Fewer A's at Princeton this year

Will UoP follow suit? Naw.

Click here to read the full article.

 

While grade inflation is no doubt a real problem in certain situations, I think the students quoted in this article make some excellent points regarding caps and other methods being proposed in order to rein it in. There's a huge difference between inflating a grade for a student who didn't really earn it versus withholding a top grade from a deserving student just because of some inane institutional policy. And it's no surprise that you'd find a higher concentration of "A" students at the elite schools, simply because of the selectivity of their admissions processes.

The message that you're sending to your students under these policies is to just avoid taking the harder courses. Plus I'd think they would tend to penalize the really good, effective teachers (whose students you'd expect to perform better overall) over those who are mediocre or just plain lousy.

I saw an item similar to this one last night while I was at the bookstore, flipping through a special "colleges" issue of some magazine or other (it might have been a U.S. News publication - I forget.) This special issue was chock full of articles on how colleges are supposedly harder to get into now than ever before, how prices keep going up, how it's taking longer to finish a degree program, how colleges are beginning to get stingier on things like advanced placement, how students are generally less satisfied with their schools these days, and so on. In light of these issues, and knowing that schools like UoP are growing like gangbusters, I have just one question:

When are the traditional institutions of higher learning going to wake up and realize that they are slowly but surely driving themselves out of business?

 

A political slogan during the depression-era of the 1930s was, "A chicken in every pot." Times have changed and now people are hungry for an education, so the slogan for the 21st Century could easily read, "A college degree on every office wall."

College degrees are no longer rare and the ease of obtaining one has been made available to the common citizen. Classroom walls are no longer a barrier. When dummies like "me again" start looking for a PhD program to attend, then you know that we've come a long ways, baby.

 

Across the US as of 2002, the average percentage of adults older than 25 with a bachelor's degree is 26% (stdev=4.5). While this is not statistically "rare" a degree is still not all that common. If we can assume that only a percent of these adults go on to get a masters and a smaller percentage go on to get a PhD, it may be true that adults with a graduate degree are still a rare commodity.

http://www.stateline.org/stateline/?pa=fact&sa=showAllFactsForHeader&headerId=441

 

All to the good, I think, but the problem of grade inflation is a real one.

Many, if not most, of my students come into class believing that they start out with an 'A' and they have to do something wrong not to receive that as the final grade; e.g., that I deduct points from '100' to determine their final grade. I have had graduate classes where this in fact has been the case. I have had to start explaining to my students that they start with a '0' and build points during the semester. In my courses, and I would say those of the majority of colleagues, each person gets what he/she earns at semester's end, regardless of what the dean wants. Some semesters I give out more 'As' than other semesters, but it all works out over time.

Those of us who have reached geezerhood in academia have seen recurrent but periodic spasms of attempts to combat grade inflation. They usually don't last long, but I think they do succeed in reminding faculty to take greater care in giving grades. I doubt that students are harmed by these spasms, at least in general. Faculty are a pretty independent lot: some comply with what the administration wants, others just ignore it. Most, I think, use their own judgment. The eternal credo of university faculty is: "So long as you can reasonably justify it to the dean, you can usually do it."

Ciao, baby



marilynd

 

Harvard

There is an interesting article in this month's (March 2005) Atlantic Monthly titled "The Truth About Harvard". It discusses grade inflation at some length, and also goes into a scathing critique of the Harvard educational process, at least as it relates to the Liberal Arts. Apparently the author (Ross Douthat) has just come out with a new book on the subject.

Link to a teaser for the article:
http://www.theatlantic.com/doc/prem/200503/douthat

Comments on the article:
http://forum.theatlantic.com/WebX?.4a826976

Interview with the author:
http://www.theatlantic.com/doc/prem/200502u/int2005-02-10

About the book:
http://www.amazon.com/exec/obidos/tg/detail/-/1401301126/qid=1108868209/sr=8-1/ref=sr_8_xs_ap_i1_xgl14/002-7276674-4639259?v=glance&s=books&n=507846

 

The Princeton 'solution' to the grade inflation problem isn't. Grading should be done independently of any quotas for any grade, including 'pass/fail'.

