Poll

Brett Kavanaguh: Yay or Nay?

Yay!
Nay!
Who cares? The SCOTUS doesn't matter anyways.

Author Topic: Brett Kavanaguh: Yay or Nay?  (Read 197699 times)

GuitarStv

  • Senior Mustachian
  • ********
  • Posts: 23048
  • Age: 42
  • Location: Toronto, Ontario, Canada
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1100 on: October 03, 2018, 01:31:40 PM »
So now the situation we have is everyone yelling 'Rape!' I would be 110% behind an investigation if I thought he was an evil and awful dude, but that doesn't seem the case; it is just more political haggling.

I never thought that Bill Cosby was an evil and awful dude.  I mean, it certainly didn't seem to be the case.  For an awful long time there were no allegations about him, and it certainly didn't seem to be the case that he was a serial rapist.  I'm sure glad that an investigation was done into his past regardless of my personal opinion though.

shenlong55

  • Pencil Stache
  • ****
  • Posts: 528
  • Age: 41
  • Location: Kentucky
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1101 on: October 03, 2018, 01:37:24 PM »
Quote
I'm suggesting that 100% of the population of accused rapists are accused rapists, and 5% of them are falsely accused.   I'm not saying anything at all about the population at large, but for some reason you are. 

Quote
A better and more relevant approach to determining the likelihood of guilt of a specific existing allegation is to ignore the population at large that no one is talking about, and determine what percentage of people who ARE accused of rape are falsely vs correctly accused. 

Ok, I understand where our differences are, thank you. You are not saying anything at all about the population at large, and if I understand correctly, you believe all the rape-related statistics have no bearing whatsoever on the population at large, thus one can not do the calculations I had done here.

I reject this notion completely.
1. your view implies rapists and the population at large are completely independent, its as if they exist on different planes of existence or live on two different planets. But we know this is not true, rapists unfortunately form a subset of the population at large.

2. When you ignore the population at large completely, you are effectively treating the false accusation rate regarding the population as 0. We also know this is not true, it is definitely lower than 2-10%, but it is not zero.

3. Given that the two groups coexist with known composition, my method is correct. What we need is to find an appropriate false positive rate so that the output is 2-10%. Interestingly, in laserlady's calculation, it was achieved with a false positive rate of 0.1%. So now we need to ask ourselves, is this rate reasonable? I have no clue if it is, if you have studies that assert this I would be happy to see them.

Quote
No one is accused of rape by random chance.

This is both true and false. Because conventionally, to be named a suspect by the victim, one must already know (or usually have some form of prior contact) the victim beforehand. But one needs not to be accused randomly. I am not claiming one is accused of rape by random chance.

My method details that given only the fp and fn rates, the likelihood of some random person I see on the news being actually guilty is y in the absence of hard physical evidence. It is very important to understand this part.

Quote
Does testimony carry no weight with you?  What if one person's story checks out and the other person is shown to be a dirty rotten liar?  Regardless of which side is which, I don't think you need physical evidence of a crime to establish the credibility of a witness.

I have seen enough people poking holes to both versions of the stories, with both parties changing their versions overtime. If I had to pick a side, then yes, I found Dr. Ford's testimonies to be much more credible, but I can not condemn BK on this ground alone. People are naturally unreliable narrators, especially under pressure. I will fully accept the final FBI result, whatever the outcome, because I believe in due process. I believe in due process so much that even IF, against all odds, Mueller came out and said there were no collusion by Trump, I would still accept it.
 
I am sorry if you feel I've hijacked the thread.

How the false accusation rate is calculated: accusations "factually" determined to be false / total number of accusations.

How the false positive rate is calculated: number of false positive results / total number of negatives.

If you can't see the difference at this point I think it's because you're choosing not to.  But just in case, let me illustrate.

                              5%                             accused                          not accused
Rapists                   A                                 B                                   D
Not rapists              B                                 E                                   F
                             G                                 H                                   I

B would be the total negatives (which is the divisor in the false positive equation), whereas H would be the total accusations (which is the divisor in the false accusation equation).

Davnasty

  • Magnum Stache
  • ******
  • Posts: 2793
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1102 on: October 03, 2018, 01:42:48 PM »
So now the situation we have is everyone yelling 'Rape!' I would be 110% behind an investigation if I thought he was an evil and awful dude, but that doesn't seem the case; it is just more political haggling.

I never thought that Bill Cosby was an evil and awful dude.  I mean, it certainly didn't seem to be the case.  For an awful long time there were no allegations about him, and it certainly didn't seem to be the case that he was a serial rapist.  I'm sure glad that an investigation was done into his past regardless of my personal opinion though.

My first reaction to the accusation against Bill Cosby was, "No way he did that, this is some terrible woman trying to get his money and make herself famous".

My bad.
« Last Edit: October 03, 2018, 05:01:38 PM by Dabnasty »

sol

  • Walrus Stache
  • *******
  • Posts: 8433
  • Age: 47
  • Location: Pacific Northwest
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1103 on: October 03, 2018, 01:43:58 PM »
Because my son potentially did something very wrong.
...
what if he seems more like an immature idiot, then what?

I'm a little upset by the insinuation that someone shouldn't be held accountable for rape if they are a) your family, or b) an immature idiot. 

Aren't all rapists immature idiots?  How is that an excuse?
 
If my son commits rape, he goes to jail.  Ditto for murder, or embezzling corporate profits.  Why would you willingly defend a known criminal just because it's your failure as a parent that caused them to turn to a life of crime?   Part of being a good parent is teaching you kids to take responsibility for their choices in life, right?

Quote
So now the situation we have is everyone yelling 'Rape!' I would be 110% behind an investigation if I thought he was an evil and awful dude, but that doesn't seem the case; it is just more political haggling.

Just so we're clear, Judge Kavanaugh is not accused of rape, and it's not just political haggling.  He's accused of being a handsy and belligerent drunk when he was 17, of thinking it was funny to swing his dick at drunk girls at parties, of being present at parties where he knew rape was taking place, and then lying about all of it under oath.  Since then, he's tried to make the accusations into some sort of partisan conspiracy, rather than addressing the allegations themselves.

And by all other accounts, he seems to have solved his "immature idiot" problems as an adult.  He appears to have conducted himself mostly professionally since since the last known allegation in 1998, or at least we don't yet have any more recent reports of misconduct until this confirmation hearing started.  Unfortunately, he has lied repeatedly about his immature idiot phase, under oath and to congress, which is a federal crime and should probably be disqualifying for a supreme court nominee.  I think he would have been easily confirmed if he had instead owned up to his mistakes and offered sincere apologies.  The fact that he hasn't done so suggests to me that he knows he is guilty AF of these allegations and more, and that the truth can never come to light if he wants to sit on the SC.

hoping2retire35

  • Handlebar Stache
  • *****
  • Posts: 1398
  • Location: UPCOUNTRY CAROLINA
  • just want to see where this appears
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1104 on: October 03, 2018, 02:02:36 PM »
Because my son potentially did something very wrong.
...
what if he seems more like an immature idiot, then what?

I'm a little upset by the insinuation that someone shouldn't be held accountable for rape if they are a) your family, or b) an immature idiot. 

Aren't all rapists immature idiots?  How is that an excuse?
 
If my son commits rape, he goes to jail.  Ditto for murder, or embezzling corporate profits.  Why would you willingly defend a known criminal just because it's your failure as a parent that caused them to turn to a life of crime?   Part of being a good parent is teaching you kids to take responsibility for their choices in life, right?

Because my example didn't include rape, it included horseplay that got out of hand or sexually inappropriate behavior or whatever you want to call, not clothes off, not rape.

