Place Value as the Key to Teaching

Decimal Operations
Judith Sowder
Some years ago I examined several middle school
students' understanding of numbers (Threadgill
Sowder 1984). The answers that students gave me
during that study showed me that their understanding,
developed through experiences in the elemen
tary grades, was fuzzy and led me to undertake a
decade of research on children's number sense in the
elementary and middle school grades. I will set the
stage for this article by sharing two ofthe questions I
gave the students during that study and some of the
responses I received.
Question 1
The sum of 148.72 and 51.351 is approximately
how much?
One student said, "T\vo hundred point one zero
zero. Because the sum of 72 and 35 is about 1 00, and
then 148 and 51 is about 200." (Note: I have used
words for numerals where there is confusion about
how the students read the numbers.) Another said,
"One hundred fifty point four seven zero, because one
hundred forty-eight point seven two rounds to one
hundred point seven and fifty-one point three five one
rounds to fifty point four zero zero. Add those."
Fewer than half the students gave 200 as an estimate
of this sum. The others saw a decimal number as two
numbers separated by a point and considcrcd round
ing rules to be inflexible.
Question 2
789 x 0.52 is approximately how much?
One response was "789. I rounded point five two
up to 1 and multiplied." A second student said, "Zero.
This (789) is a whole number, and this (0.52) is not.
It (0.52) is a number, but it is very small. You round
789 to 800, times zero is zero." Only 19 percent of the
students rounded 0.52 to 0.5 or 1/2 or 50 percent.
Several of them said that answering this question
without paper and pencil \vas impossible and refused
to continue. The majority of students had little of
the of a decimal fraction and applied standard
rounding rules that were inappropriate for this
estimation.
Others who have studied elementary school chil
dren's understanding of decimal numbers have found
that when students are confronted with decimal num
bers, many are confused about what the symbols
mean. In a study of fourth through seventh graders by
Sackur-Grisvard and Leonard (1985), children
devised two "rules" to help them compare decimal
numbers (p. 161). These rul es worked just often
enough that students did not recognize that they were
in error. (I suspect that many teachers will recognize
them.)
Rule I: Select as smaller the number whose decimal
portion, as a whole number, is the smaller (e.g., 12.4 is
smaller than 12. because 4 is smaller than 17.
Rule 2: Select as smaller the number whose decimal
portion has more 12.94 and 12.24 are small
er than I because each have two digits and 12.7
has only one).
The first rule has its origins in the separation of a dec
imal number into two numbers; that is, children treat
the portions separately, and in this case treat the por
tion to the right of the decimal point as though it is
itself a whole number separate fl.-om the number to the
left of the decimal point. The second rule is slightly
more sophisticated; it is based on thinking that
tenths are larger than hundredths.
More recent research on decimal-number under
standing confinns that many students have a weak
understanding of decimal numbers. For a summary of
this work, see Hiebert (1992). The children in these
studies were primarily from classes where the intro
duction to decimal numbers was brief so that suffi
cient time would remain for the more difficult work
Place Value as the Key to Teaching Decima! Operations 113
.. --
of leaming the algorithms for operating on decimal
numbers. But spent on developing students'
understanding of the decimal notation is not time
wasted. Teachers with whom I have worked claim
that much less instructional time is needed later for
operating on decimal numbers if students first under
stand decimal notation and its roots in the decimal
place-value system we use. In the remainder of this
article I will discuss decimal notation and how we can
help students construct meaning for decimal number.
Giving Meaning to Decimal
Symbols
The system of decimal numbers is an extension of the
of whole numbers and, as such, contains the
set of whole numbers. For the of convenience,
this article refers to decimal numbers as those num
bers whose numerals contain a decimal point.
Decimal numbers, like whole numbers, are sym
bolized within a place-value system. Place-value
instruction is traditionally limited to the placement of
Thus, children are taught that the 7 in 7200 is
in the thousands place, the 2 is in the hundreds place,
a 0 is in the tens place, and a 0 is in the ones place.
But when asked how many $100 bills could be
obtained from a bank account with $7200 in it, or
how many boxes of 10 golf balls could be packed into
a container holding 7200 bails, children almost
always do long division, dividing by 100 or 10. They
do not read the numbers as 7200 ones or 720 tens, or
hundreds, and certainly not as 7.2 thousands. But
why not? These names all stand for the same number,
and the ability to rename in this way provides a
deal of flexibility and insight when working with the
number. (Interestingly, we later expect students to
understand such newspaper figures as $3.2 billion.)