In principle, if all candidates pass the exam they should pass and this can b extended to the proportion of cnadidates scoring in any particular grade, including Distinction.

In practice, such an outcome is unlikely, though theoretically possible. If there is a trend over time for the proportion of a class gaining an 'A' or 'First' to increase, that suggests grade inflation. When the proportion is 'A' is nearly half the class that suggests grade inflation. In any population with attributes there is a distribution of outcomes, including in a class of 'A' students.

Grade inflation outcomes are assisted by the US system of a narrow band (47 per cent scoring above 95/100); 'ranking' incentives that reward the proportion of 'A' passes - and punish below the proportions awarded by competing schools; by soft exam regimes; by the incidence of 'negotiated' grades (which seems to start at High School); by the illusion that because scarce capacity necessitates selection by previous qualifications and that the intake tends to take the top stream of students that 'A' passes are to be expected; by the lack of independent quality controls in grading outcomes; class teachers being responsible for component parts of the final grade score (marking for attendance, participation, class and group assignments); class teachers proctoring the exams - or leaving it to an 'honour' system; class teachers knowing what is in the exams and susceptible to 'review hints'; and a general lack of independent validation.

The grading score should be widened to discriminate between individual scores rather than cluster them in a ridiculously narrow range from 90-100. Imagine if we graded temperature in a range of only 10 degrees. We would all live in a 'hot' climate on the scale but we would be still experiencing a wider range of temperatures.

 

Professor Kennedy:

I have always thought of myself as pretty bright and fairly well-educated, but I have to admit that I only understood about half of your post. (I knew I shoulda staid awake it that statistics class!)

However, I agree that a quota system is no solution. It is unfair and necessarily arbitrary. When I was in university in Boston it was "common knowledge" (meaning, "rumor had it") that Harvard Law School routinely expelled the bottom 25% (or some such) of the entering class. I don't know whether this was true, but I always felt sorry for the individuals on the immediate downside of the cut off criterion. Were they that much worse than the individuals immediately above them that expulsion was the correct remedy?

One part of your post that I may be able to respond to is:

I'm not sure that I understand what you mean by "narrow range." The grading system used by many schools in the US (90-100 = A, 80-90 = B, etc.) I would describe as a rather wider range than others. Many American schools have grading systems that differentiate levels within these decades: A, A-, B+, B, B-, C+, C, C-, etc. A number of schools, however, have actually moved away from the 3-4-point range to the 10-point range, arguing that the larger the number of possible grades, the more "grade-creep" (i.e., grade inflation) occurs. Apparently, the decision is based on a perception (perhaps on actual studies, I don't know which) that it is easier when assigning grades for the individual instructor to justify a shift from C+ to B- than it is to justify a shift from C to B. Hence, grade inflation.

I actually don't think the specific grading system matters too much, so long as a reasonable range of grades is permissible. Most Americans, I think, look at a 'B-' as a 'B' anyway. The factors contributing to grade inflation are sufficiently complex, as you suggest, as to make any attempt to actually control this process exceedingly difficult. Faculty should be expected to give a range of grades, especially over a reasonable span of time. Certainly, most of the deans I have worked for have taken this approach. If an individual instructor is giving consistently high marks, this might be a result of a grading system poorly designed or implemented. More likely, in my experience, it is the result of an insufficient course regimen. If the regimen is sufficiently exacting, students will tend to fall into "insufficient," "minimum," "better," and "best" categories. The hard part of grading, regardless of the specific system, is at the intersects. I'm not sure that I understand what you mean by "narrow range." The grading system used by many schools in the US (90-100 = A, 80-90 = B, etc.) I would describe as a rather wider range than others. Many American schools have grading systems that differentiate levels within these decades: A, A-, B+, B, B-, C+, C, C-, etc. A number of schools, however, have actually moved away from the 3-4-point range to the 10-point range, arguing that the larger the number of possible grades, the more "grade-creap" (i.e., grade inflation) occurs. Apparently, the decision is based on a perception (perhaps on actual studies, I don't know which) that it is easier when assigning grades for the individual instructor to justify a shift from C+ to B- than it is to justify a shift from C to B. Hence, grade inflation.

