Polya’s Problem Solving Techniques

In 1945 George Polya published a book How To Solve It, which quickly became his most
prized publication. It sold over one million copies and has been translated into 17
languages. In this book he identifies four basic principles of problem solving.

Polya’s First Principle: Understand the Problem

This seems so obvious that it is often not even mentioned, yet students are often stymied
in their efforts to solve problems simply because they don’t understand it fully, or even in
part. Polya taught teachers to ask students questions such as:

• Do you understand all the words used in stating the problem?

• What are you asked to find or show?
• Can you restate the problem in your own words?
• Can you think of a picture or diagram that might help you understand the
problem?
• Is there enough information to enable you to find a solution?

Polya’s Second Principle: Devise a Plan

Polya mentions that there are many reasonable ways to solve problems. The skill at
choosing an appropriate strategy is best learned by solving many problems. You will
find choosing a strategy increasingly easy. A partial list of strategies is included:

*Guess and check *Look for a pattern

*Make an orderly list *Draw a picture
*Eliminate the possibilities *Solve a simpler problem
*Use symmetry *Use a model
*Consider special cases *Work backwards
*Use direct reasoning *Use a formula
*Solve an equation *Be ingenious

Polya’s Third Principle: Carry Out the Plan

This step is usually easier than devising the plan. In general, all you need is care and
patience, given that you have the necessary skills. Persist with the plan that you have
chosen. If it continues not to work, discard it and choose another. Don’t be misled, this
is how mathematics is done, even by professionals.

Polya’s Fourth Principle: Look Back

Polya mentions that much can be gained by taking the time to reflect and look back at
what you have done, what worked, and what didn’t. Doing this will enable you to
predict what strategy to use to solve future problems.

(How to Solve It by George Polya, 2nd ed., Princeton University Press, 1957)
1. Understand the Problem
• First. You have to understand the problem.
• What is the unknown? What are the data? What is the condition?
• Is it possible to satisfy the condition? Is the condition sufficient to determine
the unknown? Or is it insufficient? Or redundant? Or contradictory?
• Draw a figure. Introduce suitable notation.
• Separate the various parts of the condition. Can you write them down?

2. Devising a Plan
• Second. Find the connection between the data and the unknown. You
may be obligated to consider auxiliary problems if an immediate
connection cannot be found. You should obtain eventually a plan of the
solution.
• Have you seen it before? Or have you seen the same problem in a slightly
different form?
• Do you know a related problem? Do you know a theorem that could be
useful?
• Look at the unknown! Try to think of a familiar problem having the same or
a similar unknown.
• Here is a problem related to yours and solved before. Could you use it?
Could you use its result? Could you use its method? Should you introduce
some auxiliary element in order to make its use possible?
• Could you restate the problem? Could you restate id still differently? Go
back to definitions.
• If you cannot solve the proposed problem, try to solve first some related
problem. Could you imagine a more accessible related problem? Could
you solve a part of the problem? Keep only a part of the condition, drop
the other part; how far is the unknown then determined, how can it vary?
Could you derive something useful from the data? Could you think of other
data appropriate to determine the unknown? Could you change the
unknown or data, or both if necessary, so that the new unknown and the
new data are nearer to each other?
• Did you use all the data? Did you use the whole condition? Have you
taken into account all essential notions involved in the problem?

3. Carrying Out The Plan

• Third. Carry out your plan.
• Carry out your plan of the solution, check each step. Can you see clearly
that the step is correct? Can you prove that it is correct?

4. Looking Back
• Fourth. Examine the solution obtained.
• Can you check the result? Can you check the argument?
• Can you derive the solution differently? Can you see it at a glance?
• Can you use the result, or the method, for some other problem?
 WARN!NG
S!GNS!
Recognize three common instructional moves that are
generally followed by taking over children’s thinking.
By Victoria R. Jacobs, Heather A. Martin, Rebecca C. Ambrose, and Randolph A. Philipp

H
ave you ever finished work-
ing with a child and realized
that you solved the prob-
lem and are uncertain what
the child does or does not
understand? Unfortunately,
we have! When engaging in a problem-
solving conversation with a child, our goal
goes beyond helping the child reach a cor-
rect answer. We want to learn about the
child’s mathematical thinking, support that
thinking, and extend it as far as possible.
This exploration of children’s thinking is
central to our vision of both productive
individual mathematical conversations and
overall classroom mathematics instruction
STUDIOARZ/THINKSTOCK

(Carpenter et al. 1999), but in practice, we

find that simultaneously respecting chil-
dren’s mathematical thinking and accom-
plishing curricular goals is challenging.

www.nctm.org Vol. 21, No. 2 | teaching children mathematics • September 2014 107
Copyright © 2014 The National Council of Teachers of Mathematics, Inc. www.nctm.org. All rights reserved.
This material may not be copied or distributed electronically or in any other format without written permission from NCTM.
 In this article, we use the metaphor of travel-
ing down a road that has as its destination chil-
dren engaging in rich and meaningful problem
solving like that depicted in the Common Core
State Standards for Mathematics (CCSSM)
(CCSSI 2010). This road requires opportuni-
ties for children to pursue their own ways of
reasoning so that they can construct their own
mathematical understandings rather than
feeling as if they are mimicking their teachers’
thinking. Knowing how to help children engage
in these experiences is hard. For example, how
can teachers effectively navigate situations in
which a child has chosen a time-consuming
strategy, seems puzzled, or is going down a
path that appears unproductive?
Drawing from a large video study of
129 teachers ranging from prospective teach-
ers to practicing teachers with thirty-three

