Assessing from a CGI Perspective.pdf

Assessing from a CGI Perspective
Cheryl Ann Lubinski∗
, JoAnn Cady, Heather Bailey
Abstract
This is the story of Heather, a K–1 teacher who wanted to develop an assessment instrument to
determine the strategies her students used to solve problems involving all four basic operations–
addition, subtraction, multiplication, and division. Heather asked her mathematics education mentor
Cheryl to help her.
Keywords: Cognitively Guided Instruction, early childhood, conceptual understanding, mathematics
education
Published online by Illinois Mathematics Teacher on July 26, 2017.
1. Introduction
This article emerged from our ongoing collab-
oration as teachers and mathematics educators
who use Cognitively Guided Instruction (CGI) in
our classrooms and in staff development sessions
with preservice and in-service elementary teach-
ers. CGI is an established research-based program
that originated at the University of Wisconsin–
Madison designed, in part, to help teachers ad-
dress diverse learners. The premise underlying
CGI is that if teachers learn how children come to
understand mathematical ideas, they can provide
better instruction for the children in their classes.
According to the National Research Council:
CGI is a professional development pro-
gram for teachers that focuses on help-
ing them construct explicit models of
the development of children’s mathe-
matical thinking in well-defined con-
tent domains. No instructional mate-
rials or specifications for practice are
provided in CGI; teachers develop their
own instructional materials and prac-
tices from watching and listening to
their students solve problems. (Na-
tional Research Council, 2001, 400)
Teachers in CGI classrooms carefully create
and assign relevant problems in order to assess
∗
Corresponding author
their students’ learning strategies, and then they
plan instruction to develop those strategies along
with their students’ mathematical understanding.
The primary goal of this article is to provide teach-
ers with an instrument to assess their primary-age
students’ learning strategies.
In addition, we connect CGI to the Common
Core State Standards of Mathematical Practice
(CCSS-MP), as do many articles about CGI. For
example, the existing literature on CGI addresses
the need for teachers to:
• focus on problem solving by making sense
of problems and persevere in solving them
(Hoosain & Chance, 2004);
• prepare students for problem solving by hav-
ing them use appropriate tools strategically
(Holden, 2007);
• use problems and numbers that are appro-
priate to students’ developmental levels to
reason abstractly and quantitatively (Jacobs
& Ambrose, 2008);
• have students model with mathematics and
attend to precision by using mathematical
terms appropriately (Carpenter et al.,
2003);
• listen to and develop students’ strategies
by having them construct viable arguments
(Hiebert et al., 1997);
Illinois Mathematics Teacher 1

Cheryl Ann Lubinski, JoAnn Cady, Heather Bailey
• record symbolic representations that con-
nect to students’ thinking as they look
for and make use of structure (Hoosain &
Chance, 2004); and
• pose questions that assist students to
verbalize their reasoning process as they
look for and express regularity in repeated
reasoning (Jacobs & Ambrose, 2008).
CGI helps us to measure what each student
knows to inform how instruction should be struc-
tured to enable a particular student to learn (Cham-
bers & Lacampagne, 1994). The authors of the
articles cited above recognize that CGI teachers
encourage their students to develop the mathe-
matical practices set forth in the Common Core
State Standards. However, no article specifically
outlines an individual assessment approach to de-
termine the developmental level reflected by a stu-
dent’s strategies when teaching from a CGI per-
spective. In this article, we aim to fill this void by
providing an assessment instrument that helps to
determine a student’s developmental level as well
as to address a student’s level of achievement in
regard to the Common Core.
2. Background of the Authors
Cheryl and JoAnn are university faculty mem-
bers with prior elementary teaching experience.
They are currently working with elementary pre-
service and inservice teachers on implementing
CGI and Common Core standards and practices
in classrooms. Heather was one of the elementary
teachers participating in a series of CGI profes-
sional development workshops led by Cheryl. She
is a K–1 teacher who loops (teaches the same stu-
dents for two years—in kindergarten and in first
grade) and is new to the ideas of CGI. Fifty-two
percent of students in her school are on free or re-
duced lunch. After these workshops, Cheryl and
Heather established a two-year “co-mentoring” re-
lationship in which Cheryl routinely co-taught in
Heather’s classroom in order to better develop our
own learning about students’ thinking. As is typ-
ical in a CGI classroom, we gave students a va-
riety of problem types to solve, discussed their
strategies between us, and talked about implica-
tions for lesson planning that incorporated dif-
ferentiated instruction. After one of our collabo-
ration sessions, Heather suggested we develop an
assessment tool in order to obtain a more accurate
description of her individual student’s strategies
as well as their developmental level when solving
problems involving all four operations with whole
numbers. The remainder of this article describes
the assessment tool that we developed.
3. Background Knowledge for Understand-
ing Our Assessment Tool
In order to understand our assessment tool, it
is important for teachers to have knowledge about
Problem Types and Solution Strategies, as these
terms are used in the CGI literature. We will pro-
vide a brief explanation here. For a more detailed
description see Carpenter et al. (2015).
3.1. Problem Types
Word problems frequently found in textbooks
for the K–3 teacher are typically limited in their
structure, as indicated by the location of the un-
known, as in the following word problem:
Laura has 5 dolls and she gets 4 more.
How many dolls does she have now?
The structure of this word problem is
symbolically represented as:
5 + 4 = n,
where n represents the unknown.
Compare this with the alternatives below:
Laura has 5 dolls. How many more
does she need to have 9?
symbolically represented as
5 + n = 9.
Laura has some dolls. She gets 4 more.
Now she has 9 dolls. How many dolls
did Laura have to start?
2 Illinois Mathematics Teacher

