Language Assessment
Session 10: Chapter 11 - Grading
and student evaluation
(Brown 2004: 281-302)
• What is the difference between assessment
and evaluation?
Assessment Evaluation
• is concerned with how well • is much broader
our learners have done • is concerned with how well
• provides information for our program or course has
improving learning and served the learners
teaching.
• focuses on learning, teaching • In a course, the focus of
and outcomes evaluation is on grades
Sources:
• Nunan, D. (2015) Teaching English to speakers of other languages: An introduction, London:
Routledge, p. 6.
• https://arc.duke.edu/documents/The%20difference%20between%20assessment%20and%20e
valuation.pdf
• https://arc.duke.edu/documents/The%20difference%20between%20assessm
ent%20and%20evaluation.pdf
Philosophy of grading: what should grade reflect?
Different viewpoints on grading
Gronlund (1998) argued for using grades only for student
achievement. He stated:
Base grades on student achievement, and achievement only. Grades should
represent the extent to which the intended learning outcomes were achieved
by students. They should not be contaminated by student effort, tardiness,
misbehavior, and other extraneous factors…. If they are permitted to become
part of the grade, the meaning of the grade as an indicator of achievement is lost
(pp. 174-175)
Different viewpoints on grading
Other viewpoints:
• For Grove (1998), Pewer (1998), and Progosh (1998), other
factors should be considered in assessing and grading.
• Triangulation: all abilities of a student may not be apparent
on achievement tests and measured performances.
• Alternative assessment: Formal tests are insufficient to
capture the totality of students’ competence; other
observations are also significant indicators of ability.
➢ Improvement, behavior, effort, motivation, and attendance
might justifiably belong to a set of components that add up to
a final grade.
Guidelines for selecting grading criteria
1. It is essential for all components of grading to be consistent with
an institutional philosophy and/or regulations.
2. All of the components of a final grade need to be explicitly stated
in writing to students at the beginning of a term of study.
3. Challenge yourself to create checklists, charts, and note-taking
systems that allow you to convey to the student the basis for your
conclusions.
4. Consider allocating relatively small weights to items (c) through
(h) so that a grade primarily reflects achievement.
Calculating grades: absolute and relative grading
1. Absolute system of grading:
➢ Pre-specifying standards of performance on a numerical point
system.
➢ Clear on competencies and objectives, and on tests, tasks, and
other assessment techniques that will figure into the formula for
assigning a grade.
➢
2. Relative grading:
• is more commonly used than absolute grading.
• Advantage: allowing your own interpretation and
adjusting for unpredicted ease or difficulty of a test.
• is usually accomplished by ranking students in order of
performance (percentile ranks) and assign cut-off
points for grades.
• grading ‘on the curve’: the normal bell curve of
normative data plotted on a graph.
– Ex: A normal distribution to assign grades: A = the top 10
percent of students in a class, B = the next 20 percent, C = the
middle 40 percent, D = the next 20 percent, F = the lowest 0
percent.
• In reality, no one adhere to such an interpretation
because
– it is too restrictive
– it usually does not appropriately interpret
achievement test results in classrooms
An alternative to a normal curve is to pre-select percentiles
according to an institutional expectation
• Pre-selecting grade distributions is still arbitrary and
may not reflect what grades are supposed to ‘mean’ in
their appraisal of performance.
• A much more common method of calculate grades is a
posteriori relative grading, in which a teacher
excercises the latitude to determine grade distributions
after the performances have been observed.
(Please read more about this on page 288)
Teachers' perceptions of appropriate grade distributions
(drawn from a workshop with English teachers at the American
Language Institute at San Francisco State University):
• Most teachers bring to a test or a course evaluation an
interpretation of estimated appropriate distributions, follow that
interpretation, and make minor adjustments to compensate for
such matters as unexpected difficulty.
• However, teachers preconceived notions of their own standards
for grading often do not match their actual practice.
• Two conclusions were drawn:
1. Teachers may hypothetically subscribe to a pre-selected set of
expectations, but in practice may not conform to those expectations.
2. Teachers all agreed that they were guilty of grade inflation.
INSTITUTIONAL EXPECTATIONS AND
CONSTRAINTS
Philosophies of grading and procedures for
calculating grades:
- Letter grades (A, B, C, D, F)
- Point systems (100 point or percentages)
- Grade point averages (GPAs) A= 4, B= 3, C= 2,
D= 1
- Narrative evaluations of students instead of a
letter grade or a numerical system
Cross-Cultural Factors and the Question of Difficulty
• Every learner of English comes from a native culture that may
have implicit philosophies of grading at wide variance with those
of an English-speaking cultures.
• Grading variation: In many cultures,
– It is unheard of to ask a student to self-assess performance
– The teacher assigns a grade, and nobody questions the
teacher’s criteria
– The measure of a good teacher is one who can design a test
that is so difficult that no student could achieve a perfect
score. The fact that students fall short of such marks of
perfection is a demonstration of the teacher’s superior
knowledge.