I believe you are thinking of doggyfizzle, reply #1094



Just so we're clear, Judge Kavanaugh is not accused of rape, and it's not just political haggling.  He's accused of being a handsy and belligerent drunk when he was 17, of thinking it was funny to swing his dick at drunk girls at parties, of being present at parties where he knew rape was taking place, and then lying about all of it under oath.  Since then, he's tried to make the accusations into some sort of partisan conspiracy, rather than addressing the allegations themselves.

And by all other accounts, he seems to have solved his "immature idiot" problems as an adult.  He appears to have conducted himself mostly professionally since since the last known allegation in 1998, or at least we don't yet have any more recent reports of misconduct until this confirmation hearing started.  Unfortunately, he has lied repeatedly about his immature idiot phase, under oath and to congress, which is a federal crime and should probably be disqualifying for a supreme court nominee.  I think he would have been easily confirmed if he had instead owned up to his mistakes and offered sincere apologies.  The fact that he hasn't done so suggests to me that he knows he is guilty AF of these allegations and more, and that the truth can never come to light if he wants to sit on the SC.
Because most of what we are talking about centers on Ford's testimony.

Let's be honest, Democrats and people on the this forum(holding steady at 3-to-1) and aren't going to just throw their hands up and say "We tried!" once the FBI says later this week they got nothing. This is partisan politics using the implication of rape. Shameful.
« Last Edit: October 03, 2018, 02:05:47 PM by hoping2retire35 »

anisotropy

  • Pencil Stache
  • ****
  • Posts: 681
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1105 on: October 03, 2018, 02:10:52 PM »
Shenlong,

The standard 2x2 errors table in this case looks like this

True Positive       False Negative
False Positive      True Negative

False positive is the proportion of negatives that still yield positive test outcomes, ie FP.

Thus, False positive rate = FP / (FP+TN). Instead, you wrote

Quote
How the false positive rate is calculated: number of false positive results / total number of negatives.

ie, False positive rate = FP / (FN+TN)

Surely you made a typo? What you are suggesting is wrong. https://en.wikipedia.org/wiki/False_positive_rate

The contention is regarding my treatment of the 2-10% as FP, some believe it's not appropriate. I've avoided using jargons thus far, but fundamentally this is an inference problem using multiple comparisons.

Sorry if I am being rude, but if you can't even define FP rate correctly, I see very little merit in discussing the technical side of this with you. Again, statistics is often counter intuitive. Most people can tell you the odds of flipping a fair coin is 50%, but it can get really ugly real fast.

As the inference remains valid given the relations of the two subsets. The goal now is to justify IF 0.1% serves as a realistic false positive for the not-rapist population so that the final output gets to 5%. Personally I have seen no literature that points to this direction.
« Last Edit: October 03, 2018, 03:01:29 PM by anisotropy »

sol

  • Walrus Stache
  • *******
  • Posts: 8433
  • Age: 47
  • Location: Pacific Northwest
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1106 on: October 03, 2018, 02:11:22 PM »
This is partisan politics using the implication of rape. Shameful.

My dad tried this same argument with me a few days ago.  He said "the real problem here is the way the democrats have handled these allegations" and I was basically dumbfounded.  Just maybe, the "real problem" is sexual assault of a minor and not how a bunch of geriatric politicians argue with each other?  Why is everyone on the right suddenly in such a rush to defend the rights of the accused?  Where was this advocacy when it came to the war on drugs, or prison reform, or immigration hearings? 

And if you're genuinely upset about partisan politics influencing the way these accusations impacted the nomination, how do you feel about the Senate refusing to uphold it's constitutionally defined obligation to hold a hearing for Merrick Garland?  That was purely partisan shenanigans.  At least in this case, partisan liberals are correctly handling allegations of sexual assault (in a way that happens to align with their partisan interests) instead of just blatantly saying "no, we don't like the President so we're going to refuse this nominee or any other nominee from this President."  I think it speaks volumes about the status of our politics that the very people who cheered the Garland debacle are now crying crocodile tears over Kavanaugh, who genuinely appears unfit for the court.
« Last Edit: October 03, 2018, 02:16:06 PM by sol »

Kris

  • Walrus Stache
  • *******
  • Posts: 7306
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1107 on: October 03, 2018, 02:23:41 PM »
As a sexual assault victim, this thread sucks balls.
Just FYI.

No shit.

I'm sorry, @hoping2retire35 , but every time you use the word "horseplay" to casually dismiss sexual violation it makes me want to freaking flip tables.

nereo

  • Senior Mustachian
  • ********
  • Posts: 17472
  • Location: Just south of Canada
    • Here's how you can support science today:
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1108 on: October 03, 2018, 02:25:02 PM »
(Never) Scenario; 10 years from now my son is accused by another teenage girl of pushing her down and possible attempted rape. The next day my daughter tells me a male, approximately their age, did something similar to her. What do I do?


Why not choose to involve the legal system in both cases?  Hire a defense attorney for you son if he is going to be charged criminally, or take your daughter to the police ASAP and have her go on the record so the attempted rapist does not go unpunished.  No need for vigilante justice...
Because my son potentially did something very wrong.

Sure, I'll hire a defense attorney if necessary. But more importantly I want to be sure my son understands, is sorry, and never does it again.

 Or what if one or both boys flat deny it even happened. "Hey son, just tell them you are sorry and we will forget about this"-a dad. Either boy-"Nope, I never did that." It is easy when everything is clear and the girls motivation for telling a lie would be low but what if there is only circumstantial evidence; they know each other, were in the same area/building at the same time. What if it is dark and she couldn't see him but recognized his voice.

In the immediate aftermath of it happening to your daughter it is probably best to go to the police; but then what? If the kid is a real SOB, then sure lock him up. But what if he seems more like an immature idiot, then what?

What you are describing is the desire to protect your child at all costs vs your societal obligation to report all crimes.
It's a horrible place to be in, and I hope the outcomes are as good as possible for everyone involved.
My only advice for you is to hire good legal council and make sure that you are doing everything you can as a parent to teach right from wrong.
If the kid is an immature idiot that genuinely didn't understand what harm was being done, that will be factored in during a plea, particularly the kid shows remorse and now understands what he/she did wrong.  Our judicial system really will go out of its way to not lock up young, first-time offenders who try to cooperate.

PoutineLover

  • Handlebar Stache
  • *****
  • Posts: 1567
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1109 on: October 03, 2018, 02:28:13 PM »
As a sexual assault victim, this thread sucks balls.
Just FYI.

No shit.

I'm sorry, @hoping2retire35 , but every time you use the word "horseplay" to casually dismiss sexual violation it makes me want to freaking flip tables.
Same.

wenchsenior

  • Magnum Stache
  • ******
  • Posts: 3779
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1110 on: October 03, 2018, 02:41:25 PM »
As a sexual assault victim, this thread sucks balls.
Just FYI.

No shit.

I'm sorry, @hoping2retire35 , but every time you use the word "horseplay" to casually dismiss sexual violation it makes me want to freaking flip tables.
Same.

Why apologize? The past few days of posts (not just h2r's) actually make me feel like our collective IQ is dropping as we read them. 


sol

  • Walrus Stache
  • *******
  • Posts: 8433
  • Age: 47
  • Location: Pacific Northwest
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1111 on: October 03, 2018, 02:42:04 PM »
I've avoided using jargons thus far, but fundamentally this is an inference problem using multiple comparisons.

No, it's not.  We keep explaining that you're using the wrong tool.

