- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- Fantasy Sports
- Golf Board
- Soccer Board
- O-T Lounge
- Tech Board
- Home/Garden Board
- Outdoor Board
- Health/Fitness Board
- Movie/TV Board
- Book Board
- Music Board
- Political Talk
- Money Talk
- Fark Board
- Gaming Board
- Travel Board
- Food/Drink Board
- Ticket Exchange
- TD Help Board
Customize My Forums- View All Forums
- Show Left Links
- Topic Sort Options
- Trending Topics
- Recent Topics
- Active Topics
Started By
Message
re: "Half the schools are below average" - not always true
Posted on 10/1/14 at 10:41 am to SpidermanTUba
Posted on 10/1/14 at 10:41 am to SpidermanTUba
quote:
You'll still assert the distribution of differences in performances of different schools is due entirely to random selection of students.
Distribution of differences in performance is NOT the same thing as the distribution of performances.
You can't even correctly describe what is being measured and analyzed.
It's pathetic.
You'll have to tell me when your defense is...I'll enjoy coming and peppering your idiocy with questions you won't be able to answer
1
Posted on 10/1/14 at 5:05 pm to CptBengal
quote:
You can't even correctly describe what is being measured and analyzed.
Sure I can. We're measuring how many schools fall below average. Look at the OP. Its not that hard to grasp.
You're asserting that for large enough numbers of schools, half will fall below average. You are basing this on the central limit theorem which applies to random samples - however, the samples are not random, meaning you are wrong. (Not surprising from a guy who thinks ice has no mass.)
Clearly - if a state desired to - it could create a system whereby 90% of the schools performed below average - simply by putting 100% of the resources into 10% of the schools and putting 0% into the remaining 90%. It can do this regardless how many schools. You are asserting this isn't mathematically possible to do given a large enough number of schools. That's hilarious.
This post was edited on 10/1/14 at 5:11 pm
Posted on 10/1/14 at 5:17 pm to SpidermanTUba
quote:
Sure I can. We're measuring how many schools fall below average. Look at the OP. Its not that hard to grasp.
You're asserting that for large enough numbers of schools, half will fall below average. You are basing this on the central limit theorem which applies to random samples - however, the samples are not random, meaning you are wrong. (Not surprising from a guy who thinks ice has no mass.)
wow. the stupidity here is on par with runningtiger from the rant.
good luck fatso.
Posted on 10/1/14 at 5:23 pm to SpidermanTUba
quote:
Actually nature is usually more skewed than normal. The stock market learned that the hard way in 1987. Simply presuming a normal distribution without evidence is a serious folly.
Look I think The Black Swan is a great book as you seem to be using Taleb's argument, but you're misapplying his argument. He differentiated between that do not have an underlying normal distribution, and instead have fatter tails due to extreme, disproportionate outliers and variables that do have an underlying distribution. Test scores would fall in the latter groups for a number of reasons that I explained in an earlier post.
Posted on 10/1/14 at 5:25 pm to CptBengal
quote:
wow. the stupidity here is on par with runningtiger from the rant.
good luck fatso.
Hey I think I found something you should read
quote:LINK /
Central limit theorem (CLT) is considered an important topic in statistics, because it serves as the basis for subsequent learning in other crucial concepts such as hypothesis testing and power analysis. There is an increasing popularity in using dynamic computer software for illustrating CLT. Graphical displays do not necessarily clear up misconceptions related to this theorem. Many interactive computer simulations allow users to explore the programs in a "what-if" manner. However, users may further build up other misconceptions when they start with unclear concepts of the components that contribute to CLT. This paper analyzes common misconceptions in each component of CLT and evaluates the appropriateness of use of computer simulation. CLT states that a sampling distribution, which is the distribution of the means of random samples drawn from a population, becomes closer to normality as the sample size increases, regardless of the shape of the distribution. Misconceptions are found about the following areas: (1) randomness and random sampling; (2) relationships among sample, population, and sampling distribution; (3) normality; (4) parameters of the sampling distribution; and (5) relationships between the sampling distribution and hypothesis testing. (Contains 31 references.) (Author/SLD)
Its right up your alley
Posted on 10/1/14 at 5:26 pm to buckeye_vol
quote:never read it
Look I think The Black Swan is a great book
quote:
Test scores would fall in the latter groups for a number of reasons that I explained in an earlier post.
