What constitutes sound testing and grading strategies? Here is one set
of guidelines that principals might find useful when evaluating teachers.
Holding teachers more accountable for their grading practices should have
a positive effect on the overall school climate.
Hills' (1991) article on "Apathy Concerning Grading and Testing"
contained an interesting idea. After describing a number of inappropriate
practices used by teachers, he proposed that principals base their evaluations
of teacher performance, in part, on testing and grading practices used by
teachers. The idea makes sense to me, but depends on the principal having
a set of criteria in mind.
Over the years, there has been a good deal of interest in appropriate grading
practices; much of what follows is based on the collective wisdom of experts
as reflected in textbooks, and some common sense. It is important to realize,
however, that there is little that compels teachers to be knowledgeable
about these practices, let alone use them.
1. Understand the teacher's perspective.
Teachers tend to find a way to have their grading policy reflect their own
deeply held beliefs (not necessarily knowledge) about how students should
be graded, even if building or district-level grading policies exist. Often,
this is a reflection of how their own teachers graded them and is characterized
by a strong sense of ownership. Principals should not be too surprised if
raising questions about grading practices evokes some fairly powerful reactions
from teachers.
2. Ask for a copy of the teacher's grading policy
Teachers should submit a copy of their grading policy to the principal.
I'm not talking about the classroom rules: Be in your seat when the bell
rings, no gum chewing, etc. Some research indicates that many teachers see
grading and behavior management as the same thing; they are not. What principals
need to see is a detailed explanation of how the student will be graded-what
pieces of student work will contribute to the final grade; what percentage
each one counts; how often grading information will be gathered; how the
pieces are combined to get a final grade; what grading scale will be used,
and so forth.
It is a plus if the written policy includes a course description with the
general instructional objectives the teacher hopes to accomplish. These
objectives should be stated in terms of what students will be able to know
or do by the end of the course. It is also helpful if teachers include their
philosophical perspective on grading-an indication that they have thought
deeply about the process. Principals should regard this as a contract between
the teacher and students.
3. Beware of traits that do not measure achievement.
Work that counts toward the final grade should include such factors as tests,
quizzes, assignments, and projects. Most of us probably have a clear idea
of how a test results in some kind of a score that could be used for grading
purposes.
But what if a teacher lists something like "attitude" or "effort"?
Principals should question how such traits are assessed. What do they mean?
Does the teacher keep track of the attitude displayed by each student each
day and assign a score? Is each student's level of effort similarly measured
each day? What about other traits like student ability or behavior?
I believe that attitude, effort, ability, behavior, and the like should
not affect grades at all, because grades should reflect the level of student
achievement in each course. If there is interest in non-achievement traits,
the report card should have a place where this information can be shared,
separate from the course grade. If teachers are allowed to mix something
like attitude with course grades, there is a real danger that grades will
be strongly influenced by teacher opinions as opposed to work that can be
graded objectively.
4. Consider the quality of the grading information.
When a test is given, a fundamental expectation is that the students have
been taught what is to be tested. Do teachers routinely articulate their
objectives and then develop an assessment that measures whether or not the
objectives have been achieved? Classroom activities should revolve around
ensuring students perform well on the assessment, thereby indicating that
objectives have been mastered. Often, however, teachers develop instructional
strategies or materials without a clear vision of their objectives or the
assessment. Or worse, the end-of-unit test found in the textbook dictates
the instruction.
On the grading policy, principals should expect teachers to state their
general instructional objectives for the course. Then, as teachers are observed
as part of the evaluation cycle, the principal should ask the teacher to
produce the specific learning outcomes that are the focus for the lesson
being observed and the assessment that will be used, to see if the outcomes
have been achieved. Principals should be concerned if the teacher cannot
produce the assessment. Having the outcomes and the assessment gives principals
a context within which they can judge the quality of the instruction being
observed.
5. Think about the role of homework.
Teachers commonly grade almost everything students do. Some have reported
to me they have 40-50 items in their gradebook for each quarter. Some run
out of room. Teachers have also admitted to me they would rather not score
all that homework, but if they don't, students will not do it. Something
may be gained, but there is also something lost.