ERPRODUCTIONS LTD/THINKSTOCK
years of experience, we found that even those
who are committed to pointing students to the
rich, problem-solving road often struggle when
trying to support and extend the thinking of
individual children. After watching teachers and
children engage in one-on-one conversations
about 1798 problems, we identified three com-
mon teaching moves that generally preceded a
teacher’s taking over a child’s thinking: Three warning signs
Consider the following interaction in which
1. Interrupting the child’s strategy Penny, a third grader, is solving this problem:
2. Manipulating the tools
3. Asking a series of closed questions The teacher wants to pack 360 books in boxes.
If 20 books can fit in each box, how many
When teachers took over children’s thinking boxes does she need to pack all the books?
with these moves, it had the effect of transport-
ing children to the answer without engaging Penny pauses after initially hearing the prob-
them in the reasoning about mathematical lem, and the teacher supports her by discussing
ideas that is a major goal of problem solving. the problem situation, highlighting what she is
We do not believe that any specific teaching trying to find:
move is always productive or always problem-
atic, because, to be effective, a teaching move Teacher [T]: So, she has 360 books and 20 books
must be in response to a particular situation. in each box. So, we’re trying to find how many
However, because these three teaching moves boxes 360 books will fill.
were almost always followed by the taking over Penny [P]: Hmm …
of a child’s thinking, we came to view them as T: So, you have 360 books, right? And what do
warning signs, analogous to signs a motorist you want to do with them?
might see when a potentially dangerous obsta- P: Put them in each boxes of 20.
cle lies in the road ahead. By identifying these T: Boxes of 20; so you want to separate them
warning signs, we hope that teachers will learn into 20, right?
to recognize them so that they can carefully P: Mmm-hmm.
examine these challenging situations before T: Into groups of 20. So, what are you trying to
deciding how to proceed. find?

108 September 2014 • teaching children mathematics | Vol. 21, No. 2 www.nctm.org
 Warning! Even with the
best of intentions, some
teacher efforts to move
students’ thinking forward
can actually stifle it.

under it (see fig. 1). When Penny pauses briefly

before writing the next number, the teacher
interrupts Penny’s strategy to introduce her
own by asking, “Do you know how many times
two goes into thirty-six?”
Here we see the first warning sign: interrupt-
ing the child’s strategy. The teacher then picks
up a pen and writes the problem 36 ÷ 2 as the
standard division algorithm, and we see the
second warning sign: manipulating the tools.
Penny responds, “Twenty,” and the teacher
invites her to follow the steps to complete the
algorithm (e.g., “How many times does two go
into three?”) but then changes the conversa-
tion slightly to consider the original numbers
in the problem, writing the division problem
360  ÷ 20 as the standard division algorithm.
The teacher completes the first part of the algo-
rithm for this problem herself and then guides
Penny through the rest of the steps by asking a
series of closed questions, requiring only agree-
ment (“Mmm-hmm”) or short answers (e.g.,
“Eight”)—illustrating the third warning sign:
asking a series of closed questions.
P: Trying to find how many go in each—well,
you already finded out that, but you need to T: Do you know how many times 20 goes into
find how … 160? [Penny does not respond.] Do you know
T: How many boxes, right? how many times 2 goes into 16?
P: Right. P: Two times sixteen? Times?
T: So, you’re trying to find out how many
groups of 20 there are?
Penny’s strategy was to count by twenties.
F IG URE 1

P: Mmm-hmm.
T: In 360?
(a) She recorded each number, placed a mark under it, and
then tallied the marks.
After discussing the problem situation,
Penny develops an approach, writes 360, and
starts incrementing by twenties, writing 20 and
40. At this point, she whispers, “It’s gonna take
too long,” but the teacher encourages Penny
to continue by asking about her strategy. “Are
you counting by twenties? Is that what you’re
doing there?”
(b) When Penny paused,
Penny confirms and resumes her strategy, her teacher interrupted and
writing multiples of 20 through 140. Then, from introduced a different approach.
the beginning of her list of numbers, she makes
a mark under each one, apparently tallying the
number of boxes she has made so far. At the
end of her list, she resumes her strategy by writ-
ing the next number, 160, and making a mark

www.nctm.org Vol. 21, No. 2 | teaching children mathematics • September 2014 109
 MANIPU
L
THE T O ATING
understanding of the problem) and then helped
her reach a correct answer. However, we share

OLS this illustration because it also highlights the

three moves that should serve as warning signs
because they often, and in this case did, lead to
taking over the child’s thinking: interrupting the
child’s strategy, manipulating the tools, and ask-
ing a series of closed questions.
T: Well, if you go, how many 2s are in 16—so, 2,
ALEXANDER BEDRIN/THINKSTOCK

