My inside source at a major southern public university informs me that over half of all LD course grades in the departments of Religion, Women's Studies, and English are As.... DOES Political Correctness lead to grade inflation? Or do we blame the rise of Post-Modern ethical relativism in these disciplines? --Orson
Depends on the goal of teaching? I'm not sure what the fuss is about "grade inflation"? There are essentially two competing beliefs in scoring education. One is that the goal is to establish social and academic pecking orders by assigning grades following a bell curve. The other orientation is that there is a standard of information or skills that need to be mastered in a course. Attainment can be measured following a rubric. The measure is as much an indicator of the teaching techniques used as the student performance. I tend to fall on the side that there should be a standard of information for most courses. Therefore every student could get an "A" or every student could get a "F" based on actual student product or performance. It seems to me that if the overall student product is low, it doesn't make sense to give the top 10% "A"s. At the same time, if you have met all the rubric criteria, why should you be given a "B" just to keep the number of As at 10%? I think every student should find out what the "academic contract" is for the college or class. I prefer the institutions which grade on a curriculum standard. But that is my bias. All the best!
I attended probably the only university in the world with a nine point grading system. I am not smart enough to twig the results to a system with 5 possible grades so you get raw data. I took one math class - class average 4.2 - 4 was conditional pass. A friend took a native language class - class average 8 - 7.5 was honours. Some students excelled in prof shopping and probably gained a grade point by doing so. In my business program almost all classes had precisely a 6 average - 6 being defined as average. I think that any prof who deviates from the curve in classes of 20 or more should have to write a detailed explanation as to why the particular class is smarter or stupider than the norm. I don't remember much of stats but I do remember that it doesn't take a very big sample size for the chance of error to decrease in signicance. There are no absolutes in testing so I prefer the curve.
Why is the "bell-curve" desirable Why was it decided that the "bell-curve" was "educational"? How does having more "C"s prove a course was better taught than more "A"s? The Elementary and High Schools are being pushed towards standards based criteria. If a student meets the rubric criteria for an "A", they deserve an "A". Don't they? I always find it discrimminatory when the "bell curve" proponents admit that some years an "A" would be an "B" and vice versa based on the students in the class. Isn't that fraudulant grading? The use of the bell curve also makes it harder to evaluate the caliber of teaching. How do colleges/universities evaluate the teaching ability of it's instructors? If a student has a choice between a college which bases grades on curriculum attainment and referenced standards, versus one that bases grades on a bell-curve, regardless of knowledge, Why would anyone willingly opt for the bell curve approach to teaching? Is the bell curve a teaching technique, or just a statistical tool? Criterion based, starndard referenced curriculums do promote teaching techniques. All the best!
irat It's just that the liklihood of a randomly selected class of 50 not fitting a normal curve is very small. I would suggest that instructor bias would be much more significant than any statistical error. A poor professor may be more likely to give higher grades to compensate for lack of ability. Who is going to complain if they are clearly graded above their normal performance level? There are no absolutes in testing, nor can there be without repeating the same test. I also believe that if an instructor can show that his class is more or less talented than average, then the grades should be adjusted accordingly.
Dennis why a normal curve? Dennis why do you feel there has to be a normal curve in grading? Where was it written that a normal curve has anything to do with learning? Do you believe in criterion and standards based instruction? All the best!
Re: Depends on the goal of teaching? I agree, Irat. HOwever, the problem for the internal governance of this 24,000 student institution--where lare intro level classes predominate--is that an "A" earned in LD English or Women's Studies or Religion simply doesn't comport with an "A" truely earned in an LD science or engineering course there. Therefore the local National Association of Scholars will expose this charade and profer the remedy that LD courses include a class ranking. That way outsiders can judge--or not judge--the worthiness of the 50th "A" in a class of 100, versus a 10th ranked "B" in the same class! I too perfer a grade based on mastery. I recall a truly great prof, now deceased and a founder of the academic study of American legal history, who admitted being forced to sunder grades on students--and asked my class in not just rhetorical desperation, "'what does measuring learning' really mean?" He embraced the notion that any such measure is inherently imperfect, and to some large degree, consequently unfair. However, students typically hold greater respect for profs who not only have high expectations for their students, but show some willingness to enforce such discipline. As much as enjoy collegiality in a seminar led by a prof, I also resent those who--embraced by increasing numbers of (especially baby-boomer) profs--disclaim ANY hierarchies of authority! THAT's fundamentally dishonest! Besides, if "we" are all equal learners, then I'm not getting my monies worth in paying to be (ultimately) judged on my level of learning! That would be fradulent--just as fraudulent as any Worldcom accounting scheme. One can embrace variance and admit to a range of ignorance while still being honest about just who has authority and why! We all differ in our ignorance, but not infinitely, and not indefinitely. --Orson
why a normal curve? Thanks Orson. I guess my question is more simple. Where was it written, what god decreed, why does anyone believe, why is it so ingrained, that a normal curve has anything to do with learning? People sometimes cite "discipline" as being tied to a normal curve. How? Is "discipline" not there for a t-distribution? What does a "high standard" have to do with a normal curve? Aren't mastery based courses using strict curriculums really setting "high standards"? There are many "curves" in mathematics. Why is the "bell curve" somehow "educational" but other curves not? I sit on a Workforce Investment Board. The business people don't give a hoot about "grades". The business people want to know what someone can actually do. What skills are guranteed by the learning process. They want standards based education. What does meeting a standard, on a rubric, have to do with assigning grades based on a mathematical curve? All the best!
Grade assigment has everything to do with comparing students. Does mastering a subject rate an A or does excelling rate an A? If mastering rates an A then 2 grades A and F would suffice - pass/fail. Perhaps percentiles would be a better grade. Other than testing errors, it is without bias as far as reporting is concerned. The user of the data would have to make value judgements such as a 40th percentile at Harvard equalling a 70th percentile at Acne State University. Note that there is no direct comparison using letter grades. There would be great resistance to percentiles as a B at Harvard sounds a lot better than the 20th percentile.
Okay, so what's the deal when the sole standard for a grade is one comprehensive, final exam? What if there are, say, 50 students in the class, the (essay) exam has a possible total of 300 points, the highest score is 240, and the average is 200? Does that indicate that the students failed to "master" the material (for whatever reason) or that the exam was either extremely difficult or otherwise flawed?
What if the average score was 100 with a high of 149. Is everyone unworthy enough to fail. If the curve fits use it.
what does the bell curve have to do with "excellence?" I don't believe that when I grade a student product, it has anything to do with other students. I don't tell a student, well this would have been an "A", but I gave Sally an "A" so you get a "B". My idea is to compare a student product to a standard rubric for a quality. This creates standards which can be the same year after year. This means I can tell an employer that my student has specific skills. To me that is excellence. I don't know how saying a mathematical curve determines a grade has anything to do with excellence? Why not award grades based on student height. Height pretty much can be traced on a normal curve. It sounds like others believe you cmpare one student's work with other students. To me that creates relative grades without standards. When I take a college class, the contract is between myself and the college. I want my work to stand on it's own merits. The other students are irrevalent in the grading of my product. When I teach a class my students want the quality of their work assessed on it's own merits. Would you apply a mathematical curve to the award of doctorates. Well now, we expect 10% of our doctoral candidates to fail, regardless of how good their work is. All the best!