Multiple comparison test are only valid when comparing simultaneous application of independent probabilities.  In the case of the problem you've posed, those are 1) the chance a random person is accused of rape, and 2) the chance that a random person is actually a rapist.  You've correctly identified the tiny overlap between two probabilities that have nothing to do with whether or not a specific person who stands accused is actually a rapist.

Random people are not accused of rape.  If you are accused of rape, you are not a random person.  You are accused because there is some evidence beyond random probability that you are guilty of rape.  Rapists get accused of rape.  As a general rule, nonrapists do not get accused of rape. 

Your two tests are not independent of each other (unless everyone is accused of rape).  Your methodology is not applicable.  Tweaking the numbers will not help, because you're answering the wrong question.  We don't care about the chance that a random person will be accused of rape and will also be guilty of rape, we care about the probability that a person who is already accused of rape as a defined precondition of any further tests will then turn out to be guilty, aka fp.  The non-accused population should not appear in your equations in any form.  They don't matter, in determining how many accused rapists are guilty.

As a sexual assault victim, this thread sucks balls.

Sorry.  Imagine how much better you'll feel after republicans put a second sexual abuser on the Supreme Court and then use that majority to curtail women's rights. 

This whole stats discussion feels like a targeted refutation of the #metoo movement.  What ever happened to "I believe the women"?  This is like something cooked up in a dark corner of the internet by red-pill loving math grad students.  #notallstatisticians
« Last Edit: October 03, 2018, 02:55:36 PM by sol »

Malaysia41

  • Magnum Stache
  • ******
  • Posts: 3311
  • Age: 51
  • Location: Verona, Italy
    • My mmm journal
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1112 on: October 03, 2018, 02:56:51 PM »
#notallstatisticians

lol

anisotropy

  • Pencil Stache
  • ****
  • Posts: 681
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1113 on: October 03, 2018, 03:00:53 PM »

Random people are not accused of rape.  If you are accused of rape, you are not a random person.  You are accused because there is some evidence beyond random probability that you are guilty of rape.  Rapists get accused of rape.  As a general rule, nonrapists do not get accused of rape. 


WOw I wrote "multiple comparison"? SORRY! that's super embarrassing! I will correct it now. Now everyone knows what I was thinking of doing next.

Your way of interrupting it is wrong. As long a non-zero fp exists, a population that can be broken down in subsets with known compositions, and fp with fn that are independent of the results, you don't need "everyone is accused of rape". It's "suppose someone is accused of rape given the fp".

So really, the problem is to find appropriate fp

« Last Edit: October 03, 2018, 03:07:47 PM by anisotropy »

Unique User

  • Pencil Stache
  • ****
  • Posts: 718
  • Location: NC
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1114 on: October 03, 2018, 03:03:43 PM »
As a sexual assault victim, this thread sucks balls.
Just FYI.

No shit.

I'm sorry, @hoping2retire35 , but every time you use the word "horseplay" to casually dismiss sexual violation it makes me want to freaking flip tables.
Same.

Yep. 

Beyond the shock that sexual assault seems to be okay with some on this forum because (1) it was a long time ago and (2) there are no witnesses, I'm also truly shocked that they think this is the only reason Kavanaugh does not belong on any court.  He lied prior to this last hearing about things unrelated to Dr. Ford.  He lied in his written answers to Chuck Grassley.  He lied in his 2006 hearing, it took 3 years to get him confirmed because he was considered "too partisan".  He lied and advocated for sleazy tactics against Clinton when he was on Ken Starr's commission.   I absolutely agree that Clinton lied, but Kavanaugh went a bit too far with his Vince Foster conspiracy nonsense.  Just the lies alone should be enough to keep him off any court. 
 

runbikerun

  • Pencil Stache
  • ****
  • Posts: 539
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1115 on: October 03, 2018, 03:05:36 PM »

Random people are not accused of rape.  If you are accused of rape, you are not a random person.  You are accused because there is some evidence beyond random probability that you are guilty of rape.  Rapists get accused of rape.  As a general rule, nonrapists do not get accused of rape. 


WOw I wrote "multiple comparison"? SORRY! that's super embarrassing! I will correct it now. Now everyone knows what I was thinking of doing next.

Your way of interrupting it is wrong. As long a non-zero fp exists, you don't need "everyone is accused of rape". It's "suppose someone is accused of rape given the fp".

You are starting with the fact that about 5% of rape allegations are found to be false.

You go from there to a point where you find that 73% of rape allegations are false.

It is a logical impossibility for both of these to be the case.

Glenstache

  • Magnum Stache
  • ******
  • Posts: 3493
  • Age: 94
  • Location: Upper left corner
  • FI(lean) working on the "RE"
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1116 on: October 03, 2018, 03:10:00 PM »
Can we just stop with the statistics. As much as I LOVE talking about statistics (really), this is getting way off topic and even if resolved/reconciled isn't really going anywhere.

hoping2retire35

  • Handlebar Stache
  • *****
  • Posts: 1398
  • Location: UPCOUNTRY CAROLINA
  • just want to see where this appears
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1117 on: October 03, 2018, 03:17:58 PM »
As a sexual assault victim, this thread sucks balls.
Just FYI.

No shit.

I'm sorry, @hoping2retire35 , but every time you use the word "horseplay" to casually dismiss sexual violation it makes me want to freaking flip tables.
Same.
How about you not get mad at one thing you read and nothing else I wrote?

How more clear can I be other than...(hypothetical scenario)
"My son potentially did something very wrong."

The scenario describes a situation going from(wait for it) horseplay to, possibly a sexual violation; the problem is no one really knows, not the victim not the potential perps.


anisotropy

  • Pencil Stache
  • ****
  • Posts: 681
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1118 on: October 03, 2018, 03:19:26 PM »

Random people are not accused of rape.  If you are accused of rape, you are not a random person.  You are accused because there is some evidence beyond random probability that you are guilty of rape.  Rapists get accused of rape.  As a general rule, nonrapists do not get accused of rape. 


WOw I wrote "multiple comparison"? SORRY! that's super embarrassing! I will correct it now. Now everyone knows what I was thinking of doing next.

Your way of interrupting it is wrong. As long a non-zero fp exists, you don't need "everyone is accused of rape". It's "suppose someone is accused of rape given the fp".

You are starting with the fact that about 5% of rape allegations are found to be false.

You go from there to a point where you find that 73% of rape allegations are false.

It is a logical impossibility for both of these to be the case.

rbr, try this

recall the link I provided regarding Sam Gross article

"How many other Glenn Fords are still behind bars? How many will die there? Just how often does our venerated justice system fail? Rarely, at least according to U.S. Supreme Court Justice Antonin Scalia. In a 2006 opinion he cited an approximate error rate of 0.027 percent, based on back-of-the-envelope calculations by an Oregon district attorney in a fiery op-ed for the New York Times. The op-ed was in response to a report by Samuel Gross, a law professor at the University of Michigan, cataloguing 340 exonerations between 1989 and 2003. “Let’s give the professor the benefit of the doubt,” the op-ed read. “Let’s assume that he understated the number of innocents by roughly a factor of 10, that instead of 340 there were 4,000 people in prison who weren’t involved in the crime in any way. During that same 15 years, there were more than 15 million felony convictions across the country. That would make the error rate .027 percent — or, to put it another way, a success rate of 99.973 percent.”

This incredibly small error rate, becomes 4% when Sam gross did it regarding potentially innocent inmates wrongly convicted. This is the same technique I used.

