differentiation & grading: can they co-exist?caroltomlinson.com/presentations/grading_di.pdf ·...
TRANSCRIPT
Copyright 2010 Carol Tomlinson 1
Differentiation & Grading:
Can They Co-exist?
Qikiqtani Teachers’ Conference
Iqaluit, Nunavut
February 24, 2010
Carol Tomlinson
William Clay Parrish, Jr. Professor
University of Virginia
What is a Grade????“…a grade (is)…an inadequate report
of an imprecise judgmentof a biased and variable judgeof the extent to whicha student has attainedan undefined level of masteryof an unknown proportionon an indefinite amountof materials.
Paul Dressell,Michigan State University
Copyright 2010 Carol Tomlinson 2
To Untangle the Grading Knot
We need to consider two elements:
1) Grading issues of particular concern in a differentiated
classroom
2) Best practices in assessment
and grading
3) Whether or not best practices in
assessment & grading would
adequately address the issues
related to grading &
differentiation.
Unless we understand both the issues related to academically diverse classrooms and best practices in assessment & grading (and their interrelationship) we’ll stay tied in a knot! And unless the former leads us to solve the latter, we have a problem as well.
Copyright 2010 Carol Tomlinson 3
All learners need
a balanced success
to effort ratio
Struggling
Learners:
Heavy Effort
Little Success
Copyright 2010 Carol Tomlinson 4
Advanced
Learners:
Great Success,
Little Effort
The Unspoken Effect of Grades
For some students, the certainty of praise
and success in school has become a drug;
they continually need more.
For many other students, year upon year of
“not good enough” has eroded their
intellectual self-confidence and resulted in
a kind of mind-numbing malaise.
Earl, L. (2003). Assessment as learning. Thousand Oaks, CA: Corwin, p. 15.
Copyright 2010 Carol Tomlinson 5
If the question is, “Do rewards
motivate students?”
the answer is,
“Absolutely! They motivate students
to get rewards.”
Zen And The Art of Public School Teaching by John Perricone • Publish America • p. 68
In grading, as in all other teaching decisions, we need to
claim a share of the Hippocratic Oath…
Grading practices should support student learning.
They certainly should not undermine it—nor harm student
development as learners.
Copyright 2010 Carol Tomlinson 6
Essential Question: To what degree would adhering to the key principles of
effective grading address the differentiation-related concerns about grading
while eliminating error and communicating clearly to students & parents?
We should seek to ensure that grades:
1)Keep the Success to Effort Ratio in Balance
That is, grades (in conjunction with school
tasks/experiences) contribute to the student’s
sense that when they work hard, something
good generally comes of it.
2) Ensure that students develop a growth mindset
That is, grades (in conjunction with school
tasks/experiences) contribute to student trust
that sustained effort and hard work make
most things possible.
Copyright 2010 Carol Tomlinson 7
We should seek to ensure that grades:
1)Communicate Clearly
That is, a parent can trust that the grade is a
very accurate representation of just what a
student knows, understands, and can do in a
given subject, at a given time, based on clearly
understood criteria.
2) Support the parent in supporting learning
That is, grades guide parents in what to do
next to encourage student growth.
We should seek to ensure that grades are:
1)RELIABLE—
That is, that if we were to use the same
measure a couple of days, weeks, or months
later, the results would be relatively the same
for a given student.
2) VALID—
That is, we actually measured what we meant
to measure.
3) TRUE SCORE—
That is, the grade really represents what a
student knows, understands, and can do.
Copyright 2010 Carol Tomlinson 8
X = T + E
The Observed Score = the True Score + Error
The Grade We Give Really Indicates What the Student
Knows, Understands, and Can Do + Extraneous Factors
that Get in the Way of Indicating Precisely What the
Student Knows, Understands and Can Do.
For Grades to Be Valid, We Have to Do All We Can Do
to Eliminate Error.
That’s the Game Plan for Grading: To Ensure that Our
Grades are as Close as Possible to the Student’s
“True Score”—to eliminate “error.”
Let’s examine six key principles
of best-practice grading
to see if they would
help us address concerns that
arise in regard to grading &
differentiation, balance the
success to effort ratio, help
develop a growth mindset, communicate
clearly, and eliminate error.