I actually don't think the specific grading system matters too much, so long as a reasonable range of grades is permissible. Most Americans, I think, look at a 'B-' as a 'B' anyway. The factors contributing to grade inflation are sufficiently complex, as you suggest, as to make any attempt to actually control this process exceedingly difficult. Faculty should be expected to give a range of grades, especially over a reasonable span of time. Certainly, most of the deans I have worked for have taken this approach. If an individual instructor is giving consistently high marks, this might be a result of a grading system poorly designed or implemented. More likely, in my experience, it is the result of an insufficient course regimen. If the regimen is sufficiently exacting, students will tend to fall into "insufficient," "minimum," "better," and "best" categories. The hard part of grading, regardless of the specific system, is at the intersects, it seems to me.

Regards,

marilynd

 

My apologies for the post above. I'm not sure how I succeeded in duplicating my paragraphs. (I have sometime been accused of repeating myself, so . . . )

Please ignore that post, and read this post.

________________________________
Professor Kennedy:

I have always thought of myself as pretty bright and fairly well-educated, but I have to admit that I only understood about half of your post. (I knew I shoulda staid awake it that statistics class!)

However, I agree that a quota system is no solution. It is unfair and necessarily arbitrary. When I was in university in Boston it was "common knowledge" (meaning, "rumor had it") that Harvard Law School routinely expelled the bottom 25% (or some such) of the entering class. I don't know whether this was true, but I always felt sorry for the individuals on the immediate downside of the cut off criterion. Were they that much worse than the individuals immediately above them that expulsion was the correct remedy?

One part of your post that I may be able to respond to is:

I'm not sure that I understand what you mean by "narrow range." The grading system used by many schools in the US (90-100 = A, 80-90 = B, etc.) I would describe as a rather wider range than others. Many American schools have grading systems that differentiate levels within these decades: A, A-, B+, B, B-, C+, C, C-, etc. A number of schools, however, have actually moved away from the 3-4-point range to the 10-point range, arguing that the larger the number of possible grades, the more "grade-creep" (i.e., grade inflation) occurs. Apparently, the decision is based on a perception (perhaps on actual studies, I don't know which) that it is easier when assigning grades for the individual instructor to justify a shift from C+ to B- than it is to justify a shift from C to B. Hence, grade inflation.

I actually don't think the specific grading system matters too much, so long as a reasonable range of grades is permissible. Most Americans, I think, look at a 'B-' as a 'B' anyway. The factors contributing to grade inflation are sufficiently complex, as you suggest, as to make any attempt to actually control this process exceedingly difficult. Faculty should be expected to give a range of grades, especially over a reasonable span of time. Certainly, most of the deans I have worked for have taken this approach. If an individual instructor is giving consistently high marks, this might be a result of a grading system poorly designed or implemented. More likely, in my experience, it is the result of an insufficient course regimen. If the regimen is sufficiently exacting, students will tend to fall into "insufficient," "minimum," "better," and "best" categories. The hard part of grading, regardless of the specific system, is at the intersects.

Regards,

marilynd

 

Hi Maryland

Compressing arguments into a paragraph or two inevitably causes inexactitude. Comparing education processes is fraught with difficulty and can be a blunt comparison.

Of course we can use four divisions which sounds narrower than a 100 point scale. But when over half the class scores 95+ per cent and others string out down toward 70 or less there is still something wrong with the discriminatory assessment at the top end.

What is the difference between 95 and 96? (I am considering social science, humanities and arts subjects, not mathematics.)
Now answer the same question when there are more than two students being compared, say, 30 to 40? Can somebody really 'hold' stable the comparator criteria between a 95 versus 96 for the time it takes the read through and grade the students in that set?

Well, of course, somebody could because it is done in the US system many times over. But suppose the range is widened from a single point difference to 5 points? It is easier to accurately discriminate the wider the range. At 10 or 20 points in a range, previously set at a single point, a grader can more accurately assess a difference between two students. And as important, can credibly explain the difference, in writing, between any two or more students to the External Examiner (senior faculty from another University in the British system), or the student on Appeal and to the University Academic Quality Assurance auditors and other colleagues. The same people can read the reasons for the grade being awarded too, by reading the examination answer book and comparing it with the grading guide and model answer written by faculty.