Not of the reasons you explained in an earlier post are correct reasons.
The central limit theorem only applies when the sample is randomly selected. In this case the sample is chosen by school.
Ignoring the political impossibility of it - are you seriously telling me its mathematically impossible for a state to allocate 100% of its education resources to 10% of the schools and 0% to 90% of the schools, thus creating a system with 90% of the schools below average - regardless of the number of schools involved?
This post was edited on 10/1/14 at 5:32 pm
Posted on 10/1/14 at 5:32 pm to SpidermanTUba
quote:
You're asserting that for large enough numbers of schools, half will fall below average. You are basing this on the central limit theorem which applies to random samples - however, the samples are not random, meaning you are wrong. (Not surprising from a guy who thinks ice has no mass.)
Actually your use of a sample is flawed since we are actually talking about the population as a whole. Regardless you seem to be arguing that if a sample doesn't meet each assumption of the CLM that the scores won't be normally distributed; however, the underlying distribution is normal because the variables are theoretically normal and raw test scores are "transformed" into a scale that is normal (e.g., t-scores, standard scores, normal curve equivalents, etc).
Posted on 10/1/14 at 5:40 pm to SpidermanTUba
quote:Are you attempting some sort of gottcha based on mean vs median performance?
NGT's so called lies aside - the statement "half the schools are below average" is NOT automatically true
Our primary and secondary educational system is subpar. You can average it, statistically weight it, sort it by mean, median, any mechanism you choose. It remains the same overall underperformer.
Why the focus solely on domestic cost and performance?
Why not instead focus on international comparisons?
Ours is the most expensive system on a per student basis in the entire world.
The MOST EXPENSIVE!
Student performance comes nowhere near matching the funding.
US students badly trail their international counterparts, despite US expenditures. US teachers are paid more than their global peers.
Our 1° and 2° educational woes won't be changed by throwing more money at the problem.
Posted on 10/1/14 at 5:44 pm to SpidermanTUba
quote:
Not of the reasons you explained in an earlier post are correct reasons.
The central limit theorem only applies when the sample is randomly selected. In this case the sample is chosen by school.
Tests are designed to have an underlying normal distribution. Furthermore, test scores are then converted from raw scores to another scale so that the scores are truly normal.
quote:
Ignoring the political impossibility of it - are you seriously telling me its mathematically impossible for a state to allocate 100% of its education resources to 10% of the schools and 0% to 90% of the schools, thus creating a system with 90% of the schools below average - regardless of the number of schools involved?
Then maybe the score that represents the average will be different but since these are norm-referenced tests, they usually adjust accordingly when converted (e.g., Flynn Effect for IQ stores). Although the distribution may be slightly impacted, it should still be relatively normal because there will be ceiling and floor effects (i.e., usually can't go lower than a 200 on SAT composite or higher than an 800. Not to mention, there are so many other variables at play in your hypothetical that aren't accounted for (e.g., demographics, developmental limitations, cognitive limitations).
Posted on 10/1/14 at 5:53 pm to buckeye_vol
He's not worth the effort.
He doesn't understand what's even being measured by the distribution, let alone the underlying assumptions.
He doesn't understand what's even being measured by the distribution, let alone the underlying assumptions.
Posted on 10/1/14 at 6:01 pm to SpidermanTUba
quote:
The central limit theorem only applies when the sample is randomly selected. In this case the sample is chosen by school
If we are using the entire population then we aren't worrying about randomization of a sample because it's not a sample. Furthermore, it is often difficult to get a truly random sample of an entire population and instead use convenience sampling. Guess what? Most of the time the distribution is approximately normal. The generalization of results may be impacted though. Besides the variables themselves are still random and there are statistical methods that are used to account for the non-independence of participants (HLM) if that is necessary.
