It's easier than you'd think. You take the vote counts from different areas and by using the recorded demographics of the subject areas a statistician can parse out the voting patterns by race and gender.
He was a very good teacher. It's the data that can be skewed and then the way you look at it too...at best they are only good for a general idea. The entire class we learned how this can and does happen even with the best intentions. I do believe that some number of busy people ask friends or family how to vote and that could give you blocks of one opinion. I don't believe that racial groups are all voting one way though. We did ask the teacher why he was teaching us how to cheat. He said he wasn't advocating cheating but wanted us to know what we're up against. Each agenda's faction can and do get hugely different numbers from the same data. I guess you could put them all together and average them but it would still be a guess with an opinion.
You say he's a very good teacher, but after his course could you actually do analysis?
I'm not trying to be mean about this, but you are making a very broad statement that to me makes very little sense: "ah, I see... first lesson in my statistics class, don't trust statistics." There is a difference between truisms and knowing how to actually use the tools you are given.
I understand the gist of what you are trying to convey, that bad data can lead to inaccurate conclusions, and it's possible to load the proverbial dice. However, in a professional setting statistical analysis is deadly accurate. The techniques which are used to derive voting patterns hold up to scrutiny.
No offense taken. I appreciate the questions and input.
The class was 20+ yrs ago. And I pretty much hated it. It was a highly rated JC that try as I may to find it, had no political bias. Yes, I could do analysis well enough to get a B. I promptly forgot just about all of it, never to use those skills again. Computers were just becoming in home products, so there's that time line to consider.
I'm unsure what broad statement. That statistics isn't an exact science? I'm unsure how that couldn't still be true.. I believe you can get deadly accurate results but how often? For voting stats.... the corruption in our voting system is so bad how can you trust any voter roll? 95% just screams something isn't right.
I'm not saying it's not possible. But for me there are too many factors stacked up against this claim. This article claims a racial group and reveals (supposedly) that 95% of black women in a county voted the same way. Which screams HYPE and trouble for those women no matter what the truth is...and what ever happened to your vote being private? Why did they think Trump was never going to win ... stats and corruption?
I admit to being mildly curious why people in stats are so sure of results... I'm aware that a machine may be able to predict patterns pretty precisely and learn human behavior especially with the data gathered on every citizen but I have some questions there about social engineering and brain washing. There can be so many variables. If your population is brain washed you could get pretty exact stats, for sure!
When are stats used as propaganda and how is that done? I don't mean to challenge or offend just to say that, hey, don't believe everything you read. This one kinda stinks IMO. The more questions we ask the more we learn. I am interested how you come to believe in statistics as much as you do. They do have a place and are very useful. I don't mean to disregard them. Just this particular story doesn't add up for me.
view the rest of the comments →
15243088? ago
How the heck do you prove that?
15244241? ago
It's easier than you'd think. You take the vote counts from different areas and by using the recorded demographics of the subject areas a statistician can parse out the voting patterns by race and gender.
15254082? ago
ah, I see... first lesson in my statistics class, don't trust statistics.
15256210? ago
That's a strange thing to tell a class, statistics is just data analysis.
15258335? ago
He was a very good teacher. It's the data that can be skewed and then the way you look at it too...at best they are only good for a general idea. The entire class we learned how this can and does happen even with the best intentions. I do believe that some number of busy people ask friends or family how to vote and that could give you blocks of one opinion. I don't believe that racial groups are all voting one way though. We did ask the teacher why he was teaching us how to cheat. He said he wasn't advocating cheating but wanted us to know what we're up against. Each agenda's faction can and do get hugely different numbers from the same data. I guess you could put them all together and average them but it would still be a guess with an opinion.
just food for thought...
15260530? ago
You say he's a very good teacher, but after his course could you actually do analysis?
I'm not trying to be mean about this, but you are making a very broad statement that to me makes very little sense: "ah, I see... first lesson in my statistics class, don't trust statistics." There is a difference between truisms and knowing how to actually use the tools you are given.
I understand the gist of what you are trying to convey, that bad data can lead to inaccurate conclusions, and it's possible to load the proverbial dice. However, in a professional setting statistical analysis is deadly accurate. The techniques which are used to derive voting patterns hold up to scrutiny.
15263175? ago
No offense taken. I appreciate the questions and input.
The class was 20+ yrs ago. And I pretty much hated it. It was a highly rated JC that try as I may to find it, had no political bias. Yes, I could do analysis well enough to get a B. I promptly forgot just about all of it, never to use those skills again. Computers were just becoming in home products, so there's that time line to consider.
I'm unsure what broad statement. That statistics isn't an exact science? I'm unsure how that couldn't still be true.. I believe you can get deadly accurate results but how often? For voting stats.... the corruption in our voting system is so bad how can you trust any voter roll? 95% just screams something isn't right.
I'm not saying it's not possible. But for me there are too many factors stacked up against this claim. This article claims a racial group and reveals (supposedly) that 95% of black women in a county voted the same way. Which screams HYPE and trouble for those women no matter what the truth is...and what ever happened to your vote being private? Why did they think Trump was never going to win ... stats and corruption?
I admit to being mildly curious why people in stats are so sure of results... I'm aware that a machine may be able to predict patterns pretty precisely and learn human behavior especially with the data gathered on every citizen but I have some questions there about social engineering and brain washing. There can be so many variables. If your population is brain washed you could get pretty exact stats, for sure!
When are stats used as propaganda and how is that done? I don't mean to challenge or offend just to say that, hey, don't believe everything you read. This one kinda stinks IMO. The more questions we ask the more we learn. I am interested how you come to believe in statistics as much as you do. They do have a place and are very useful. I don't mean to disregard them. Just this particular story doesn't add up for me.