4, 6, 8, 10, 12, 14, 16 [writing the numbers while 1. Interrupting the child’s strategy
she counts by twos]. How many is that? [The When a teacher interrupts a child’s strategy
teacher points along the list of numbers while to suggest a different direction, the teacher’s
she counts aloud.] 1, 2, 3, 4, 5, 6, 7, 8, right? thinking becomes privileged because the child’s
P: Mmm-hmm. thinking—which was “in process”—is halted.
T: So, 20 goes into 160, which is just [attaching] This interruption may involve talking over a
a zero. [The teacher points at the appropriate child who is already speaking, or jumping in
spot on the paper for Penny to write.] when a child is working silently. In both cases,
P: [writing] Eight. this warning sign generally accompanies the
T: Mmm-hmm. Twenty times 8. Yes, ’cause 20 hazard of breaking the child’s train of thought—
times 8 is 160, so this would be an 8, right? the child may struggle to regain momentum
P: Mmm-hmm. in solving the problem or may lose the thread
of his or her idea altogether. Additionally, the
With the answer of 18 now written, the teacher teacher may introduce a strategy that does not
checks Penny’s understanding of what they make sense to the child. In the example above,
have just done with another series of closed Penny had a viable strategy and was in the
questions. process of executing it when her strategy was
interrupted with a different approach proposed
T: So, how many boxes do we need? [When by her teacher. Perhaps the teacher thought
Penny does not respond, the teacher points to that Penny’s strategy of counting up by twenties
the answer of 18.] What does this represent? Do would take too long or that she would struggle
you know? too much to find each multiple. Or perhaps the
P: Eighteen. teacher had expected (or hoped) that Penny
T: Mmm-hmm, but do you know like in this would use the standard division algorithm. In
problem how we would … any case, Penny had no opportunity to return to
P: Eighty-one? I mean … her original strategy and complete it. Further-
T: Do you know what this [18] represents? Like more, Penny was making sense of the problem
this 20 represents the 20 books that can fit in situation with her original strategy, but this
each box. sense making disappeared when the teacher
P: Mmm-hmm. introduced the algorithmic strategy.
T: And 360 represents the total number of In our larger study, we observed that some
books. So, 18 represents … children, like Penny, had viable strategies for
P: The boxes. solving their problems, whereas other children’s
T: How many boxes? strategies and intent were unclear. However,
P: Eighteen. in all cases, their thinking was “in process” in
T: There you go. Does that make sense? that they were writing, counting aloud, moving
P: Mmm-hmm. fingers while working silently, and so on. The
T: ’Cause you just have to divide them into the teachers’ interruptions sometimes introduced
different boxes. completely new strategies (as in Penny’s case)
and other times pushed children to engage
In this example, the teacher began the with their partial strategies in specific ways that
interaction with moves that supported Penny’s changed children’s problem-solving approaches
thinking (e.g., probing her initial strategy and and were inconsistent with their reasoning. In

110 September 2014 • teaching children mathematics | Vol. 21, No. 2 www.nctm.org
each case, teachers risked impeding or aborting who owned the thinking. Teachers also risked
children’s thinking by inserting and privileging altering problem representations to representa-
their own ideas while halting the children’s in- tions unclear to children—teachers and children
process thinking. may be thinking differently, even when looking
at the same manipulatives or written represen-
2. Manipulating the tools tations (Ball 1992).
Another warning sign teachers should notice is
when they visibly take control of the interaction 3. Asking a series of closed questions
by manipulating the pen, cubes, or other tools. This third warning sign highlights a situation
In the example above, Penny had a written that may begin nonhazardously—when the
recording of her strategy in progress at the top of teacher asks a question with a simple and often
the page when the teacher’s writing of the stan- obvious answer. The danger arises when this
dard division algorithm shifted Penny’s focus to question is followed by another and another and
the teacher’s strategy. The teacher then retained another such question. The net effect of a series
control of the pen for much of the interaction of closed questions is that the problem gets
while she wrote and talked her way through broken down for the child into tiny steps that
this algorithm. In doing so, she changed the require minimal effort and little understanding
representation of the problem from Penny’s of the problem situation. Such was the case for
written recording of the multiples of twenty Penny after the standard division algorithm was
and the accompanying tallying of boxes to an introduced because the teacher asked questions
approach that was abstract for Penny and not that required little more than Penny’s agreement
a good match for her thinking—as evidenced (“Mmm-hmm”). Penny did not have to think
in Penny’s struggles to make sense of both the about the underlying ideas of division, and the
calculation and the result. problem-solving endeavor was instead reduced
In our larger study, we observed teach- to following directions.
ers writing things or moving manipulatives, In our larger study, we observed teachers
although sometimes they did so without chang- giving directions that were sometimes phrased
ing the course of conversations so completely. as questions and other times as steps to fol-
However, taking over tools was inherently risky low. In either case, when the answer was finally
because doing so sent children a message about reached, the children had often forgotten the

Become aware of teaching moves and of potentially taking over students’ thinking.
TABL E 1

Warning signs for taking over children’s thinking

Warning signs Questions to consider before proceeding Potential alternative moves

1. Interrupting Do I understand how the child is thinking and will • Slow down: Allow the child to finish
the child’s my ideas interfere with that thinking? before intervening.
strategy
Will the child be able to make sense of my ideas? • Encourage the child to talk about his
or her strategy so far.
2. Manipulating Will the child still be in control of the problem solving?
the tools • Ask questions to ensure that the
Will my problem representation make sense to child understands the problem
the child? situation and how the strategy
relates to that situation.
3. Asking Will my questions be about the child’s thinking or
a series my thinking? • Ask whether trying another tool or
of closed strategy would help.
Will the child still have an opportunity to engage
questions
with substantive mathematics, or will my questions
prevent him or her from doing so?

www.nctm.org Vol. 21, No. 2 | teaching children mathematics • September 2014 111
 ED
original goal and were could have left Penny in control of the pen and
rarely able to relate the posed some open-ended questions to explore

OS solution to the prob-

lem situation. We
saw this confusion
Penny’s understanding of the algorithm and its
connection to the problem situation. Another
option would have been to ask Penny to con-
CL
with Penny when sider efficiency while she was still solving the
she guessed, problem with her original strategy. After Penny
“Eighty-one?” had completed 160 books (8 boxes) by count-
i n re s p o n s e ing by 20s, the teacher could have asked her to
to a question reflect on what she had done so far and if that
about how work could help her proceed more quickly. (This
many boxes question might prompt Penny to recognize that
were needed. This apparent stab in the dark was doubling 160 books [and 8 boxes] would be close
STUDIOARZ/THINKSTOCK