Many teachers like to point out to students that they can learn from their
mistakes. Homework gives students a chance to see if they have mastered
certain knowledge or skills. It is a chance to practice. I would like to
see teachers (with the support of principals) explore other ways to motivate
students to do their homework. At some point, teachers will administer a
test of some kind to see if student have learned from the mistakes they
were free to make while doing homework. These are the scores that should
be used in assigning final grades. Certainly some work/projects done at
home could be used for grading purposes, but many teachers grade every assignment,
which pressures students to get it right the first time.
6. Expect teachers to use numbers.
If the teacher is entering letter grades-i.e., A's, B's, etc.-into the gradebook,
there is cause for concern. Many teachers have grown accustomed to assigning
letter grades to everything. This is probably because the standard letter
grade system is easily understood by students and parents or guardians.
Perhaps they even demand this system.
But what do teachers do with a string of letters at the end of a grading
period? How do they arrive at the final grade? Well, many teachers change
the letters back to a number (an A = 4 points, a B = 3 points, etc.), average
those numbers, and assign a final grade. But much information about student
performance is lost in the process. Worse yet, they may simply peruse the
series of letters and estimate the level of performance. The greatest precision
is maintained by keeping the information in its original form-a number-until
a letter grade must be assigned.
7. Know how numbers should be combined.
How should teachers combine the numbers in their gradebooks? Usually, teachers
count things differently-e.g., tests will be worth more than quizzes. Be
sure the teacher figures a separate score for each part of the final grade.
For example, if the student has taken nine tests during the quarter, a final
score just for tests should be figured. Once this is done for all parts
of the final grade-tests, quizzes, assignments, projects, etc.-it is important
that teachers count each part as much as they said they would (check the
written grading policy). If, for example, the tests were to count 60 percent,
they must affect the final grade by that much.
This can be done in various ways, but the important thing is that teachers
should be able to explain the process to both students and parents or guardians.
They might practice by seeing if they can explain it to the principal. If
teachers seem reluctant to do this, it may be because they have a string
of numbers (or worse yet, letter grades) for each student that are not grouped
by tests, quizzes, etc. It may also be that teachers are not able do the
necessary calculations, even though grading software makes this job fairly
easy. As a result, teachers often do not actually follow what is written
in their own grading policies.
8. Encourage consumerism.
Students have a right to information that helps them understand their level
of competence, and principals should support this right. When teachers go
over tests with students, some description of how the group performed should
be made available. I encourage my own students to keep track of their scores
during the semester and check with me if something doesn't seem right when
the final grade is assigned. Their scores and the overall performance of
the class on any test/assignment are always available to them on the spot.
I have found that teachers are sometimes not receptive to requests of this
nature; perhaps they see it as a challenge to their authority.
After each test, I share with students information about the group's performance,
usually a list of the scores and the average score. I also help them interpret
their score by posting the range of scores for each letter grade. I remind
them, however, that it is their score that will be recorded in the gradebook
and weighted as described in the grading policy. Great care is taken to
protect each student's right to privacy with respect to his or her performance.
I would like to see students (and their parents or guardians), with the
support of the principal, feel comfortable asking for grading information
when questions arise.
9. Demand fairness.
To me, the foundation of grading is fairness. Principals should be open
to the possibility that an individual teacher might not agree. I was recently
involved in a situation where a teacher who, without grounds, raised the
final grade of one student because he had vociferously complained. What
about the rest of the students in the class who might be less vocal (or
more fearful)? Or consider teachers who count attitude, effort, ability,
or behavior as a major part of the course grade. How do teachers begin to
fairly assess such traits in all their students? These should be regarded
as serious threats to fairness. The key is to have each teacher's policy
in writing and not hesitate to discuss fairness issues with teachers if
a problem arises. If we do not begin with the expectation that all students
will be treated fairly (to the extent this is humanly possible), it will
be impossible to know what grades mean.
10. Hold teachers accountable
Teachers should know that grading competently is perhaps the most important
service they provide their students. Principals should not hesitate to send
that message. Generally, teachers have not been held accountable for the
grading practices they use. Principals who communicate their expectations
will help ensure greater accountability in grading.
These guidelines can be used to help principals grade a teacher's grading
policy. But we can't have it both ways. It seems clear that there is great
interest in grading among principals, but is there the will to hold teachers
accountable for their grading practices? This would mean that the "wiggle
room" that is often inherent in a teacher's grading policy would be
reduced, and greater precision would be expected.