symbolically represented as:
n + 4 = 9.
In a CGI classroom, word problems vary in
structure, and teachers initially begin to use those
from the problem types chart found in Appendix
A. Problems in this chart are categorized as Join,
Separate, Part-Part-Whole, Compare, Grouping,
and Partitioning. Each of these six categories has
variations based on the location of the unknown.
For example, in the Join row we have Result Un-
known, Change Unknown, and Start Unknown,
abbreviated as JRU, JCU, and JSU. We see that
the unknowns are located at the end, in the mid-
dle, and at the beginning, respectively. The Sep-
arate problems follow a similar structure. The
Part-Part-Whole problems have either the whole
or the part unknown. The Compare problems
have a difference, a compare quantity, or refer-
ent unknown. Grouping problems have the to-
tal number unknown. Partitioning problems are
of two types, either the number of groups is un-
known (Measurement Division) or the number in
a group is unknown (Partitive Division).
3.2. Strategies
CGI research provides compelling evidence that
even young students have strategies to solve word
problems involving whole numbers with all four
operations. Joining, Separating, Grouping, and
Partitioning problems involve actions; thus the
Direct Modeler (defined in Table 1) can solve these
more easily than problems that don’t have ac-
tions, such as the Compare and Part-Part-Whole
problems. The Start Unknown problems are the
exception because Direct Modelers have no num-
ber with which to start and can only solve these
problem types with a Trial and Error strategy.
It is important for teachers to not connect the
problem type category to an operation. How stu-
dents solve a problem indicates how they think
about the situation and provides developmental
insight into their strategy level. For example, con-
sider the JCU problem below:
Direct
Modeling
Student represents all of the numbers in a word
problem with counters, fingers, or pictures, in
the order that the numbers are given.
Counting Student relies on a counting strategy rather
than modeling each number.
Derived
Facts
Student uses an already known fact to deter-
mine an unknown fact.
Number
Facts
Student uses any fact about one of the four
operations with the numbers 0–12.
Table 1: Problem-solving strategy categories
Anyae has 5 bows, and Gloria gives
her some more. Now Anyae has 11
bows. How many bows did Gloria give
Anyae?
A student responding to this problem could:
• count out five cubes and join cubes to the
five cubes until the number eleven is reached.
Then the student would count the cubes
joined and say, “six.” (Student Directly Mod-
els.)
• say, “five,” count on her fingers, “six, seven,
eight, nine, ten, eleven,” count the fingers
she used, and say, “six.” (Student uses a
Counting strategy.)
• say, “eleven, because five plus five is ten and
one more is eleven, and five plus one is six.”
(Student uses a Derived Fact.)
• start by saying, “eleven,” and count back
until the number five is reached, using fin-
gers or counters to count back, then count
how many were taken away and say, “six.”
(Student uses a Counting strategy.)
• say, “six, because I just know five plus six
is eleven.” (Student uses recall of Number
Facts.)
Thus either a joining or separating action can be
connected to a solution strategy. All of the above
strategies provide a correct answer, but how the
student solves the problem provides information
about the student’s strategy level. In Appendix
B, we provide Individual Assessment Sheets for
documenting both the strategies a student uses

