Re: Steerzing in a New Direction...
Posted: Wed Nov 17, 2021 1:30 pm
Steer's if I have a stroke right now I will blame you! (or Matt for tricking me into reading that shitty Collapse book)!
Exposing the stupidity, lies, and hypocrisy of Social Justice Warriors since July 2012
http://slymepit.com/phpbb/
:)
Ok; I stand corrected - more or less ... ;)
Kinda think you're missing the point. That some samples are not perfectly normal does not mean that the whole population is likewise not normal - sampling theorems and all that:
How much more or less? I go back to John D's statement. The curve isn't normal, it's "sort of bell shaped". Often we care so little about accuracy that we call "sort of bell shaped" normal and act as if it was, but it isn't normal.
Generally speaking, it doesn't greatly matter that "the curve isn't normal". As long as it has the rough shape of normality - as is clearly the case for IQs and heights and many other physical and psychological traits ["gender"] - one can still make useful inferences. For example, the "agreeableness" of men versus women:
https://en.wikipedia.org/wiki/Sampling_distributionSampling distributions are important in statistics because they provide a major simplification en route to statistical inference. More specifically, they allow analytical considerations to be based on the probability distribution of a statistic, rather than on the joint probability distribution of all the individual sample values.
Steers - my friendSteersman wrote: ↑Generally speaking, it doesn't greatly matter that "the curve isn't normal". As long as it has the rough shape of normality - as is clearly the case for IQs and heights and many other physical and psychological traits ["gender"] - one can still make useful inferences. For example, the "agreeableness" of men versus women:
PMC_Frontiers_Weisberg_DeYoung_Hirsh_GenderDiffs2B.jpg
Definitely not normal distributions, but one can still see that, on average (the peaks) women have a higher "agreeableness" score than men.
But that's the whole principle behind sampling - we can't possibly measure everyone's heights, weights, IQs, and agreeableness; we have to take a sample - the bigger the sample, the more random the selection, the more likely that the sample will accurately reflect the population. See:
https://en.wikipedia.org/wiki/Sampling_distributionSampling distributions are important in statistics because they provide a major simplification en route to statistical inference. More specifically, they allow analytical considerations to be based on the probability distribution of a statistic, rather than on the joint probability distribution of all the individual sample values.
hung.... hung... hungMatt Cavanaugh wrote: ↑ No verdict today. Jury back in the morning.
Dunno how long Schroeder will let this play out, cuz there will be no verdict ever.
BP is just one variable I monitor. I’m not going to grapple with bulbs and stethoscopes. I use a cuff and machine.John D wrote: ↑Wed Nov 17, 2021 1:26 pmWas Socrates married?... and if he was married to Steers would he be a philosopher?
One should never use a normal distribution to prove the fairness of dice. There are better statistical tests for this. This is an example of my point. That someone would use a normal distribution to prove the fairness of a pair of dice is using the wrong hammer.
And has anyone seen actual large sample IQ data and tested it for normalcy? It doesn't look very normal to me. And how do these IQ testers get their population? So many questions.... so much bullshit.
Hmmm... anyway... most of may rants are just rants. Please ignore if you are so inclined.
Brive - do you take you own bp and if you do how do you do it? Do you use a machine or are you able to use a manual cuff? My life is in the balance!
Anything to get away from the wife, eh? ;)
https://vultureofcritique.files.wordpre ... vipers.pdfMOM IS THE END PRODUCT OF SHE. She is Cinderella, the creature I discussed earlier, the shining-haired, the starry-eyed, the ruby-lipped virgo aeternis, of which there is presumably one, and only one, or a one-and-only for each male, whose dream is fixed upon her deflowerment and subsequent perpetual possession. This act is a sacrament in all churches and a civil affair in our society. The collective aspects of marriage are thus largely compressed into the rituals and social perquisites of one day. Unless some element of mayhem or intention of divorce subsequently obtrudes, a sort of privacy engulfs the union and all further developments are deemed to be the business of each separate pair, including the transition of Cinderella into mom, which, if it occasions any shock, only adds to the huge, invisible burthen every man carries with him into eternity. It is the weight of this bundle which, incidentally, squeezes out of him the wish for death, his last positive biological resource.