Note this 4% is wrongful CONVICTION rate (accused, charged, then convicted), it is entirely in-line with my conclusion given we know low conviction rate on rape in general, to have a much higher wrongful rate when we consider accusation alone.

shenlong55

  • Pencil Stache
  • ****
  • Posts: 528
  • Age: 41
  • Location: Kentucky
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1119 on: October 03, 2018, 03:19:51 PM »
Shenlong,

The standard 2x2 errors table in this case looks like this

True Positive       False Negative
False Positive      True Negative

False positive is the proportion of negatives that still yield positive test outcomes, ie FP.

Thus, False positive rate = FP / (FP+TN). Instead, you wrote

Quote
How the false positive rate is calculated: number of false positive results / total number of negatives.

ie, False positive rate = FP / (FN+TN)

Surely you made a typo? What you are suggesting is wrong. https://en.wikipedia.org/wiki/False_positive_rate

The contention is regarding my treatment of the 2-10% as FP, some believe it's not appropriate. I've avoided using jargons thus far, but fundamentally this is an inference problem using multiple comparisons.

Sorry if I am being rude, but if you can't even define FP rate correctly, I see very little merit in discussing the technical side of this with you. Again, statistics is often counter intuitive. Most people can tell you the odds of flipping a fair coin is 50%, but it can get really ugly real fast.

As the inference remains valid given the relations of the two subsets. The goal now is to justify IF 0.1% serves as a realistic false positive for the not-rapist population so that the final output gets to 5%. Personally I have seen no literature that points to this direction.

No, no mistake on my part.  Just more attempts by you to muddy the waters.  Total negatives include both false positives and true negatives, as both are 'actual' negatives.  From your link...

Quote
...where FP is the number of false positives, TN is the number of true negatives and N=FP+TN is the total number of negatives.

nereo

  • Senior Mustachian
  • ********
  • Posts: 17472
  • Location: Just south of Canada
    • Here's how you can support science today:
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1120 on: October 03, 2018, 03:21:21 PM »

How more clear can I be other than...(hypothetical scenario)
"My son potentially did something very wrong."

The scenario describes a situation going from(wait for it) horseplay to, possibly a sexual violation; the problem is no one really knows, not the victim not the potential perps.

There is no ambiguity with sexual assault.  I do not understand what you mean by the term "sexual violation". 

Malloy

  • Bristles
  • ***
  • Posts: 403
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1121 on: October 03, 2018, 03:23:58 PM »
As a sexual assault victim, this thread sucks balls.
Just FYI.

No shit.

I'm sorry, @hoping2retire35 , but every time you use the word "horseplay" to casually dismiss sexual violation it makes me want to freaking flip tables.
Same.

Why apologize? The past few days of posts (not just h2r's) actually make me feel like our collective IQ is dropping as we read them.

Right here with all of you.

anisotropy

  • Pencil Stache
  • ****
  • Posts: 681
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1122 on: October 03, 2018, 03:24:33 PM »
Shenlong,

The standard 2x2 errors table in this case looks like this

True Positive       False Negative
False Positive      True Negative

False positive is the proportion of negatives that still yield positive test outcomes, ie FP.

Thus, False positive rate = FP / (FP+TN). Instead, you wrote

Quote
How the false positive rate is calculated: number of false positive results / total number of negatives.

ie, False positive rate = FP / (FN+TN)

Surely you made a typo? What you are suggesting is wrong. https://en.wikipedia.org/wiki/False_positive_rate

The contention is regarding my treatment of the 2-10% as FP, some believe it's not appropriate. I've avoided using jargons thus far, but fundamentally this is an inference problem using multiple comparisons.

Sorry if I am being rude, but if you can't even define FP rate correctly, I see very little merit in discussing the technical side of this with you. Again, statistics is often counter intuitive. Most people can tell you the odds of flipping a fair coin is 50%, but it can get really ugly real fast.

As the inference remains valid given the relations of the two subsets. The goal now is to justify IF 0.1% serves as a realistic false positive for the not-rapist population so that the final output gets to 5%. Personally I have seen no literature that points to this direction.

No, no mistake on my part.  Just more attempts by you to muddy the waters.  Total negatives include both false positives and true negatives, as both are 'actual' negatives.  From your link...

Quote
...where FP is the number of false positives, TN is the number of true negatives and N=FP+TN is the total number of negatives.

YOU were the one that said total number of negatives,
Quote
How the false positive rate is calculated: number of false positive results / total number of negatives.

total negatives are defined as TN+FN. I am really not sure how this is not getting through.

What then did you mean by total negatives? and how would you write out the equation? 

runbikerun

  • Pencil Stache
  • ****
  • Posts: 539
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1123 on: October 03, 2018, 03:33:42 PM »

Random people are not accused of rape.  If you are accused of rape, you are not a random person.  You are accused because there is some evidence beyond random probability that you are guilty of rape.  Rapists get accused of rape.  As a general rule, nonrapists do not get accused of rape. 


WOw I wrote "multiple comparison"? SORRY! that's super embarrassing! I will correct it now. Now everyone knows what I was thinking of doing next.

Your way of interrupting it is wrong. As long a non-zero fp exists, you don't need "everyone is accused of rape". It's "suppose someone is accused of rape given the fp".

You are starting with the fact that about 5% of rape allegations are found to be false.

You go from there to a point where you find that 73% of rape allegations are false.

It is a logical impossibility for both of these to be the case.

rbr, try this

recall the link I provided regarding Sam Gross article

"How many other Glenn Fords are still behind bars? How many will die there? Just how often does our venerated justice system fail? Rarely, at least according to U.S. Supreme Court Justice Antonin Scalia. In a 2006 opinion he cited an approximate error rate of 0.027 percent, based on back-of-the-envelope calculations by an Oregon district attorney in a fiery op-ed for the New York Times. The op-ed was in response to a report by Samuel Gross, a law professor at the University of Michigan, cataloguing 340 exonerations between 1989 and 2003. “Let’s give the professor the benefit of the doubt,” the op-ed read. “Let’s assume that he understated the number of innocents by roughly a factor of 10, that instead of 340 there were 4,000 people in prison who weren’t involved in the crime in any way. During that same 15 years, there were more than 15 million felony convictions across the country. That would make the error rate .027 percent — or, to put it another way, a success rate of 99.973 percent.”

This incredibly small error rate, becomes 4% when Sam gross did it regarding potentially innocent inmates wrongly convicted. This is the same technique I used.

Note this 4% is wrongful CONVICTION rate (accused, charged, then convicted), it is entirely in-line with my conclusion given we know low conviction rate on rape in general, to have a much higher wrongful rate when we consider accusation alone.

This is madness, and makes absolutely no sense in any direction. You started off with the 5% figure in your first post on this. Now you're working off a seemingly entirely random 4% figure that bears no relation to anything you wrote above. There's nothing further to be gained from discussing this.

anisotropy

  • Pencil Stache
  • ****
  • Posts: 681
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1124 on: October 03, 2018, 03:40:03 PM »

Random people are not accused of rape.  If you are accused of rape, you are not a random person.  You are accused because there is some evidence beyond random probability that you are guilty of rape.  Rapists get accused of rape.  As a general rule, nonrapists do not get accused of rape. 


WOw I wrote "multiple comparison"? SORRY! that's super embarrassing! I will correct it now. Now everyone knows what I was thinking of doing next.

Your way of interrupting it is wrong. As long a non-zero fp exists, you don't need "everyone is accused of rape". It's "suppose someone is accused of rape given the fp".

You are starting with the fact that about 5% of rape allegations are found to be false.

You go from there to a point where you find that 73% of rape allegations are false.

