Academic grading in North America

The following is a summary of the academic grading systems in North America.

Canada

In the Canadian province of Ontario, another system is placed that replaces the A–F system. This system was instituted by the provincial government in around 2001 . It is very much the same as the A–F system but uses numbers instead of letters. It goes like this:
* Level 4 = A or excellent (exceeds provincial standard, 80%–100%)
* Level 3 = B or good (meets provincial standard, 70%–79%)
* Level 2 = C or average (approaches provincial standard, 60%–69%)
* Level 1 = D or poor (well below provincial standard, 50%–59%)
* Level R = F or failing (remedial action necessary, 0%–49%)

The system also adopts the +s and −s of the A–F system. So a 4− is about equal to an A−. Some teachers may also attribute the +s and −s to the R, meaning that an R+ is an almost fail, and an R− meaning no work or work of inferior quality. Some teachers have been known to become overzealous and give students 5s for spectacular achievement and −1s for below what is possible. These are usually converted to 4+s and R−s on the report card. The students' marks in Canada are also weighed differently, the marks are divided in four categories, "Knowledge", "Thinking and Inquiry", "Communication" and "Application". The categories are worth different amounts depending on the course. For example, a knowledge-heavy course such as math would have "Knowledge" worth more than "Communication" while an English class would be the opposite. Lastly, in secondary school, the categories are equal to around 70% with the exam and culminating performance task worth the other 30% of the mark. Also a student may not get lower than a 20% in a class as long as the student hands in work.

The English system is very similar to the A–F system, but with the inclusion of the grades G, U and at GCSE level A*. A* (at GCSE) or A (at A level) is the highest, C is average, E is the minimum pass mark and U being unacceptable. Modifiers such as B− or B+ are not used as extensively as in the US and final qualification grades are never expressed as such. FF is never used.

Percentage-based grading

In objective subjects such as mathematics, grades are normally computed according to percentages such as class attendance, homework completion, and test averages. A weighted average of these variables is used to compute one percentage, which is the index from which grades are determined.

In subjective disciplines where essay exams and papers are more common, grades are sometimes represented numerically, other times with letter grades.

The specific conversion of percentages to letter grades varies according to the class. In classes with very difficult problem sets, it's not unheard of for the cutoff for passing to be 20%, and that for an A grade to be given at 50%.

Usually, though, primary and secondary schools use fixed systems. The traditional system is the "Tens System", written as (90/80/70/60). In other words, the lowest A (or A/B line) is at 90%, while the lowest D (or D/F line) is at 60%. In order either to set a higher standard or correct for grade inflation, however, some schools use the "Nines System" (92/83/74/65) or "Eights System" (either 93/85/77/70 or 94/86/78/70). Usually, the system employed may affect grading, since difficulty of exam questions may be calibrated to the grading system; indeed, exams in a school using the Tens System are often more difficult than those in schools using the other systems.

The Tens System is used in Canada but the A–F system (or in the case of Ontario, the 0–4 system) values are different from those of the United States. It goes as follows:

* A = 80 or higher
* B = 70–79
* C = 60–69
* D = 50–59
* F = 0–49

The pluses and minuses are taken into account also, so a plus is closer to the higher end of the score or the minus is at the lower end of the score. The percentage system is not used in primary schools, as all marks shown on tests, assignments and on the report card are shown with the A–F or 0–4 system depending on province. Percentage may also be provided along with tests. In senior elementary or secondary schools, tests and assignments are provided with both the mark on the present system in the province and also with the percentage. On the report card, only the percentage is shown on the final mark.

Various rubrics exist for assigning pluses and minuses, usually assigning them to roughly the top and bottom third of a grade level, with the base grade (that is to say, the one that does not carry either a plus or minus sign) being widest if the number of points in the entire letter grade are not evenly divisible by three. For example, under the Tens System, the plus grades will most commonly end in 7, 8 or 9 while the minus grades will end in 0, 1 or 2, with the base grades ending in 3, 4, 5 or 6. In the Nines System the plus, base and minus bands are typically equal (spanning three points each) while plus and minus formats under the Eights System vary widely; often the base grade will consist of four points and the plus and minus grades will consist of two points each, but this is far from universal. If A+ is omitted (and it often will be if the institution does operate under the grade-point system), the A and A− grades may cover the same number of points (or the A will contain one more point if their sum is an odd number), or the A− range may not be larger than those of the plus and minus grades found elsewhere along the grading scale, and the A range will be twice a large as that of the other base grades.