Therefore, before we begin instruction on decimal
numbers, we need to provide more instruction on
place value as it is used for whole numbers, by asking
such questions as the bank question and the golf-ball
question, and we need to practice reading numbers in
different ways. Problems that require working with
powers and multiples of 10, both mentally and on
paper, give students a flexibility useful with whole
nmnbers. and this flexibility makes it to extend
instruction to decimal numbers.
The naming of decimal numbers needs special
attention. place-value name for 0.642 is six hun
dred forty-two thousandths. Compare this form with
642, where we simply say six hundred forty-two, not
642 ones. This source of confusion is compounded by
the use of the dths (thousandths, hundredths) or nths
(tenths) with decimal numbers and the use of d (thou
sand, hundred) or n (ten) with whole numbers. The
additional digits in the whole number with a similar
name is another source of confusion. Whereas 0.642
is read 642 thousandths, 642 000 is read 642 thou
sand, meaning 642 thousand ones.
In a number containing a decimal point, the units
place, not the decimal point is the focal point of the
number, as shown in figure 1. The decimal point
identifies where the units, or ones, place is located; it
is the first place to the left of the decimal point. The
decimal point also tells us that to the right the unit one
is broken up into tenths, hundredths, thousandths, and
so on. So really, 0.642 is 642 thousandths of 1. Put
another way, 0.6 is six-tenths of 1, whereas 6 is 6 ones
and 60 is 6 tens, or 60 ones. But just as 0.6 is six
tenths of 1, 6 is six-tenths of 10, 60 is six-tenths of
100, and so on. These relationships can be more
clearly seen in the base-ten-blocks representations
shown in figure 2.
Similarly, starting with the smaller numbers, 0.006
is six-tenths of 0.0 1, whereas 0.06 is six-tenths of 0.1.
1 2 3 4. 5 6 7
thousands thousandths
hundreds hundredths
tens tenths
ones
Fig. I, Ones as the focal point of the decimal system
114 .R.esearch into Practice in the Elementary Grades' from Journals of the NCTM
Moving in the opposite direction, 6000 is 60 hun
dreds, 600 is 60 tens, 60 is 60 ones, 6 is 60 tenths, 0.6
is 60 hundredths, 0.06 is 60 thousandths, and so on.
Although this convention might seem confusing at
first, it becomes less so with practice. These issues
are discussed more fully in Sowder (1995).
Students who try to make sense of mathematics
must become very confused when they are told to
"add zeros so the numbers are the same size" when
comparing numbers such as 0.45 and 0.6. This strate
gy does not develop any sense of number size for dec
imal numbers. Instead of annexing zeros, it would be
more appropriate to expect students to recognize that
another name for 6 tenths is 60 hundreths, which is
more than 45 hundredths, or that 45 hundredths has
only 4 tenths and what is left is less than another
tenth, so it must be less than 6 tenths. Students do
come to think this way when comparing decimal
numbers if they have had sufficient opportunities to
explore and think about place value, using manipula
tives as representations for numbers. Heibert (1992)
discusses research showing that if students do not
have a sound understanding of place value when they
learn to add and subtract decimal numbers, they make
many errors that are very difficult to overcome
because they are reluctant to relearn how to operate
on decimal numbers in a meaningful way.
An Instructional Unit on Decimal
Numbers
The unit summarized here was developed for a
research study (Markov its and Sowder 1994) and
resulted in students' performing much better on later
decimal topics in their textbook. This unit has also
been used by teachers who asked me for a way to
teach decimals meaningfully. These teachers later
Number 60 0.6 6
Base-ten-number
six tens six ones six tenths
name
Base-ten-block
representation
Alternative
base-ten
representation
six-tenths
Alternative six-tenths six-tenths
of 1
base-ien name of 100 of 10
Fig. 2. Alternative number names and representations when a long represents one unit
Place Value as the Key to Teaching Decimal Operations 115
told me that they thought the students who completed
this instructional unit had a much better grasp of dec
imal numbers than did their students in previous
years.