It is a logical impossibility for both of these to be the case.

rbr, try this

recall the link I provided regarding Sam Gross article

"How many other Glenn Fords are still behind bars? How many will die there? Just how often does our venerated justice system fail? Rarely, at least according to U.S. Supreme Court Justice Antonin Scalia. In a 2006 opinion he cited an approximate error rate of 0.027 percent, based on back-of-the-envelope calculations by an Oregon district attorney in a fiery op-ed for the New York Times. The op-ed was in response to a report by Samuel Gross, a law professor at the University of Michigan, cataloguing 340 exonerations between 1989 and 2003. “Let’s give the professor the benefit of the doubt,” the op-ed read. “Let’s assume that he understated the number of innocents by roughly a factor of 10, that instead of 340 there were 4,000 people in prison who weren’t involved in the crime in any way. During that same 15 years, there were more than 15 million felony convictions across the country. That would make the error rate .027 percent — or, to put it another way, a success rate of 99.973 percent.”

This incredibly small error rate, becomes 4% when Sam gross did it regarding potentially innocent inmates wrongly convicted. This is the same technique I used.

Note this 4% is wrongful CONVICTION rate (accused, charged, then convicted), it is entirely in-line with my conclusion given we know low conviction rate on rape in general, to have a much higher wrongful rate when we consider accusation alone.

This is madness, and makes absolutely no sense in any direction. You started off with the 5% figure in your first post on this. Now you're working off a seemingly entirely random 4% figure that bears no relation to anything you wrote above. There's nothing further to be gained from discussing this.

I am SHOWING you how using the similar inference tool, you can get from 5% false positive to 75% innocent rate, just like how you can get from 0.02% error rate to 4% wrongful conviction rate. Did you even read the article(s)? otherwise you would not have called 4% figure i cited from article as random.

It is completely relevant because it's a similar idea/tool, if you can't understand one I don't expect you to understand the other.

shenlong55

  • Pencil Stache
  • ****
  • Posts: 528
  • Age: 41
  • Location: Kentucky
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1125 on: October 03, 2018, 03:41:31 PM »
Shenlong,

The standard 2x2 errors table in this case looks like this

True Positive       False Negative
False Positive      True Negative

False positive is the proportion of negatives that still yield positive test outcomes, ie FP.

Thus, False positive rate = FP / (FP+TN). Instead, you wrote

Quote
How the false positive rate is calculated: number of false positive results / total number of negatives.

ie, False positive rate = FP / (FN+TN)

Surely you made a typo? What you are suggesting is wrong. https://en.wikipedia.org/wiki/False_positive_rate

The contention is regarding my treatment of the 2-10% as FP, some believe it's not appropriate. I've avoided using jargons thus far, but fundamentally this is an inference problem using multiple comparisons.

Sorry if I am being rude, but if you can't even define FP rate correctly, I see very little merit in discussing the technical side of this with you. Again, statistics is often counter intuitive. Most people can tell you the odds of flipping a fair coin is 50%, but it can get really ugly real fast.

As the inference remains valid given the relations of the two subsets. The goal now is to justify IF 0.1% serves as a realistic false positive for the not-rapist population so that the final output gets to 5%. Personally I have seen no literature that points to this direction.

No, no mistake on my part.  Just more attempts by you to muddy the waters.  Total negatives include both false positives and true negatives, as both are 'actual' negatives.  From your link...

Quote
...where FP is the number of false positives, TN is the number of true negatives and N=FP+TN is the total number of negatives.

YOU were the one that said total number of negatives,
Quote
How the false positive rate is calculated: number of false positive results / total number of negatives.

total negatives are defined as TN+FN. I am really not sure how this is not getting through.

What then did you mean by total negatives? and how would you write out the equation?

Seriously, you didn't even read the post your responding to or the link that you posted?  Here you go...

False positive rate = number of false positive / total negatives
Total negatives = false positives + true negatives

Same as is described at the link you posted.


anisotropy

  • Pencil Stache
  • ****
  • Posts: 681
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1126 on: October 03, 2018, 03:50:33 PM »
Shenlong,

The standard 2x2 errors table in this case looks like this

True Positive       False Negative
False Positive      True Negative

False positive is the proportion of negatives that still yield positive test outcomes, ie FP.

Thus, False positive rate = FP / (FP+TN). Instead, you wrote

Quote
How the false positive rate is calculated: number of false positive results / total number of negatives.

ie, False positive rate = FP / (FN+TN)

Surely you made a typo? What you are suggesting is wrong. https://en.wikipedia.org/wiki/False_positive_rate

The contention is regarding my treatment of the 2-10% as FP, some believe it's not appropriate. I've avoided using jargons thus far, but fundamentally this is an inference problem using multiple comparisons.

Sorry if I am being rude, but if you can't even define FP rate correctly, I see very little merit in discussing the technical side of this with you. Again, statistics is often counter intuitive. Most people can tell you the odds of flipping a fair coin is 50%, but it can get really ugly real fast.

As the inference remains valid given the relations of the two subsets. The goal now is to justify IF 0.1% serves as a realistic false positive for the not-rapist population so that the final output gets to 5%. Personally I have seen no literature that points to this direction.

No, no mistake on my part.  Just more attempts by you to muddy the waters.  Total negatives include both false positives and true negatives, as both are 'actual' negatives.  From your link...

Quote
...where FP is the number of false positives, TN is the number of true negatives and N=FP+TN is the total number of negatives.

YOU were the one that said total number of negatives,
Quote
How the false positive rate is calculated: number of false positive results / total number of negatives.

total negatives are defined as TN+FN. I am really not sure how this is not getting through.

What then did you mean by total negatives? and how would you write out the equation?

Seriously, you didn't even read the post your responding to or the link that you posted?  Here you go...

False positive rate = number of false positive / total negatives
Total negatives = false positives + true negatives

Same as is described at the link you posted.



When you say total negatives, you are implicitly saying total negatives=True Neg + False Neg, this is not right. Note how the page explicitly defined N being (FP+TN). The correct way is to include ONLY True negatives with FP, NOT false negatives.

So, FP rate = FP / (FP+TN)

Again you said
Quote
False positive rate = number of false positive / total negatives

I am not sure what else to add here. I am going to conclude and end here. Yes I am going to pull a run and hide, provided no one provokes me again. Like Sol said, arguing about stat is too much work, especially against novices. Basically we have 3 kinds of replies here

1. How do you go from 5% to 75%???
These include all the novices that cant seem to wrap their heads around inference/conditional probabilities.

2. Your method doesn't apply because of independence / wrong formulation
It certainly does, this is how inference works.

3. You are misusing 2-10% as fp
I can buy that, makes some sense, but is there any evidence suggesting the fp being as small as 0.1% for the general population?
« Last Edit: October 03, 2018, 04:01:37 PM by anisotropy »

shenlong55

  • Pencil Stache
  • ****
  • Posts: 528
  • Age: 41
  • Location: Kentucky
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1127 on: October 03, 2018, 04:05:02 PM »
Seriously, you didn't even read the post your responding to or the link that you posted?  Here you go...

False positive rate = number of false positive / total negatives
Total negatives = false positives + true negatives

Same as is described at the link you posted.



When you say total negatives, you are implicitly saying total negatives=True Neg + False Neg, this is not right. Note how the page explicitly defined N being (FP+TN). The correct way is to include ONLY True negatives with FP, NOT false negatives.

So, FP rate = FP / (FP+TN)

Again you said
Quote
False positive rate = number of false positive / total negatives

I am not sure what else to add here.

One last time, from your link...

Quote
N=FP+TN is the total number of negatives.

Glenstache

  • Magnum Stache
  • ******
  • Posts: 3493
  • Age: 94
  • Location: Upper left corner
  • FI(lean) working on the "RE"
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1128 on: October 03, 2018, 04:28:43 PM »
Can we just stop with the statistics. As much as I LOVE talking about statistics (really), this is getting way off topic and even if resolved/reconciled isn't really going anywhere.
again...