and the mathematical practices a student demon-
strates.
3.3. The Strategies a Student Uses
On the first Individual Assessment Sheet in
Appendix B, there is a list of strategies a stu-
dent could use. Each of the strategies listed in
Appendix B are explained in more detail in Ap-
pendix C. The strategies students use provide in-
sight to their developmental level, as described in
the CGI literature. Initially, a student Directly
Models solution strategies, using one of the four
Direct Modeling strategies listed. Then a student
moves to using one of the eight Counting strate-
gies. Finally, students will routinely use Derived
Facts or known Number Facts. It is important
to note that even Direct Modelers and Counters
use some Derived Facts and Number Facts as they
solve problems.
In this article, we refer to Beginning Problem
Solvers, Developing Problem Solvers, and Mature
Problem Solvers. We provide a brief description
of each in the next paragraphs.
Beginning Problem Solvers are students who
can successfully find a solution to a problem using
a Direct Modeling solution strategy. Some need to
hear the problem one step at a time. Some need
to work on counting with one-to-one correspon-
dence. As they practice “telling the story” (say-
ing the word problem in their own words), they
can eventually recount the complete problem and
attempt to solve it. They have the most success
with JRU, JCU, PPWwu, SCU, and SRU prob-
lems for the Addition and Subtraction problem
types. They can often solve Grouping and Par-
titioning problems because these problem types
have actions students can easily model.
Developing Problem Solvers use more abstract
strategies than Beginning Problem Solvers. They
do not need to represent all numbers in the prob-
lem with counters and often use Counting strate-
gies as well as Derived Facts and Number Facts.
They plan ahead when solving problems, i.e., they
think before doing.
Mature Problem Solvers use more abstract strate-
gies routinely, often in their heads using Derived
Facts and Number Facts. Additionally they can
solve all of the problem types using strategies that
sometimes rely on inverse number relationships
and the Commutative Property. Their flexible use
of strategies helps them to be successful with op-
erations involving multi-digit numbers as well as
with JSU and SSU problems—problems that Be-
ginning and Developing Problem Solvers can only
solve with a Trial and Error strategy.
3.4. The Mathematical Practices a Student Demon-
strates
The Common Core State Standards recom-
mends that a teacher help students develop math-
ematical practices (Council of Chief State School
Officers, 2011). The resources cited in this report
go into detail about what can be developed for
each practice. We have included an assessment
sheet for the Mathematical Practices in Appendix
B so that teachers have a form on which to take
notes about an individual student’s progress with
these practices.
4. One Option for Assessing
Using the flow chart in Appendix D, we can as-
sess an individual student’s strategy level, which
is developmental. The flow chart provides one ex-
ample of how to assess. A teacher may choose
a different order in which to present the problem
types depending on the teacher’s objective for as-
sessment. During assessment, the teacher chooses
number sets that will provide information on the
student’s most typical strategies. The teacher
observes and takes notes on how the student is
solving the problem, using the Individual Assess-
ment Sheets from Appendix B and referring to
Appendix C to help with strategy identification.
It is recommended that teachers give individual
students between five and seven different problem
types in order to fully describe how that student
characteristically solves problems (Fennema et al.,
2000).
To begin an assessment, we recommend us-
ing a JCU problem because this problem type
can be solved at all developmental levels with Di-
rect Modeling, Counting, Derived Facts, or known
Number Facts. Additionally, if a student solves a