United States

The most commonly used index in the U.S. educational system uses five letter grades. Historically, the grades were A, B, C, D, and F—A being the highest and F, denoting failure, the lowest. In the mid-twentieth century, many American educational institutions—especially in the Midwest (particularly the State of Michigan)—began to use the letters A, B, C, D, and E. The only difference here is that failure is denoted by E instead of F, which is not used by these schools. By comparison, the grade E is sometimes used in Canada as a conditional failing grade. No grades awarded on American quality indices are conditional, except special grades like I (Incomplete) and Y (course on non-traditional calendar, assigned to regular term in which the student enrolled in the course).

The A–F (A–E) quality index is typically quantified by correlation to a five-point numerical scale as follows:

*A = 4.0
*B = 3.0
*C = 2.0
*D = 1.0
*F = 0.0

Chromatic variants, represented by + and −, are commonly used. They are most commonly quantified as x.3 and y.7, e.g., B = 3.0, so B+ = 3.3 and B− = 2.7). A few institutions use only a single midpoint between the major points on the scale; that is, they regard an A− as effectively the same grade as B+. In those cases, an AB replaces the options of A- and B+ and is quantified as 3.5; a BC replaces B− and C+, with a value of 2.5; and a CD replaces C−/D+, worth 1.5. This approach is unusual and is most notably typified by institutions in the state of Wisconsin.

The grade A+ is a novelty in American education. The minority of institutions that use it may quantify the grade as 4.3 or 4.5, but many of them quantify A+ as 4.0 on the theory that a 4.0 scale cannot go higher than 4.0. By convention, quantitative scales are called by the highest whole number, so there is—at least, conventionally—no such scale based on 4.3 or 4.5, but it is still a 4.0 or 4-point scale because the fraction is ignored in naming the scale. D- is also rarely found, under the assumption that anything less than a D is by definition failure.

American high schools and universities sometimes weight their GPAs.

In the late 1970s and early 1980s, many primary schools began to employ quasi-eccentric quality indices in which E, historically a failing grade, was recast to represent "Excellent." Similarly, the graduate business school at the University of Michigan awards the grade X to represent "Excellent." (Please see the section on the The E-S-N-U system.)

American high schools typically require a 1.0 grade point average to qualify to take a diploma. The industry standard for undergraduate institutions is a minimum 2.0 average. Most graduate schools have required a 3.0 grade point average since 1975 (the transition began two decades earlier), but some schools still have 2.75 as their pass standard. Some doctoral programmes do not have a formal pass standard. For example, the Michigan Doctorate, conferred by the Rackham School of Graduate Studies at the University of Michigan, is awarded solely on the basis of competence in research. It is unlikely, however, that the University of Michigan would retain a student who is doing work below 'B' quality, notwithstanding the grade point average "is" technically irrelevant to conferral of the degree.

American law schools are notoriously out of step with mainstream graduate-level education. Most of them still require no more than a 2.0 grade point average to qualify for the professional doctorate in law. A few require 2.3 or 2.5 for post-doctoral degrees, such as the American LL.M. or S.J.D. degrees. Law schools also typically continue to award the grade D whereas the industry standard is to eliminate it from the graduate-level quality index.

The [http://www.rackham.umich.edu/ Rackham School of Graduate Studies] , for example, uses the following 9.0 scale:
*A+ = 9.0
*A = 8.0
*A− = 7.0
*B+ = 6.0
*B = 5.0
*B− = 4.0
*C+ = 3.0
*C = 2.0
*C− = 1.0
*E = 0.0Apart from law schools, graduate schools in some states (California among them) continue to award the grade D in graduate school, despite having a 3.0 degree pass standard—measured against which a D (1.0) is normally considered superfluous, because even B− (2.7 or 2.5) is a failing grade in most graduate schools.

Standards based grading

With the evolution of Outcomes Based Education to standards, most states have created examinations in which students are compared to a standard of what educators, businessmen, parents, and other stakeholders have determined to be what every student should know and be able to do 10 years into the future. Students are graded as exceeding, meeting, or falling below the standard. The advantage is that students are not compared against each other, and all have the opportunity to pass the standard. However, the effect is that the standard is typically set at a level that fails 50 to 80 percent of students in the first year, and nearly all non-college bound students, so the failure rate can exceed the 50 percent of students who are defined to score below grade level. Though the passage rates for all groups rises, the failure ratio of minorities typically remains at 2 to 4 times higher than majority students.

As an instrument of systemic reform, the tests are targeted to items and skills not currently in the curriculum to promote adoption of methods such as constructivist mathematics, inquiry based science, and problem solving.

As the curriculum is aligned to match the test, more difficult problems are removed, and passing percentages are changed, these tests generally show rising grades each year. Passing levels however are not comparable to other states, or even between years, and typically rise much faster than standardized tests given over the same time period. After the goal is reached, the standards and tests are changed again. In most states, the diploma is or will be attached to meeting standard as a minimal requirement, which has met some resistance. Some critics have charged that the standards are fuzzy and not content-based. One person who has graded such exams found college level writing that was graded as below standard in a 10th grade test because it did not meet a detailed rubric, while some 5th grade level papers did meet the rubric for content. Fewer than 1 percent of papers were permitted in the perfect category, a standard higher than admission to Harvard. Mareen DiMarco testified that the first year of the CLAS test, graders were forbidden to give any 4 exceeding standard grades.

