tag:blogger.com,1999:blog-9430835.post2540207267193618652..comments2024-03-28T16:22:14.888-07:00Comments on Steve Sailer: iSteve: Vulcan Society v. Fire Department of New YorkUnknownnoreply@blogger.comBlogger15125tag:blogger.com,1999:blog-9430835.post-21595764217790747762009-07-28T19:30:41.468-07:002009-07-28T19:30:41.468-07:00Bring back quotas! At least whites would be hired ...Bring back quotas! At least whites would be hired and promoted on merit.<br /><br />"Leadership" and whatever the eff else used to benefit NAMs fool people, and whites get selected on bullshit too.robnoreply@blogger.comtag:blogger.com,1999:blog-9430835.post-23986073901000205112009-07-28T06:33:47.827-07:002009-07-28T06:33:47.827-07:00"and hire on a racially balanced basis from t...<i>"and hire on a racially balanced basis from that larger pool."</i><br /><br><br />What's the bet that blacks, were they to make up the majority, yould still insist on AA aka quotas?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9430835.post-28024772855810113442009-07-28T01:52:02.896-07:002009-07-28T01:52:02.896-07:00Steve,
What's the deal with "statistical...Steve,<br /><br />What's the deal with "statistical significance?" Inferential stats are used on SAMPLE DATA to determine the probability of a fluky difference. These tests don't involve SAMPLE DATA. Every guy who wanted to be a fireman took the test. It's not a sample! And BTW, it's not a random sample, either.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9430835.post-9225473311057349822009-07-28T00:30:56.487-07:002009-07-28T00:30:56.487-07:00I guess the issue with strict racial quotas as use...I guess the issue with strict racial quotas as used effectively by eg US Universities and the Navy is that it creates obvious, visible, gaps in ability between favoured and disfavoured groups, which will undermine confidence in members of the favoured groups. Hiring lots of incompetent whites can greatly reduce the visible difference in ability, although of course it also lowers ability overall.Simonnoreply@blogger.comtag:blogger.com,1999:blog-9430835.post-8200089428492311462009-07-27T20:20:22.981-07:002009-07-27T20:20:22.981-07:00The expert report is not online....
The relevant ...<i>The expert report is not online....</i><br /><br />The relevant portions are online, obviously, as they had to be authenticated and included as an exhibit to the MSJ if they were to be considered. See docket no. 264.<br /><br />Get yourself a PACER account. It's free, and you pay just 8 cents a page for Public Access to the Courts' Electronic Records. <br /><br />And when you read it, you will see that the "expert" is an utter nediocrity whose opinions have never been exposed to intelligent cross-examination. Look at this:<br /><br /><i>In practical terms, the effect of the disparate impact...upon Hispanics was to eliminate 292 Hispanics from any possibility of appointment as entry-level firefighters in the FDNY.</i><br /><br />Is this a joke? Why would the "disparate impact" eliminate Hispanics unfairly instead of including whites unfairly? Why isn't the result of the "disparate impact" that a large number of whites were wrongly afforded the "possibility of appointment as entry-level firefighters in the FDNY"? Or that some Hispanics were wrongly excluded from, and some whites were wrongly included in, the group of candidates who had the "possibility of appointment as entry-level firefighters in the FDNY"?<br /><br />These are the mistakes of an extremely lazy 115-IQer, and he's extremely lazy because, when he's involved, the fix is always in.ben tillmannoreply@blogger.comtag:blogger.com,1999:blog-9430835.post-79308723579125894422009-07-27T19:54:07.553-07:002009-07-27T19:54:07.553-07:00Regarding statistical significance and innumeracy,...Regarding statistical significance and innumeracy, on the New Haven captain exam 16 of 25 whites passed, 3 of 8 blacks, and 3 of 8 Hispanics (according to Wikipedia). A chi-square test of homogeneity gives a p-value of over 1/2, so significance wouldn't be attained even at level 0.50, let alone 0.01. <br /><br />That is, even with no differences between the three groups, over half the time you would expect to see larger racial pass rate disparities than in New Haven.<br /><br />However, promotions weren't based on pass rates, but (again according to Wikipedia), who got the top nine scores (no blacks). Once more, even if scores for the 41 applicants were total noise, such as if an applicant's score was his social security number, then over 7% of the time you would still see no blacks among the top nine scorers. Usually you need to be under 5% to impress a judge.