Figure 1: A Developing Problem Solver uses a Counting
strategy to solve his problem: If a student has 51 trucks,
how many more does he need to have 54 trucks? The
teacher decided to list the numerals so the student can
focus on the strategy.
JCU problem the student usually can solve a JRU
or SRU problem.
Number suggestions are provided for each of
the problem types in Appendix A so that teachers
can make decisions based on their objectives for
assessment. For example, for JCU, the numbers
in the problem could be 5 and 8 (the first numer-
als in each of the two number sets) to assess how
a student determines more; 37 and 47 to assess
how a student determines a group of 10 more; or
89 and 109 to assess how a student determines
20 more and bridges the number 100, i.e., “does
the student Count On by ones or by groups of
tens?” Using groups of tens indicates the student
uses knowledge of place value to solve problems
involving multi-digit numbers.
5. Implementing a Lesson Based on the As-
sessment Results
Using the assessment instrument we developed,
Heather assesses her students’ strategies, both in-
dividually and during group instruction; then she
plans for instruction using her findings from the
assessments. During instruction she notes which
students need probes to encourage more abstract
strategies, such as going from Direct Modeling
strategies to Counting strategies, or from Count-
ing strategies to Derived Fact strategies, and which
students need practice at their current develop-
mental level. Every lesson provides Heather with
opportunities to refine her assessments of individ-
uals.
Heather’s purpose for instruction is to develop
meaningful strategies that the students can use
with understanding. She carefully plans probes,
both prior to and during instruction, that encour-
age her students to use their most sophisticated
strategies based on the results of her assessment.
For the students who need to Directly Model, she
encourages pictures as a representation or has the
students tell her about their thinking using ma-
nipulatives they select to help them explain. For
students who use Counting strategies, she expects
them to use those strategies rather than draw
pictures, thus encouraging them to reason more
maturely or abstractly about quantities. To en-
courage the use of Derived Facts, she might ask
students what they know about “five and five”
to help them to think about “five and six,” thus
encouraging them to reason quantitatively about
number relationships.
An example of how this might look in a
classroom is given using the Dalmatian Puppy
Problem presented to Heather’s entire class:
There are four Dalmatian puppies.
Each puppy has five large spots. How
many large spots are there?
This problem was chosen because it is
appropriate for a diverse group of problem solvers.
It can be Directly Modeled or solved using
more mature strategies such as Skip Counting or
Number Facts. After the problem is presented,
many students use their fingers to solve it.
Anyae, however, appears to be stuck. Anyae
is a student who needs to Directly Model the
structure of a mathematics problem and she
struggles to explain her reasoning process. To
encourage Anyae to persist to solve the problem,
the teacher uses helpful probes so Anyae can,
“make sense of the problem and persevere”
(Common Core State Standards Initiative, 2010,
CCSS.MP.1) as she solves it at her developmental
level of Direct Modeling. Following is an

episode that is representative of how CGI’s
philosophy encourages mathematical thinking
involving Anyae’s work with that of the rest of
the class:
Teacher: There are four Dalmatian
puppies. Each puppy has 5 large
spots. How many large spots are
there? [Pause]
Teacher to Anyae: Tell me the story.
Anyae: There are four puppies. . . .
Teacher: What do they have?
Anyae: I don’t remember.
Teacher: They each have five large
spots. I want you to figure out how
many large spots there are. Tell me
the whole story please.
Anyae: There are four puppies. They
have five large spots. How many large
spots are there?
Teacher: What could you do to figure
this out?
Anyae shrugs.
Teacher: Draw a picture of the
puppies and their spots on the white
board.
(Note: It is appropriate for a teacher
to suggest what a student could do
or think about if the teacher has
assessed the student’s developmental
level and knows what the student
needs to think about or do to solve
a problem. The teacher has assessed
that Anyae needs to Directly Model
the problem situation.)
Teacher (while Anyae is drawing):
Nora, what did you do?
Nora: I got 19. I counted, “1, 2, 3, 4,
5. . . ” [Nora extends one finger to keep
track of the number of groups of five
she has counted] “6, 7, 8, 9, 10. . . ”
[two fingers] “11, 12, 13, 14, 15. . . ”
[three fingers] “16, 17, 18, 19” [four
fingers].
Teacher: Show me the last part again.
[As Nora repeats the last part, she self
corrects, “16, 17, 18, 19, 20. It’s 20.”
The teacher then asks Luke what he
did.]
Luke: I counted by fives, “5, 10, 15,
20.”
The teacher goes back to Anyae’s picture and
asks her:
• How many puppies do you have?
• How many large spots are on each?
• Can you count them for the class?
Anyae answers each question and then counts
the spots by ones and gets twenty. During this
question-and-answer time, Heather is assessing
how well Anyae makes sense of the problem.
Heather then decides to see if Nora can
connect her strategy to Luke’s; thus encouraging
Nora to reason more abstractly about quantities.
Heather: Nora, why do you go “1, 2,
3, 4, 5,” and then pause?
Nora: Because there are five large
spots on the first puppy.
Heather: Do you know why Luke
went, “5, 10, 15, 20?”
Through discussion, Nora realizes that she
doesn’t have to count by ones and pause after
every fifth number, instead, she can just say every
fifth number, i.e., “5, 10, 15, 20.” (Nora still
uses fingers to represent each group of five.) It
is important to note here that Heather needs to
go back to what Nora says to make connections
to what Nora could learn to do. Heather makes a
note to check to see if Nora uses this Skip Count
strategy for a similar problem in the future.
Next, Heather asks Luke about his counting
by fives. Heather asks Luke how many fives
he counted and why. Through discussion,
using his strategy of counting by fives, Luke