Rank-based grading

Articleissues
section = y
criticisms = December 2007
weasel = December 2007

Informally, grading on the curve refers to any system wherein the group performance is used to moderate evaluation — grading need not be strictly or purely rank-based. In the most extreme form, students are ranked and grades are assigned according to a student's rank, placing students in direct competition with one another.

The following is an example of a grade distribution commonly used when this sort of grading is employed.

These percentages derive from a normal distribution model of educational performance. An A is given here for performance that exceeds the mean by +1.5 standard deviations, a B for performance between +0.5 and +1.5 standard deviations above the mean, and so on.

Grade-rationing

Rank-based grading is popular among some American educators, usually under the euphemism of grade-rationing. The arguments for grade-rationing are that

* grade inflation represents a serious problem in education, that can only be counteracted by the enforcement of rank-based standards. As technology and general public knowledge increase over time (i.e. what is known about calculus, subatomic particles, enzymes, etc.) there is a constant requirement that in the higher levels of education there be a modification of the difficulty and content of courses to reflect both the increased knowledge of each incoming class compared to previous classes, as well as the increased knowledge of the scientific community in relation to the topic being taught. For example, many topics which in past years would have been generally considered upper division college courses are today taught at the high school level, and so this is evidence that over time there has been a transition occurring in the general amount of knowledge possessed by the public about such topics. If non rank-based methods of grading are used to evaluate newer students, then educators will be left to decide the grades for the students based on historic/past measures of performance in the subject matter, but this invokes the problem that progressive classes will necessarily be less and less challenged over time than they would otherwise be challenged, not because they are not capable of learning more, but rather because the method of measuring their performance is based on students of the past, who in most cases did not possess the same level of basic background to the material. Rank-based grading can therefor arguably push classes to their greatest performance potential as conditions change over time.
*In terms of measuring the student ability to learn rather than simply what the student has learned, rank-based grading is uniquely advantageous. By requiring that the top 30% (or any other given percent) of students receive a separate distinction than their peers, the grading system becomes a relevant measure of student performance in relation to their peers. While this does have the disadvantage that the sample-size only includes a certain small group of students, such a disadvantage does not disappear with any other method of measuring student ability to learn, because even if a nationally standardized test were used to determine student performance, the outcome would be based on such local factors as who taught the students, which does not say anything about student's ability or inability to learn or learn to perform, but rather about the raw information the student has learned. If grades are meant to represent the student's ability to learn, rather than the mere abilities learned, then rank-based grading shows clear superiority in methodology to non-curved methods of grading, because it incorporates evaluation of students who all had the same instructor with the same lessons and homework during the same time period, which non-rank-based grading cannot do. Admittedly, however, if the purpose of grading is purely to indicate abilities learned, rather than the students ability to learn, then it may be argued that a non-rank system might show some advantages.
* As many corporations used rank-based evaluation measures, sometimes even related to termination (see: rank and yank) such grading prepares students for the real world, in that they better understand where they are at academically in relation to their peers, who will later be their competitors in the job market. In this way, rank-based grading prevents the disillusion that students are competitive in areas in which they are actually only competent.

Case against rank-based grading

Despite near-universal concerns about grade inflation, rank-based grading systems are mostly out-of-favor in the contemporary United States. When rank-based systems are used, in education or employment situations, cutthroat behavior and cheating become rampant. In some situations, high-scoring students are disliked by their classmates for raising the curve.Cited as a case against rank-based evaluation specific to employment, Enron used a rank-based evaluation scale; it was argued that the cutthroat environment created there resulted in the disgrace and downfall of the corporation. Fact|date=February 2007 Microsoft in 2006 announced abandoning a similar disliked system. Some predict that analogous problems, on a more minor scale (cheating, theft of reserved materials) will occur in schools that use rank-based grading.

Critiques of this justification of rank-based grading assert that it is simply a way for universities and colleges to maximize their revenues. Implementation of stringent rank-based evaluations, where only 30% of the entering class are permitted to graduate, are a very clever way to collect tuition from a vastly larger student population than they ever intend to graduate.Applicants are often baited into a program that promises job security and then switched into one of the institution's less desirable majors due to poor academic performance. However, if all of one's classmates are in the top 30%, then only the top 10% would be allowed to graduate.