My argument is that with an exceptionally narrow band, particularly in the 90+ range - a somewhat nebulous concept for the courses I am referring to and which I am experienced in (except Multiple Choice Questions that have been psychometrically proven) - such ranges and 'high marking habits' fuel grade inflation because the narrow range does not discriminate shades of difference all along the range.

Moreover, close to 100 per cent final grades imply being close to perfection, a wholly unwarranted illusion which makes students unhappy, to say the least at receiving a more realistic grade. When I have quoted a 'good student' answer around '65' I have been told by students used to the US system that a '65' is really a 'fail', yet a 70+ would put that person in the 'Honours' pass grade in most British Universities. I am prepared to accept that someone could get a 90+ in the UK but that would be so exceptional as to be remarked upon (it would definitely have to be shown to the External Examiner for approval). If near half of the entire class was at 90+ I doubt if the University Academic Registry would approve the grades without an inquiry.

When we read a student's examination answer book, the student starts at zero marks and is awarded marks only for the correct parts of their answers. We do not start at 100 and then deduct marks for errors. The final mark has to be earned by the quality of the answer. There is no presumption that they have 100 per cent to begin with. The outcome is a distribution, though not often a normal distribution.

In our MBA final exams usually about 20 to 25 per cent score under 50 per cent and thereby fail. About 7 per cent are awarded Distinctions (70 plus) and a few - one or two per Diet out of several thousand candidates - are awarded for scoring over 80. The bulk score between 50 and 69 (hence 65 is a 'good student performance').

I hope this clarifies the points I was trying to make.

 

Actually, I supervised a final year project at Oxford a few years ago where we ended up awarding somebody that sort of mark!

The student was a final year Engineering and Computation student who wanted to do a computation final project. The joint marker (all final year projects are double marked, just like exams) and I went through the mark breakdown sheet (does the project work? has it got documentation? etc) provided for us by engineering twice in detail and could find absolutely no grounds for docking the student more than a few marks . . . and a fair bit of that was because he'd ummed and ahed a bit in his final presentation.

But yes, this was an exceptional student even by Oxford standards.

During the rest of my academic career on the dais side of the lecturn (ahem, currently on hold), I've ventured into the 80s once (also an exceptional student at Glasgow) and down into the 20s once, but the rest of the marks have been pretty much scattered along the continuum from about 35 to 75.

Like Professor Kennedy I consider 65 to be a pretty decent mark. I come from a humanities background and 65 is a good solid 2:1 in my book, and anything over 70 is a first. However, I know that the US system equates a 2:1 with a GPA of between 3.3 and 3.7, which would suggest that an equivalent student in the US would be getting a much higher mark for the same work.

Angela

 

The points here are interesting, but I think there should be some clarification (or perhaps) defense of the US system. Professor Kennedy makes many valid points, but some of them are really non-issues.

First, the professor’s touting of the UK (read EBS) system of invigilated exam regimes, no choice of questions on exams etc. etc. is all well and good. It makes perfect sense in a DL environment where all measures to avoid cheating and the like must be employed. It is not, however, a guarantee of the best mode of learning, or even a guarantee of the most rigorous system. It minimizes cheating is all we can really say for sure. For a student that is honest and is motivated to do his or her best work, it can certainly be argued that the many alternative modes of assessment are more rigorous. Exams, combined with written papers, group projects and the like can more fully test an honest student’s abilities. That is the key – the honest student.

I fully appreciate that a large-scale DL endeavor like EBS must use very strictly invigilated exams, unseen questions, no choice of questions etc., because they must ensure the integrity of the program. Again, for an honest student, however, that concern is irrelevant (except in the sense that a program with failing integrity adversely affects the student’s qualification).

An analogous situation is the practice of partly basing a school’s ranking on dropout frequency (US News does this). If you are not one to drop out, this criterion means nothing. (Again, apart from whatever may be inferred about the quality of a program with a very high dropout rate.)

So let’s be clear here. DL institutions and DL students have somewhat diverging interests. Schools need to maintain integrity and minimize cheating. They try to design the most rigorous program that keeps this goal paramount. Students want a program that is the most enriching and fosters learning and fair assessment (presumably). EBS seems to have stuck a nice balance, but that is not to say that other systems that may not minimize cheating don’t also provide a demanding program.

