Thread: "What The Bleep Do We Know?"

Originally Posted by thegnome54

No, there really isn't

How would you go about doing this?

I haven't been following the context of this thread completely. You would look at a graph which displays scores against frequency, and examine the shape, or skewness of the dataset relative to the position of the mean.

Originally Posted by really

I haven't been following the context of this thread completely. You would look at a graph which displays scores against frequency, and examine the shape, or skewness of the dataset relative to the position of the mean.

OH!

I wasn't using the term 'skewed' as a statistical description of the shape of the data, I was referring the effects of a possible bias inherent in the experimental design (it's not blinded). My bad.

I meant to say "You can't tell if a data set is biased just by looking at it".

Originally Posted by thegnome54

OH!

I wasn't using the term 'skewed' as a statistical description of the shape of the data, I was referring the effects of a possible bias inherent in the experimental design (it's not blinded). My bad.

I meant to say "You can't tell if a data set is biased just by looking at it".

Okay.

You can tell the way data can be biased by the way it is collected though.

Originally Posted by really

Okay.

You can tell the way data can be biased by the way it is collected though.

Precisely! I was trying to say that it's possible for little nuances in the way such data is collected to massively influence the results - these influences wouldn't be apparent just by looking at the data.

You just have to have a very watertight experimental design for these sorts of things.

Yeah with that water thing the guy who did it actually said that he simply selected those pictures which seemed to reflect the thought being directed at them and discarded all the others. So way to go Jeff, the water thing just shows that such things are easily very biased and that this documentary was a pile of bull.

"Classical computers do not have the ability to generate true random numbers. The random number generators on today’s computers are pseudo-random generators—there is always a cycle or a trend, however subtle. Pseudo-random generators cannot simulate natural random processes accurately for some applications, and can not reproduce certain random effects. Quantum computers can generate true randomness, thus give more veracity to programs that need true randomness in their processing. Randomness plays a significant part of applications with a heavy reliance on statistical approaches, for simulations, for code making, randomized algorithms for problems solving, and for stock market predictions, to name a few."

Thought this might be relevant in the arguments above regarding data and stuff (and the shuffle feature on the ipod). Excerpt was taken from this article.

Any decent random electronic number generator will use a pattern which is far too complex for the human mind to deduce and extrapolate, however. I think it's kind of irrelevant.

There doesn't seem to be any significant evidence for people influencing random number generators, and if there was there would be a field day. Scientists do not conspire to hide controversial evidence; quite the opposite in fact. But this and the fact that there is no possible mechanism currently apparent by which the positions of the diodes in the microchip are detected by the human brain lead to the inevitable conclusion that it's ridiculous, frankly.