The first lessons focus on gaining familiarity with
base-ten materials. which can be ordered from most
catalogs of mathematical aids. The materials consist
of individual centimeter cubes, long blocks that are
marked to look as though ten cubes have been glued
in a row; 11at blocks that are marked to look as though
ten longs have been glued into a ten-by-ten block, and
large blocks that are marked to look as though ten 11at
blocks have been glued to form a ten-by-ten-by-ten
cube. Note that we do not call the smallest bloek a
unit as is commonly done, because in this instruction
we change the naming ofthe unit so that other blocks
can represent one, that is, one unit.
Students must play with the blocks and leam rela
tionships to answer such questions as the following:
How many longs are in a 11at?
How many small blocks are in 3 longs?
Where do you think there will be more longs, in 3
flats or in 1 big block?
I have 6 longs and 3 small blocks. What do I have
to add in order to have a flat?
Which is bigger, that more wood, 1 block or
10 flats? Four flats or 48 longs?
In the next lesson we begin to use the blocks to rep
resent numbers. The small blocks are used to represent
the number 1. Students then are asked what numbers
are represented by various sets of blocks: two big
bloeks, three flats, and two little blocks; one flat and
two longs; and so on. They must also represent num
with blocks; for example, they show 404 with
blocks. Two-dimensional drawings can later be used
the blocks, and these drawings can be used on
assignments for problems like the one in figure 3.
Put <, >, or", in the circle to show how the two sets
on either side
<'0,
"1/
A
, ~ ! . . . . ?
GJ
(iJ
Altematively, students can be asked to show with
blocks the of99 and Ill, which ofthe numbers
204 and 258 is closer to 235, and so on. Students
should be asked questions that indicate the limitations
of bloek representations of numbers when the small
cube represents 1:
Can you represent 46,321 with the blocks you
have? Why or why not?
Can you represent 8 112 with the blocks you have?
Why or why not?
The next lessons should focus on changing the
unit. First let one long represent one whole, or one
unit Students can then be asked to represent 76.
(They would do so with seven flats and six longs.)
After many such questions, they can again be asked,
"Can you represent 8 112 with the blocks you have'?
Why or why not?" (Yes, with eight longs and five
small cubes.) It is then worthwhile to ask a few ques
tions-remaining in the whole-number system
where the l1at represents one unit.
It is then natural to decimal instruction. If
the flat represents one whole, then what does a long
represent? It is obviously than 1. What pmi of 1
is it? Since ten longs are in a flat, one long represents
0.1. Several questions should follow:
How would you represent 0.3? 4.3? (See fig. 4.)
How many tenths are in four wholes?
What do you have to add to 0.9 to have one whole?
is _ ones and _ tenths, or tenths.
Which of the following are equivalent to one flat
and four longs: 14? 1.4'1 140? 14 longs? 41 longs?
41?
Likewise, children can come to understand that a
small block in this context represents one-hundredth,
3. Substituting a two-dimensional for bloeks 4. Representing lllimbers with base-ten bloeks
Putting Research into Practice in the Elementary Grades.' Readinas from Journals of the NUM 116
-
and many questions similar to the previous questions
can be asked. Teachers can also present such prob
lems as the following:
In 6.40 are tens and ones and tenths and
hundredths.
In 6.4 are tens and ones and tenths and
hundredths.
In 6.04 are tens and ones and tenths and
hundredths.
Are any of these numbers the same? Why?
A great deal of practice is needed in each of the
lessons deseribed here; the questions indicated are
only a small sample. Try using the big block as the
unit and through all the exercises again, this
time introducing one-thousandth. Ask students to
describe how they could cut up the blocks to repre
sent one ten-thousandth and one hundred-thousandth.
When students feel very secure with the blocks, with
changing units, and with problems involving deci
mals, it is time to switch to another representation. A
day or 1:\'10 spent with money-dollars and cents
\'lill work well. Finally, a lesson or two should focus
directly on decimal numbers without using another
representation (although many children will naturally
answer in terms of "blocks" or "wood"). Questions
like the following can be asked:
Is 0.1 closer to 0 or to I?
Is 1.72 closer to I or to 27
I am a number. I am bigger than 0.5 and smaller
than 0.6. Who am I?
Are there decimals between 0.3 and 0.4? How
many do you think
Are there decimals between 0.35 and 0.36? How
many?
Are there decimal numbers between 0.357 and
0.358? How many?