a signal that the teacher’s sequence of closed to the needed 360 books, but she would also
questions did not help Penny make sense of the have the option of continuing with her original
teacher’s algorithmic strategy or relate it to the strategy.) Although there is no perfect move in
original problem. any situation, these types of alternative moves
might have increased the likelihood that the
Heeding the warning signs teacher would have supported and extended
The warning signs exemplified in Penny’s Penny’s thinking without taking over that think-
interaction arose often in our study, sometimes ing. (See Jacobs and Ambrose [2008–2009] for
in isolation and sometimes as a set. So, what more on alternative moves.)
can teachers do? When possible, we encourage
teachers to heed the warning signs by choos- Are these moves ever productive?
ing alternative moves that are more likely to Our data convinced us that the warning signs
preserve children’s thinking. The questions in were generally unproductive moves, but we
table 1 are designed to help teachers consider wondered if these same moves could ever be
alternative moves. We do not suggest that these productive. After all, teaching moves need to
alternative moves are foolproof—unfortunately, be considered in context because the same
no moves are. Engaging with children’s thinking move can be productive in one situation but
is a constant negotiation, fraught with trial and unproductive in another. We found that the
error, as teachers work to find ways to elicit and three warning signs were occasionally used
respect children’s thinking while nudging that productively but, to us, they almost seemed like
thinking toward reasoning that is more sophis- different moves because, although they looked
ticated. However, in analyzing our data, we were similar on the surface, they were coupled with
struck with how often the three warning signs the preservation of children’s thinking.
were unproductive in achieving this goal, thus For example, teachers sometimes produc-
prompting us to consider alternative moves. tively interrupted a child going far off track or
For example, how might the interaction engaging in an extremely inefficient strategy
have been different if Penny had not been by discussing with the child how he or she was
interrupted and had been able to complete her thinking. This move was not, as we saw with
initial strategy? The teacher could have probed Penny, used to immediately suggest a differ-
Penny’s completed strategy, validating and elic- ent direction but instead deepened the child’s
iting her ways of thinking about the problem. (and teacher’s) understanding of how the child
If the teacher still wondered about efficiency, was thinking about the problem. Similarly,
she might have asked if Penny could think of teachers sometimes productively manipulated
another way of solving the problem, perhaps the tools to help organize the workspace by
in a way that was more efficient. This approach removing “extra” cubes after ensuring that they
would have built on Penny’s ways of thinking were considered “extra” by the child (versus,
about the problem while still preserving the for example, removing cubes to ensure that the
goal of efficiency. Alternatively, if the teacher did correct quantities were represented). This move
choose to suggest the division algorithm, she provided some organizational scaffolding while

112 September 2014 • teaching children mathematics | Vol. 21, No. 2 www.nctm.org
preserving the child’s way of thinking about the Carpenter, Thomas P., Elizabeth Fennema,
problem. Finally, teachers sometimes produc- Megan Loef Franke, Linda Levi, and Susan
tively asked a series of closed questions to check B. Empson. 1999. Children’s Mathematics:
on their understanding of a child’s strategy. This Cognitively Guided Instruction. Portsmouth,
move kept the focus on the student’s thinking NH: Heinemann.
by putting the child in position to confirm or Common Core State Standards Initiative (CCSSI).
deny what he or she had already done, said, 2010. Common Core State Standards for
or thought. Thus, we are not suggesting that Mathematics. Washington, DC: National
the three warning signs can never be used pro- Governors Association Center for Best
ductively. However, our data overwhelmingly Practices and the Council of Chief State School
showed that these moves typically led to taking Officers. http://www.corestandards.org
over children’s thinking and thus should be used /wp-content/uploads/Math_Standards.pdf
with caution. Jacobs, Victoria R., and Rebecca C. Ambrose.
2008–2009. “Making the Most of Story
Good intentions Problems.” Teaching Children Mathematics 15
All four authors have had the experience of solv- (December/January): 260–66.
ing a problem for a child without gaining any
idea what the child does or does not understand. This research was supported in part by a
We always begin these interactions with good grant from the National Science Foundation
intentions, but other pressures (e.g., shortness (ESI0455785). The opinions expressed in this
of time) or goals (e.g., desire to see the child use article do not necessarily reflect the position,
a more sophisticated strategy) often derail our policy, or endorsement of the supporting agency.
efforts. Our data also showed that taking over a
child’s thinking was not linked to any particular Victoria R. Jacobs,
tone or interaction style. In other words, in any [email protected],
given situation, any of us can be tempted to take is a mathematics
over a child’s thinking. educator at the
In summary, avoiding the impulse to take over University of North
a child’s thinking in one-on-one conversations Carolina at Greens-
(either inside or outside the classroom) is chal- boro. Heather A.
lenging. We also recognize that the task becomes Martin, hmartin@
even more challenging in social situations like ncbb.net, and
small-group work or whole-class discussions. Rebecca C. Ambrose,
Nonetheless, in all these instructional situations, [email protected], are mathematics educators
the same goals exist: eliciting, supporting, and at the University of California–Davis. Randolph A.
extending children’s thinking. Further, the moves Philipp, [email protected], is a mathematics
identified as warning signs are likely to thwart educator at San Diego State University in California.
efforts to achieve these goals because children They all collaborate with teachers to explore children’s
get transported to the answer without actually mathematical thinking and how that thinking can
engaging in problem solving. In identifying the inform instruction.
warning signs, our hope is that teachers will be
more likely to pause and consider alternative
moves to avoid the dangers of taking over chil-
dren’s thinking. As a first step, we invite readers
to go online (see the More4U box to the right) to
practice recognizing these warning signs in an
interaction with a first grader.
Download one of the free
apps for your smartphone
R EFER E N CE S
to scan this code, or go to
Ball, Deborah Loewenberg. 1992. “Magical Hopes: www.nctm.org/tcm067 to
Manipulatives and the Reform of Math Educa- access an appendix.
tion.”American Educator 16 (2): 16–18, 46–47.

www.nctm.org Vol. 21, No. 2 | teaching children mathematics • September 2014 113
 Equity:

Save the Date!

All Means ALL
Washington DC • April 22–25, 2009

NCTM 2009 Annual

Meeting & Exposition

Visit www.nctm.org/washingtondc for the most up-to-date information.