The list of prosecution infractions is growing. Latest rumour is that Jump Kick tried to get existing charges dropped in exchange for testifying, which would be very inconvenient if proven true because the prosecution deny knowing his identity. Judge should have knocked proceedings on the head with the right to silence violation so I don't see how he can honestly deny an updated motion. My current theory is that Little Binger bet the farm with an eye on the DA spot and a future in Democrat politics. I can't explain the rotund sidekick's suicide attempt though.Matt Cavanaugh wrote: ↑Wed Nov 17, 2021 9:25 amJudge's hate hung juries, and will lean pretty hard on a jury to avoid one. But pretty much everyone but him and the defense would love one at this point.
Although an acquittal would give Joy Reid more material, and rioters a better excuse.
I'd bet on either an acquittal following the dismissal of one or more jurors, or a mistrial with prejudice. By end of day.
Source?ThreeFlangedJavis wrote: ↑ Latest rumour is that Jump Kick tried to get existing charges dropped in exchange for testifying, which would be very inconvenient if proven true because the prosecution deny knowing his identity.
Binger ran for DA a few years ago in Racist Racine County next door, as a Dem, and lost. He flailed for a while in private practice before joining Kenosha DA office just a few months before Kyle was charged. Kraus seems to have worked there for ever.My current theory is that Little Binger bet the farm with an eye on the DA spot and a future in Democrat politics. I can't explain the rotund sidekick's suicide attempt though.
How so? Which graphs? This one?John D wrote: ↑Steers - my friendSteersman wrote: ↑ <snip>
Generally speaking, it doesn't greatly matter that "the curve isn't normal". As long as it has the rough shape of normality - as is clearly the case for IQs and heights and many other physical and psychological traits ["gender"] - one can still make useful inferences. For example, the "agreeableness" of men versus women:
PMC_Frontiers_Weisberg_DeYoung_Hirsh_GenderDiffs2B.jpg
Definitely not normal distributions, but one can still see that, on average (the peaks) women have a higher "agreeableness" score than men.
<snip>
You just proved my point. You can make many conclusions from these graphs.
Where the fuck have I said that that one is a normal distribution? All I've said is that it's a close approximation to a normal distribution and that, as is typical with sampling distributions, one can still make useful inferences from such samples.
So fucking what? Is it true or not that the averages - the peaks - suggest that females are, on average, more agreeable than men? That that particular sample justifies concluding that there are more men - in that sample - with low agreeableness (between 2 and 4) than there are women? That, in that sample, there are more women in the high agreeableness range (between 4 and 5) than there are men?
"normalizing" is an entirely different kettle of fish from "normal distribution":John D wrote: ↑ Normalizing this data wouldn't help in many cases. Now... you could claim the average is different.... you don't need to normalized the data for that. But to put a standard deviation on this. well... I don't even know what it would help with. And part of the problem is the discrete nature of this data. It can't be normal. You can never have a score over 5... right? Even this graph suggests you can go over 5.... but I suspect that is an error in the plot. I doubt this is even a plot of the raw data. What in the fuck is this really?
You're either being sloppy or you don't understand the difference; nobody is trying to "normalize" that data. All that is being attempted is to infer something about the two populations being considered. Is the average, for the sample, for females higher than for males? Is it likely or not that the entire population will show pretty much the same distribution and means?Normalize; mathematics: multiply (a series, function or item of data) by a factor that makes the norm or some associated quantity such as an integral equal to a desired value (usually 1).
"both sets of data have been normalized such that the lowest value is equal to 1"
You're really not succeeding very well at not being pedantic. Or you really don't understand the statistics or their applications to "psychometrics" or even various physiological traits as well as you think you do.John D wrote: ↑I am not trying to be pedantic.... but this has the risk of really misrepresenting this data set. Can you score more than 5? If not... you should try to avoid a normal distribution model. The tails are fucked.
I don't know. I guess... I am just tired of smart people making everything they work with normal.... and then making some fucked up claim (especially about the tails). IT MUST BE STOPPED!!!!!
The judge … suggested the possibility of putting the prosecution on the witness stand themselves to testify under oath as to how this video mix-up occurred.
