierHawk
Moderator
Title: Warrior-Poet
Joined: Jan 15 2009
Location: San Diego, CA
Posts: 6113
|
Well and remember the point of this whole story isn't that they've found a cure or anything. The point is that up until now they have had ZERO success with ANYTHING they have ever done when it comes to fighting this virus. That this shows even the smallest hint of being useful IS the huge breakthrough, and will hopefully be the foothold we need to eventually finding a way to combat it for real.
|
| William Shakespeare wrote: |
| Love all, trust a few, do wrong to none. |
|
|
    |
|
Pandajuice
Title: The Power of Grayskull
Joined: Oct 30 2008
Location: US and UK
Posts: 2649
|
| SoldierHawk wrote: |
| Well and remember the point of this whole story isn't that they've found a cure or anything. The point is that up until now they have had ZERO success with ANYTHING they have ever done when it comes to fighting this virus. |
SH, as was said before, that statement is far from true. There have been numerous breakthroughs in the fight against AIDS over the last 15 years. Haven't you heard of the "triple cocktail" (discovered in 1996) that's successfully treated a lot of people, and all the experimental stuff that's been working to keep Magic Johnson alive?
There's been a ton of progress on HIV and AIDS treatments to the point where now, it's not nearly the automatic death sentence it was in the 80s.
|
|
|
  |
|
FNJ
2010 SLF Tag Champ
Joined: Jun 07 2006
Posts: 12294
|
| Dr. Jeebus wrote: |
| FNJ wrote: |
| hey guys, remember when reed richards used statistics and algorithms to come to the conclusion that superhero registration was a good idea, and it led to the civil war? |
Yeah, but Stark and SKRULL Pym were helping him. |
you know, I stopped reading before secret invasion, but pym being a skrull kinda irks me now that you've told me.
fucking skrulls. they fuck everything up.
|
|
|
  |
|
Lady_Satine
Title: Head of Lexian R&D
Joined: Oct 15 2005
Location: Metro area, Georgia
Posts: 7287
|
THURSDAY, July 8 (HealthDay News) -- Scientists report they've discovered possible new weapons in the war against HIV: antibody "soldiers" in the immune system that might prevent the AIDS virus from invading human cells.
According to the researchers, these newly found antibodies connect with and neutralize more than 90 percent of a group of HIV-1 strains, involving all major genetic subtypes of the virus.
That breadth of activity could potentially move research closer toward development of an HIV vaccine, although that goal still remains years away, at best, experts say.
The findings "show that the immune system can make very potent antibodies against HIV," said Dr. John Mascola, a vaccine researcher and co-author of two new studies published online July 8 in the journal Science.
"We are trying to understand why they exist in some patients and not others. That will help us in the vaccine design process," said Mascola.
Antibodies are warriors in the body's immune system that work to prevent infection. "Neutralizing" antibodies bind to germs and try to disable them, explained Ralph Pantophlet, an immunologist and assistant professor at Simon Fraser University in Burnaby, British Columbia, Canada.
With HIV, the antibodies are in a continual race to adjust to the virus, which evolves to escape detection. "The reason the antibodies generally do not work so well is because they're always playing catch up," said Pantophlet, who is familiar with the findings of the new studies.
However, some people's antibodies are known to cope especially well with HIV, although even these rare patients can't get rid of the virus entirely, Pantophlet said.
In the new studies, researchers report on three antibodies that appear to have major powers to fight off HIV. In a sense, the antibodies gum up a lock that the virus tries to pick to get into healthy cells, said Mascola, deputy director of the Vaccine Research Center at the U.S. National Institutes of Allergy and Infectious Diseases.
However, making antibodies in large enough quantities to boost the immune system remains a challenge, said Pantophlet.
While researchers haven't given up on that prospect, some think it's more feasible to use the new findings as another avenue to an AIDS vaccine. The idea would be to teach the body to produce the antibodies so the person is protected when exposed to the virus, Mascola said.
But that won't happen for some time, if at all. "Developing a vaccine always takes a fairly long period of research with some trial and error," Mascola said.
"The goal is to vaccinate individuals and have their own immune systems make an antibody like this," he said. "To do that, we have to design a new vaccine, study it first in animal models, and then try it in small scale human studies, and see if it does what we expect it to do. That takes a quite a bit of time and effort."
|
 "Life is a waste of time. Time is a waste of life. Get wasted all the time, and you'll have the time of your life!" |
|
   |
|
Drew Linky
Wizard
Joined: Jun 12 2009
Posts: 4209
|
So... in other words, this cure is--instead of being several miles closer to the goal--more like ONE mile. Still, I have to admit, this whole business has given me confidence in it. GO SCIENCE!
|
https://discord.gg/homestuck is where you can find me literally 99% of the time. Stop on by if you feel like it, we're a nice crowd. |
|
   |
|
sidewaydriver
2010 SLF Tag Champ
Title: ( ͡� 
Joined: May 11 2008
Posts: 6160
|
Fuck it, no more condoms for me, this shit's as good as cured.
|
 Shake it, Quake it, Space Kaboom. |
|
  |
|
|
|
|