Secondly, the whole issue of grading scales is a non-issue. The issue of grade inflation is really a separate issue, and it has almost nothing to do with grading scales. It’s a political, sociological, and economically entrenched issue that one could write a book upon. Moreover, it is not an issue at all US schools, or even all elite US schools. Princeton has it. Berkeley does not. One could crudely cite the reason as having to do with money. Princeton gives a damn if you drop out. Berkeley does not (ok, I did say crude, but the issue is relevant).

Professor Kennedy implies that social sciences and humanities grades in the US are awarded on a numeric scale. I’m sure this is true in some places, but certainly not all (or even the majority). Generally, essays are scored with letter grades. You write an A paper or a B paper, or a D paper. Rarely have I seen an essay graded as a 96, versus a 95. This, as the professor points out, is mostly nonsense. It’s also rarely done, so not often does this nonsensical situation arise.

Moreover, a 95 in the US system does not imply a score of “near perfection” – at least not in the humanities or social sciences. Far from it. The tacit understanding is that an A paper (or a 94, say, in those rare instances where a numeric grade is awarded) means that you have written an excellent paper – for a student! Your paper is not compared to academic research papers, or those written with PhD’s in the subject.

This is where the UK and US systems involve only a scale translation. Angela and the professor have implied that scores above 75 or 80 in the UK are almost unheard of. So what is the result? The UK has, for all intents and purposes, a scale that runs from 0 to 80. The US has a scale that runs from 0 to 100. Everyone on this board can figure out the mapping that takes a score from one to the other. Apparently, a score of 95 in the UK (for a non-quantitative paper) would imply that it was of publishable quality. How often does this happen? Almost never. Guess what? It almost never happens here either. If it did, you’d receive an A+ and loads of lofty compliments from your professor, and that would be the end of it. The idea that a score in the nineties (again if we are to accept that numeric scores are handed out for humanities-type classes) means the professor is assessing a grade of perfection is nonsense. The grade means you have written an excellent STUDENT paper. You are not ready for the publishing world, or to skip straight to the PhD.

Grade inflation exists, and it is a problem. Perhaps it happens more in the US than in the UK, but it has nothing to do with grading scales. Perhaps the UK system is designed to subtly remind the student that he or she IS a student, and thus a fine score of 70 is really still very far from perfection. Perhaps students in the US mistakenly believe that their 95 is really a mark of perfection, and the US system may be misinterpreted to foster that belief. However, you will find no discernible difference between a Princeton student’s 90, an Oxford student’s 70, and a Berkeley student’s 85.

 

I would argue that the US scale did not actually run from 0-100. Given than anything below 70 (or indeed often below 80 at Graduate level) is seen as failing, are there really 70 different levels of failure and only 30 different levels of success?

Is there really a need to so finely differentiate failure? "You messed up - your paper was a disaster, but it was ever so slightly better than that other chap's".

Is a failure with a 40 much better than a failure with a 20? If anything, the grading system is non-linear (and was actually designed backwards - greater sensitivity should have been designed for the higher end).

 

Practically, you're quite right. In the US, scores below 60 are essentially lumped into the failure bucket. I guess the point I'm trying to make is that, in theory, in the US one can get a grade from 0 to 100 (or, in practice, from 60-100). In the UK, perhaps the scale is 0 to 80 (or, in practice, 40 to 80). Translating between the two shouldn't be a problem (though translating between honors categories is more problematic as these aren't standardized). A superb paper in the UK gets you a 70; in the US you get a 90. They mean the same thing. The US is not handing out awards for near perfection by awarding scores close to 100.

This is a source of confusion for folks in the UK, and they associate this issue with the issue of grade inflation. They don't necessarily have anything to do with one another. Princeton handing out too many A's (scores in the 90's let's suppose) has nothing to do with professors evaluating all their students as near perfection. It probably has more to do with internal pressure to make students look successful. That's a whole other ball of wax.

 

Hi Tom

Thank you for your thoughtful contribution on US and UK practice. I think you summarise our views neatly and I am not sure that we disagree fundamentally.