The NCTM 2009 Annual • Learn from more than 800 • Develop your mathematics
Meeting and Exposition in presentations in all areas of resource library with books
Washington D.C. will be mathematics and products from the
the mathematics teaching • Network with other educators NCTM Bookstore
event of the year. This is one from around the world • Enjoy Washington D.C.
professional development • Explore the NCTM Exhibit and all the nation’s
opportunity you can’t afford to Hall and experience the latest capitol has to offer
miss. Conference attendees will: products and services
Making the Most of

Story Problems
S
Honoring students’ solution tory problems are an important component of
the mathematics curriculum, yet many adults Photograph by Rebecca C. Ambrose; all rights reserved

approaches helps teachers capitalize shudder to remember their own experiences

with them, often recalling the elusive train prob-
on the power of story problems. lems from high school algebra. In contrast, research
shows that story problems can be powerful tools for
No more elusive train scenarios! engaging young children in mathematics, and many
students enjoy making sense of these situations
(NCTM 2000; NRC 2001). Honoring children’s
By Victoria R. Jacobs and Rebecca C. Ambrose
story problem approaches is of critical importance so
Victoria R. Jacobs, [email protected], is a mathematics educator at San Diego State that they construct strategies that make sense to them
University in California. Rebecca C. Ambrose, [email protected], is a mathematics edu-
cator at the University of California–Davis. They collaborate with teachers to explore children’s
rather than parrot strategies they do not understand.
mathematical thinking and how that thinking can inform instruction. To explore how teachers can capitalize on
the power of story problems, we chose to study

260 Teaching Children Mathematics / December 2008/January 2009

Copyright © 2008 The National Council of Teachers of Mathematics, Inc. www.nctm.org. All rights reserved.
This material may not be copied or distributed electronically or in any other format without written permission from NCTM.
teacher-student conversations in problem-solving Table 1
interviews in which a K–3 teacher worked one-on-
one with a child. The skills needed for productive Teacher Moves to Support a Child’s Thinking before a Correct
interviewing are the same as those needed in the ­Answer Is Given
classroom: Teachers must observe, listen, question,
Category Sample Teacher Moves
design follow-up tasks, and so on. We focused our
investigation on interviews because interviews iso- Ask him to explain what he knows about the
problem.
late these important teacher-student conversations Ensure that the child
from other aspects of classroom life. Rephrase or elaborate the problem.
understands the problem.
Use a more familiar or personalized context in
the problem.
Supporting and Extending Change the mathematics
Mathematical Thinking in the problem to match
Change the problem to use easier numbers.

the child’s level of Change the problem to use an easier math-

After analyzing videotaped problem-solving inter- ematical structure.
understanding.
views conducted by 65 teachers interviewing 231
Ask him to explain a partial or incorrect
children solving 1,018 story problems, we identified
strategy.
eight categories of teacher moves (i.e., intentional Explore what the child has
Ask specific questions to explore how what he
actions) that, when timed properly, were produc- already done.
has already done relates to the quantities and
tive in advancing mathematical conversations. We relationships in the problem.
separately considered (a) the supporting moves that
Ask him to consider using a different tool.
a teacher used before a student arrived at a correct
Remind the child to use Ask him to consider using a different strategy.
answer and (b) the extending moves that a teacher
other strategies­.
used after the child gave a correct answer. We want Remind him of relevant strategies he has
used before.
to be clear that the eight categories of teacher moves
we present are not intended to be a checklist that
a teacher executes on every problem. Instead, we Ensure that the child understands the problem.
consider these moves to be a toolbox from which a A teacher can provide support by helping a child
teacher can draw, after considering the specific situ- develop an understanding of the problem to be
ation and instructional goals. In the midst of instruc- solved. Typical teacher moves include rereading a
tion, the most effective moves arise in response to problem multiple times and asking a child about
what a child says or does and, therefore, cannot be specific quantities in a problem (e.g., “How many
preplanned. Because strategically responding to puppies are in the park?”). A twist on this repetition
children’s mathematical thinking is challenging, we is to ask children to explain problems in their own
identified our eight categories of teacher moves in words. In listening to them describe a story problem
an effort to assist teachers in this decision making. in its entirety, a teacher can pinpoint what children
do and do not understand.
Before a correct answer is given Rephrasing or elaborating on a story can also
When a child struggles or has the wrong answer, help to engage a child. Often, this elaboration
a teacher must determine how and when to inter- involves using a more familiar context or personal-
vene in order to facilitate moving the child forward ization so that the child and her friends are charac-
without taking over the child’s thinking. Support- ters in the story. For example, a kindergartner was
ing a student’s mathematical thinking requires the asked to solve the following problem:
teacher to “enter the child’s mind” (Ginsburg 1997)
as much as possible to determine what the source of The teacher has twelve pencils and three baskets.
difficulty might be. Then the teacher’s hypotheses If she wants to put the same number of pencils
about a child’s thinking should drive the choices in each basket, how many pencils should she put
made. Because “entering the child’s mind” can be in each basket?
quite difficult, a teacher needs to be flexible and
prepared to explore various supportive approaches. The child made a pile of fifteen cubes and kept
In our analysis, we identified four categories of rearranging them. In response, the teacher, Mr.
moves that teachers regularly used to support a Reynolds, decided to elaborate and personalize the
child’s thinking before the student arrived at a cor- problem by involving their classroom and making
rect answer (see table 1). himself the teacher in the story:

Teaching Children Mathematics / December 2008/January 2009 261

 Change the mathematics to match the child’s
level of understanding. When children do not
understand a problem, even after attempts to
rephrase or elaborate it, changing the problem
itself can be productive. One type of change is
to use easier numbers. Specifically, using smaller
or friendlier numbers (e.g., decade numbers) can
help them gain access to the mathematics under-
lying a problem. After making sense of an easier
Photograph by Rebecca C. Ambrose; all rights reserved

problem, students generally gain confidence and,

in many cases, can then make sense of the original
problem.
Similarly, because research shows that children
have more difficulty with some problem structures
than others, another type of change is to use an eas-
ier mathematical structure (Carpenter et al. 1999).
For example, a first grader was asked to solve this
problem:

Twelve mice live in a house. Nine live upstairs.

