Posted by cccp1014 on 3/14/2018 1:06:00 PM (view original):

Posted by wylie715 on 3/14/2018 12:37:00 PM (view original):

Posted by cccp1014 on 3/14/2018 12:08:00 PM (view original):
My parents are also Russian. LOL.

Have you been drinking again? Difference is I was raised Jewish, not some other religion. Tang was raised Christian. Was likely baptized, etc.

Do you really not see the difference?

How does how you were raised make a difference? You are an agnostic. Tangplay is a Catholic. You were both born Jewish, but neither practices the religion now. I don't really see much of a difference.

WRONG!!

Tang's mom is Jewish. He did not convert. He was raised a Christian. I was raised Jewish. Temple and all. When I got older I rationally decided that I need proof of God's existence. I still follow all the traditions and such but deep down I do it for traditions not because I believe in a higher power. Tang is a Polish Christian with a Jewish mom. He is as Jewish as Obama is Muslim.

okay. You believe what you want to believe. I don't know why you care so much about what he calls himself.

my favorite song about stats:
https://www.youtube.com/watch?v=rT9y1PtENsU

Posted by cccp1014 on 3/14/2018 5:05:00 PM (view original):

Posted by bad_luck on 3/14/2018 2:53:00 PM (view original):

Posted by cccp1014 on 3/14/2018 2:47:00 PM (view original):
32 is statistically significant. I had the same issues with it when I learned it too.

What’s the margin of error and confidence level?

Margin of Error between 8-10.
Confidence Level - ~90%

Are we being civil now? Did you take your meds?

A 32 person sample out of a 20,000 person population with a 90% confidence level has a 15% margin of error.

That's horrible.

Going back to the John/Steve election, if the true position of the 20,000 voters was John 55% and Steve 45% and you ran a 32 voter pre-election poll 100 times, 90% of the polls would come back with John getting somewhere between 40% and 70% of the vote and Steve getting between 30% and 60% of the vote.

The other 10% would come back with fluky results outside those ranges. And you wouldn't know which ones were the flukes.

It's a worthless sample that tells you nothing.

cccp believes everything his friends tell him.....or he would if he had any friends!

Posted by bad_luck on 3/14/2018 7:08:00 PM (view original):

Posted by cccp1014 on 3/14/2018 5:05:00 PM (view original):

Posted by bad_luck on 3/14/2018 2:53:00 PM (view original):

Posted by cccp1014 on 3/14/2018 2:47:00 PM (view original):
32 is statistically significant. I had the same issues with it when I learned it too.

What’s the margin of error and confidence level?

Margin of Error between 8-10.
Confidence Level - ~90%

Are we being civil now? Did you take your meds?

A 32 person sample out of a 20,000 person population with a 90% confidence level has a 15% margin of error.

That's horrible.

Going back to the John/Steve election, if the true position of the 20,000 voters was John 55% and Steve 45% and you ran a 32 voter pre-election poll 100 times, 90% of the polls would come back with John getting somewhere between 40% and 70% of the vote and Steve getting between 30% and 60% of the vote.

The other 10% would come back with fluky results outside those ranges. And you wouldn't know which ones were the flukes.

It's a worthless sample that tells you nothing.

Nope it would tell you that more likely John would win. It would tell you SOMETHING. Your posts tell me nothing.

Posted by cccp1014 on 3/14/2018 8:36:00 PM (view original):

Posted by bad_luck on 3/14/2018 7:08:00 PM (view original):

Posted by cccp1014 on 3/14/2018 5:05:00 PM (view original):

Posted by bad_luck on 3/14/2018 2:53:00 PM (view original):

Posted by cccp1014 on 3/14/2018 2:47:00 PM (view original):
32 is statistically significant. I had the same issues with it when I learned it too.

What’s the margin of error and confidence level?

Margin of Error between 8-10.
Confidence Level - ~90%

Are we being civil now? Did you take your meds?

A 32 person sample out of a 20,000 person population with a 90% confidence level has a 15% margin of error.

That's horrible.

Going back to the John/Steve election, if the true position of the 20,000 voters was John 55% and Steve 45% and you ran a 32 voter pre-election poll 100 times, 90% of the polls would come back with John getting somewhere between 40% and 70% of the vote and Steve getting between 30% and 60% of the vote.

The other 10% would come back with fluky results outside those ranges. And you wouldn't know which ones were the flukes.

It's a worthless sample that tells you nothing.