Interestingly, a photo purportedly of ADA Kraus’ laptop screen being projected onto the large 4k TV screen in the courtroom shows the presence of an imaging software named Handbrake. I lack any technical expertise or experience with Handbrake software, but apparently, a core purpose is to generate low-definition versions of high-definition videos. This has led to speculation that the provision of the low-resolution video to the defense was not an accident at all, but rather an intentional act of the prosecution. If so, this would be prosecutorial misconduct sufficiently grievous to not only warrant a mistrial with prejudice but certainly misconduct perhaps even malicious prosecution charges.
Game over, man. Game over.Amendment VI wrote:In all criminal prosecutions, the accused shall enjoy the right to ... be confronted with the witnesses against him....
Hard-hearted response... not necessarily literal advice... just attempting to commiserate:
This may be a bit outside-of-the-box, but have you considered pimping her out until it's paid off?John D wrote: ↑ Jesus fucking christ.... my daughter just told me she can't make her RV payments (the RV I am cosigned for).... and wants me to pay the next two $800 payments. This is the daughter that one year ago told me she would rather sell herself on the street than ask for help. God damn. What a fucking week.
:) I was thinking of something along the same line, of even "honouring her offer" - so to speak - if it weren't for the distance at least ... ;)Matt Cavanaugh wrote: ↑This may be a bit outside-of-the-box, but have you considered pimping her out until it's paid off?John D wrote: ↑ Jesus fucking christ.... my daughter just told me she can't make her RV payments (the RV I am cosigned for).... and wants me to pay the next two $800 payments. This is the daughter that one year ago told me she would rather sell herself on the street than ask for help. God damn. What a fucking week.
Sad to see. But think you need to take up meditation or something - you look to be going around in ever diminishing circles.John D wrote: ↑ Jesus fucking christ.... my daughter just told me she can't make her RV payments (the RV I am cosigned for).... and wants me to pay the next two $800 payments. This is the daughter that one year ago told me she would rather sell herself on the street than ask for help. God damn. What a fucking week.
When you're up to your ass in alligators it's often rather difficult to think rationally about much of anything. Much less that one DID set out to drain the swamp ... ;)
Seems to me that John is talking out of both sides of his mouth. Or wants to have his cake and eat it too. He has clearly and unambigously asserted the general usefulness of "the tool":Service Dog wrote: ↑Nope. I am saying the tool is defective... in exactly the way John said it was.
However, he seems rather reluctant to even consider that there are significant differences - on average - between men and women, not just physiological but psychological as well. Which even Fafnir accepts:John D wrote: ↑
I imagine that height is sort of normally distributed... but it is a broad approximation.
<Begin rant on normal distributions>
To start... remember... "When all you have is a hammer, everything looks like a nail"
THE VERY USEFUL NORMAL DISTRIBUTION - Many sets of data are roughly normally distributed. It is a super useful tool. I use it in my work on a regular basis and it can help predict the nature of a full data set from a smaller sample. I am not saying it is not a really great hammer. It is a most excellent hammer. But... it should be used properly.
Do you yourself accept those differences - on average - or not? Try thinking about the "joint probability distributions" I've provided for both heights and "agreeableness"; let me know if you need some assistance to get over those hurdles ... ;)fafnir wrote: ↑Tue Nov 02, 2021 3:42 pmIt isn't a question of denying that males and females have, on average, psychological differences. That's obvious and sane people have known it since the dawn of time. ....Steersman wrote: ↑
<snip>
But I'll more or less readily agree with you about "linguistic games". But that does not mean that there isn't some value in the concept of gender as personalities and the stereotypes which are derived from them. Why there's some value in trying to define the term as accurately as possible, why there's some value in clearly differentiating between the biological and the psychological.
https://media.patriots.win/post/lYAaaB9O.jpegMatt Cavanaugh wrote: ↑Sat Nov 13, 2021 2:54 pmBut I wasn't paying much attention to the testimony, cuz she was absolutely gorgeous.
Hmm 2 out of the required 3 for a diagnosis of metabolic syndrome.
“More than if you did it properly. Less than if you just did a 5 mile run”Service Dog wrote: ↑I'd be curious to know-- how much variance you get if you intentionally do it 'wrong'.
Like-- take your bp while flat in bed. Or right after 2 cups of coffee.