Posted on 10/1/14 at 6:04 pm to CptBengal
quote:
He's not worth the effort.
He doesn't understand what's even being measured by the distribution, let alone the underlying assumptions.
It's just frustrating because he presented himself as an expert in statistics, then chastises people who point out his errors and misrepresentations of statistical theory and the applications of statistics.
Posted on 10/1/14 at 6:06 pm to NC_Tigah
quote:
Our primary and secondary educational system is subpar. You can average it, statistically weight it, sort it by mean, median, any mechanism you choose. It remains the same overall underperformer.
Although I see obvious flaws in our system, I think that people overstate (not saying you are) how extreme these flaws truly are.
Posted on 10/1/14 at 6:08 pm to SpidermanTUba
quote:
are you seriously telling me its mathematically impossible for a state to allocate 100% of its education resources to 10% of the schools and 0% to 90% of the schools, thus creating a system with 90% of the schools below average - regardless of the number of schools involved?
Where is this being done?
If you can't point a link to a state where this policy is implemented, then why would it be wrong to assume a normal distribution?
Posted on 10/1/14 at 6:12 pm to buckeye_vol
quote:
It's just frustrating because he presented himself as an expert in statistics, then chastises people who point out his errors and misrepresentations of statistical theory and the applications of statistics.
He isnt.
He claims to be a PhD in astrophysics. He isnt.
Don't worry. If you'd ever like to rap about stats I'm down, and do have a MS in it....there are several other posters here that know their shite too. Just Ignore him.
Posted on 10/1/14 at 8:04 pm to CptBengal
quote:So was this a discussion related to educational quality disguised as statistics, or was it a discussion about statistics under superficial guise of something related to the US educational system?
He isnt.
He claims to be a PhD in astrophysics. He isnt.
Don't worry. If you'd ever like to rap about stats I'm down, and do have a MS in it....there are several other posters here that know their shite too. Just Ignore him.
This post was edited on 10/1/14 at 8:05 pm
Posted on 10/1/14 at 8:07 pm to NC_Tigah
quote:
So was this a discussion related to educational quality disguised as statistics,
This. The simpleton tried to use an incorrect depiction of stats to make some point.
The thread went to stats because he was wrong.
Posted on 10/1/14 at 8:09 pm to CptBengal
quote:A good while ago, I discovered he was cutting and pasting answers from global warming sites that basically have lists of "if they say this, you say that" stuff I mean, verbatim cut and paste full of technical jargon etc etc. He was posting the answers as his own but when called out on it(and linked to it), he simply gave the excuse of "why write it out myself if it's already written out" or something along those lines.
He claims to be a PhD in astrophysics. He isnt.
Basically, at that point, I knew he was a joke and pretty much haven't even discussed AGW with him since. He doesn't know anything more than how to use Google to find responses.
Posted on 10/1/14 at 8:13 pm to SettleDown
quote:
I discovered he was cutting and pasting answers from global warming sites that basically have lists of "if they say this, you say that" stuff I mean, verbatim cut and paste full of technical jargon etc etc
I should have figured as much. His understanding of topics is surface level shallow.
I won't engage his bs again.
Posted on 10/1/14 at 9:09 pm to SpidermanTUba
Your assertion is seriously flawed. If you have a population of 1,000,000 and the disease rate is 0.1%, then only 1000 people will get the disease. The biggest number you left out is "what percentage of the population are suspected of having the disease and are tested?" "What percentage of those tested, test negative?" "What percentage of those who test negative are falsely negative?"
Let's assume it is .15 percent being tested (1500 people). Of those, 1000 have tested positive for the disease, 15 people had a false positive and 485 tested negative. Therefore, the chance that someone who tested positive actually has the disease is 98.5%.
Let's assume it is .15 percent being tested (1500 people). Of those, 1000 have tested positive for the disease, 15 people had a false positive and 485 tested negative. Therefore, the chance that someone who tested positive actually has the disease is 98.5%.
Page 3 of 4
Popular
Back to top
Follow TigerDroppings for LSU Football News