<p>@sdapkid: sorry, I really don’t have a feel for that … haven’t read those questions before, so I’m not sure if there’s a penalty or not. </p>
<p>@ang920: again, I haven’t seen the question. Using variables you haven’t declared might be no-penalty, if you properly declared other local variables. If you didn’t declare any local variables, it’s a half-point penalty. The bigger problem, though, is the lack of initialization: that sets up problems with the correctness of the algorithm. (E.g. if you don’t initialize your loop counter variable, you can’t be sure it executes the proper number of times.) That can be a more significant problem … 0.5-1.0 points, depending on the rubric itself.</p>
<p>@ajeck513: each FRQ is graded independently as an integer 0-9. Half-point scores like x.5 are rounded up to the next whole integer (x+1).</p>
<p>Non-standard and/or inefficient code is eligible for full credit. (No harm, no foul.) Of course, you run a risk with non-standard solutions that you might end up inadvertently confusing the reader, thereby getting a lower score. But most readers are pretty good at looking at unusual approaches. And with 20,000 students taking the exam, certain non-standard solutions actually become pretty common.</p>
<p>I can’t reveal confidential information. But I have no confidential information about the 2012 exam. All I’ve said tonight is based on the scoring rubrics from past years FRQs, which are all public (though not necessarily easy to figure out), and my experiences as a reader. I’m anxiously awaiting Thursday, when I can download the FRQs and try them myself.</p>
<p>Also, I have to make it clear that I’m not speaking on behalf of College Board. I work for them, not the other way around. </p>
<p>But other than that, as long as I play nice, I’m good to go …</p>
<p>YES. speaking <em>generally</em> about the choice structure, I found myself filling consecutive E’s or D’s which made me second-guess and third-guess myself >.></p>
<p>do you guys think about half the multiple choice and half the free response correct would get a 3? this exam didn’t exactly matter a whole lot to me but it would be great to at least pass!</p>
<p>Assume someone used ArrayList methods on an Array (for instance, .add(obj, 3))</p>
<p>The rubric states that’s 1/2 off. My question is, even though the question is 9 points, the question is split up into parts. If I used .add/.remove for my arrays on both parts, is that 1 point off or still only 1/2?</p>
<p>Second, it was about creating something alphabetically, so whenever I compared and found one that was larger I used .add(obj, i) in order to make it alphabetical.</p>
<p>Would I be looking at point loss much greater? (Obviously the logic works but the methods really do change what you have to do. (Arr* = vs List.add(obj, i)) As well as .remove not being a method for arrays at all.</p>
<p>We have to be careful about general statements here.</p>
<p>Confusing Arrays and ArrayLists for <em>access</em> (e.g. [] versus get()) is a general scoring guideline deduction of 0.5 points. General scoring guideline deductions are only made once for an entire FRQ, regardless of the number of times the error occurs, or in how many parts the error occurs in.</p>
<p>Confusing Arrays and ArrayLists for <em>modification</em> (e.g. calling add() on an array) could be a more fundamental problem. Deductions will be made from the rubric in each part … basically, any point in the rubric that calls for an array modification will lose points, because the add() call is invalid. The number of points deducted will depend on how the rubric. I note for example, that the 2011 FRQ 1 only assessed a 0.5 point penalty for using set() on an array. On the other hand, the 2011 FRQ 4 (which I read) didn’t have a specific penalty, which meant that people who treated the 2D arrays like ArrayLists lost a lot of points. It all depends on how the rubric gets written.</p>
<p>Alright. Would you happen to know about using .getNumRows (and .getNumCols) instead of the actual way to get the length of rows and columns? The regular rubric has no mention, so if the 2012 rubric had no mention of this error, would that ruin the entire code or would it be more like a 1/2 point thing?</p>
<p>Very rarely does a single error “ruin the entire code”. Most of the rubrics are constructed in a way to isolate errors like this. So, for example, if getting the proper number of rows and columns was needed in order to set the proper boundaries for a loop, the ‘natural’ place to take the deduction would be in the point(s) dealing with loop correctness. So … it could be something like a 0.5 deduction, but perhaps larger (1.0-1.5 points), depending on what parts of the algorithm it affects. Certainly it wouldn’t invalidate the entire problem.</p>