PDXTabs

  • Walrus Stache
  • *******
  • Posts: 5160
  • Age: 40
  • Location: Vancouver, WA, USA
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1129 on: October 03, 2018, 04:44:11 PM »
Quote
N=FP+TN is the total number of negatives.

I just want to say that I really like the turn that this thread has taken.

anisotropy

  • Pencil Stache
  • ****
  • Posts: 681
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1130 on: October 03, 2018, 04:52:05 PM »
Quote
N=FP+TN is the total number of negatives.

I just want to say that I really like the turn that this thread has taken.

Haha glad you liked it PDXT, much more fun to take a break from partisan politics ya?

Shenlong, that definition (the N=FP+TN equation) was missing in your original post that started this, so don't parade what I told you from the link I provided to pretend that's what you meant all along.

When you mentioned total number of negatives WITHOUT explicitly stating N=FP+TN, you are implicitly saying that
total negatives = true negatives + false negatives. Hope this helps.

partgypsy

  • Walrus Stache
  • *******
  • Posts: 5196
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1131 on: October 03, 2018, 05:07:26 PM »
Just thinking about the entire Ford situation and why I feel conflicted; I have a very good personal, hopefully never, example of its complexity.

I am the father of boy-girl twins.

(Never) Scenario; 10 years from now my son is accused by another teenage girl of pushing her down and possible attempted rape. The next day my daughter tells me a male, approximately their age, did something similar to her. What do I do?

The two extreme responses;
1."Boys will be boys!"
2. Send my son to boarding school in a third world country and never speak to him again. Sue, bury, and destroy the entire family of the other male, who potentially assaulted my daughter, until they are utterly, socially, financially and any other way obliterated.

Many here, on this forum, seem to be adamant to do the second. I think that seems, well, extreme.

In the heat of the moment, dealing with an older teenage boy it would difficult to not do something physical. I would probably take away anything that gave him social access; phone, internet, cars. I might even homeschool him for a semester. But it could come down to the facts of how it happened. Were they horseplaying with other people there, as part of a game or something? Did she not want to be apart of what was going on or did the boy in both situations take their part too far? I would be mad but still try to remember they are teenagers who did something stupid not 25 y.o. I would want to know that the other boy is punished and I would reassure the girl's parents that our son is being punished. Which kind of leads to another problem.

What do you do if the boy's parents do not think anything needs to be done(or just a stern talking to and him promising it won't happen again)? What if the girl's parents say they want harsher punishment?

What if years later(for argument's sake he was given proper punishment) he applies for a job at the business you own?

I have thought of all this before but have had to think about it a little more in the last week or so.

please tell me you don't have children. What you have written is really offensive. I'm imagining a daughter telling a father what happened if she is sexually assaulted, and him telling her that it was possibly horseplay? Possibly a sexual violation? But no one really knows. No. Two people know: the assaulter and the assaulted. I'm thinking from your views, you would side with the boy over your own daughter.

And where in the hell is ANYONE saying in the forum this is what should happen to Kavanaugh "Sue, bury, and destroy the entire family of the other male, who potentially assaulted my daughter, until they are utterly, socially, financially and any other way obliterated." When you create these straw men and misrepresent other people's views, not only is it disrespectful it really looks like you aren't actually reading or understanding what people are saying, and discredits whatever you are attempting to say.
« Last Edit: October 03, 2018, 06:39:01 PM by partgypsy »

sol

  • Walrus Stache
  • *******
  • Posts: 8433
  • Age: 47
  • Location: Pacific Northwest
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1132 on: October 03, 2018, 05:21:19 PM »
Men who want to be allies?
Stop engaging.
Please.
It’s NOT HELPING.

I have difficultly letting counterfactual attacks be used to cause people offense, without speaking up.  Would you prefer the entire world just let the red pill people do their own thing as loudly and publicly as they want to?  From my perspective, arguing with anisotropy about this is like being a counterprotestor at a white supremacy rally.  Man or woman doesn't matter, someone needs to stand up and say "this is not okay." 

But not everyone.  For anyone who is uncomfortable with this topic, by all means please protect yourselves by limiting your exposure to ugly content. 

In fact, for struggling survivors of sexual assault, I recommend you just totally avoid national news for at least the next week.  Shit's about to go down.

anisotropy

  • Pencil Stache
  • ****
  • Posts: 681
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1133 on: October 03, 2018, 05:26:52 PM »
Men who want to be allies?
Stop engaging.
Please.
It’s NOT HELPING.

I have difficultly letting counterfactual attacks be used to cause people offense, without speaking up.  Would you prefer the entire world just let the red pill people do their own thing as loudly and publicly as they want to?  From my perspective, arguing with anisotropy about this is like being a counterprotestor at a white supremacy rally.  Man or woman doesn't matter, someone needs to stand up and say "this is not okay." 

But not everyone.  For anyone who is uncomfortable with this topic, by all means please protect yourselves by limiting your exposure to ugly content. 

In fact, for struggling survivors of sexual assault, I recommend you just totally avoid national news for at least the next week.  Shit's about to go down.

Sol, for what it's worth, I applaud you for this. We may not agree on the way I formulated the case or that what I presented are counterfactual, but your behavior is admirable.

You know what, just to settle this, I will reach out to experts in this field, whether that be Statistics or Sexual violence. Give me sometime, if I am wrong I will surely apologize to you and everyone else. Thanks.

PDXTabs

  • Walrus Stache
  • *******
  • Posts: 5160
  • Age: 40
  • Location: Vancouver, WA, USA
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1134 on: October 03, 2018, 05:28:24 PM »
Men who want to be allies?
Stop engaging.
Please.
It’s NOT HELPING.

I have difficultly letting counterfactual attacks be used to cause people offense, without speaking up.  Would you prefer the entire world just let the red pill people do their own thing as loudly and publicly as they want to?  From my perspective, arguing with anisotropy about this is like being a counterprotestor at a white supremacy rally.  Man or woman doesn't matter, someone needs to stand up and say "this is not okay." 

I would add that while I want to be an "ally" and will literally help you in real life if I get the chance, I also want to live in a world where facts matter. I want to have an administration that believes in facts (like what has Kavanaugh actually done). Math would seem to be an important part of that world.

Glenstache

  • Magnum Stache
  • ******
  • Posts: 3493
  • Age: 94
  • Location: Upper left corner
  • FI(lean) working on the "RE"
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1135 on: October 03, 2018, 05:39:05 PM »
Meanwhile, it looks like a timeline for the FBI report is solidifying along with how the report will be distributed. Bottom line is that one copy of the report will be made available to the entire senate and that it will be available for viewing in one hour increments.
https://thehill.com/homenews/senate/409786-senators-will-view-fbi-report-on-kavanaugh-thursday?__twitter_impression=true&__twitter_impression=true

This process strains credulity and is absolute bullshit. One copy for a body of 100 senators with one hour viewing intervals something on the order of a day before McConnell wants to call the vote. This setup makes it an absolute impossibility for senators to process the information before voting.

Want people to believe that Kavanaugh has nothing to hide? Don't hide the information!

Davnasty

  • Magnum Stache
  • ******
  • Posts: 2793
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1136 on: October 03, 2018, 05:48:31 PM »
Meanwhile, it looks like a timeline for the FBI report is solidifying along with how the report will be distributed. Bottom line is that one copy of the report will be made available to the entire senate and that it will be available for viewing in one hour increments.
https://thehill.com/homenews/senate/409786-senators-will-view-fbi-report-on-kavanaugh-thursday?__twitter_impression=true&__twitter_impression=true

This process strains credulity and is absolute bullshit. One copy for a body of 100 senators with one hour viewing intervals something on the order of a day before McConnell wants to call the vote. This setup makes it an absolute impossibility for senators to process the information before voting.

