Transcript Document 7315535

ERLC Webinar Series
Fall 2009
A Repair Kit for Grading 15 Fixes for Broken Grades
Webinar Session 5
With Ken O’Connor
WELCOME
Thank you for participating
in the Webinar
“15 Fixes for Broken Grades”
Presented by
Ken O’Connor
A Repair Kit For Grading:
15 Fixes for Broken Grades
Session #5
Fixes 11 & 12
Presented by Ken O’Connor
Assess for Success Consulting
[email protected]
www.oconnorgrading.com
HANGOVERS
Fixes 10 & 13
2-4
5-6
The Essential Question:
How confident are you that the grades students
get in your school are:
• consistent
• accurate
• meaningful, and
• supportive of learning?
If grades do not meet these four conditions of quality
they are “broken” i.e. ineffective.
5-7
Grading Issues
•
•
•
•
Achievement (only)
Evidence (quality)
Calculation
Learning (support)
5-8
Grades are broken when they …
• include ingredients that distort achievement
• arise from low quality or poorly organized evidence
• are derived from inappropriate number
crunching, and when they
• do not support the learning process
5-9
Fix # 11
Don’t rely on the mean; consider other
measures of central tendency and use
professional judgment.
5-10
Fix #11
“Averaging falls far short of providing
an accurate description of what students
have learned. . . . If the purpose of grading
and reporting is to provide an accurate
description of what students have learned,
then averaging must be considered
inadequate and inappropriate”.
Guskey, Thomas R. (Editor), Communicating Student Learning:
The 1996 ASCD Yearbook, ASCD, Alexandria, VA, 1996, 21
5-11
Fix #11
“Grades based on averaging have meaning only
when averaging is done on repeated measures of
similar content. Teachers average (marks for)
tests on fractions, word problems, geometry and
addition with marks for attendance, homework
and notebooks - and call it Mathematics. (Similar
examples could be given for other subjects.)
In Mathematics we teach that you cannot average
apples, oranges and bananas but we do it in our
grade books!”
R. Canady, Workshop presentation, ASCD Annual Conference, Washington,
D.C., April 1993
5-12
Fix #11
“Grades based on averaging have meaning only
when averaging is done on repeated measures of
similar content. Teachers average (marks for)
tests on fractions, word problems, geometry and
addition with marks for attendance, homework
and notebooks - and call it Mathematics. (Similar
examples could be given for other subjects.)
In Mathematics we teach that you cannot average
apples, oranges and bananas but we do it in our
grade books!”
R. Canady, Workshop presentation, ASCD Annual Conference, Washington,
D.C., April 1993
5-13
Fix #11
“Educators must abandon the average,
or arithmetic mean, as the
predominant measurement of student
achievement.”
Reeves, D., “Standards are Not Enough: Essential Transformations
for School Success,” NASSP Bulletin, Dec. 2000, 10
5-14
Fix #11
“Most fifth-grade students learn the difference
between mean, median, and mode, and thus gain
the insight that the arithmetic mean or average,
may not be the best representation of a set of data.
Yet the teachers of those students remain
stubbornly allegiant to the average.”
Reeves, D., Ahead of the Curve, Solution Tree, 2007, 230
5-15
Fix #11
Letter to the Editor
- Toronto Globe and Mail
October 15, 2003
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.
GORDON McMANN
Campbell River, B.C.
5-16
Fix #11
Total
89
89
89
20
89
89
89
20
89
89
752
Mean or Average = 75.2
Median =
89
5-17
Fix #11
"Grading by the median provides
more opportunities for success by
diminishing the impact of a few
stumbles and by rewarding hard
work."
