WSJ: Standards Tighten for AP Courses

Parents Forum
21

<p>"A's don't require a 90% anymore? Grade inflation?"</p>

<p>I attended UVa and University of Florida as an undergrad and MIT in grad school about 20 to 25 years ago, and even then, in math, engineering and science courses, an A was usually much lower than 90 to 100. And as a faculty member at Michigan in the early 1990's, I tried to design tests so the average was about 60 to 70 percent. This doesn't mean that average students only learned about 60 to 70 percent of the important material, but rather that I had designed a difficult test so the top students would not be killed by minor math errors and would be rewarded by answering some questions on the subtle aspects of the material. At UofM I was instructed for my course to have an average grade of C+/B-, which was easily achieved using a curve. I always wanted to write tests that were relevant to the latest news (e.g. , engineering problems relevant to the Persian Gulf War), and it would have been too difficult to come up with new exams that were exactly scaled properly and allowed the students to see the relevance of what they were learning. Class averages varied from about 60 to 75 on my exams, and frequently I would have one student who had at or near 100 percent on each test, but students knew they would be primarily ranked relative to their peers. If I sensed that the whole class has mastered the material particularly well relative to previous terms, then I could have shifted all grades, but I never encountered this.</p>

<p>In terms of the AP exams, I do not know whether or not a 5 really represents mastery of the material, but if you really wanted different schools to be able to use this info for admissions and class credit, couldn't the College Board release the RAW scores as well? Then individual colleges could assign credit based on something other than the 3/4/5 scale.</p>

22

<p>EllenF- Have you looked lately at the breadth of material covered by some AP exams? When the percentage of students who achieve 5s hovers around 10% for many exams (much higher for Calc BC- but then again, they are an extremely self selective group), you can't say that the cutoff is too low. If 67% correct is a 5 (your data) and 10% of test takers can get it, and you propose to raise the passing mark to 70%, do you really think it's sensible to have less than 10% of test takers even PASS?</p>

<p>On a three hour test designed to test knowledge of a broad subject in a reasonable amount of depth, you've got to leave some room for error.</p>

23

<p>So, is it a newflash that the UC is at it again? Why is it surprising that the UC system is looking for yet another scapegoat to mask its own glaring ineptitude in designing an admission system that would have a minimal correlation with the state population? The UC system might be large but it is quite an oxymoron to use "leadership" and UC in the same sentence -although the second part of oxymoron offers an unmistakable fit. </p>

<p>The problem is not as much with the AP system itself as with the interpretation of the system by high schools and colleges.</p>

<p>The surest way to correct the problem is by stopping the practice of rewarding students the CREDITS and stopping the misleading marketing that AP classes are acceptable substitutes. It would be nice to see a drastic cap on AP (1 to 4) with the added restriction that they are only used for placement purposes only.</p>

<p>The current system is totally out of control: high schools pretend to be able to teach at the college level while the majority of colleges have to add remedial classes to overcome the deficiencies of high schools. The biggest scam is perpretrated on the students who are shortchanged a full and complete college education and experience. Let the high schools focus on PREPARING the students for college and not pretend they can be a substitute for entry classes. School districts that offer 25+ AP but fail to reach the 1000 (or 1500/2500) mark for the SAT should be scrutinized. </p>

<p>The AP exams do provide a good measure of students' and schools' achievement and the AP program does have a real purpose in providing a comparative yardstick. Other than that, it usefulness is way overstated.</p>

24

<p>I could agree with placement only. I definitely agree that colleges have been forced to add remedial courses and at the expense of students who have come to campus prepared. I believe that the remedial courses should occur in community college or continuing ed courses.....NOT the regular campuses. It is a tremendous waste of resources.</p>

25

<p>xiggi, I couldn't agree with you more. Well said.</p>

26

<p>They are indeed a "comparative yardstick" - the thing is, we don't really know of what? The fact that the courses were even offered? the quality of the teaching? the quality of the students? the ability to take tests? the income of the students in the geographic area? The elimination of honors courses? All of the above? None?</p>

