<p>Isn’t the possibility of eliminating wrong answers different for different kinds of tests? I agree that it’s not very likely in a math test, but it seems to me that it’s quite common on a reading test to come down to two answer that seem as though they might be correct.</p>
<p>My verbal teaching colleague tells me that weaker readers FREQUENTLY rule out the correct answer, especially when it involves a word being used in a way other than the “first” definition.</p>
<p>Men guess more… duh, because they have the balls to.</p>
<p>Xiggi, could you explain why you think guessing is harmful for nearly everyone? Outside of the rounding-down case for those shooting for a 760, random guessing is an even bet. Most people would advise against even bets, but any information supporting or opposing an answer tilts the bet in guesser’s favor. And, if you consider the fact that it is the norm to take the SAT more than once, it could be argued that risk is valuable just for the purpose of increasing randomness. </p>
<p>People in the middle or low range will sometimes recognize answers that are completely out of place in any section. At the high end, it’s often possible to spend the time to make a mental argument for more than one of the answers, or just forget what two vocab words mean. One who “guessed” in this way on about four questions in the Reading/Writing sections could expect to get two or so wrong, and might end up with a 2400.</p>
<p>So then, we can conclude that any test that males score higher in is proof positive of gender bias. </p>
<p>And any test that females score higher in is proof positive of no gender bias. </p>
<p>Can you say: PC nazi’ish?</p>
<p>
</p>
<p>As a female entirely inclined to leave blanks instead of guessing, it has actually impacted me severely on my SAT IIs. I get so freaked out that even if I am only 99% sure of an answer, I get worked up and can’t bring myself to put down the answer, instead leaving it blank. This has led to me leaving horribly large numbers of questions blank on SATs IIs, like 15 blank on one (even though I later saw the answers and had the correct answer for 12 of them) and similar numbers and outcomes for other tests. It reduced my potential scores by more than 100 points in some cases.</p>
<p>I think that there is a good case to be made that one gender is more likely to engage in bad test-taking strategy. We can’t automatically call the results of that “bias”, or at least not in the politically-loaded sense. If males are more likely to get bored and stop reading carefully in the AP English exam, and this causes them to score badly, that is not a fault of the test.</p>
<p>Lockn, I have explained why in prior posts. See 20.</p>
<p>xiggi, I saw your explanation in #20, and I’m not sure what you mean by it. Is the problem with the definition of “guessing?” I’m not sure I really consider it guessing if you can narrow down the potential answers. Are you perhaps suggesting that people really can’t do that?</p>
<p>
</p>
<p>Guessing by males could just as easily lead to lower scores, considering that there is a penalty for guessing. </p>
<p>“Bias” is such an overused and vacous word, and seems to only be negative if it is slanted against females. In Newsweek (or Time–I forget which,) a collegeboard official touted the addition of the writing section as helping females since they tend to be better at the verbal. So essentially you have two verbal sections and one math, and one of the main reasons for the change according to the College boardwas to equalize the scores between boys and girls. There was no handwringing about potential bias. '</p>
<p>It seems like any piece of data comes out, and the goals seems to be to fit this into the “bias against females” theme. </p>
<p>For instance, MIT officials like to cite that women have slightly higher GPAs than men at MIT despite having slightly lower stats (on average.) This, according to them, is evidence that the there is either bias against females in high school or that the SAT is biased. You could just as easily say that MIT serves the female population better than the male population. If the situations were reversed, wouldn’t we be saying “the females come in with higher stats and end up with lower GPAs at MIT. How is MIT failing the females?” The different distribution of majors between the genders also complicates the evaluation of GPA as an indicator of relative performance, so of course this factor is ignored as well.</p>
<p>A little bit of critical thinking is in order before one goes announcing that something is biased.</p>
<p>Here is a link to an interesting article that should give pause to those who think that the research reported by Freakonomics blog referenced by the OP actually means anything significant. The article is entitled: False-Positive Psychology - Undisclosed Flexibility in Data Collection and Analysis Allows Presenting Anything as Significant</p>
<p>[False-Positive</a> Psychology](<a href=“http://pss.sagepub.com/content/22/11/1359]False-Positive”>http://pss.sagepub.com/content/22/11/1359)</p>
<p>From the Abstract: We present computer simulations and a pair of actual experiments that demonstrate how unacceptably easy it is to accumulate (and report) statistically significant evidence for a false hypothesis.</p>
<p>There is also a good article on why some (much?) social science research is bogus in the 11/18 issue of the Chronicle of Higher Education, entitled Some Social Sciences Question the Fields Research Practices. </p>
<p>So to add my final 2 cents to this discussion, I suppose one could conclude that, because I happen to be female, my preference is not to guess at the meaning of things unless I have clear and convincing evidence. :)</p>
<p>I’m a girl and I hate to leave a question blank! lol. I always eliminate at least 2 answers before I guess though :P</p>
<p>Boys tend to be more risk-taking in general, so I’m not really surprised.</p>
<p>@heps1996 yeah same lol
I’m a girl and I pretty much never leave a question blank. If there’s a question I don’t know, I can pretty much always eliminate it down to 2, and then I guess
I’d rather not leave it blank when you have the change of getting it right
but yeah, not surprising, since girls in general seem to be less decisive and willing to take risks :P</p>
<p>Hmm, interesting. Probably not true in all cases. Just a generalization, and is it even accurate? I am a girl, and I ALWAYS guess.</p>
<p>I always guess (male)… got a 2300</p>
<p>Too bad. Guessing is always better.</p>
<p>@MiroKaz
I totally agree with you. The uncertainty in SAT is simply so great that someone who scored a 2350 may score well below 2200 on his next try. But it all comes back to the conclusion - admission is not decided only on SAT scores. Oh well.</p>
<p>I’m a female and I have no issue in guessing. </p>
<p>I can’t just leave the answer blank- it kills me. I have to fill it with SOMETHING even if I have not an inkling of an idea (in which case I guess B- don’t ask me why, I guess I just like the letter).</p>
<p>I find the idea that females don’t guess rather ludicrous, to be honest.</p>
<p>Haha @TheASP same! I usually go for B if I have no idea & I hate leaving it blank</p>
<p>a complete guess is a waste. but an educational guess (say you’re stuck with 2 choices) is a very good guess, considering that getting right (which is 50% chance) will earn you 1, but getting it wrong (which is also a 50% chance) will deduct 1/4</p>