realizes that four fives are twenty, and he can
represent his thinking by the expression 4 ×
5 = 20. Making connections from Luke’s
thinking to symbols encourages him to model his
mathematical thinking with symbols. Through
discussion with the teacher, he described his
understanding of the problem, and she helped
him to connect this understanding to the
structure of the problem as illustrated with having
a multiplication representation. Additionally,
Heather could have asked the class:
• How do all three solution strategies
compare?
• How do Anyae’s groups of five spots connect
to Luke’s skip counting strategy?
• How does Anyae’s groups of five spots
connect to Nora’s counting the individual
spots using her fingers to keep track of the
number of dogs?
These teacher moves allow students to develop a
deeper understanding of the mathematics.
Both Nora and Luke have been assessed as
students who use Counting strategies routinely.
Additionally, they are part of a group of students
who are the most mature problem solvers in the
class. Heather has encouraged them to use what
they know about addition and subtraction facts
to solve problems involving multi-digit numbers.
This reflects CCSS Mathematical Practice 8,
“Look for and express regularity in repeated
reasoning,” or, using repeated reasoning that can
encourage students to reason more abstractly and
quantitatively (Common Core State Standards
Initiative, 2010). With this one problem, Heather
has individualized instruction in a whole class
setting as she encourages CCSS Mathematical
Practices.
6. Heather’s Most Mature Problem Solvers
An example of Heather’s instruction with
her most mature problem solvers is illustrated
when she presented the following problem (which
initially appeared in CGI materials for teacher
leaders):
The baby elephant at the zoo weighed
488 pounds. During the summer she
gained 145 pounds. How much does
she weigh now?
Several of the first graders thought about it and
then shared their thinking. Heather has to listen
carefully and ask questions in order to understand
the strategies her students use.
• Jude said, “400 plus 100 is 500 plus 80 is 580
plus 20 from the 40 [thinking 45 = 20 + 25].
You have 600 plus 25 plus 5 plus 3 [8 =
5 + 3], which is 633.”
• Luke shared a different strategy: “488 plus
100 is 588 plus 12 is 600 plus 33 is 633.”
Heather asked Luke, “Where did the 12
come from?” Luke replied, “It was inside
the 45 [45 = 12 + 33].”
• Whitney provides yet another strategy:
“400 plus 100 is 500 and 80 plus 20 is 100
and 500 plus 100 is 600. 600 plus 20 plus 8
is 628 plus 5 [she counted up by ones using
her fingers] is 633.”
Heather’s most advanced Mature Problem Solvers
were able to achieve success because Heather
carefully assesses what each student in her class
can do and plans individual questions for them as
she teaches in a whole class or small group setting.
In this way she truly attends to the diversity
of her students’ needs. As a consequence, each
student in her class is challenged to reason and
problem-solve at a level that is in alignment with
their developmental levels, as indicated by the
strategies they use most often.
7. Discussion of the Assessment Process
Teachers who have knowledge of problem
types and students’ strategies can choose
problems and numbers for their students that
encourage more sophisticated strategies and
increase mathematical understandings.
Additionally, using the lists in Appendix B,
teachers can keep track of students’ progress
throughout the year. Progress is determined by