Copyright 2010 Carol Tomlinson 9
Principles of Effective Grading & Reporting
Principle #1
• It’s unwise to
over-grade
student work
Copyright 2010 Carol Tomlinson 10
Principles of Effective Grading & Reporting
Principle #2
Grades should be
based on clearly
specified learning
goals
Principles of Effective Grading & Reporting
Principle #3
Grades should be
criterion-based,
not norm-based
Copyright 2010 Carol Tomlinson 11
Principles of Effective Grading and
Reporting
Principle # 4
Data used for grading
must be valid (measure
what we intend to measure).
That is, the data must be free
of “Grade Fog.”
Common Sources of Bias and Distortion
Problems that can occur with the student
Lack of reading skill/language
Emotional upset
Poor health
Lack of testwiseness
Evaluation anxiety
Problems that can occur with the setting
Physical conditions – light, heat, noise, etc.
Problems that can occur with the assessment itself
Directions lacking or unclear
Poorly worded questions/prompts
Insufficient time
Misaligned with goals and/or instruction
Based on the ideas of Rick Stiggins
Copyright 2010 Carol Tomlinson 12
Grades are broken when zeros are used:
Zeros distort the actual achievement record
and can decrease student motivation to
learn.
There are, however, many fixes in the form of grading alternatives.
Schools/districts develop policies regarding these alternatives, then indicate to
their teachers which alternative(s) they can or should use in their classrooms.
A zero has an underserved and devastating influence so much so that
no matter what the student does, the grade distorts the final grade as a
true indicator of mastery. Mathematically and ethically this is
unacceptable. Wormeli, 2006, pp. 137-138
A Repair Kit for Grading by Ken O’Connor • Educational Testing Service, • p. 91-92
Zeros Cause Grade Fog
Whenever I hear statistics being quoted, I am
reminded of the statistician who drowned
while wading across a river with an average
depth of three feet.(McMann, 2003, np)
The mean can be very well named -- it is truly “mean” to students because it
overemphasizes outlier scores, which are most often low outliners. As we see
in the following case, the calculation of the mean can distort the final grade.
Ten assessments have been converted to percentage scores to calculate a final
grade:
91, 91, 91, 91. 91. 91. 91. 70. 91. 91
Total; = 889. Mean = 88/9. Final grade = B
This student performed at an A level, 9 times out of 10 and the 70 is clearly an
anomaly. But the grade as calculated in most schools would be a B.
A Repair Kit for Grading by Ken O’Connor • Educational Testing Service • p. 81-82
Assigning a Mean Can Cause Grade Fog
Copyright 2010 Carol Tomlinson 13
Principles of Effective Grading and
Reporting
Principle # 5
Grade later in the
learning cycle
rather than
earlier.
Some Possible Solutions
What About Report Cards?
Copyright 2010 Carol Tomlinson 14
Principles of Effective Grading and
Reporting
Principle # 6
When it’s time for
report cards,
practice 3-P
grading.
Achievement
on clearly
delineated
content goals
Habits of
mind and
work
Growth in
achievement
on clearly
delineated
content goals
Copyright 2010 Carol Tomlinson 15
1. Attach an explanatory checklist with the 3 Ps
2. Talk with parents during conferences about
the 3 Ps.
3. Send an e-mail or a note with a report on the 3 Ps.
4. Have students keep records of their 3 Ps and
write a summary to parents.
5. Work to change the report card.
In any case, talk with your students consistently about the 3 Ps, their importance
in student development, and their interrelationships. Give 3P feedback!
What Might My Grade Book
Look Like?
Copyright 2010 Carol Tomlinson 16
Remember:
the learning
goals
typically
don’t change
when a
task is
differentiated.
It’s the
support
system,
route of
access,
mode of
expression,
personal
connection,
etc. that
generally
constitutes
differentiation
Curriculum Assessment Instruction Reporting
What is the target? (KUDs)
Who’s with me as we make the journey toward the KUDs?
(Pre-and Formative Assessment)
What do I do along the way to ensure that each student masters
--&, if possible, moves beyond the KUDs?
(Differentiation—and best practice assessment)
How do I communicate the student’s status
relative to the KUDs?
(Grading/Reporting)
Copyright 2010 Carol Tomlinson 17
Grade Inflation: Sense and
NonsenseTeachers should set criteria for grades and determine
what should set criteria for grades and determine
what achievement level equals mastery.
The number of students who get A’s should equal the
number who master the criteria set
for an A. If our instruction (i.e. treatment)
is effective, that number should be high!!
Phi Delta Kappan – Thomas F. Kelly – May 2009 – p. 696
Too often, educational tests, grades, and report cards are treated by teachers as autopsies when they should be viewed as physicals.
(Reeves 2000, p. 10)