"John Knight" <johnknight at usa.com> wrote:
>"Bob LeChevalier" <lojbab at lojban.org> wrote in message
>news:n53viuso190dvhbk2o6c2m5fu7gvmvranh at 4ax.com...
>> >Let's use a simple example of how wrong you are, cary.
>> >
>> >Question K09 on the 12th Grade TIMSS Math test given to 12th graders
>around
>> >the world in 1995 reveals an astounding difference in math skills between
>> >the sexes in all the countries who participated. The average difference
>in
>> >all countries was 10.5%, with 47.3% of boys and 36.8% of girls answering
>> >correctly, but the difference in the US was 22.1% (28.6% of girls and
>50.7%
>> >of boys). In countries like Cyprus where 60.1% of the boys answered
>> >correctly, guesses on the test would not have influenced the scores by
>that
>> >much, but where only 28.6% of American girls answered correctly, guesses
>> >must be taken into account.
>>>> But did you not say just above that "Zero percent of American 12th grade
>> girls correctly solved TIMSS math problems." Now you are admitting that
>> 28.6% solved question K09, thereby admitting your falsehood.
>>You obviously didn't understand any part of the above, so why bother to
>explain it to you?
>>Oh, well, let's try, just one more time.
>>It's disingenuous for you to ignore the effects of the multiple choice
>questions on this test. On this particular question, because there are four
>choices, 25% of the answers will be correct JUST because students guessed.
That is a false statement. You do not know that ANY of the students guessed,
and only a student who guesses randomly will be subject to your 25% logic.
It is perfectly possible for 100% of the students to get a multiple choice
problem wrong with none of them guessing. All it takes is that they have
been taught erroneous information or incomplete information.
>So when the scores are reported and it shows that 28.6% of girls got this
>question correct, you cannot claim that 28.6% of them understood the
>question.
You cannot claim that any percentage understood the question. None might
have understood it and just guessed. Or 100% might have understood it, but
71.4% of them applied principles incorrectly to the solution and therefore
got the wrong answer.
At BEST (after subtracting the 25% of the questions which were
>correct because of guesses) you COULD claim that 3.6% of them understood the
>question.
Completely fallacious.
>No, that's not zero percent, but when this is averaged along with all the
>questions where girls actually scored lower than if they had just GUESSED,
>the sum total of their scores is a statistial ZERO percent.
>>The number of times that they scored lower than if they had just guessed was
>uncanny
And that fact shows that in fact "guessing" does not explain the pattern.
Oh, and your 3% error range is inappropriate and incorrect. You seem to
think that all statistical data comes with a 3% error range. It doesn't.
But we've torn you a strange orifice before because of your incompetence of
statistics, and you persist in pulling numbers out of your strange orifice.
lojbab