<br /><br />Significance is very hard to reach with small sample sizes, and easy to reach with big samples.<br /><br />Maybe small town firefighter unions should embrace Garaufis.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9430835.post-20912040814774613412009-07-27T19:46:03.837-07:002009-07-27T19:46:03.837-07:00This judge may not even be as smart as Sotomayor. ...This judge may not even be as smart as Sotomayor. Look at the opinion:<br /><br /><i>In recent years, black and Hispanic residents of New York City (the “City”) have come to<br /><b>comprise</b> a substantial portion of the City’s <br />population....</i><br /><br />"Comprise" means to be made up of, not to make up; the word is "constitute". Or "be". <br /><br /><i>In 2002, the New York City<br />Department of City Planning identified 25% of the City’s residents as black and 27% of its<br />residents as Hispanic.</i> <br /><br />And, without explanation, the population within the city limits is judged to be the relevant population, even though the financial sector and all sorts of other economic endeavors draw large portions -- or even majorities -- of their workers from outside NYC. <br /><br /><i>These numbers stand in stark contrast to some of the<br />nation’s other large cities, such as Los Angeles, Chicago, Philadelphia, and Houston, where<br />minority firefighters have been represented in significantly higher percentages.</i><br /><br />Just four sentences earlier, this judge informed us that blacks and "Hispanics" were 52% of the city's population, which means that there is no ethnic majority in NYC. And, we have been informed, the composition of the population outside the city is irrelevant. Accordingly, no matter whom the NYFD hires, its work force will be 100% "minority", and it is an absolute impossibility that there is any place in the country "where minority firefighters have been represented in significantly higher percentages". <br /><br /><i>In this case, Plaintiff United States of America (the “Federal Government”) as well as the<br />Vulcan Society, Inc., Marcus Haywood, Candido Nuñez and Roger Gregg (the “Intervenors”),<br />have sued to enforce....</i><br /><br />So the first sentence of the second paragraph of the opinion treats us to a subject/verb disagreement. "USA" has been a singular noun since circa 1865. <br /><br />Next sentence:<br /><br /><i>Specifically, the Federal Government and the Intervenors (“Plaintiffs”)6 challenge the City’s reliance on two written examinations that <b>are</b> used<br />to appoint entry-level firefighters to classes at the New York City Fire Academy (“Academy”).<br />These examinations — Written Examination 7029 and Written Examination 2043 — were<br />administered from 1999 to 2007....</i><br /><br />The judge doesn't know what year it is. The word is "were", not "are".<br /><br />God help us!ben tillmannoreply@blogger.comtag:blogger.com,1999:blog-9430835.post-88960263356226311712009-07-27T19:36:43.026-07:002009-07-27T19:36:43.026-07:00Some recent commenters have said that the Obama ad...Some recent commenters have said that the Obama administration is working hard to get government jobs for black people. <br /><br />Here we see a judge trying to push that same hidden agenda. All of his words are just misdirection and nonsense aimed at concealing a hidden agenda.JWnoreply@blogger.comtag:blogger.com,1999:blog-9430835.post-21141706696013951902009-07-27T18:50:52.710-07:002009-07-27T18:50:52.710-07:00Something is seriously fishy with the statistical ...Something is seriously fishy with the statistical analysis. It looks like the "units of standard deviation" mentioned in the opinion really are standard deviations (seems outrageous, I know)<br /><br />Check out page 26: <br /><br /><i>(quoting Smith, 196 F.3d at 365); <br />Waisome, 948 F.2d at 1376 (“[a] finding of two or three standard deviations (one in 384 chance <br />the result is random) is generally highly probative of discriminatory treatment”</i><br />Indeed, 3 standard deviations from the mean corresponds to approximately a 1 in 384 chance. <br /><br />But on page 18, there is talk of 21.8 standard deviations when the black pass rate is 87.8% of the white. He says that the odds this is by chance are "less than 1 in 4.5 million-billion", when it's actually something like 1 in 10e+105. This is without even getting into how they came up with the 21.8 standard deviations line.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9430835.post-37944625756089690692009-07-27T18:48:14.582-07:002009-07-27T18:48:14.582-07:00Steve:
