For first time, a vaccine helps prevent HIV

[Josquius]

How do they test this? They give 16,000 the vaccine then tell them to go fuck around and report back?
Sounds very....not the done thing.

If you read the article, you'd find out.

[q

Not exactly proof still, just correlation.

[Razgovory]

That really doesn't sound very effective.

[Barrister]

How do they test this? They give 16,000 the vaccine then tell them to go fuck around and report back?
Sounds very....not the done thing.

If you read the article, you'd find out.

[q

Not exactly proof still, just correlation.

That's all you can do in medical research though.

[Josquius]

That's all you can do in medical research though.

Semi-true.
True you can never know 100% but this is a very vague study. Over a period of years and living their normal life otherwise...

[Neil]

This has the potential to be disastrous to the gay community.

[Barrister]

That's all you can do in medical research though.

Semi-true.
True you can never know 100% but this is a very vague study. Over a period of years and living their normal life otherwise...

The only other option is to deliberately infect people with HIV.  No ethical researcher would do that.

31% difference is enough to be statistically significant, so it's important.

[Viking]

This has the potential to be disastrous to the gay community.

you mean it might survive? 

[Josquius]

The only other option is to deliberately infect people with HIV.  No ethical researcher would do that.

Yeah, that's where my how on earth do they do this comes from.

[DGuller]

Not exactly proof still, just correlation.

That's all you can do in medical research though.

I don't think that's actually just correlation.  Correlation is looking at a bunch of data and seeing a pattern in relationship between some variables.  Acting in some way, and then seeing whether that action has statistically significant results, is actually a way to prove causation.

[DGuller]

31% difference is enough to be statistically significant, so it's important.

No, not necessarily, I can easily think of a scenario where it isn't.  However, any competent researcher would use an appropriate statistical test, and would make sure that the result is indeed statistically significant before claiming that it is.

[Barrister]

31% difference is enough to be statistically significant, so it's important.

No, not necessarily, I can easily think of a scenario where it isn't.  However, any competent researcher would use an appropriate statistical test, and would make sure that the result is indeed statistically significant before claiming that it is.

Yeah, I simplified my statement a lot.

[Admiral Yi]

No, not necessarily, I can easily think of a scenario where it isn't.  However, any competent researcher would use an appropriate statistical test, and would make sure that the result is indeed statistically significant before claiming that it is.

With a sample size of 16,000 virtually any difference is going to be statistically significant.

[DGuller]

With a sample size of 16,000 virtually any difference is going to be statistically significant.

Not necessarily.  The significance of this difference is actually pretty borderline.  If it were 55 instead of 51 infected on the vaccine, the result wouldn't be significant on 95% confidence level.

[garbon]

This has the potential to be disastrous to the gay community.

It'll be like a throwback!

[Viking]

This has the potential to be disastrous to the gay community.

It'll be like a throwback!

good god, don't bring back Studio 54