how students are developing their strategies to
more abstract levels. Assessment sheets could
be kept in a folder for individual students along
with student work that supports the teacher’s
comments. These assessment sheets can be
used for both planning throughout the year
and reporting student progress during conference
time. These sheets are a record of what the
students need in order to be successful as well as
of their individual goals for the year.
When teachers assess students’ strategies
while solving word problems, they can plan
questions to ask during instruction, as Heather
does, that provide opportunities for diverse
problem solvers to progress from Directly
Modeling, to Counting, to Derived Facts, and
finally to Number Facts. Most students know
some Number Facts, and encouraging students
who are ready to derive new Number Facts from
those that they already know can help them
attain fact mastery. Therefore, recording known
facts involving all four operations on a student’s
individual assessment sheet as well as realizing
what facts students can derive could aid teachers
when planning individual and group instruction.
Teachers can help students move through
the developmental levels by selecting appropriate
problem types and numbers that help students
to develop more abstract strategies over time.
Knowing how a student solves problems is
important for planning whole-group lessons. For
example, during instruction a teacher who has
knowledge of the students’ strategies can plan to
ask individual students questions appropriate to
each of their developmental levels.
We have used all of these problem types in our
work in pre-K–3 classrooms. The introduction
of multiplication and division problem types is
beneficial to understanding place value concepts.
Thus work with these problem types should begin
early, ideally in kindergarten. Beginning problem
solvers need to practice their strategies until they
can plan ahead and solve problems involving all
four operations with whole numbers. Mature
problem solvers need challenging problems
to further develop place-value concepts and
relational thinking. All problem solvers need
practice with strategies that encourage more
abstract reasoning.
8. Reflecting on Our Collaborations
As we reflect on our work, we realize that
using CGI and continually focusing on assessing
students’ mathematical thinking not only informs
our instructional decisions, but also helps us to
develop the Mathematical Practices described in
the CCSS-M documents. Additionally, the co-
mentoring relationship I had with Heather helped
Heather attend to precision with her students
(Common Core State Standards Initiative, 2010,
CCSS.MP.6), when she asks probing questions
and uses appropriate mathematical terminology
and mathematically correct representations. Our
collaborative work was successful because we had
a desire to learn and to develop within each
of our areas of expertise—be it conducting staff
development sessions, working with preservice
teachers, or learning about K–3 students’ strategy
development. Our assessment instrument contrib-
uted to our success.
References
Carpenter, T. P., Fennema, E., Franke, M. L., Levi, L., &
Empson, S. B. (2015). Children’s mathematics: Cogni-
tively guided instruction. (2nd ed.). Portsmouth, NH:
Heinemann.
Carpenter, T. P., Franke, M. L., & Levi, L. (2003). Think-
ing mathematically: Integrating arithmetic and algebra
in elementary school. Portsmouth, NH: Heinemann.
Chambers, D. L., & Lacampagne, C. B. (1994). Cogni-
tively guided instruction. Teaching Children Mathemat-
ics, 1(2), 116–117.
Common Core State Standards Initiative (2010). Common
core state standards for mathematics. Washington, DC:
National Governors Association Center for Best Prac-
tices and the Council of Chief State School Officers.
Council of Chief State School Officers (2011). Com-
mon core state standards mathematics curriculum ma-
terials analysis project. URL: http://tinyurl.com/
y8oceyaq. Accessed June 6, 2017.
Fennema, E., Carpenter, T., Levi, L., Franke, M., &
Empson, S. (2000). Children’s mathematics: Cogni-
tively guided instruction: A guide for workshop lead-
ers. Washington, DC: National Council of Teachers of
Mathematics.

Hiebert, J., Carpenter, T., Fennema, E., Fuson, K.,
Wearne, D., Murray, H., Olivier, A., & Human, P.
(1997). Making sense: Teaching and learning mathe-
matics with understanding.. Portsmouth, NH: Heine-
mann.
Holden, B. (2007). Preparing for problem solving. Teach-
ing Children Mathematics, 14(5), 290.
Hoosain, E., & Chance, R. H. (2004). Problem-solving
strategies of first graders. Teaching Children Mathe-
matics, 10(9), 474–480.
Jacobs, V. R., & Ambrose, R. C. (2008). Making the
most of story problems. Teaching Children Mathemat-
ics, 15(5), 260–266.
National Research Council (2001). J. Kilpatrick, J. Swaf-
ford, and B. Findell (Eds.). Mathematics Learning
Study Committee, Center for Education, Division of
Behavioral and Social Sciences and Education. Wash-
ington, DC: National Academy Press.
Cheryl Ann Lubinski
ILLINOIS STATE UNIVERSITY
NORMAL, IL
E-mail: calubinski@ilstu.edu
JoAnn Cady
UNIVERSITY OF TENNESSEE
KNOXVILLE, TN
E-mail: jcady@utk.edu
Heather Bailey
MRH EARLY CHILDHOOD CENTER
MAPLEWOOD, MO
E-mail: heather.bailey@mrhschools.net