Do we want teachers to have sufficient "wiggle room" so a student
who earns a B for a course can be assigned a higher grade because he had
a good attitude, tried hard, didn't have very high ability to begin with,
or behaved really well? How do we know students are trying hard? Even if
they are, shouldn't actual competence in the subject matter be reflected
in a course grade as opposed to actions that might please the teacher? In
the end, long after grades are forgotten, won't the real achievements of
students matter most? I would encourage principals to demand a grading policy
from each teacher that ensures a fair evaluation of what students know and
can do.
Students' Rights
Principals might be interested in whether students or their parents or guardians
have a legal right to question teachers' grading practices. Court rulings
have been mixed in cases where grades have been reduced as a punishment
for student misbehavior and/or nonattendance. In Wermuth v Bernstein (1965),
the New Jersey commissioner of education wrote:
The use of marks or grades as deterrents or as punishment is likewise usually
ineffective in producing the desired results and is not educationally defensible.
Whatever system of marks and grades a school may devise will have serious
inherent limitations at best, and it must not be further handicapped by
attempting to serve disciplinary purposes also.
In Dorsey v Bales (1976), the court of appeals upheld a lower court ruling
that reducing the grades of a student who had been absent due to a suspension
for misbehavior was invalid. The suspension negated the school board's authority
to impose additional sanctions (grade reductions).
I am not encouraging legal action in cases involving grading and certainly
do not presume to have provided a complete legal history here. What is probably
most important to realize is that there have been rulings like those described
above, and teachers' grading practices can be expected to fall within certain
guidelines. Generally, though, if grading can be characterized as arbitrary,
capricious, or in bad faith, then perhaps grounds for legal action exist.
But should it even come to that? I don't see much that is unreasonable in
the expectations I have outlined above. On the other hand, I have seen some
pretty unreasonable grading policies. I would prefer to see teachers adhere
to professional standards like those developed by the National Council on
Measurement in Education (1995) when grading students. But there is no mechanism
for peer oversight within the field of teaching. Thus, I would argue for
administrative oversight, and principals can be the key.
Teachers have told me they do not include many details on their written
grading policy because students would not understand them. Some have also
admitted that a vaguely stated grading policy makes it easier for them to
modify it for individual students. As a result, students are sometimes left
with little recourse if a problem arises. Perhaps principals have been kept
at a disadvantage, also. I would like to see the entire policy, in writing,
out on the table so everyone-principals, students, and parents/guardians-knows
what to expect.
I believe the overall climate within the school would be enhanced by expecting
teachers to grade students in accordance with the guidelines presented here.
If written grading guidelines that measured up to a standard became the
norm in schools, I believe we would have more students who feel confidence
in the way that they are evaluated. Expectations would be clear, and students
would worry less about having their grades affected by factors not germane
to their competence in the subject matter. Outcomes like these warrant the
principal's commitment to upgrading teachers? grading practices.
References
Dorsey v Bales 521 S W 2d 76 (Ky 1976)
Hills JR "Apathy Concerning Grading and Testing." Phi Delta Kappan
7(1991): 540-45.
National Council on Measurement in Education. Code of Professional Responsibilities
in Educational Measurement Washington. SC.: NCMF 1995.
Wermuth v Bernstein S.L.D 128 (NJ 1965)
Stephen J. Friedman is associate professor of educational measurement and
statistics, University of Wisconsin-Whitewater; readers may continue the
dialogue on the Internet at friedmas@uwwvax.uww.edu.
What do grades really mean? Do they translate from one school or classroom
to the next? What do we hope to accomplish by giving grades? Should a school
have a consistent grading policy? Do the traditional A, B, C, D, and grades
truly reflect what we want to convey?
I believe that each of these questions should be answered by the educators
at every school in America if we are to regain the confidence of a our constituents.
The Way We Grade
Many educators today are questioning the way we grade students. Mastery
learning and outcome-based education (OBE) have helped to focus a great
deal of attention on grading, which may have contributed to what some have
called "dumbing down," or the lowering of standards and expectations.
We have questioned the role of competition, honor rolls, tracking, and standardized
tests. These trends have caused non-educators to question our motives and,
in some cases, our sanity.
We need to analyze where these trends in education have led us, and a if
we are where we want to be. Should we constantly jump on every bandwagon
without thinking of the ramifications that the approach may have, or should
we take all these new ideas and show some moderation as we implement them?