– As a corollary, grades of A are reserved for a highly select
few, and students are delighted with Bs.
– One single final examination is the acceptable determinant
of a student’s entire course grade.
– The notion of a teacher’s preparing students to do their
best on a test is an educational contradiction.
• Philosophies of grading, along with attendant
cross-cultural variation, must address the
issue of gauging difficulty in tests and other
graded measures
e.g. in some cultures, a ‘hard’ test is a good test.
To gauge difficulty in tests and other graded
measures: combining some possible factors:
• Experience as a teacher (with appropriate intuition)
• Adeptness at designing feasible tasks
• Special care in framing items that are clear and relevant
• Mirroring in- class tasks that students have mastered
• Variation of tasks on the test itself
• Reference to prior tests in the same course
• A thorough review and preparation for the test
• Knowledge of your students’ collective abilities
• A little bit of luck
What do letter grades ‘mean’?
• Descriptors of letter grades:
❖ A = excellent
❖ B = good
❖ C = adequate
❖ D = inadequate/unsatisfactory
❖ F = failing/Unacceptable
• Every teacher who uses letter grades or a percentage score to provide
an evaluation should:
• Every teacher who uses letter grades or a
percentage score to provide an evaluation
should:
a. use a carefully constructed system of grading
b. assign grades on the basis of explicitly stated
criteria
c. base the criteria on objectives of a course or
assessment procedure(s)
Alternatives to letter grading
• Letter grades – and along with them numerical
scores – are only one form of student
evaluation.
• The principle of triangulation cautions us to
provide as many forms of evaluation as are
feasible.
Summative alternatives to grading: include self-
assessment, narrative evaluations, checklists, and
conferences.
1. Self-assessment:
➢Checklists
➢A guided journal entry that directs the student
to reflect on the content and linguistic objectives
➢An essay that self – assesses
➢A teacher – student conference.
2. Narrative evaluations
An example of narrative evaluation
• Advantages:
– Individualisation
– Evaluation of multiple objectives of a course
– Face validity
– Washback potential
• Disadvantages: (overriding these advantages)
– Narratives cannot be quantified easily by admissions and transcript
evaluation offices
– They take a great deal of time for teachers to complete
– Students have been found to pay a little attention to them
– Teachers have succumbed to formulaic narratives that simply follow a
template with interchangeable phrases and modifiers.
3. Checklists
3. Checklists
Advantages:
– Increased practicality and reliability while maintaining washback
– Minimizing teacher time
– Uniform measures are applied across all students
– Some open-ended comments from the teacher are available
– The student responds with his or her goals
– When the checklist formant is accompanied, virtually none of
the disadvantages of narrative evaluation remain, with only a
small chance that some individualization might be slightly
reduced.
– In the end-of-term chaos, students are also more likely to
process checked boxes than to labour through several
paragraphs of prose.
Conferences
• Advantages:
– washback benefits
– can accomplish much more than can a simple grade
• Disadvantages: Difficult to schedule the time for
the conference
PRINCIPLES AND GUIDELINES FOR
GRADING AND EVALUATION
Principles:
• Grading is not necessarily based on a universally
accepted scale
• Grading is sometimes subjective and context-
dependent
• Grading of tests is often done on the “curve”
• Grades reflect a teacher’s philosophy of grading
• Grades reflect an institutional philosophy of grading
• Cross – cultural variation in grading philosophies
needs to be understood
• Grades often conform, by design, to a teacher’s
expected distribution of students across a continuum
• Tests do not always yield an expected level of
difficulty
• Letter grades may not “mean” the same thing to all
people
• Alternatives to letter grades or numerical scores are
highly desirable as additional indicators of
achievement.
Guidelines
1. Develop an informed, comprehensive personal philosophy of grading
that is consistent with your philosophy of teaching and evaluation.
2. Ascertain an institution’s philosophy of grading and, unless otherwise
negotiated, conform to that philosophy (so that you are not out of
step with others).
3. Design tests that conform to appropriate institutional and cultural
expectations of the difficulty that students should experience.
4. Select appropriate criteria for grading and their relative weighting in
calculating grades.
5. Communicate criteria for grading to students at the beginning of the
course and at subsequent grading periods (mid-term, final).
6. Triangulate letter grade evaluations with alternatives that are more
formative and that give more washback.
• As you develop your own philosophy of grading, make some
attempt to conform that philosophy to your approach to
teaching.
• In a communicative language classroom, that approach usually
implies
– meaningful learning,
– authenticity,
– building of student autonomy,
– student-teacher collaboration
– A community of learners
– The perception that your role is of a facilitator or coach rather than a
director or dictator
• Let your grading philosophy be consonant with your teaching
philosophy