How many live downstairs?

A teacher’s most effective teaching arises in response to what a child says or does. Because part-whole problems such as this do not
have an explicit joining or separating action, chil-
Let me change it a little bit. Let’s try this. Mr. dren often do not know how the quantities relate.
Reynolds has three baskets. I have three baskets, This student made a set of nine cubes and a set of
and I have twelve pencils in my hand, and I say, twelve cubes and joined them to get twenty-one.
“I’ve got to do something with these pencils. I After several unsuccessful attempts to help the
can’t walk around with them all day! What am I child understand the problem, the teacher chose
going to do with these pencils? Oh, here’s what to change the problem to include an explicit sepa-
I’ll do. I’ll put some in each basket so the kids rating action. Specifically, the teacher explained,
can come get them.” But then I think, “I’d better “Nine of those mice are going to go upstairs and
put the same number in each basket. Because if I watch TV.” In response, the girl separated nine
put, like, two in one basket and ten in one basket, mice from her set of twelve, leaving a group of
that’s not fair. So I have to put the same number three. This change in mathematical structure did
of pencils in each basket.” How many pencils more than allow the student to solve a problem
would I put in each one of those baskets so that correctly. By providing her access to an easier but
all the baskets would have the same number of related problem, the teacher created opportunities
pencils inside? for discussing the quantities and relationships in
both problems. Thus, with further skilled question-
This elaborated story did not change the math- ing, the teacher could use the child’s understanding
ematical structure of the problem but did make the of the second problem to help her understand the
problem more real for the child, and in this case, she original problem and, more generally, problems
solved the problem correctly by using trial and error with a part-whole structure.
to create three piles of four cubes each. Elaborating
a story may seem counterintuitive because it goes Explore what the child has already done. When
against the traditional approach of helping children struggling with a problem, children can sometimes
identify keywords or irrelevant information in story determine what went wrong if they are encouraged
problems. However, when elaboration is designed to articulate partial or incorrect strategies. General
to make a problem more meaningful, children are questions, such as “Can you tell me how you solved
more likely to avoid mechanical problem-solving it” or “What did you do first?” can be helpful for
approaches and instead work to make sense of the starting conversations, but follow-up questions
problem situation. require a teacher to ask about the details of a child’s

262 Teaching Children Mathematics / December 2008/January 2009

strategy and thus cannot be preplanned. For exam- Table 2
ple, a first grader was asked to solve this problem:
Teacher Moves to Extend a Child’s Thinking
One cat has four legs. How many legs do seven after a Correct Answer Is Given
cats have?
Category Sample teacher moves
The child (C) put out seven teddy-bear counters. Ask her to explain her strategy.
He saw teddy-bear counters as having two legs and Promote reflection on Ask specific questions to clarify how the
two arms and, therefore, counted only two legs on the strategy the child just details of her strategy are connected to the
completed. quantities and mathematical relationships in
each teddy bear, answering ”Fourteen.” The teacher
the problem.
(T) recognized that his confusion was linked to the
counters he had chosen, and she posed questions Ask her to try any second strategy.
to clarify how his work related to the problem Encourage the child to
Ask her to try a second strategy connected
context: to her initial strategy in deliberate ways (e.g.,
explore multiple strategies
more efficient counting or abstraction of work
and their connections.
with manipulatives).
T: How many legs on a bear?
Ask her to compare and contrast strategies.
C: Two.
T: How many legs on a cat? Ask her to write a number sentence that
Connect the child’s thinking “goes with” the problem.
C: Four. to symbolic notation.
T: How many did you count? How many legs each Ask her to record her strategy.
did you count? Ask her to solve the same or a similar prob-
C: Two. Generate follow-up lem with numbers that are more challenging.
T: Is that how many legs cats have? problems linked to the Ask her to solve the same or a similar
problem the child just problem with numbers that are strategically
C: No, cats have four, and bears have two.
completed. selected to promote more sophisticated
T: OK, could you do that again for me?
strategies.
C: First I get one cat [puts out one teddy-bear
counter], and then I get a bear [puts out another
teddy-bear counter], and this cat has four legs, and may also find that suggesting a particular tool or
the bear has two legs. reminding a child of strategies used in the past is
T: Are there bears in the story? beneficial. For example, a first-grade student was
C: No, there’s cats. asked to solve the following problem:

This dialogue continued for some time before the Let’s pretend we’re out at the snack tables, and
child solved the problem correctly by counting four seagulls come to the snack tables. And then
four legs on each bear and then again by using a seven more seagulls come to the snack tables.
different tool. The support the teacher provided How many seagulls are at the snack tables?
began with what the child had already done, and
through specific questioning, she helped him make The child first counted to four, raising one finger
sense of how his initial strategy was related (and with each count. She then put those four fingers
not related) to the problem. Note that she could not down. Next, she counted to seven, raising one finger
have preplanned this conversation, because it grew with each count. At this point, the child was baffled,
out of her careful observation of his way of using staring at her fingers. The teacher suggested, “Want
the teddy-bear counters. to try it with cubes?” The child immediately made
a stack of four Unifix cubes and a stack of seven
Remind the child to use other strategies. Some- Unifix cubes and then counted them altogether to
times students get lost in a particular strategy, and get an answer of eleven. She was confident and
instead of abandoning that strategy for a more efficient once she started using the Unifix cubes.
effective one, they persist in using it in unproductive The teacher did not tell the child how to solve the
ways. A teacher can help by nudging them to think problem but did encourage her to consider using
more flexibly and to try alternative approaches. A a tool that was more conducive to representing
simple suggestion to try a different tool or a differ- both sets; the child did not have enough fingers to
ent strategy can sometimes give a child permission show seven and four at the same time. This support
to move on and self-redirect. At times, a teacher reflected the teacher’s understanding of children’s