Wright, Russell. G., "Success for All: The Median is the Key",
Kappan, May 1994, 723-725
5-18
Fix #11
First attempt
Second attempt
Third attempt
Fourth attempt
Fifth attempt
Sixth attempt
Source: an Alberta High School teacher
5-19
Fix #11
First Test
30/100
Second Test 30/100 + 50/100 = 80/200 = 40%
Third Test
30/100 + 50/100 + 60/100 = 140/300 = 46.66%
Fourth Test 30/100 + 50/100 + 60/100 + 70/100 = 210/400 = 52.5%*
Fifth Test
30/100 + 50/100 + 60/100 + 70/100 + 80/100 = 290/500 = 58%
Sixth Test
30/100 + 50/100 + 60/100 + 70/100 + 80/100 + 90/100 = 380/600 = 63.66%**
Seventh Test 30/100 + 50/100 + 60/100 + 70/100 + 80/100 + 90/100 + 100/100 = 480/700
= 68.66%
Eighth Test 30/100 + 50/100 + 60/100 + 70/100 + 80/100 + 90/100 + 100/100 + 100/100
= 580/800 = 72.5%***
* Pass if pass/fail cut is 50% - 3 attempts at passing level
** Pass if pass/fail cut is 60% - 5 attempts at passing level
*** Pass if pass/fail cut is 70% - 7 attempts at passing level including 2 with perfect scores
5-20
142
Fix #11
155
O’Connor, K., How to Grade for Learning, Second Edition, Skylight, 2002, 142
5-21
Fix #11
Think about this
• What grade should each student receive? Why?
• Is it accurate to use the same approach for
each student?
• What additional information would help you
decide the grade for each student?
• Change the 90s and 100s to 4s and the 62s and 63s
to 1s.
How does this change your view of the grade each
student should receive and the role of number
crunching in the determination of grades?
5-22
Fix #11
Other Calculation Issues
Weighting
- avoid weighting unless it is obviously needed
Variability of Scores
-recognize that this is an issue when combining
grades for awards, scholarships, etc
5-24
Fix #11
“Data should be used to INFORM
not determine decisions”
Management Consultant, The Hay Group, personal conversation,
January 2002
5-25
For Fix #11
What do you think?
check mark
X
5-26a
For Fix #11
Where are you/school/district now?
Implementation
A complete
B almost complete
C partial
D limited
E none
5-26b
For Fix #11
Where do you want to go you/school/district now?
Implementation
A complete
B almost complete
C partial
D limited
E none
5-26c
Fix #12
Don’t include zeros in grade
determination when evidence is
missing or as punishment; use
alternatives, such as reassessing to
determine real level of achievement
or use “I” for Incomplete or
Insufficient evidence.
5-27
Fix #12
Problems with zeros
• “Zeros give a numerical value to something that
has never been assessed and that therefore has
no basis in reality.
• They can have a counterproductive effects on
student motivation.
• They involve inappropriate mathematics.
. . . zeros in the record (thus) render grades
ineffective as communication.”
O’Connor, K., A Repair Kit for Grading, ETS/ATI, Portland, 2007, 86
5-28
Fix #12
“Assigning a score of zero to work that is
late, missed, or neglected does not
accurately depict students’ learning . Is
the teacher certain the student has learned
absolutely nothing, or is the zero assigned
to punish students for not displaying
appropriate responsibility?”
Guskey, Thomas R. (Editor), Communicating Student Learning:
The 1996 ASCD Yearbook, ASCD, Alexandria, VA, 1996, 21
5-29
Fix #12
“Most state standards in mathematics require that
fifth-grade students understand the principles of
ratios - for example, A is to B as 4 is to 3; D is to F
as 1 is to zero. Yet the persistence of the zero on
the 100-point scale indicates that many people with
advanced degrees, . . . have not applied the ratio
standard to their own professional practices.
Reeves, D.B., “The Case Against the Zero,” Kappan, December 2004, 324-325
5-30
Fix #12
A 90-100; B 80-89; C 70-79; D 60-69; F <60
‘the interval between grades through A-D is 10
points, yet the potential interval from D to F is 60
points. The result is . . . the 0 grade has a
disproportionate impact on the average grade. If
educators must use a numerical scale then the
lowest possible number on the scale should be one
grade value lower than a D.’