Want people to believe that Kavanaugh has nothing to hide? Don't hide the information!

Politics under Trump continue to become more and more like reality TV. This sounds like some sort of challenge to the contestants.

nereo

  • Senior Mustachian
  • ********
  • Posts: 17472
  • Location: Just south of Canada
    • Here's how you can support science today:
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1137 on: October 03, 2018, 05:50:46 PM »

This process strains credulity and is absolute bullshit. One copy for a body of 100 senators with one hour viewing intervals something on the order of a day before McConnell wants to call the vote. This setup makes it an absolute impossibility for senators to process the information before voting.


Maybe Mitch is trying to save money on the photocopying budget?

What baffles me is the basic math.  100 senators.  1 copy, 1 hour per senator = 100 hours.  24 hours in a day, 20(ish) hours before the vote is called.... anyone else seeing the problem?

Laserjet3051

  • Pencil Stache
  • ****
  • Posts: 904
  • Age: 95
  • Location: Upper Peninsula (MI)
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1138 on: October 03, 2018, 05:56:15 PM »
Meanwhile, it looks like a timeline for the FBI report is solidifying along with how the report will be distributed. Bottom line is that one copy of the report will be made available to the entire senate and that it will be available for viewing in one hour increments.
https://thehill.com/homenews/senate/409786-senators-will-view-fbi-report-on-kavanaugh-thursday?__twitter_impression=true&__twitter_impression=true

This process strains credulity and is absolute bullshit. One copy for a body of 100 senators with one hour viewing intervals something on the order of a day before McConnell wants to call the vote. This setup makes it an absolute impossibility for senators to process the information before voting.

Want people to believe that Kavanaugh has nothing to hide? Don't hide the information!

Is that really true? I dont know the exact technology they will employ for viewing the report, but it is most certainly not impossible for all to read. Could not all dem senators gather in a single room with a video screen where all eyes can view and thereafter, republican senators do the same? I'm not saying this is how it will work, but it seems like this is a reasonable approach to allow all eyes to read it. If the report is 500 pages, then it is a totally different story. My guess, on the time limitation, is that republicans don't want to risk dems deploying additional delay tactics that might jeopardize the final vote timeline/schedule.

GuitarStv

  • Senior Mustachian
  • ********
  • Posts: 23048
  • Age: 42
  • Location: Toronto, Ontario, Canada
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1139 on: October 03, 2018, 05:57:45 PM »
As a sexual assault victim, this thread sucks balls.
Just FYI.

No shit.

I'm sorry, @hoping2retire35 , but every time you use the word "horseplay" to casually dismiss sexual violation it makes me want to freaking flip tables.
Same.
How about you not get mad at one thing you read and nothing else I wrote?

How more clear can I be other than...(hypothetical scenario)
"My son potentially did something very wrong."

The scenario describes a situation going from(wait for it) horseplay to, possibly a sexual violation; the problem is no one really knows, not the victim not the potential perps.

You kinda keep writing offensive things though.

Like when you earlier wrote that you didn't want Kavenaugh investigated because you felt he didn't look like "an evil and awful dude" - the implication being that you can tell at a glance if a man is a rapist/guilty of sexual assault (which I pointed out didn't work very well for known rapists like Bill Cosby).  Or like in this post where you seem to be implying that neither the victim or aggressor know when sexual assault has occurred.

Kris

  • Walrus Stache
  • *******
  • Posts: 7306
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1140 on: October 03, 2018, 06:00:22 PM »
.

annnnnnooooon

  • 5 O'Clock Shadow
  • *
  • Posts: 7
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1141 on: October 03, 2018, 06:00:54 PM »
Men who want to be allies?
Stop engaging.
Please.
It’s NOT HELPING.

I have difficultly letting counterfactual attacks be used to cause people offense, without speaking up.  Would you prefer the entire world just let the red pill people do their own thing as loudly and publicly as they want to?  From my perspective, arguing with anisotropy about this is like being a counterprotestor at a white supremacy rally.  Man or woman doesn't matter, someone needs to stand up and say "this is not okay." 

But not everyone.  For anyone who is uncomfortable with this topic, by all means please protect yourselves by limiting your exposure to ugly content. 

In fact, for struggling survivors of sexual assault, I recommend you just totally avoid national news for at least the next week.  Shit's about to go down.

Sol, for what it's worth, I applaud you for this. We may not agree on the way I formulated the case or that what I presented are counterfactual, but your behavior is admirable.

You know what, just to settle this, I will reach out to experts in this field, whether that be Statistics or Sexual violence. Give me sometime, if I am wrong I will surely apologize to you and everyone else. Thanks.

Sol, I hear that you are doing what you can to try to address a huge issue, and at the same time, as a woman, what you say here does not feel like allyship at all. Your arguments upthread did, but this does not.

Please listen to this. I realize it sucks, as you are very actively trying to be an ally. I would not have sleuthed through my email trying to find one of my old logins if I did not respect your opinions and your intentions. I even occasionally chose to engage a troll or two to test my communication skills, if I've got the bandwidth and the time. I understand the temptation, and there can be a time and place for it - you often hit that line, and I appreciate your voice. Up until this last post I even felt you were mostly pretty solidly doing awesome.

Women don't get to chose to walk away from what is happening with Kavenaugh. We live it, every day of our lives. We also live the experience of being told to go away in online contexts. I know it is not your intention, and at the same time what I felt when I read your comment, and the thread that came before it, was the following:

1) almost all of the outspoken women dropped out of the conversation several pages ago. This was not a coincidence.
2) The statistics debate is a deliberate derail and consists of a couple red pill bros deliberately reframing the debate to silence it. I wish that the ally men had enough experience with what it is like to be a woman on the internet to see that.
3) YAY! several women directly called it out, now the smart/not awful men will stop feeding it!
4) HOLY FUCK one of the more well spoken non-awful men just accidentally told us women to get off the internet and leave it for the boys to argue so he can argue and rescue all us damsels in distress from the feels and the bad math monsters of the deep, and he has no idea what he just said/implied/tapped into/added to our fuck this shit 2018 overwhelm.

I stopped reading the statistics pages ago, but was increasingly upset by how many allies were letting the derail happen. Your response to being asked to stop feels worse then the red pill nonsense that crept in here, because it comes from someone who is generally worth respecting. #listentowomen. We're trying to tell you something you need to hear - stop engaging with the math, in this place, in this time. There will be another time, probably in this thread, where your voice will be welcome, and supportive. The stats are no longer it. And when women tell you to stop, stop.

shenlong55

  • Pencil Stache
  • ****
  • Posts: 528
  • Age: 41
  • Location: Kentucky
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1142 on: October 03, 2018, 06:03:14 PM »
Haha glad you liked it PDXT, much more fun to take a break from partisan politics ya?

Shenlong, that definition (the N=FP+TN equation) was missing in your original post that started this, so don't parade what I told you from the link I provided to pretend that's what you meant all along.

When you mentioned total number of negatives WITHOUT explicitly stating N=FP+TN, you are implicitly saying that
total negatives = true negatives + false negatives. Hope this helps.

Anisotropy,

I apologize for assuming that you would be familiar with the terminology of the subject that you've been talking about for pages now.  Just FYI, Wikipedia is where I originally got the formula for false positive rate.