What has being able to make inferences based on the shape of the curve got to do with justifying pretending it's normally distributed?Steersman wrote: ↑Generally speaking, it doesn't greatly matter that "the curve isn't normal". As long as it has the rough shape of normality - as is clearly the case for IQs and heights and many other physical and psychological traits ["gender"] - one can still make useful inferences. For example, the "agreeableness" of men versus women:
PMC_Frontiers_Weisberg_DeYoung_Hirsh_GenderDiffs2B.jpg
Definitely not normal distributions, but one can still see that, on average (the peaks) women have a higher "agreeableness" score than men.
As we both know, they almost certainly count the infertile in the data, and so clumsily include them in those graphs. The data is therefore hopelessly compromised.
It is helpful... it moves my thinking forward a couple babysteps. But it's not helpful-enough to quench my curiosity.Brive1987 wrote: ↑“More than if you did it properly. Less than if you just did a 5 mile run”Service Dog wrote: ↑
I'd be curious to know-- how much variance you get if you intentionally do it 'wrong'.
Like-- take your bp while flat in bed. Or right after 2 cups of coffee.
This may not be a helpful response.
You found a source then. I'd forgotten where I heard the rumours.Matt Cavanaugh wrote: ↑Thu Nov 18, 2021 8:52 amThe 'hero' Jump Kick Man: YA career criminal and domestic abuser:
https://www.dailymail.co.uk/news/articl ... ealed.html
You're grabbing at straws; you're a better man than that - Gunga Din ;)
FFS, I'm not "pretending those curves (& distributions) are normally distributed". I'm saying that they're a sufficiently close approximation to normal distributions to make reasonable inferences about the entire populations. There ARE some reasons for measuring populations, although some are maybe questionable or the results are put to unjustified ends or untenable objectives.What has being able to make inferences based on the shape of the curve got to do with justifying pretending it's normally distributed?
Note the "roughly normally distributed", "broad approximation", "predict the nature of a full data set", "excellent hammer", and "should be used properly". But you can't really say whether someone is using that tool properly, whether someone is trying to pull the wool over your eyes - lies, damned lies, and statistics - if you don't understand anything about the process or anything about the reasons behind it.John D wrote: ↑
I imagine that height is sort of normally distributed... but it is a broad approximation.
<Begin rant on normal distributions>
To start... remember... "When all you have is a hammer, everything looks like a nail"
THE VERY USEFUL NORMAL DISTRIBUTION - Many sets of data are roughly normally distributed. It is a super useful tool. I use it in my work on a regular basis and it can help predict the nature of a full data set from a smaller sample. I am not saying it is not a really great hammer. It is a most excellent hammer. But... it should be used properly.
I haven't seen you make any inferences that have anything to do with the distribution being normal, or approximately normal. I think you should chuck out the whole normal claim. All you want to say is covered by things that can simply be observed from the distribution curves.Steersman wrote: ↑FFS, I'm not "pretending those curves (& distributions) are normally distributed". I'm saying that they're a sufficiently close approximation to normal distributions to make reasonable inferences about the entire populations. There ARE some reasons for measuring populations, although some are maybe questionable or the results are put to unjustified ends or untenable objectives.
Oh. If I knew a belt wank was required then I’d have found some spoons.Service Dog wrote: ↑It is helpful... it moves my thinking forward a couple babysteps. But it's not helpful-enough to quench my curiosity.Brive1987 wrote: ↑“More than if you did it properly. Less than if you just did a 5 mile run”Service Dog wrote: ↑
I'd be curious to know-- how much variance you get if you intentionally do it 'wrong'.
Like-- take your bp while flat in bed. Or right after 2 cups of coffee.
This may not be a helpful response.
So now you gotta drink a pot of coffee, run 5 miles, then pepper-spray yourself in the face, & measure your bp while wanking-off with a belt around your neck.
(or heck... just walk 5 miles from your house/ let the cops pepper spray you/ then chase you back home.)
You're the healthiest guy we got here. So whatever numbers you report... as long as the rest of us stay under-that... while sitting around doing nothing & nibbling bon-bons...
we should be fine.
Right?
science.