Another criticism of rank-based grading is that it only measures performance in a given group, but not the real potential or abilities of a given student. There is no actual evidence that a given group really performs along the curve, the overall level of the group might be better or worse, or the distribution does not match the pattern at all.For example in a generally good class the pressure to assign grades along the curve would produce an artificial 7% of F-students, although all students actually performed quite well. This also works the other way round: in a class with generally bad performance the students whose performances are not totally bad would be singled out to form an artificial group of A-students, although in another context they would never get these grades.Essentially, this criticism means that rank-based grades become meaningless when taken out of the context of a given class or school. This would mean that these grades are useless to compare students of different institutions, as it is done with of college applicants, job applications etc. The most important reason for issuing grades at all would therefore become meaningless.

Grade point average

Grade point average (GPA) is a number that represents the average of a student's grades during his or her time at an institution. Usually it is weighted by number of credits given for the course.

Most high schools and nearly all colleges in the United States use a four-point system. Universities in Hong Kong and Canada, as well as some schools in Singapore also use this system. Numerical values are applied to grades as follows:

* A = 4
* B = 3
* C = 2
* D = 1
* F = 0
* F-/G (if given) = –1

This allows grades to be easily averaged. Additionally, many schools add .3 for a + grade and subtract .3 for a − grade. Thus, a B+ yields a 3.3 whereas an A– yields a 3.7. A+s, if given, are usually assigned a value of 4.0 (equivalent to an A) due to the common assumption that a 4.00 is the best possible grade-point average, although 4.3 is awarded at some institutions. In some places, .25 or .33 instead of .3 is added for a + grade and subtracted for a − grade.

Weighted GPA

Some high schools, to reflect the varying skill required for different level courses and to discourage students from selecting easy 'A's, will give higher numerical grades for difficult courses, often referred to as a weighted GPA. For example, two common conversion systems used in honors and advanced placement courses are:

* A = 5 or 4.6
* B = 4 or 3.5
* C = 3 or 2.1
* D = 1
* F = 0

Another policy commonly used by 4.0-scale schools is to mimic the eleven-point weighted scale (see below) by adding a .33 (one third of a letter grade) to an honors or advanced placement class. (For example, a B in a regular class would be a 3.0, but in an honors or AP class it would become a B+, or 3.33).

Sometimes the 5-based weighting scale is used for AP courses and the 4.6-based scale for honors courses, but often a school will choose one system and apply it universally to all advanced courses. A small number of high schools use a 5 point scale for Honors courses, a 6 point scale for AP courses, and/or a 3 point scale for courses of below average difficulty.

Eleven-point system

A few high schools in the United States use an eleven-point system. In this system, one point is usually added to weight a more challenging course. Numerical values are applied to grades as follows:
* A = 11
* A− = 10
* B+ = 9
* B = 8
* B− = 7
* C+ = 6
* C = 5
* C− = 4
* D+ = 3
* D = 2
* D− = 1
* F = 0

The E-S-N-U system

At one time (until roughly the mid-20th Century), the most popular grading system in the United States used four letters, which ranked, in descending order:

* E (excellent)
* S (satisfactory)
* N (needs improvement; NI was also used interchangeably)
* U (unsatisfactory)

This system has largely been replaced by the A–F system dealt with previously, but is still encountered quite often at the elementary school level, particularly in kindergarten and Grades 1 through 3 (this educational level being frequently referred to as primary school). It is also occasionally used at schools for older children, including high schools, especially in the issuance of conduct or citizenship grades.

There are a few variations to this system, including the use of an O (for "outstanding") grade, which is even higher than the E; a G (for "good") placed between the E and the S; the use of a G (again for "good") instead of the E; and the lack of a U grade. Some high schools also use this system for grading classes as an alternative to the traditional grading system. In this version, E stands for "exemplary" and P proficient, with AE and AP for work that approaches the E and P levels. "Credit" is equivalent to the D level and "No Credit" is equivalent to F. [http://www.csupomona.edu/~ceis/ipoly/I-Poly%20Handbook%202004-2005.pdf]

The S grade may be so modified with an S+ or S−, but otherwise plus and minus are seldom used.

A similar system is used to rank practical work in the certain science department of Oxford University; however only with the grades S (Satisfactory) S+ (more than satisfactory, and may be used in the allocation of degree grades) and NS (Not Satisfactory).

De-emphasis of grades

A number of reputable liberal-arts colleges in the U.S. either do not issue grades at all (such as Antioch College, Bennington College,The Evergreen State College, Prescott College, New College of Florida, and Hampshire College) or de-emphasize them (St. John's College, Reed College, Sarah Lawrence College). In all cases, the rationale is that grades often do not provide a clear picture of academic aptitude or of potential for success, and that learning, not achieving the highest score, should be the goal of a liberal education. In many cases, narrative evaluations are used as an alternative measurement system.