Teaching Children Mathematics / December 2008/January 2009 263

 direct-modeling strategies in which they represent tive when they were specific and in response to the
both sets before combining them. details of what a student had already said or done.
For example, a second grader was asked to solve
After a correct answer is given this problem:
Solving a story problem correctly using a valid
strategy is an important mathematical endeavor. This morning I had some candy. Then I gave you
However, we view problem solving as a context five pieces of candy. Now I have six pieces of
for having mathematical conversations, and this candy left. How many pieces of candy did I have
conversation need not end when the correct answer this morning before I gave some to you?
is reached. Instead, a teacher can pose additional
questions to help students deepen their understand- The student quickly solved this problem mentally
ing of completed work and connect it to other math- and explained, “Five plus five, if you took one
ematical ideas. We have identified four categories away, is ten and then one more is eleven, so you
of moves that teachers regularly used to extend had eleven.” Children often provide correct answers
children’s thinking after arrival at a correct answer to problems with this structure, in which the initial
(see table 2). quantity is unknown, without really understand-
ing what they are finding. In this case, the teacher
Promote reflection on the strategy just com- probed the child’s thinking in relation to this issue:
pleted. Once a student has correctly solved a prob-
lem, a teacher can ask for a strategy explanation or T: So how did you know to add them together?
for clarification about how the use of a particular C: I don’t know. I just added them, I guess.
strategy makes sense with the quantities and math- T: Well, think about it. Why does that make sense
ematical relationships expressed in the problem. for the problem?
Articulating these ideas can reinforce a child’s
understanding and give a teacher a window into The child thought about this question for some
that understanding. Again, attention to detail mat- time and eventually used Unifix cubes to act out
ters. Similar to the supporting questions intended the story and convince himself (and the teacher)
to explore children’s partial or incorrect strategies, that eleven was the correct answer and made sense
teachers’ extending questions were most produc- with the story. By asking him to reflect further on
his strategy, the teacher ensured that he was making
sense of the mathematics.

Encourage the child to explore multiple strate-

gies and their connections. Children need oppor-
tunities to not only solve problems but also explore
the mathematical connections among multiple
strategies for the same problem. One approach is
to ask them to generate a second strategy—any
Photograph by Rebecca C. Ambrose; all rights reserved

strategy—to a problem they have already solved.

Another approach is to ask for a second strategy
that is connected to their initial strategy in deliber-
ate ways. For instance, a third grader using base-ten
blocks to represent 12 pages of 10 spelling words
per page put out 12 tens rods but counted all 120
blocks by ones! The teacher built on this initial
strategy by asking her to count the blocks another
way. The child responded by counting by tens and
even shared that this second strategy was easier.
Another way a teacher can deliberately build on
an initial strategy is to ask for a mental strategy that
is an abstraction of work with manipulatives. For
Although a teacher may initially support a student’s step-by-step recording of a example, a third grader was asked to solve the fol-
strategy, he should diligently support the child’s thinking. lowing problem:

264 Teaching Children Mathematics / December 2008/January 2009

 There are 247 girls on the playground and there sively more abstract until they are completely
are 138 boys on the playground. How many chil- symbolic.
dren are on the playground? Generating a symbolic representation of a strat-
egy can help children develop meaning for, and
The student initially represented both quantities facility with, mathematical symbols because the
with base-ten flats, rods, and single cubes. Next representation is linked with their interpretation of
he combined the hundred flats (3), combined the the problem. For example, a second grader solved
ten rods (7), combined some of the single cubes to the problem about the number of legs on seven cats
make 10, traded the 10 single cubes for 1 tens rod, by first putting out seven tiles (cats). Next he moved
making a total of 8 tens rods, and finally counted two tiles to the side and said, “Four plus four equals
the remaining single cubes (5) to answer 385. The eight.” He then moved another tile to the side and
teacher then asked, “Doing just what you did with said, “Eight plus four equals twelve.” He continued
the materials, could you solve that problem in moving one tile at a time until he had used them
your head?” The child looked at the numbers and all, each time adding four more to his running total.
abstracted what he had just done with the cubes. When asked to write a number sentence to show
Specifically, he explained that he could add 100 to what he had done, he wrote the following:
200 to get 300 and then add 30 to 40 to get 70. Next
he put 2 from the 7 with the 8 to get another 10, 4 + 4 g 8 + 4 g 12 + 4 g 16 + 4 g 20 + 4 g
which made 80, and had 5 ones left, so the answer 24 + 4 g 28
was 385. When executing this mental strategy, the
child articulated the underlying mathematical idea Unlike the number sentence that ‘“goes with” the
of both strategies: combine like units and, when problem (7 × 4 = 28), his symbolic representation
necessary, regroup (i.e., decompose the 7 into 5 and reflects how this student thought about and solved
2 so that the 2 can be combined with the 8 to make the problem. Note that his use of arrows instead of
a new 10). equal signs avoids the incorrect use of the equal
Through experiences with multiple strate- sign between expressions.
gies, children can gain the ability and flexibility Requesting links between strategies and sym-
to change strategies when one is unsuccessful. A bolic notation is important so that children see the
teacher can also use multiple strategies to highlight mathematics done on paper as connected to solving
underlying mathematical ideas by asking students story problems. Moreover, once children become
to explicitly compare and contrast strategies. At facile with symbolic notation, the notation itself
times, a teacher may even ask a child to compare can become a tool for problem solving and reflec-
a successful strategy to a previously unsuccessful tion. We offer a final note of caution: A teacher
attempt, because, in many cases, the child will dis- may initially need to support a student in recording
cover the reason the strategy failed. each step of a strategy so that parts are not omitted.
However, a teacher needs to be vigilant in providing
Connect the child’s thinking to symbolic nota- support to record the child’s—not the teacher’s—
tion. When solving a story problem by drawing, ways of thinking about a problem.
using manipulatives, or computing mentally, stu-
dents may not use any symbolic notation. A teacher Generate follow-up problems. By carefully
can encourage students to connect their work with sequencing problems, a teacher can create unique
mathematical symbols by asking them to either opportunities for mathematical discussions. Although
generate a number sentence that “goes with” the we recognize the importance of practice, we are sug-
problem or record the strategy used to solve the gesting something beyond simply assigning addi-
problem. Although requesting a number sentence tional problems to solve. We advocate that, in the
that “goes with” the problem is perhaps the more midst of instruction, a teacher can consider a child’s
typical request, asking for a strategy representa- existing understanding and then modify the initial
tion can be powerful. Young children often begin problem or create a new problem to add challenge or
recording their strategies in unconventional ways to encourage use of more sophisticated strategies. A
that include a mix of symbols and drawings. They first grader was asked to solve this problem:
might draw pictures of manipulatives they used and
then add number labels to parts of those pictures. The Kumyeey woman was collecting acorns.
Over time, children’s recordings become progres- She had nine baskets, and she put ten acorns in