The only thing you would apparently allow me to say is that the distribution curves describe only the samples on which they're based. But the whole point of sampling is to be able to make inferences about the whole population:fafnir wrote: ↑I haven't seen you make any inferences that have anything to do with the distribution being normal, or approximately normal. I think you should chuck out the whole normal claim. All you want to say is covered by things that can simply be observed from the distribution curves.Steersman wrote: ↑FFS, I'm not "pretending those curves (& distributions) are normally distributed". I'm saying that they're a sufficiently close approximation to normal distributions to make reasonable inferences about the entire populations. There ARE some reasons for measuring populations, although some are maybe questionable or the results are put to unjustified ends or untenable objectives.
https://en.wikipedia.org/wiki/Sampling_(statistics)In statistics, quality assurance, and survey methodology, sampling is the selection of a subset (a statistical sample) of individuals from within a statistical population to estimate characteristics of the whole population. Statisticians attempt to collect samples that are representative of the population in question. Sampling has lower costs and faster data collection than measuring the entire population and can provide insights in cases where it is infeasible to sample an entire population.
Nonsense. You were talking about where the peaks were, what shape the distributions were, that the distribution for the men's heights was to the right of the women's. I'm not trying to restrict you. I'm telling you that nothing you said about the distribution curves requires you to claim they are normal. I don't know why it keeps being claimed.
In what way do I, in particular, resemble the virulent anti-science, anti-intellectual sentiment driving many of the professors staff & students of various Women's Studies programs?Steersman wrote: ↑ You and Service Dog, the latter in particular, remind me of the "virulent anti-science, anti-intellectual sentiment driving many of the professors, staff and students" of various Women Studies programs that were the subject of the book Professing Feminism:
https://www.feministcritics.org/blog/20 ... inism-noh/
Do you understand what it means to say that a distribution curve is normal?fafnir wrote: ↑Nonsense. You were talking about where the peaks were, what shape the distributions were, that the distribution for the men's heights was to the right of the women's. I'm not trying to restrict you. I'm telling you that nothing you said about the distribution curves requires you to claim they are normal. I don't know why it keeps being claimed.
You insisted earlier that "the tool is defective... in exactly the way John said it was". And refused to answer my questions and address my arguments that John was more or less blowing smoke out of his arse. And/or that you were seriously misinterpreting what he was saying.Service Dog wrote: ↑In what way do I, in particular, resemble the virulent anti-science, anti-intellectual sentiment driving many of the professors staff & students of various Women's Studies programs?Steersman wrote: ↑ You and Service Dog, the latter in particular, remind me of the "virulent anti-science, anti-intellectual sentiment driving many of the professors, staff and students" of various Women Studies programs that were the subject of the book Professing Feminism:
https://www.feministcritics.org/blog/20 ... inism-noh/
Yea, I did. In some detail:Service Dog wrote: ↑Your claim is impossible to evaluate or heed, because you haven't actually stated it.
Of particular relevance is the question of whether men and women show, on average, some significant differences for various physiological and psychological traits. You might try answering that one.Steersman wrote: ↑Seems to me that John is talking out of both sides of his mouth. Or wants to have his cake and eat it too. He has clearly and unambigously asserted the general usefulness of "the tool":Service Dog wrote: ↑Nope. I am saying the tool is defective... in exactly the way John said it was.
<snip>
Do you yourself accept those differences - on average - or not? Try thinking about the "joint probability distributions" I've provided for both heights and "agreeableness"; let me know if you need some assistance to get over those hurdles ... ;)
But they're really not all that difficult if you give them a bit of thought.
Thos graphs doesn't show normally distributed data. They aren't symmetrical. They seem to have a longer tail on the left than the right and the male one seems to have two peaks.Steersman wrote: ↑ Do you understand what it means to say that a distribution curve is normal?
https://en.wikipedia.org/wiki/Normal_distribution
It has a particular equation that describes it. The properties of which justify particular conclusions.
Do those ones on agreeableness for "males" and "females" have more or less the same shape?
The height graph is calculated from a normal distribution? So it isn't actually a distribution of real heights? As I was saying, intuitively height isn't going to be normally distributed. I don't think you are making any claims that require you to insist the distribution is normal anyway. That's my point.Steersman wrote: ↑But the ones for heights are more or less calculated from that equation so they're normal to begin with. But I seem to recollect I created them based on fairly common demographic measurements that have apparently been repeated many times; it's not like I'm pulling those graphs out of my nether regions.