Each of the trends we have embraced in education has merit. Each of them,
if we had shown moderation, might have helped us gain the confidence of
the public and slowed the roller coaster we have ridden for the last several
decades. There are many good points in OBE, mastery learning, cooperative
learning, multiple intelligences, etc., but we have come under criticism
for the way we dove in to implement each trend. Had we taken the time to
analyze what is useful in this trend and then moved forward cautiously,
we might have come under less criticism.
According to Stephen Friedman, associate professor of educational measurement
at the University of Wisconsin-Whitewater (NASSP, 1995), educators need
to design an approach to grading that supports rather than discourages student
learning. When developing a grading strategy teachers need to think about
the values they and their institution share.
If a teacher or school feels that grades should reflect content mastery,
an absolute scale should be used. If a teacher or school feels that grades
should compare students against each other, a flexible scale should be used.
Are attendance, behavior, tardies, homework, parent expectations, and class
participation factors in grades? This complex and critical decision--how
to grade students--is often not properly thought through, and even less
often explained to parents and students at the beginning of the school year.
The Way We Should Grade?
Many schools are exploring block scheduling. If your school is considering
or has adopted a block scheduling model, it has taken a philosophical turn
that should lead to rethinking your grading policy. If your school is using
block scheduling, you are taking a new look at the way you do business.
A grading scale is an arbitrary range that sends a message to students and
their parents; be sure that message is the one you want to send.
In block scheduling students may have more time to achieve outcomes. If
that is the case (Canady and Rettig, 1995), you will most certainly want
to consider using grades of "I" for "incomplete," "NY"
for "not yet," or "E" for "exposed" rather
than restricting yourself to A, B, C, D, or F.
In a math class, for example, a student who has mastered some of the course
outcomes, let's say at the 60 percent level, is not the same as a student
who has mastered none of the course outcomes. With a traditional grading
scale they would both receive a grade of "F"; there is no way
to tell the difference. There is, however, a real difference in learning.
A grade is a vehicle for communicating with students, parents, colleagues,
and colleges. To communicate well we must make sure that the grade we give
reflects the student's degree of mastery of the course outcomes.
If an algebra course has 12 outcomes for the year and a student demonstrates
mastery of 7, a traditional grader could be tempted to give that student
an "F." According to Canady and Rettig (1995) that student should
receive an I (Incomplete or NY (Not Yet) under a block scheduling model
that allows students to have a reasonable amount of additional time to successfully
master the course standards. They would advocate giving a grade of E (Exposed)
when the student has gained some knowledge, but not enough to continue with
the next course in the sequence.
This student will probably need to repeat the course the next semester or
during summer school, if he or she wishes to continue in the sequence. In
my algebra example, this student might be able to demonstrate mastery of
3 or 4 of the course outcomes, certainly not enough to continue in algebra
with any chance for success.
Keep in mind that a grade is meant to be a form of communication and a grade
of F says the student has failed. That is not exactly a ringing endorsement
for the student's self-esteem. If the student deserves an F because he or
she has not tried to learn, as evidenced by lack of work or attendance,
that should be indicated differently than a student who deserves the F because
he or she was lying hard but did not learn. If the student can be successful
with more time, try the I, NY, or E. If it is unlikely that the student
will be successful, the F is appropriate.
While enhancing student self-esteem is not a primary goal in grading, we
should not needlessly hurt the student's self-esteem. In many block scheduling
models there is room to provide students with extra time. If we don't utilize
that time to help our students perform better, we are missing one of the
major benefits in block scheduling. While self-esteem is not our primary
goal, we do not want to have a needless negative impact on a student's self-esteem.
The Purpose of Grading
When looking at these possibilities, we must remember our purpose for grades,
i.e., to communicate progress to students and parents. If we want to communicate
accurately, we need to expand the choices available for teachers to use
so the purpose can be realized. If we have high expectations for our students,
we need to establish the expectations early and demand that students meet
the expectations. One way we can make demands on our students is to motivate
them to achieve by working with parents to establish the relevance and importance
of what we are doing in school.
What should be the relationship between the grades and national tests? There
should be some consistency between the grades given in your high school
and national tests such as the ACT and SAT. Do your honor roll students
perform at levels that translate to well above average on the college entrance
tests? If there is not a relationship between grades at your school and
the national tests, you are not adequately communicating to your parents
and students about their academic progress relative to the real world.