Cheryl
Ann
Lubinski,
JoAnn
Cady,
Heather
Bailey
Appendix
A
Problem
Types
Chart
∗
Join (Result Unknown, JRU)
(Student’s name) has (2, 18, or 98) jelly
beans and I give you (5, 10, or 10) more.
Now how many do you have?
(Change Unknown, JCU)
(Student’s name) has (5, 37, or 89)
M&Ms. I give her some more and now
she has (8, 47, or 109). How many did I
give her?
(Start Unknown, JSU)
(Student’s name) has some trucks. I
give you (3, 10, or 35) more. Now you
have (7, 27, or 90) trucks. How many
trucks did you have to start?
Separate
(Result Unknown, SRU)
(Student’s name) has (5, 21, 84, or 101)
cookies and I take (2, 10, 20, or 35).
Now how many does he have?
(Change Unknown, SCU)
(Student’s name) has (9, 30, or 102) pret-
zels. I take some and now you have (7,
20, or 80). How many did I take?
(Start Unknown, SSU)
(Student’s name) has some Legos. I
take (3, 10, or 35) from you. Now you
have (4, 17, or 55). How many Legos
did you have to start?
Part-Part-
Whole
(Whole Unknown, PPW-WU)
(Student’s name) has (3, 8, 20, or 38) red beads and (5, 4, 30,
or 49) blue beads. How many beads do you have?
(Part Unknown, PPW-PU)
(Student’s name) has (8, 12, 50, or 87) beads. (3, 8, 20, or 38)
are red the rest are blue. How many are blue?
Compare
(Difference Unknown, CDU)
matchbox cars. (Friend’s name) has (5,
50, or 57) matchbox cars. Who has
more? How many more?
(Compare Quantity Unknown, CQU)
matchbox cars. (Friend’s name) has (2,
20, or 21) more than (Student’s name).
How many matchbox cars does (Friend’s
name) have?
(Referent Unknown, CRU)
matchbox cars. She has (2, 20, or
21) more matchbox cars than (Friend’s
name). How many matchbox cars does
(Friend’s name) have?
Multiplication
and
Division
(Multiplication, M)
a) There are 4 firefighters wearing their
firefighting uniforms. How many boots
are there?
b) There are (5 or 12) lizards in a pet
store. How many lizard legs are there?
(Measurement Division, MD)
If there are (10 or 50) suckers and each of
(Teacher’s name)’s friends gets (2 or 10)
suckers, how many friends does (Teacher’s
name) have?
(Partitive Division, PD)
You have (12, 39, or 90) candy bars and
you give them to 3 friends (names here)
so they each get the same number. How
many do they each get?
∗
Adapted
from
Carpenter
et
al.
(2015)
10
Illinois
Mathematics
Teacher

Appendix B
Individual Assessment Sheets
Not
Yet
Sometimes
Usually
Teacher’s comments
Strategies
that
Student
Uses
ONE-TO-ONE
CORRESPONDENCE
DIRECT MODELING
Joining All
Separating From
Joining To
Matching
Trial and Error
Grouping
Measurement
Partitive
COUNTING
STRATEGIES
Counting On from First
Counting On from Larger
Counting Down
Counting On To
Trial and Error
Skip Counting
Counting strategies with
number sets 20–100
Counting strategies with
number sets 0–20
DERIVED FACTS
NUMBER FACTS

Needs
Help
Sometimes
Usually
Teacher’s comments
Mathematical
Practices
for
Student
to
Develop
1) Makes sense of problems and per-
severes in solving them by attempt-
ing a solution strategy.
2) Reasons quantitatively and ab-
stractly by making sense of quan-
tities and the relationships among
them such as realizing 5 + 7 = 5 +
(5 + 2).
3) Constructs viable arguments and
critiques the reasoning of others
such as having students know that
when classmates’ explanations are
unclear, students are to ask ques-
tions to make sense of the reasoning
of others.
4) Models with mathematics such
as using pictures and numerical ex-
pressions to represent understand-
ing of the problem.
5) Uses appropriate tools strategi-
cally such as knowing when and
how to use base-ten blocks.
6) Attends to precision by com-
municating precisely using correct
terms and by calculating accu-
rately.
7) Looks for and makes use of struc-
ture such as using the inverse, com-
mutative, or distributive property
appropriately.
8) Looks for and expresses regular-
ity in repeated reasoning, such as
knowing that 3 × 1 + 4 × 1 connects
to 3 × 10 + 4 × 10.

Appendix C
Strategies Students Could Use for Each of the Problem Types
Descriptions of strategies that go with each Problem Type are given below. These
descriptions connect to Appendix B as listed on the Individual Assessment Sheets.
For more details of the strategies refer to Carpenter et al. (1999) or Carpenter et
al. (2015). For the strategies listed below, refer to Appendix A when examples are
provided using specific numbers. This list is not all-inclusive; students may use other
strategies than those listed.
Join Result Unknown (JRU)
• Join All by joining two sets of counters.
• Count On From First by using the first number given.
• Count On From Larger by starting with the larger of the two numbers and
Counting On.
• using knowledge of Number Facts and place value, e.g., grouping by tens.
Separate Result Unknown (SRU)
• Separate From by taking away objects from the initial set given.
• Count Down 2, 10, 20 or 35 by ones.
• using knowledge of Number Facts and place value by Counting Down by groups
of tens when given multi-digit numbers.
Join Change Unknown (JCU)
• Join To with counters having to make the initial set.
• Count On To with fingers or counters, not having to make the initial set.
• using a strategy that does not connect to the structure of the problem which in-
dicates flexibility of thinking, e.g., since the structure of the problem is addition,
recognizing and using knowledge of the inverse relationship of subtraction.
Separate Change Unknown (SCU)
• Separate To with counters or fingers by starting with 9 counters and separating
counters until 7 is reached, the answer being the counters in the set being
separated.
• Count Down To with fingers or counters and mentally keeps track of the first
number and the number counted down.
• using a strategy that does not match the structure of the problem, thus indicat-
ing flexibility of thinking, e.g., since the structure of the problem is subtraction,
recognizing the inverse relationship of addition indicates maturity of thinking:
the student adds 2 to 7 to get 9 and knows the 2 represents the unknown.
Join Start Unknown (JSU)
• These problems can only be solved with understanding by students who recog-
nize inverse relationships. Often students memorize these relationships rather
than think about them. Solving this problem by thinking about relationships
reflects maturity of thinking.