Thanks. So much for my reading comprehens...Steve:<br /><br />Thanks. So much for my reading comprehension. You have a link in the article leading to the tests in totam.<br /><br />After skimming them, the term "dumbed down" SAT is an extreme overstatement.<br /><br />"How many emergency vehicles are in the picture?"<br /><br />"What's the address of the building in the picture?"<br /><br />What a joke. How can someone graduate Jr. High much less High School and have difficulty with this test?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9430835.post-45647723485119651512009-07-27T18:39:28.551-07:002009-07-27T18:39:28.551-07:00"he says that disparate impact is established..."he says that disparate impact is established any time the racial gap is more than 3 standard deviations"<br /><br />A dozen or so years ago I asked poker author Mason Malmuth a question on his forum, can't remember what. He replied something like "first, assume everything in the universe is withing 3 standard deviations". <br /><br />It's been a long time since I took math. <br /><br />I used to work in a casino as a pit boss and we'd do hourly action counts, add up what is being wagered, assume an x percent house edge, and estimate what we should be making. It was explained to me by the boss, a legit math guy, that anything south of two standard deviations over a significant period, you're probably getting ripped off by dealers and/or players, and if it's 3 standard deviations you *are* getting ripped off. <br /><br />Sort of the same principle I'm guessing is being applied here, that 3 standard deviations precludes luck being a factor in the disparity. Math guys feel free to elaborate.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9430835.post-20123151169025465782009-07-27T17:48:13.385-07:002009-07-27T17:48:13.385-07:00"I have a friend who tells me the test is sim..."I have a friend who tells me the test is simply a "dumbed down" version of the SAT with no prior knowledge of firefighting and related topics necessary. At the same time I keep hearing "fire buffs" are skewing the scores for "everyone else." There's a definite disconnect here."<br /><br />The explanation is that you can pass the test in one of two ways:<br /><br />1. All the information needed to answer the questions is in reading passages above. So, anybody with good reading comprehension can walk in cold and pass the test.<br /><br />2. If you aren't that good at reading, you can study ahead of time and learn much of the material asked about so that your poor reading skills won't be a problem.<br /><br />However, if you are bad at reading and you aren't interested enough in the vocation of firefighting to study ahead of time, you probably won't do well on the test.Steve Sailerhttps://www.blogger.com/profile/11920109042402850214noreply@blogger.comtag:blogger.com,1999:blog-9430835.post-27390901811294685532009-07-27T17:40:45.410-07:002009-07-27T17:40:45.410-07:00"The expert report is not online, and I don’t..."The expert report is not online, and I don’t know what a “unit of standard deviation” is. Obviously the black result is not 34 standard deviations below the white result."<br /><br />This probably means a t-statistic of 34. Given the large sample size, virtually all t-statistics will be highly significant.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9430835.post-58986812083398794462009-07-27T17:40:20.248-07:002009-07-27T17:40:20.248-07:00Steve:
It would be interesting to see actual samp...Steve:<br /><br />It would be interesting to see actual sample test questions.<br /><br />I have a friend who tells me the test is simply a "dumbed down" version of the SAT with no prior knowledge of firefighting and related topics necessary.<br /><br />At the same time I keep hearing "fire buffs" are skewing the scores for "everyone else." There's a definite disconnect here.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-9430835.post-69065993804217717092009-07-27T17:20:47.779-07:002009-07-27T17:20:47.779-07:00This cannot be real.This cannot be real.AmericanGoyhttps://www.blogger.com/profile/00865892490752172185noreply@blogger.com