<p>What we do know, without question, is that high schools, and the students within them, are subject to the social science variant of Heisenberg Principle - the choice of the measurement tool impacts the the organization and delivery of services (and not necessarily for the better, as many of the colleges are now noting.)</p>

27

<p>Mini:</p>

<p>Unlike the SAT, the AP exams are supposed to test the AP curriculum which is described in great detail. Most colleges, except perhaps Caltech, do believe that students who do well on AP exams are capable of performing well in more advanced classes.
I don't buy the SES/geographic area argument. It does not really matter if the student has been attending an inner-city school that offers no calculus. However bright s/he may be, if the student has never had calculus, s/her should not be placed into Multivariable Calculus.
It does not matter if the student was bright but slacked off throughout the year, the teacher was lousy, or some other factor that affected performance on the test. If the student is not prepared, the student should not be placed in a more advanced class. If the student consistently did well in class but had a bad day on the test, there should be some mechanism in the college to allow him/her to demonstrate preparedness. Most colleges begin with placement tests for incoming freshmen.</p>

28

<p>The Heisenberg Social Science variant comes into play in that it has pushed interesting, exciting honors programs, specialized courses, and opportunities for students to organize their own study/thinking/writing independently virtually out of existence. Or at least it has in my town.</p>

<p>My d.'s school has all but eliminated the use of APs for credit or placement. In math and languages, they have their own placement test, and could care less about how you acquired the skill. In the sciences, they have decided not to exempt ANYONE from first-year courses, but rather to set up "enriched sections" for high AP and/or SAT II-scoring students - they found that the APs -- regardless of the students' scores -- simply DON'T prepare students well enough for advanced classes. You do get to use the AP credits if for some reason you get sick and can't complete the required number of credits, but that's pretty much it.</p>

29

<p>Mini:</p>

<p>The schools that are not offering AP classes are able to offer exciting courses and their students do well on APs. The AP curriculum is very helpful to schools and teachers who would like to offer more challenging classes; AP exams validate the quality of the course. I have no problem with it.
I do think that APs in the humanities and social sciences do not prepare students well for college level work. In the sciences, with a better defined curriculum, they generally do.</p>

30

<p>Well, it's a good theory. Smith doesn't think so. Actually that is unfair. It "may" prepare students for college level work - they just don't find from experience that it is a substitute for it.</p>

31

<p>Smith must see things differently from MIT. It's quite happy to grant credit for BC Calc and Physics C (probably other sciences, too, but these are the two I remember).
You will find that a lot of college introductory sciences courses use the same textbook as the AP courses. It's true of Physics, Biology and Calculus.
My S's introductory biology at the Extension School used exactly the same textbook as the AP-Biology class in his high school. The only difference is that it covered the same material in less time. That, and the fact that we had to shell out $100+ for the textbook which was distributed free to the students in his high school, plus the tuition for the class.</p>

32

<p>From the MIT Admissions website:
College Board Advanced Placement Program</p>

<p>For most secondary school subjects that closely parallel the College Board Advanced Placement guidelines, the only method for generating credit at MIT is through the regular College Board Advanced Placement tests. If you wish to receive such credit, you should take those AP tests for which you are prepared. Only one test in a given subject area will be recognized.</p>

<p>Below is a general list of College Board tests currently recognized by MIT.</p>

<pre><code>* Biology
* Calculus (BC exam only)
* Physics (C exam only, both parts)
* Humanities and Social Sciences (most exams accepted for elective units only)
</code></pre>

33

<p>"I do think that APs in the humanities and social sciences do not prepare students well for college level work. In the sciences, with a better defined curriculum, they generally do."</p>

<p>That is very true. Some subjects are a mile wide but an inch deep, and seem to reward the memorization of a ton of facts over a deeper understanding. </p>