Malkynn (and anyone else hurt by this math distraction),

I apologize for any pain I caused you by engaging him, although I do think it is important to fight back against misinformation like this.

Gin1984

  • Magnum Stache
  • ******
  • Posts: 4928
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1143 on: October 03, 2018, 06:19:35 PM »
Men who want to be allies?
Stop engaging.
Please.
It’s NOT HELPING.

I have difficultly letting counterfactual attacks be used to cause people offense, without speaking up.  Would you prefer the entire world just let the red pill people do their own thing as loudly and publicly as they want to?  From my perspective, arguing with anisotropy about this is like being a counterprotestor at a white supremacy rally.  Man or woman doesn't matter, someone needs to stand up and say "this is not okay." 

But not everyone.  For anyone who is uncomfortable with this topic, by all means please protect yourselves by limiting your exposure to ugly content. 

In fact, for struggling survivors of sexual assault, I recommend you just totally avoid national news for at least the next week.  Shit's about to go down.

Sol, for what it's worth, I applaud you for this. We may not agree on the way I formulated the case or that what I presented are counterfactual, but your behavior is admirable.

You know what, just to settle this, I will reach out to experts in this field, whether that be Statistics or Sexual violence. Give me sometime, if I am wrong I will surely apologize to you and everyone else. Thanks.
Dr David Lisak, one of the premier professors on rapists who be someone to listen to.  I wonder what experts you will chose to contact.

sol

  • Walrus Stache
  • *******
  • Posts: 8433
  • Age: 47
  • Location: Pacific Northwest
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1144 on: October 03, 2018, 06:28:54 PM »
Women don't get to chose to walk away from what is happening with Kavenaugh.

Neither do I. 

But when I get exposed to triggering content, and I assure you it happens to all trauma survivors, I look away to protect my own sanity.  I know it's still out there.  The President tweeting about it doesn't help.

Quote
4) HOLY FUCK one of the more well spoken non-awful men just accidentally told us women to get off the internet and leave it for the boys to argue so he can argue and rescue all us damsels in distress from the feels and the bad math monsters of the deep, and he has no idea what he just said/implied/tapped into/added to our fuck this shit 2018 overwhelm.

That was definitely not my intent.  I didn't tell anyone to go away, I suggested that we all self-limit our exposure to awful things if we feel it necessary to do so for our own mental health. 

I'm not trying to rescue anyone, there are no distressed damsels here.

And perhaps most importantly, I did exactly I was asked.  I stopped engaging the stupid stats discussion and instead engaged the self-identified victim who was asking to be heard.  And then someone shouted HOLY FUCK at me anyway.

Gin1984

  • Magnum Stache
  • ******
  • Posts: 4928
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1145 on: October 03, 2018, 06:42:17 PM »
Women don't get to chose to walk away from what is happening with Kavenaugh.

Neither do I. 

But when I get exposed to triggering content, and I assure you it happens to all trauma survivors, I look away to protect my own sanity.  I know it's still out there.  The President tweeting about it doesn't help.

Quote
4) HOLY FUCK one of the more well spoken non-awful men just accidentally told us women to get off the internet and leave it for the boys to argue so he can argue and rescue all us damsels in distress from the feels and the bad math monsters of the deep, and he has no idea what he just said/implied/tapped into/added to our fuck this shit 2018 overwhelm.

That was definitely not my intent.  I didn't tell anyone to go away, I suggested that we all self-limit our exposure to awful things if we feel it necessary to do so for our own mental health. 

I'm not trying to rescue anyone, there are no distressed damsels here.

And perhaps most importantly, I did exactly I was asked.  I stopped engaging the stupid stats discussion and instead engaged the self-identified victim who was asking to be heard.  And then someone shouted HOLY FUCK at me anyway.
Honestly Sol, as a woman and a scientist I appreciated that you spoke up and argued the misuse of the statistics.  I know not all women would/did but some of us did. And just a side note, not all women were not involved in that conversation, I just was at work and could not argue as much as I wanted to.

Sent from my SPH-L720 using Tapatalk


anisotropy

  • Pencil Stache
  • ****
  • Posts: 681
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1146 on: October 03, 2018, 07:02:49 PM »
Men who want to be allies?
Stop engaging.
Please.
It’s NOT HELPING.

I have difficultly letting counterfactual attacks be used to cause people offense, without speaking up.  Would you prefer the entire world just let the red pill people do their own thing as loudly and publicly as they want to?  From my perspective, arguing with anisotropy about this is like being a counterprotestor at a white supremacy rally.  Man or woman doesn't matter, someone needs to stand up and say "this is not okay." 

But not everyone.  For anyone who is uncomfortable with this topic, by all means please protect yourselves by limiting your exposure to ugly content. 

In fact, for struggling survivors of sexual assault, I recommend you just totally avoid national news for at least the next week.  Shit's about to go down.

Sol, for what it's worth, I applaud you for this. We may not agree on the way I formulated the case or that what I presented are counterfactual, but your behavior is admirable.

You know what, just to settle this, I will reach out to experts in this field, whether that be Statistics or Sexual violence. Give me sometime, if I am wrong I will surely apologize to you and everyone else. Thanks.
Dr David Lisak, one of the premier professors on rapists who be someone to listen to.  I wonder what experts you will chose to contact.

Gin,

First I need to address the criticism that my framing and formulation is flawed, I will reach out to local college professors that do researches in Bayesian statistics.

If it turns out I was wrong, then I would stop there. If it turns out my method was correct, then I would talk to sexual violence professionals to determine if I misused the 2-10%, and if so, what would be the appropriate number.

Does that sound like a good plan to you?

Kris

  • Walrus Stache
  • *******
  • Posts: 7306
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1147 on: October 03, 2018, 07:34:52 PM »
Guys, I know the real issue is whether a demonstrably dishonest, partisan politician is being fast-tracked to an incredible position of influence despite a demonstrably credible allegation (or four) of being, at the very least, a skeezball.

But I started a distracting argument about statistics, and it's really important that I convince some internet strangers that I'm smart. So can we work on that first rather than discuss the big picture like adults?

Thank god SOMEONE said it!

Seriously.

former player

  • Walrus Stache
  • *******
  • Posts: 8723
  • Location: Avalon
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1148 on: October 03, 2018, 07:35:08 PM »
Apparently the Republican staff of the Senate Judiciary Committee released a tweet stating that there was no "whiff" in the previous 6 background reports on Kavanaugh of any concerns about alcohol use or sexual impropriety.

Democrats on the Committee have now said 1) given that background reports are confidential this tweet should not have been published, and 2) its contents are inaccurate and should be withdrawn.

That means that there are previous indications in FBI reports about either Kavanaugh's drinking or his sexual behaviour.

The latest FBIi report?  Christopher Wray, Trump appointee as FBI Director, was at Yale Law School at the same time as Brett Kavanaugh  (two years behind, but it's a small school).  I wonder how hard the FBI has been pushing back against any constraints on its investigation.

Oh, and there's a letter in the New York Times from 650 (and counting) law professors saying that Kavanaugh does not have the judicial temperament to sit on the Supreme Court.  Including some from Yale.


« Last Edit: October 03, 2018, 08:16:07 PM by former player »

nereo

  • Senior Mustachian
  • ********
  • Posts: 17472
  • Location: Just south of Canada
    • Here's how you can support science today:
Re: Brett Kavanaguh: Yay or Nay?
« Reply #1149 on: October 04, 2018, 05:06:49 AM »
early reporting indicates the FBI probe was severely curtailed in scope.  It appears they were not allowed to probe the frequency of his drinking and interviewed a relatively limited number of people.