By way of making this clear I append some comments (much as I would if replying to a contribution at a seminar).

The basis of ascribing a grade of A, B, or C, or their + and - variants) at root has to be a scoring system in a monitored procedure. These numerical scores may not be identified to the students but they are the only means by which quality assurance can be satisifed. Hence, in an exam paper with several questions the candidate's individudal score in each question will be given a numerical score. This ensures that a quality check by the External Examiner and a second, or even third, marker can be compared properly.

To leave the first marker free to place a letter range on the paper as a whole, without individual scores on each component part, expose the letter grading to unsafe variations (what constitutes an 'A' to marker Smith may not be the same as for marker Jones, but a question with 16/20 against it can be compared if Jones thinks it is worth 9/20). As the number of questions increases variations on the letter scale become unreliable for comparison.

Hence, behind the award of a 'A', 'B, or 'C' lies a numerical scoring range. Numbers are additive and shades of difference cumulate to comepnsate each other whereas letter grades do not. This is true in post-graduate degrees where it is 'pass'/'fail'. The system requires all questions are graded indiviudally by number score, even if by the third question out of five the candidate is a clear pass or a clear fail. At EBS we record the mark for every question in every paper and the total, transformed later into letter grades (for international and inter-university comparisons).

I have seen papers with marks showing 9, or 11, or such out of 100. The lowest I can remember had 4/100! Clear fails. I am not sure why the range is 40-80 as you suggest. Or why it is not really out of 80 and not 100. 90+ is perfection, i.e., what I would expect the faculty answer (written by faculty when sending the exam paper into the system) to reach. Whether it would be publishable is not an issue because that involves grammar, style and spelling not usually associated with writing under time pressure. Many faculty in universities cannot write to a publishable standard without a great deal of effort lasting much longer than the 3 hours allowed in an examination.

Grade inflation is a form of corruption of an exam regime. It is present in both the US and the UK wherever the exam regime is softened and it squanders the value of a University's attestation of the fitness of the candidate. Our exam regime is not just a product of our DL programme. It is rooted in what I consider to be the appropriate standard for examinations in a university whether on campus or DL.

 

Hi Professor,

Thank you for your detailed thoughts on grading. I think we are essentially in agreement. The reason I insist that the UK system is effectively 40-80 is that scores below 40 are failing, and scores above 80 are so rare as to be virtually inconsequential. So I'm not sure that the UK system really gives more latitude compared to the US system.

One thing for sure in my experience with UoL's external program is that there is no grade inflation! The grading is very tough. From the sounds of it, so is the mba grading at EBS. My own sense is that perhaps the average school in the UK tends to be a little more stingy than the average school in the US. I can't really back this up - it's just a sense.

Tom

 

Marilynd,

It is irrelevant whether you start me out with a 0 and start adding points or whether you start me out with 100 and lop off points. In short, if you start me off with 100 points and lop off 10 or 20 as I screw up occasionally, I will have the same A or B at the end of the term that would have been the case if you started me out with 0 and added 80 or 90. So, what's the diff?
It seems to me intuitively obvious to the most casual observer that this whole grade inflation rigmarole is nothing more than a worthless act of intellectual masturbation cooked up by a few brainless EdD's in need of a few brownie points from Publish or Perish University.

 

Marilynd,

It is irrelevant whether you start me out with a 0 and start adding points or whether you start me out with 100 and lop off points. In short, if you start me off with 100 points and lop off 10 or 20 as I screw up occasionally, I will have the same A or B at the end of the term that would have been the case if you started me out with 0 and added 80 or 90. So, what's the diff?
It seems to me intuitively obvious to the most casual observer that this whole grade inflation rigmarole is nothing more than a worthless act of intellectual masturbation cooked up by a few brainless EdD's in need of a few brownie points from Publish or Perish University.

 

"It seems to me intuitively obvious to the most casual observer that this whole grade inflation rigmarole is nothing more than a worthless act of intellectual masturbation cooked up by a few brainless EdD's in need of a few brownie points from Publish or Perish University."

You sure have a chip (maybe a forest?) on your shoulder.