Adapted from Reeves, D., “Standards are Not Enough: Essential
Transformations for School Success,” NASSP Bulletin, Dec. 2000, 11-12
5-31
Fix #12
The Effect of Zeros
4
3
2
1
0
2
5 pt scale
(A)
(B)
(C)
(D)
(F)
(C)
101 point scale
90-100
11
95
80-89
10
85
70-79
10 75
60-69
10
65
<60
60
0
64 (D)
95
85
75
65
50
74 (C)
5-32
Fix #12
101 point scale
95
0
0
0
85
0
0
80
0
0
260
Mean
26
Letter Grade
F
What grade should this student get?
5 point scale
4
0
0
0
3
0
0
3
0
0
10
1.0
D
5-33
Fix #12
“The use of an I or “Incomplete” grade is
an alternative to assigning zeros that is both
educationally sound and potentially quite
effective.”
Guskey and Bailey, Developing Grading and Reporting Systems for Student
Learning, Corwin Press, 2001, 144
5-34
Fix #12
No Zero Policy
At George P. Vanier Junior High, students receive
percentage grades in their core subjects and letter
grades in their Option Classes. We have a “NO
ZERO” policy in term evaluation. This means
teachers do not calculate a “zero” into the overall
average for an assignment that was not completed
by a student. The mark that the student receives is
a reflection of the curriculum outcomes that have
been achieved that term, and the student’s
understanding of the curriculum concepts.
Source: George P. Vanier School, School Handbook, 2008-09, 7
5-35
Fix #12
No Zero Policy 2
“Because we have the “NO ZERO” policy, students
are required to complete their assignments. We
keep the library open after school until 4:00
Monday to Thursday and there is a teacher
available to help students with their assignments.
When students do not complete their work they are
we require that they stay after school till the work is
submitted. If they are unable to stay after school, or
if the work is still not completed, they may be
removed from option classes till the work is done.”
Source: George P. Vanier School, School Handbook, 2008-09, 7
5-36
152
Fix #12
Completion Contract
165
Student Name:
Course:
Missed Work - The following work has not been handed in:
Original Due Date:
Reason – Please indicate why the work is late.
Next Steps – What will you now do to get this work completed?
New Date for Submission:
Once this new date is negotiated, the student agrees to submit this work on that date or
receive a mark of I for Incomplete. The student and parent acknowledge that I’s may lead
to the teacher determining that there is insufficient evidence for a grade and that this is the
equivalent of a failing grade.
Student Signature:
Parent Signature:
Teacher Signature:
Figure 6.7
Adopted by Ken O’Connor from original work by Jennifer Perkin, Catholic School Board of Eastern Ontario
5-37
Fix #12
The Last Words on Zeros
“A zero has an undeserved 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.”
Rick Wormeli quoted in
O’Connor, K., A Repair Kit for Grading, ETS/ATI, Portland, 2007, 92
5-39
For Fix #12
What do you think?
check mark
X
5-40a
For Fix #12
Where are you/school/district now?
Implementation
A complete
B almost complete
C partial
D limited
E none
5-40b
For Fix #12
Where do you want to go you/school/district now?
Implementation
A complete
B almost complete
C partial
D limited
E none
5-40c
An ASSESSMENT PLAN should start with the
• desired results (learning goals, standards, etc), then the
• summative assessments that are going to be used to
determine whether the student ‘knows and can do,’ next should be
the
• diagnostic assessment(s) that are going to help to
determine the what and how for teaching and learning,
then should come the
• formative assessments that are going to help students
achieve the learning goals and that are going to cause
the teacher to adjust teaching and learning activities.
- homework, quizzes
- practices
- first draft, second draft
tests
performances
product(s)
5-41
A vital part of the ASSESSMENT PLAN is
how much evidence and
which assessments
are critical to being able to determine student
achievement/grades, e.g., there will be 9 summative
assessment opportunities, of which at least six,
(including the third, fifth and ninth) must be done.
5-42
5-43
Fixes #11 & 12
Think about this
“we have to see grading not as simply a numerical,
mechanical exercise, but as primarily an exercise in
professional judgment. It calls for teachers to
demonstrate two key aspects of professional
behaviour - the application of craft knowledge of
sound assessment practice, and the willingness and
ability to make and defend one’s professional
judgement.”
O’Connor, K., A Repair Kit for Grading, ETS/ATI, Portland, 2007, 83
5-44