Nope it would tell you that more likely John would win. It would tell you SOMETHING. Your posts tell me nothing.

This is tax credits/deductions all over again.

Think of it this way, you run your 32 voter poll. It comes back John 41%, Steve 59%.

What does it tell you?

Posted by cccp1014 on 3/14/2018 1:38:00 PM (view original):
But it actually is but it is many standard deviations out.

Anything over 2 deviations is not significant.

Posted by bad_luck on 3/14/2018 10:18:00 PM (view original):

Posted by cccp1014 on 3/14/2018 8:36:00 PM (view original):

Posted by bad_luck on 3/14/2018 7:08:00 PM (view original):

Posted by cccp1014 on 3/14/2018 5:05:00 PM (view original):

Posted by bad_luck on 3/14/2018 2:53:00 PM (view original):

Posted by cccp1014 on 3/14/2018 2:47:00 PM (view original):
32 is statistically significant. I had the same issues with it when I learned it too.

What’s the margin of error and confidence level?

Margin of Error between 8-10.
Confidence Level - ~90%

Are we being civil now? Did you take your meds?

A 32 person sample out of a 20,000 person population with a 90% confidence level has a 15% margin of error.

That's horrible.

Going back to the John/Steve election, if the true position of the 20,000 voters was John 55% and Steve 45% and you ran a 32 voter pre-election poll 100 times, 90% of the polls would come back with John getting somewhere between 40% and 70% of the vote and Steve getting between 30% and 60% of the vote.

The other 10% would come back with fluky results outside those ranges. And you wouldn't know which ones were the flukes.

It's a worthless sample that tells you nothing.

Nope it would tell you that more likely John would win. It would tell you SOMETHING. Your posts tell me nothing.

This is tax credits/deductions all over again.

Think of it this way, you run your 32 voter poll. It comes back John 41%, Steve 59%.

What does it tell you?

I explained the tax deductions already. You just refuse to listen. You would rather beat a dead horse.

That’s not an answer. What does the 41/59 poll tell you?

Posted by bad_luck on 3/15/2018 9:12:00 AM (view original):
That’s not an answer. What does the 41/59 poll tell you?

That depends: was it conducted by the same polling group that told us HRC would win by double-figures?

Posted by bad_luck on 3/15/2018 9:12:00 AM (view original):
That’s not an answer. What does the 41/59 poll tell you?

It tells me that Steve would win. Similarly how many polls stated that HRC would win?

My statement to tangplay should have stated that my sample size was MORE statistically significant than his was. Is that better? You butt into other conversations and get off on another tangent. Stop stalking me you weirdo. Whats next? Calling me at work? I think you are insane as are your Leftist views. The fact that you called Ben Shapiro a "white supremacist" speaks volumes to me. Go away.

Posted by cccp1014 on 3/15/2018 9:44:00 AM (view original):

Posted by bad_luck on 3/15/2018 9:12:00 AM (view original):
That’s not an answer. What does the 41/59 poll tell you?

It tells me that Steve would win. Similarly how many polls stated that HRC would win?

My statement to tangplay should have stated that my sample size was MORE statistically significant than his was. Is that better? You butt into other conversations and get off on another tangent. Stop stalking me you weirdo. Whats next? Calling me at work? I think you are insane as are your Leftist views. The fact that you called Ben Shapiro a "white supremacist" speaks volumes to me. Go away.

Exactly. It tells you that Steve is maybe winning at 59%. The problem is that John is actually going to get 55% of the vote.

So your 32 person sample was misleading and you needed a larger sample in order to rely on the result.

Yes. It could be misleading. Polls can be misleading. We agree.

I still say my poll of ~100 persons was more statistically significant than his poll of 15 people.

Posted by cccp1014 on 3/15/2018 9:56:00 AM (view original):
Yes. It could be misleading. Polls can be misleading. We agree.

I still say my poll of ~100 persons was more statistically significant than his poll of 15 people.

Do we agree that a sample of 32 is not good enough for large populations?

Posted by bad_luck on 3/15/2018 10:06:00 AM (view original):

Posted by cccp1014 on 3/15/2018 9:56:00 AM (view original):
Yes. It could be misleading. Polls can be misleading. We agree.

I still say my poll of ~100 persons was more statistically significant than his poll of 15 people.

Do we agree that a sample of 32 is not good enough for large populations?

Define large

But yes it can be misleading when you have over 20,000 persons. "Not Good Enough" is not a term I would use. I would say "could be misleading".