'casual observer' - who they?; 'grade inflation rigmarole' - what is that?; 'worthless act of intellectual masterbation' - how is that done?; 'cooked by a few brainless EdDs' - how do they get up in the morning and function without brains?; 'a few brownie points' - how are these measured?; 'publish or perish university - where is this located?

The serious point you make - why does it make a dfference - is lost in your verbage. This is supposed to be a Board to which reasonably educated people seek information, exchange knowledge and challenge concepts. And you have two MBAs from a reputable British University!

Your point would be reasonable if the two systems of grading were the same. If they were the same it would not make a difference. But they are not the same in practice.

Consider and examination system that tests how well you have understood the subject's concepts and how well you demonstrate their application to business problems - the MBA, for instance. Now what is the assumption of the standard?

Where faculty, subject experts, prepare 'faculty answers' to the questions they set and these are used as grading guides they complete this task in their own time and not under exam conditions. Their answers will become public after the exam. It is appropriate they be expected to get their answers correct (near perfect even, because students who fail will have grounds for an appeal if the professor's answers are shown to be incorrect, incomplete, or too general).

It takes longer to write answers for a three hour exam than is available to the students. Hence, the correct standard is already beyond the average student, even the better ones. So marking for the 'correctness' of a student's MBA paper means even the best student is going to fall short of the high 90s; the same should be true for the 'deduct for errors' method. But is it?

If the graders are (consciously or sub-consciously) grading a student's paper to the standard achievable in a third or even a tenths of the time taken by faculty to write the 'perfect' answer by a student, then a whole range of potential, though unobtainable' marks are discounted in the 'correct answer' method and too few are deducted in the 'take marks off for errors' method and this leads the graders into 'grade inflation'.

They are not grading against excellence, though the marks they give suggest they are. They are grading top students' performance against a 'softer' average standard based on what we may expect average students to obtain given the difficulties of writing exams. Meanwhile the 'no marks for errors but marks for correct points' graders are marking in a lower range than the other method. The mark against excellence standard leads to lower numerical grading in the mid-60s; the mark against average student capability (which translates in time to exams dominated by mere knowledge not application testing) leads to higher numerical grading in the high 90s. It is this phenomenon that is known as grade inflation, irrespective of the publishing regime in a university.

I won't develop this theme, though it is implicit in what I have contributing to this and similar debates here for some years. I am sure that if you were an experienced university grader either in the UK or US, you would appreciate these points and realise that what may be 'intuitively obvious to a casual observer', in this and other occasions, is not a reliable guide to uncovering the best policies for the serious issues under discussion.

 

Prof. Kennedy,

1. Chip on my shoulder? Quite possibly. But a "forest (?)" on my shoulder ?

2. Casual observer? Anyone with half a brain.

3. Grade inflation rigmarole? By using the term "rigmarole" I intended to convey the idea that it seemed like a lot of hot air.

4. W******** A** of I*********** M***********? Something that's fun to do but doesn't really lead to much

5. A few brainless EdDs? Probably it is an overstatement to refer to EdDs (or education professors or anyone else in the grade inflation debate) as brainless. But the grade inflation argument just seems to me like grasping at straws.

6. A few brownie points? Call them publication credits or whatever

7. Publish or Perish University? Any university that insists that quantity of publications should be considered to quality of both teaching and publications in faculty hiring, tenure, and retention decisions.

8. Two MBAs from a reputable British university? Actually, the City University that my two MBA degrees are from is a fairly respectable university in Washington state.

9. My point lost in verbiage? I suppose that this is an objection to certain certain choice phraseologies that just might be construed as a bit rough by some. I am certainly sorry if I have offended you or anyone else on this list. On the other hand, there was an off-topic discuission re use of "rough language" and education or the lack thereof; it was concluded that there is no relationship between the two inasmuch as our language habits, shall we say, are formed much earlier in life than our adult educational accomplishments.

All in all, Prof. Kennedy, I would expect a large portion of A's and B's at certain schools, to-wit: (1) schools with extremely high reject ratings (80%-95% of applicants for admission rejected, e.g., Harvard, Yale) and (2) schools where the average student is a mature adult who knows what he or she wants (usually having to pay one's own way serves as a motivator to do better than the kid that thinks that college is a Daddy-funded four-year beer party).