Let me deal specifically with the math portions of the SAT.
Supposedly there are a total of 54 math questions spread out in a few sections,
with five multiple choices for each question. (I am not sure if this is the set-up.)
The grading system for the raw scores is the number of questions marked
correctly (times one), minus one-fourth times the number of questions
attempted, but incorrect, and the number of questions left blank do not matter.
As I see it, an unmarked question is *also* a wrong answer, so that those
one-fourth of those should be subtracted, too.
And, to me, there looks to be more merit than testees who try to answer
as many questions as possible. I see that as trying more than a person
who leaves more questions unanswered.
Also, I see it as presumptuous to name incorrectly answered questions as
"guesses," and that testees are being punished by suctracting one-fourth
of them for forming part of the score. What I feel is more honest is that
many (some percent) of questions which were answered incorrectly were
done so by making smaller relative errors (mistakes such as typos, etc.),
that the person did not guess, but answered them to the best of their
ability at that moment.
Example:
-----------
Student A
38 - - attempted questions answered correctly
16 - - left unanswered/blank
0 - - No remaining questions attempted and are incorrect
38(1) - 0(16) - 0(0) = 38
____________________________________________________
Student B
39 - - attempted questions answered correctly
7 - - left unanswered/blank
8 - - remaining questions attempted and are incorrect
39(1) - 0(7) - (1/4)8 =
39 - 2 = 37
Conclusion:
Student A got a higher score than student B on the mathematics
portion, although he/she got fewer questions correct and left
more questions blank.
==============================================================
What are some more facts and/or opinions of any other users about this scoring
system?
Supposedly there are a total of 54 math questions spread out in a few sections,
with five multiple choices for each question. (I am not sure if this is the set-up.)
The grading system for the raw scores is the number of questions marked
correctly (times one), minus one-fourth times the number of questions
attempted, but incorrect, and the number of questions left blank do not matter.
As I see it, an unmarked question is *also* a wrong answer, so that those
one-fourth of those should be subtracted, too.
And, to me, there looks to be more merit than testees who try to answer
as many questions as possible. I see that as trying more than a person
who leaves more questions unanswered.
Also, I see it as presumptuous to name incorrectly answered questions as
"guesses," and that testees are being punished by suctracting one-fourth
of them for forming part of the score. What I feel is more honest is that
many (some percent) of questions which were answered incorrectly were
done so by making smaller relative errors (mistakes such as typos, etc.),
that the person did not guess, but answered them to the best of their
ability at that moment.
Example:
-----------
Student A
38 - - attempted questions answered correctly
16 - - left unanswered/blank
0 - - No remaining questions attempted and are incorrect
38(1) - 0(16) - 0(0) = 38
____________________________________________________
Student B
39 - - attempted questions answered correctly
7 - - left unanswered/blank
8 - - remaining questions attempted and are incorrect
39(1) - 0(7) - (1/4)8 =
39 - 2 = 37
Conclusion:
Student A got a higher score than student B on the mathematics
portion, although he/she got fewer questions correct and left
more questions blank.
==============================================================
What are some more facts and/or opinions of any other users about this scoring
system?