World is Falling Apart Thread (Locked forever)

[Pulsar_t]

[MrPopo]

We had a thread last year called, "the pseudoscience thread" wherein I caught a bit of flack for asserting that I don't believe in statistics.
http://www.racketboy.com/forum/viewtopi ... 2&start=35
People who work in the field of social science hate it when I say that.

I think it's really anyone with any scientific fluency who hates when you say that, not just the social scientists.

Yep. Every time I say it someone implies that I don't understand science just because I don't have any faith at all in statistics. I have plenty of scientific fluency thank you very much. I also have enough skepticism to know better than to believe statistics.

Don't go researching quantum physics; you won't like what you find.

[samsonlonghair]

We had a thread last year called, "the pseudoscience thread" wherein I caught a bit of flack for asserting that I don't believe in statistics.
http://www.racketboy.com/forum/viewtopi ... 2&start=35
People who work in the field of social science hate it when I say that.

I think it's really anyone with any scientific fluency who hates when you say that, not just the social scientists.

Yep. Every time I say it someone implies that I don't understand science just because I don't have any faith at all in statistics. I have plenty of scientific fluency thank you very much. I also have enough skepticism to know better than to believe statistics.

Don't go researching quantum physics; you won't like what you find.

Funny you should mention that. I was thinking about Heisenberg yesterday while I was observing a murmuration of starlings.

I understand science. I'm not one of those nuts who refuses to get vaccinated and denies climate change. I'm just skeptical of any data presented in the form of statistics because I understand how easy it is to manipulate statistics.

[Luke]

Statistics matter for a reason, and are quantified by a vaule determined the field it is used to be studied.

Just looks at the Lotto. Samson, the Lotto would love it if you bought $25,000,000 in tickets with the justification of, "I don't believe in statistics". You work against the odds, but still there is more than a chance you wouldn't win the jackpot.

But always keep in mind that seven out of two people don't understand math.


Another marketing practice you see everyday is "A Twenty Dollar Value for only Fifteen Dollars!". I love seeing this one at crappy firework packages at grocery stores. Sorry, but roman candles, snaps, and a repeater that doesn't make noise has zero value to me.

[samsonlonghair]

Statistics matter for a reason, and are quantified by a vaule determined the field it is used to be studied.

Not really sure what you mean by this.

Just looks at the Lotto. Samson, the Lotto would love it if you bought $25,000,000 in tickets with the justification of, "I don't believe in statistics". You work against the odds, but still there is more than a chance you wouldn't win the jackpot.

I never gamble; gambling is for suckers. On the other hand, I did see a story some years ago about a man who played the Virginia lottery by using a machine to fill out every possible numerical combination. He bought enough tickets to feed into the machine, and then he set back and let the machine do all the work. Seems like a clever way to ensure you win the lottery, right? Problem is that three other people also won the lottery that week by dumb luck. Once the pot was split four ways, he only made enough to break even. Easy come; easy go.

But always keep in mind that seven out of two people don't understand math.

That's a good one-liner. I think I'm going to steal that one.

Another marketing practice you see everyday is "A Twenty Dollar Value for only Fifteen Dollars!". I love seeing this one at crappy firework packages at grocery stores. Sorry, but roman candles, snaps, and a repeater that doesn't make noise has zero value to me.

Info-mercials use this one all the time.

How about, "it's such a great value, you can't afford not to buy it," or "it practically pays for itself."

[Luke]

Statistics matter for a reason, and are quantified by a value determined the field it is used to be studied.

Not really sure what you mean by this.

In some fields of psychology you have to provide a statistical significance of 5%. In other sciences, the percentage goes down to 1%. And you also have studies that go down to the absurd as "There is a .000001% of this happening". This usually takes place in movies, but also in theories of thermonuclear-dynamics.

PS- Gambling/taking a risk is not for suckers. Unless you are gambling for suckers, and I have had a poker game where candy was at stake.

[marurun]

I understand science. I'm not one of those nuts who refuses to get vaccinated and denies climate change. I'm just skeptical of any data presented in the form of statistics because I understand how easy it is to manipulate statistics.

This is not at all how you referred to things in your original "I don't trust statistics" post. Being skeptical is very important, but you simply cannot do science without statistics. Period. Proper use of statistics is just about the only way to do empirical science.

Just about every scientific study of any import has used statistics to back its conclusion in the form of a P test. A P test means little on its own. A study must be well-designed and properly undertaken. And when this is the case, the P test is usually what is used as the final determination of whether your findings were significant, or likely to be just random chance. The P test is a statistic. The P test is the most important statistic in science. And it means nothing if the underlying study is crap, and everything if the underlying study is good.

[Sarge]

Yep, statistics are very important. Now, statistics can be manipulated by withholding certain sets of data, but it's still profoundly important. In the gambling example, the reason I don't gamble is because of the statistics. The odds are stacked against me, and always will be.

Now, I do think that there are scientists that manipulate their data to try to get further research funding or push an particular agenda. I know we try to set up science as this great, impartial, fact-driven field, but that's a platonic ideal. Any time people are involved, things get messy. So that doesn't invalidate statistics. That invalidates bad science.

(My official stance on man-made climate change? It's inconclusive. We know climate change has occurred, is occurring, and will continue to occur. We just don't know how much of it is actually us, if any, based on the small sliver of maybe-passable data we're using.)

[Pulsar_t]

Cue another 20 pages of circular discussions about statistics

[dsheinem]

She may get my vote after all.


http://www.polygon.com/2015/4/20/845956 ... boy-tetris