You don't need to incorrectly state that non-symmetric distributions with two peaks are normal to make the arguments you are making.Steersman wrote: ↑Not quite sure why you balk at accepting that the "agreeableness" graphs - and those for many other traits - are more or less normal (specific equation). And that the populations they describe are more or less likewise. "Virulent anti-science and anti-intellectual sentiments"? :think: :roll:
You have entirely failed to connect any of this to "the virulent anti-science, anti-intellectual sentiment driving many of the professors staff & students of various Women's Studies programs". You assert there's some similarity, without actually saying what is supposedly similar.Steersman wrote: ↑You insisted earlier that "the tool is defective... in exactly the way John said it was". And refused to answer my questions and address my arguments that John was more or less blowing smoke out of his arse. And/or that you were seriously misinterpreting what he was saying.Service Dog wrote: ↑In what way do I, in particular, resemble the virulent anti-science, anti-intellectual sentiment driving many of the professors staff & students of various Women's Studies programs?Steersman wrote: ↑ You and Service Dog, the latter in particular, remind me of the "virulent anti-science, anti-intellectual sentiment driving many of the professors, staff and students" of various Women Studies programs that were the subject of the book Professing Feminism:
https://www.feministcritics.org/blog/20 ... inism-noh/
Yea, I did. In some detail:Service Dog wrote: ↑Your claim is impossible to evaluate or heed, because you haven't actually stated it.
Of particular relevance is the question of whether men and women show, on average, some significant differences for various physiological and psychological traits. You might try answering that one.Steersman wrote: ↑Seems to me that John is talking out of both sides of his mouth. Or wants to have his cake and eat it too. He has clearly and unambigously asserted the general usefulness of "the tool":Service Dog wrote: ↑Nope. I am saying the tool is defective... in exactly the way John said it was.
<snip>
Do you yourself accept those differences - on average - or not? Try thinking about the "joint probability distributions" I've provided for both heights and "agreeableness"; let me know if you need some assistance to get over those hurdles ... ;)
But they're really not all that difficult if you give them a bit of thought.
But if you seriously think that that tool is "defective" then you may wish to write something up proving your case and submit it to various statistical journals because its use is ubiquitous and of a great deal of value in various hard and soft sciences. I'm sure there's a Nobel prize waiting for you if you can manage that ... :roll:
You seem not to understand "more or less the same shape".fafnir wrote: ↑Those graphs doesn't show normally distributed data. They aren't symmetrical. They seem to have a longer tail on the left than the right and the male one seems to have two peaks.Steersman wrote: ↑ Do you understand what it means to say that a distribution curve is normal?
https://en.wikipedia.org/wiki/Normal_distribution
It has a particular equation that describes it. The properties of which justify particular conclusions.
Do those ones on agreeableness for "males" and "females" have more or less the same shape?
Again, not having a clue what "more or less the same shape means". You might try Googling "height distributions", this link and graph seems typical and is presumably based on real-world data:fafnir wrote: ↑The height graph is calculated from a normal distribution? So it isn't actually a distribution of real heights? As I was saying, intuitively height isn't going to be normally distributed. I don't think you are making any claims that require you to insist the distribution is normal anyway. That's my point.Steersman wrote: ↑But the ones for heights are more or less calculated from that equation so they're normal to begin with. But I seem to recollect I created them based on fairly common demographic measurements that have apparently been repeated many times; it's not like I'm pulling those graphs out of my nether regions.
FFS, I am not saying those samples are normal. I'm saying that they're a close enough approximation to make the argument that the relevant populations are more or less normal themselves. Try looking at what John D. was saying as, in his better moments, he's not saying much that's different from what I'm saying.fafnir wrote: ↑You don't need to incorrectly state that non-symmetric distributions with two peaks are normal to make the arguments you are making.Steersman wrote: ↑Not quite sure why you balk at accepting that the "agreeableness" graphs - and those for many other traits - are more or less normal (specific equation). And that the populations they describe are more or less likewise. "Virulent anti-science and anti-intellectual sentiments"? :think: :roll:
The "tool" we were talking about was normal distributions. Which you said was "defective". Maybe you can cut to the chase, can clearly and unambiguously state whether that particular tool is defective or not?Service Dog wrote: ↑You have entirely failed to connect any of this to "the virulent anti-science, anti-intellectual sentiment driving many of the professors staff & students of various Women's Studies programs". You assert there's some similarity, without actually saying what is supposedly similar.Steersman wrote: ↑
<snip>
Of particular relevance is the question of whether men and women show, on average, some significant differences for various physiological and psychological traits. You might try answering that one.