<p>It also illustrates that it is hard to discuss the AP program as an homogeneous block. Despite appealing to a smaller number of candidates (and schools), there might be some merit to develop more specialized subjects to allow students with advanced abilities to demonstrate their mastery of specific and well defined subjects.</p>

34

<p>I am rushing off to a vacation, but I wanted to say that a few years ago, a high official of the College Board admitted that the AP scores are tied to whatever standard the college grades are CURRENTLY tied to, i.e., since there has been grade inflation at colleges, so there has been grade inflation on AP scores. Merrily we roll along... </p>

<p>Easy, see?</p>

<p>If colleges aren't enforcing sensible grade distribution, why should the College Board buck the trend? At least the kids are actually taking the AP tests, unlike the ones the WSJ was referring to.</p>

35

<p>It's a chicken and egg question. William Lichten of Yale suggested that the influx of students into AP programs diluted the quality of the exams and therefore argued for raising the bar for placement and credit. His argument has been favorably received on many campuses. See for example,</p>

<p>Raising the Cut-Off Score on the College Board Advanced Placement Examinations in Composition and Literature: Justification for a First-Year Writing Curriculum</p>

<p>Richard H. Haswell</p>

<p>Texas A&M University, Corpus Christi</p>

<p><a href="mailto:rhaswell@falcon.tamucc.edu">rhaswell@falcon.tamucc.edu</a></p>

<p>24 August 2003
<a href="http://comppile.tamucc.edu/raisingAP.htm#Lichten%5B/url%5D"&gt;http://comppile.tamucc.edu/raisingAP.htm#Lichten&lt;/a&gt;&lt;/p>

<p>Lichten, William “Whither Advanced Placement?” Education Policy Analysis Archives 08.29 (June 24, 2000) <a href="http://epaa.asu.edu/epaa/v8n29.html%5B/url%5D"&gt;http://epaa.asu.edu/epaa/v8n29.html&lt;/a>. Extremely useful data substantiating the historical shift of colleges and universities toward a higher cut-off for AP advance credit in all content areas. Argues that less than half of colleges and universities allow a score of 3 or lower for credit/exemption/placement, and that the rate is dropping (this is over all subject areas). The piece documents another general trend: the more prestigious the school, the higher the cut-off score.</p>

36

<p>
[quote]
It's a bit like scoring for the IMO. A perfect score is 42, but you get a gold medal if you score higher than 36 (or is it 32?)

[/quote]
</p>

<p>Actually, the IMO cutoff for gold, silver, and bronze differs each year. It is endogenously determined after the fact by vote of the officials under a guideline that says the top 1/12 of the scorers receive gold, the next 1/6 receive silver, etc. Some years, the IMO exam may be relatively easy and/or the field of competitors may be very strong and the cutoff may be very high. (There have been some years with gold cutoffs in the 40s and other years with gold cutoffs in the high 20s.) The IMO exam can differ a lot from year to year, because the questions are outside-the-box non-routine questions. </p>

<p>On the other hand, AP scores are set by an "equating process," in which the College Board tries to ensure that a 5 represents the same level of subject mastery from year to year. There is no predetermined proportion of 5's, 4's, 3's, etc. Instead, the exam committee tries to come up with a tests that are roughly equal in difficulty each year as much as possible, so that scores will be comparable year to year. There are equating questions, which are multiple choice questions that get reused in several successive years, and the AP grading committee does statistical analysis of how students did on those questions compared to how they did on the overall exam to fine-tune the cutoffs for a 5, 4, etc. However, the cutoffs within a given subject don't differ much from year to year. The fairly cut-and-dried type of predictable questions often asked on AP exams makes it relatively straightforward to design exams of consistent difficulty.</p>