Teaching Children Mathematics / December 2008/January 2009 265

 each basket. So how many acorns did she have in mathematical discussions about story problems
altogether? is challenging; we offer three final guidelines:

The child quickly responded, “Ninety,” explain- • Elicit and respond to a child’s ideas. The most
ing that he had counted, “Ten, one,” putting up ten effective teacher moves cannot be preplanned.
fingers and then one finger. He continued, “Twenty, Instead, they must occur in response to a stu-
two,” again putting up ten fingers and this time, two dent’s specific actions or ideas. Thus, expertise is
fingers. He continued with this pattern of count- tied less to planning before a student arrives and
ing and finger use: “Thirty, three; forty, four; fifty, more to seeding conversations, finding the math-
five; sixty, six; seventy, seven; eighty, eight; ninety, ematics in children’s comments and actions, and
nine.” The teacher then decided to extend the making in-the-moment decisions about how to
child’s use of ten by posing a related problem and support and extend children’s thinking.
asking him to consider the connections: • Attend to details in a child’s strategy and
talk. Research on children’s developmental
T: So that’s how you got ninety. What if she had nine trajectories shows that subtle differences in chil-
baskets, but she put eleven in each basket instead of dren’s strategies and talk can reflect important
ten? [Child thinks for a while.] Could you use some distinctions in their mathematical understand-
of the work that you’ve already done—that we did ings (NRC 2001). A teacher can customize
during the afternoon—or would you have to start all instruction on the basis of these distinctions, and
over again? She still has nine baskets, and there are by attending to details of a child’s explanations
still ten acorns in each basket, and then she puts in and comments, a teacher also communicates
one more so that each basket has eleven. respect for a child’s ideas.
C: Ohhhhh! I get it. Well, there’s already ten in • Do not always end a conversation after a cor-
each basket, so that’s ninety. So I count up nine, rect answer is given. Important learning can
one more nine. I mean nine ones. I’m going occur after students give a correct answer if the
to add nine ones. So there’s already ninety, so teacher asks them to articulate, reflect on, and
ninety-one, ninety-two, ninety-three, ninety-four, build on their initial strategies.
ninety-five, ninety-six, ninety-seven, ninety-eight,
ninety-nine.
References
By strategically selecting numbers and by drawing Carpenter, Thomas P., Elizabeth Fennema, Megan L.
attention to the link between problems, the teacher Franke, Linda Levi, and Susan Empson. Children’s
Mathematics: Cognitively Guided Instruction. Ports-
was able to further this child’s base-ten understand- mouth, N.H.: Heinemann, 1999.
ing by helping him recognize and use the ten in the Ginsburg, Herbert P. Entering the Child’s Mind: The Clin-
number eleven. ical Interview in Psychological Research and Practice.
New York: Cambridge University Press, 1997.
Mewborn, Denise S., and Patricia D. Huberty. “Question-
Summary ing Your Way to the Standards.” Teaching Children
Mathematics 6 (December 1999): 226–27, 243–46.
Our project builds on previous work on teacher National Council of Teachers of Mathematics (NCTM).
questioning (see, for example, Mewborn and Principles and Standards for School Mathematics.
Huberty 1999; Stenmark 1991), which provide lists Reston, VA: NCTM, 2000.
of potential questions. These lists can be important National Research Council (NRC). Adding It Up: Help-
ing Children Learn Mathematics. Washington, DC:
starting points for eliciting a student’s thinking, but National Academy Press, 2001.
we hope that our eight categories of teacher moves Stenmark, Jean K., ed. Assessment Alternatives in Math-
(four designed to support children’s thinking and ematics: An Overview of Assessment Techniques That
four to extend it) can help teachers further custom- Promote Learning. Reston, VA: National Council of
ize questioning to make the most of story problems. Teachers of Mathematics, 1991.
These moves do not always lead to correct answers,
and we reiterate that not all eight are intended to This research was supported in part by National
be used in every situation. However, together they Science Foundation grant #ESI0455785. The views
form a toolbox from which teachers can select expressed are those of the authors and do not nec-
means to help students solve problems and explore essarily reflect the views of the National Science
connections among mathematical ideas. Engaging Foundation. s

266 Teaching Children Mathematics / December 2008/January 2009