Draw baskets and label them "Numbers smaller
than 0.5," "Numbers bigger than 0.5 but smaller than
1," "Numbers between 1 and 3," and "Numbers big
ger than 3." Then give the students the following
numbers and ask them to place each number in the
appropriate basket: 0, 0.03, 1.01, 5.03, 2.63, 0.49,
0.93, 0.60, 1.19, and so on. This type of problem can
be made more difficult with baskets labeled
"Numbers between 0.4 and 0.5," "Numbers between
0.7 and 0.8," and "All other numbers."
If desired, these could be interrupted
before decimal numbers are introduced, and addition
and subtraction of whole numbers could be intro
duced using the blocks. But when addition and sub
traction of decimal numbers are introduced, lessons
with the blocks should come first.
The two questions at the beginning of this article
are trivial for students who have had this instruction.
The students see the part of the number to the right of
the decimal point as a natural extension of the place
value system, and they treat the entire number as one
quantity. Such students also develop a good feeling
for the sizes of decimal numbers and can compare
them with one another. It did not occur to any of the
students who received this instruction to round 0.52
to 0 or to 1 when estimating a product-0.52 was
simply seen as "about a half."
When students understand what they are doing,
they tend to enjoy doing mathematics. It is worth the
time needed to build strong foundations. The time
will be easily made up in future lessons, and students
are much more likely to be successful.
Action Research Ideas
1. For each of the following pairs of decimal num
bers, ask students to tell which is smaller. Then
analyze their answers to see if any are making the
rule-lor rule-2 errors identified in the Sacker
Grisvard and Leonard (1985) study.
Number Pair Use of Rule 1 Use of Rule 2 Correct
3.l7 or 3.4 3.4
3.l7 3.17
14.285 or 14.19 14.19
14.285 14.19
6.43 or 6.721 6.43
6.721 6.43
11.01 or J 1.002 11.01
11.002 11.002
9.642 or 9.99 9.99
9.642 9.642
15.134 or 15.12 15.12
15.134 15.12
156.101' 156.012 156.l
156.012 156.012
If you find evidence of systematic rule-lor rule-2
errors, use some of the instructional ideas in this arti
cle and then reassess your students to determine
whether their understanding of place value has
improved. In addition to rule-lor mle-2 errors, look
Place Value as the to Oecimal Operations 117
for other systematic errors that students make. What
are the misconceptions that underlie these errors?
2. Assess your students' understanding of place value
by asking such questions as the following.
(aj The Sweet Candy Company places 10 pecan
clusters in each box they sell. The cook just
made 262 pecan clusters. How many boxes can
be filled with the fresh pecan clusters?
(bj There is $2148 in the bank, ready to be used for
prizes for the state science fair. If each prize is
$100, how many prizes can be given?
Students with good place-value understanding '.vill
not need to do any division. Some students may solve
(aj and (bj by using division. Some may not solve
them at all. In either case try numbers like 260 or
$2100 to see if easier numbers allow them to use their
more limited place-value knowledge. If you find
some students making large numbers of errors, use
some of the instructional ideas in this article. Then
reassess them using similar questions to detemline
whether their knowledge of place value has
improved.
References
Hiebert James. "Mathematical, Cognitive, and
Instructional Analyses of Decimal Fractions." In
Analvsis of Arithmetic Mathematics Teaching. edit
ed by Gaea Leinhardt, Ralph Putnam, and Rosemary A.
Hattrup, pp. 283-322. Hillsdale, N. J.: Lawrence
Erlbaum Associates. 1992.
Markovits, Zvia, and Judith T. Sowder. "Developing
Number Sense: An Intervention Study in Grade 7."
Journal for Research in Mathematics Education 25
(January 1994): 4-29.
Sackur-Grisvard, Catherine, and Fran<;ois Leonard.
"lntennediate Cognitive Organizations in the Process of
Learning a Mathematical Concept: The Order of
Positive Decimal Numbers." Cognition and Instruction
2 (1985): 157-74.
Sowder, Judith T. "Instructing for Rational Number
Sense." In Providing a Foundation Teaching
}vIathematics in the Middle Grades, edited by Judith 1.
Sowder and Bonnie P. Schappelle, pp, 15-29. Albany,
N. Y.: SlJNY Press, 1995,
Threadgill-Sowder, J. "Computational Estimation
Procedures of School Children." Journal olEducational
Research 77 (July-August 1984): 332-36,
,

Research into Practice in the Elementary Grades: Readings from Journals of the ,veTN!
III