<p>The College Board spends vast quantities of money on "psychometric analysis" to design and calibrate the AP tests so that the AP scores are comparable year to year. A college freshman who enters with a "5" on his AP transcript should, in theory, have a credential representing the same level of mastery regardless of whether he took the exam in his sophomore, junior, or senior year, for example. </p>

<p>Periodically, the College Board also calibrates their scores by asking cooperating college profs to give the AP tests to freshman enrolled in college courses at a representative sample of colleges across the country. The cutoff for the lowest 5 is supposed to be set at the level achieved by the AVERAGE college A-student, for 4 at the level achieved by the AVERAGE B-student, etc. Assuming that the College Board is defining AVERAGE as the median A-student, this means that half of the college A-students in the "representative" sample did not score at a level that would get a 5 on the AP exam.</p>

37

<p>Thanks for the clarification, Wisteria. It does support the general point that the grading scale depends on the level of difficulty of the exam. You don't necessarily need to score 90%+ to get an A if the exam is sufficiently hard.
And if a 5 is equal to an A-, it leaves room for less than perfect performance on the AP exam.</p>

<p>I believe that AP readers include both high school teachers and college profs.</p>

38

<p>
[quote]
I believe that AP readers include high school teachers and college profs

[/quote]
</p>

<p>Yes. And both groups are represented on the committee that composes the exam as well.</p>

<p>It's interesting to note that AP readers report that they regularly encounter stacks of AP exams from a single school in which half of the students from the school turned in blank exams for the Free Response--they did not even attempt to answer the questions. Calculus AP readers have also reported that some students have written limericks, stories about how students lost their virginity, and other random irrelevant compositions by students who have no interest in even making a serious attempt to pass the exam. Typically these exam books come from schools in which all students enrolled in an AP course are required to take the exam--many of these schools are schools in which the district is paying the students' fees.</p>

<p>Sending the raw scores on the multiple choice part of the AP exams would be unfair to the students, since those questions are not disclosed and the college would have no way of knowing how difficult the questions were.</p>

<p>However, in this day and age of electronic image transmission, the AP people could easily offer to send the free response (essay portion) of the exams to colleges, so academic officials at the college would be able to see which students had made a serious attempt to answer the questions. </p>

<p>College departments would also be able to form their own opinions of the quality of the students' work, based on more than just the AP readers scoring rubric.</p>

39

<p>
[quote]
Geometry has moved down into our 8th grade.

[/quote]
</p>

<p>Geometry has been an eighth-grade subject all over the world for a long time. And some of that geometry is of much higher quality than what most Americans get in higher grade levels. </p>

<p>I would be happy to see mathematics curriculum expectations, and history curriculum expectations, and science curriculum expectations, and foreign language curriculum expectations move in the general direction of the top-performing countries elsewhere in the world. Students in the United States are wealthy enough and free enough from disease to handle that.</p>

40

<p>1Down -- your interpretation about the UC statement makes no sense. Neither of my kids - at 2 separate high schools - had APs available before 11th grade - and practically speaking, given scheduling issues, its hard to get in more than a few APs. I have a feeling that by 17 "semesters" the UC rep meant that the goal was for the high schools to OFFER a full complement AP options for that many separate courses - they didn't meant that high schoolers were expected to TAKE that many. I know that UC Berkeley has capped the number of weighted grades all along, so the kids certainly don't get a particular advantage from taking more than a handful of courses. </p>

<p>My son came from a small high school that offered only APUSH, 1 year of AP English, & AP Econ -- no sciences, no AP Calc. He graduated 4 years ago, and that year 10 kids from his high school were accepted to Cal. So certainly he was in no way disadvantaged by the fact that he only had 5 semesters of AP courses in high school.</p>

<p>The context that you presented -- that parents at your high school were concerned because scheduling conflicts or arbitrary cut offs kept many kids out of APs - really seems to support my interpretation that the UC Rep was saying that ideally each school should OFFER more APs -- thereby addressing the complaint that parents were making about their kids being shut out.</p>