But if you seriously think that that tool is "defective" then you may wish to write something up proving your case and submit it to various statistical journals because its use is ubiquitous and of a great deal of value in various hard and soft sciences. I'm sure there's a Nobel prize waiting for you if you can manage that ... :roll:
That's not all there is to IQ tests:Service Dog wrote: ↑ I had zero participation in your back & forth with John.
I read about his wife's doctor's statistical ignorance. I read about the lifespan of drill bits. I read about IQ test score distributions.
I noted that IQ test questions-- are often exercises in treating imperfect metaphors-- as-if they perfectly match a discernible, abstracted, pattern template. And the more tenuous the match/ the more that's treated as evidence of higher IQ (a more difficult question); rather than evidence of the apparent match being a false positive... and the person seeing the match... being delusional.
https://en.wikipedia.org/wiki/Intellige ... factor_(g)The many different kinds of IQ tests include a wide variety of item content. Some test items are visual, while many are verbal. Test items vary from being based on abstract-reasoning problems to concentrating on arithmetic, vocabulary, or general knowledge.
You SAID that the tool was defective. And you've rather clearly indicated that the tool you had in mind was normal distributions and the inferences that can be drawn from them:Service Dog wrote: ↑In no way did I engage with your boring new attempt to prove your same old boring shit.
http://slymepit.com/phpbb/viewtopic.php ... 80#p506305Hmm... "White is even-lighter than yellow, therefore it's further-away from poisonous, if these are a normal distribution, so I will eat it!" <--Rest In Peace.
You've been pretty clear that you're looking down your nose on normal distributions and their uses. Not sure what's more of a smoking gun when it comes to anti-scientific attitudes.Service Dog wrote: ↑My only involvement has been to correct you-- when you said I was 'suggesting' something I wasn't suggesting. And to ask what the fuck you were alluding to, when you invoked my name & said I was similar-to the anti-scientific attitudes of Women's Studies... in some unstated way.
Probably unwilling or unable because you know you don't have a leg to stand on.Service Dog wrote: ↑I'm not "unwilling or unable to answer your question" about gender differences... I Wasn't Part Of That Conversation At-All, and I'm still not.
You ever do any gambling? Play the odds one way or another? That's pretty much all of what normal distributions are all about, quantifying various probabilities and the inferences that can be drawn from "the measurables of reality"; whole field of statistics has its roots in the study of gambling:Service Dog wrote: ↑And, no, I'm not gonna get a Nobel Prize for pointing-out the gap between an abstract platonic-ideal model curve... vs. measurables of reality.
https://en.wikipedia.org/wiki/Statistics#HistoryMathematical probability theory arose from the study of games of chance ...
No, it goes back at least to 1931:Service Dog wrote: ↑"The map is not the territory" is not a novel thought.
Your argument doesn't require them to be normal.Steersman wrote: ↑ FFS, I am not saying those samples are normal. I'm saying that they're a close enough approximation to make the argument that the relevant populations are more or less normal themselves. Try looking at what John D. was saying as, in his better moments, he's not saying much that's different from what I'm saying.
And your evidence and argument for that is what? You might try expending a bit of thought and effort to justify your claims. Who knows? You or I - or both of us - might learn something ... ;)fafnir wrote: ↑Your argument doesn't require them to be normal.Steersman wrote: ↑ FFS, I am not saying those samples are normal. I'm saying that they're a close enough approximation to make the argument that the relevant populations are more or less normal themselves. Try looking at what John D. was saying as, in his better moments, he's not saying much that's different from what I'm saying.
https://en.wikipedia.org/wiki/Sampling_%28statistics%29In statistics, quality assurance, and survey methodology, sampling is the selection of a subset (a statistical sample) of individuals from within a statistical population to estimate characteristics of the whole population. Statisticians attempt to collect samples that are representative of the population in question. Sampling has lower costs and faster data collection than measuring the entire population and can provide insights in cases where it is infeasible to sample an entire population.