• The Direct Modelers and Counters can only use a Trial and Error strategy to
solve this Problem Type. Thus they choose a number to start with and join
counters or count on from the first number in the problem to determine if they
get the final number in the problem. They use Trial and Error until a solution
is reached.
Part-Part-Whole, Part Unknown (PPW-PU)
• Since this Problem Type does not involve action, it is more difficult to solve for
the Beginning and Developing problem solver.
• A Mature problem solver could use a Joining To, Separating From, Counting
On To, or Counting Down strategy.
Compare
• Matching by representing both sets with counters and counting the extra coun-
ters in the set that has more.
• Join All or Count On is used by Mature Problem Solvers.
• “There is not a commonly used strategy corresponding to the action or rela-
tionship described in the problem” for these Problem Types (Carpenter et al.,
2015, p. 25).
Multiplication (M)
• Group by making 4 sets of 2.
• Skip Count by twos.
• using a Number Fact, e.g., 4 × 2 = 2 × 4 if the student understands the Com-
mutative Property.
Measurement Division (MD)
• counting out ten suckers and putting them in groups of 2, counting 5 groups to
discover there are five friends, Direct Modeling, Measurement strategy.
• double counting by putting 2 in a group and counting until 10 counters are used
and then counting the groups and saying, “five,” Direct Modeling, Measurement
strategy.
• Skip Count by 2’s until reaching 10, holding up a finger for each count, and
realizing five fingers represent five friends.
• Knowing the Number Fact 10 ÷ 2 = 5 or 10 ÷ 5 = 2.
Partitive Division (PD)
• counting out 12 manipulatives and dealing counters to 3 groups until all counters
are used and then counting how many are in each group to get the unknown 4;
or by drawing a picture of 3 friends and giving one candy bar to each, then two,
then three, then four as the student counts to 12 and then counts the number
in each group to get 4, Direct Modeling, Partitive strategy.
• Using Trial and Error by picking a number such as 2 to skip-count, and saying
“2, 4, 6” and realizing that won’t work, then trying “3, 6, 9” and then “4, 8,
12” and realizing that works so the unknown is “4.”

Appendix D
Assessment Flow Chart
Join
(Change Unknown)
Solves using a
Direct Modeling
strategy
Cannot Solve
Solves using a
Counting strategy
Join
(Result Unknown)
Separate
(Result Unknown)
Cannot Solve
Have the
student
use coun-
ters to
make sets of
numbers
between
1 and 20;
record
results
Record student’s
strategy
Grouping (M)
Record student’s
strategy
Partitioning
(MD)
Record student’s
strategy
Join (Change
Unknown)
Record student’s
strategy
Join
(Change Unknown)
No counters
Cannot
solve
Solves using
a Counting
strategy
Continue
with counters
available
Grouping
(M)
Record
student’s
strategy
Partitioning
(MD)
Record
student’s
strategy
Partitioning
(PD)
Record
student’s
strategy
Separate
(Result
Unknown)
No Counters
Record student’s
strategy and
continue with
no counters
available
Join
(Result
Unknown)
Record student’s
strategy
Grouping
(M)
Record student’s
strategy
Partitioning
(MD)
Record student’s
strategy
Partitioning
(PD)
Record student’s
strategy
Part-Part-
Whole
(Part
Unknown)
Record
student’s
strategy
Compare
Record
student’s
strategy
Separate
(Start
Unknown)
Record
student’s
strategy

Assessing from a CGI Perspective.pdf

More Related Content

Similar to Assessing from a CGI Perspective.pdf (20)

More from Nancy Ideker (20)

Recently uploaded (20)

Assessing from a CGI Perspective.pdf