If the students who receive top grades in your school cannot match that
level of performance on the national tests, you may have consistently inflated
grades. If the grades are inflated, why are they inflated? Are your expectations
too low? Despite our continued cry as educators that standardized tests
are not the be-all and end-all of education, they are a measure of how well
students perform against an acknowledged standard. If a student gets A's
in a school, that student should be able to perform well on national tests.
Many educators are frustrated by the lack of intrinsic motivation shown
by students. Rather than allowing that lack of motivation to breed failure,
we need to find ways to motivate the students. One of our primary functions
as educators is to motivate students so they will want to learn what we
have determined they should know or be able to do.
We need to stop pointing the finger at others and take steps to solve the
problems we identify. The students in our schools are the best that their
parents have. Our job is to educate them to the best of our ability. I don't
often hear medical doctors throwing in the towel simply because a patient
has done so.
There are several other areas to consider: What is the relationship between
behavior and grades? Is it a behavioral or an academic issue if a student
is not doing homework? (Dockery, 1995) Many teachers deduct points for not
formatting a paper the "right" way. Many teachers also consider
class participation in their grades. Should behavior influence grades? One
solution is to give a grade for academics and one for behavior in each class.
Did you ever get a 0 on a test, quiz, or assignment? Did you ever consider
the effect a 0 grade has on a student's average in a class? Consider the
student who gets the following grades in a class: 85, 80, 75, and 80. That
would compute to an 80 average. What if the next grade is a 0? The average
would drop from 80 to 64 based on that one 0. The student's average dropped
16 points because of the one miss. Wouldn't it be fairer to the student
if the median were used rather than the mean? Under that scenario, the average
would not be dramatically affected.
A grading policy should be developed that supports the vision, mission,
and beliefs of the school. A key to sound grading is a sound philosophical
base grounded in a school's or teacher's beliefs. What do you hope to accomplish
through grading? What does an A really mean? Those are questions you can
answer if you have that set of beliefs to support your grading policy. Your
grading philosophy should also be consistent with your other policies and
procedures, as well as your organizational design.
If your school supports academic excellence and does not believe in "dumbing
down," you need a set of policies that support that desire for excellence.
If your grades truly represent that you demand excellence, teachers will
not give grades to students who have not legitimately earned those grades.
The concept of "not yet" or "in progress" is a more
realistic assessment of where a student is than a grade that hasn't been
earned.
To hold a student accountable for successfully learning the required material
is supporting excellence. To give that student a grade he or she has not
earned is "dumbing down." By using this approach you are at no
time misleading students or parents about the student's learning. If your
policy is that at some point time is up, the interim grade should be changed
to something that will indicate the necessary learning has not occurred.
It is certainly necessary to have all grades converted to permanent ones
for transcript purposes so that universities and other schools can reasonably
interpret student transcripts.
Summary
Each school should have a philosophy of grading that is embraced by the
entire faculty. That philosophy should realistically communicate progress
to parents and students. Where appropriate, we should use additional grades
beyond the traditional A, B, C, D, and F that will indicate how much more
needs to be learned to meet our expectations. Student grades should not
be based upon the behavior of the student but on the students' learning,
nor should students be given grades they have not legitimately earned. There
should be a relationship between the grades in your school and national
tests such as the ACT and SAT. Finally, there should be some agreement among
educators as to what we can and should expect from our students.
References
Canady, Robert Lynn, and Rettig, Michael D. Unlocking the Lockstep High
School Schedule. Princeton, N.J.: Eye on Education, 1995.
Canady, Robert Lynn, and Hotchkiss, Phyliss Riley. "It's a Good Score]
Just a Bad Grade." Phi Delta Kappan, September 1989.
Dockery, E. Ray. "Better Grading Practices." Education Digest,
January 1995.
NASSP. "Grading Can and Should Encourage Student Learning." NASSP
NewsLeader, March 1995.
Olson, Margot A. "Should BAD Behavior Mean BAD Grades?" Virginia
Journal of Education, April 1991.
Sykes, Charles. Dumbing Down Our Kids. New York: St. Martin's Press, 1995.
How to get in touch
Harold London welcomes communications about his ideas. You can write him
at hlondon99(insert at symbol)yahoo.com.