The I.T.A. (Initial Teaching Alphabet) Reading Experiment
Author(s): John Downing
Source: The Reading Teacher, Vol. 18, No. 2 (Nov., 1964), pp. 105-110
Published by: Wiley on behalf of the International Reading Association
Stable URL: http://www.jstor.org/stable/20197865
Accessed: 20-04-2015 11:15 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact [email protected].
Wiley and International Reading Association are collaborating with JSTOR to digitize, preserve and extend access to The
Reading Teacher.
http://www.jstor.org
This content downloaded from 192.38.33.14 on Mon, 20 Apr 2015 11:15:06 UTC
All use subject to JSTOR Terms and Conditions
?.toa. (Initial
The
Alphabet)
Teaching
Reading
Experiment
by John
Tn
Downing
1961
September
four hundred
to
in England
children
began
a new alphabet
to read with
to give young chil
specially designed
learn
dren confidence
greater
certainty
through a
in the be
in reading
of success
stage.
ginning
Sir James Pitman
(4) is the in
ventor of the Initial Teaching
Alpha
bet (earlier known as the Augmented
Roman
Alphabet).
young
beginners
simple and more
His
idea is that
should use the more
reliable
tial Teaching Alphabet)
have
become
confident
i.t.a.
(Ini
until they
and fluent
in
reading
printed in it, and that
then
should
transfer their skill
they
to reading
confidence
and
books
books
in the
traditional
printed
alphabet
and spelling of English. The charac
ters of i.t.a. and its rules of spelling
to
have been very carefully designed
make
it easy for children to transfer
from i.t.a. to standard print. A com
plete description of i.t.a. and its spell
ing rules has been provided previous
*y(i,2).
How
i.t.a.
Helps
In the research
Reading
Research
ning reading books are being held
constant in the two groups of classes
to control
(this includes an attempt
the "Hawthorne
If the
Effect").
attainments
of
the
two
reading
differ widely,
groups
therefore, we
trace
cause
to
the
differences
in
may
the alphabet and spelling.
The progress of the two groups has
been very different indeed. After only
five months
the four- and five-year
old
who were
beginners
using i.t.a
were
materials
in the
significantly
increased
lead, and their superiority
as the months went
by. By the end
of the first school year the average
i.t.a. child was at Primer
2 of the
the average
reading program, while
t.o. pupil was still at Primer 1. Before
the middle
of the second year the
i.t.a.
child had moved
to
average
Primer 4, while
the t.o. boy or girl
had got to Primer 2. After two years
the position was beyond
Primer
5
for the average pupil
(Grade 2-ii)
in the i.t.a. classes as compared with
Primer
3 in the classes using
t.o.
1 gives more
Table
details of the
progress
conducted
Unit
the
by
of the Uni
Institute of Educa
versity of London
attainments
of children using
tion,
i.t.a. are being
the
compared with
achievements
?f pupils learning with
traditional orthography
(to.). As far
as possible all factors other than the
and spelling
in the begin
alphabet
The
Load
of
the group.
Traditional
Beginner's
of Learning
Traditionally
printed English over
loads the beginner
in three ways :
1. Too many characters. The t.o.
has to learn two or more
beginner
different characters for each letter of
as well as b, G and
the alphabet?B
105
This content downloaded from 192.38.33.14 on Mon, 20 Apr 2015 11:15:06 UTC
All use subject to JSTOR Terms and Conditions
106
THE
Distribution
Frequency
One Year,
Eighteen
Percentage,
Reader
Primer
Reached
Book 5
Beyond
At Book 5
4
3
2
1
At Introductory
Book
Nonstarter
18Mo.
10.4
4.35
9.95
16.95
20.6
24.45
8.7
4.6
35.85
7.75
13.1
16.2
11.6
11.4
3.65
0.5
413
413
of Children
Number
Months,
Experimental
Primer
Median
Position
4
the pupil using
g, etc. In contrast,
has only one single
i.t.a. materials
form for each letter of the
printed
are needed,
capitals
alphabet. Where
i.t.a. simply uses a larger version of
same
the
2.
Too
case
lower
sentations.
In
repre
there
i.t.a.
is
never
the
than one way of printing
same word. For dog the i.t.a. pupil
has to learn one visual pattern only
instead of the five or more different
more
sets
of
print,
dog
characters
conventional
in
for example:
dog
Dog
DOG.
3. Too many phonic print-signals.
In our traditional alphabet and spell
ing there is a wide variety of ways of
in print the restricted num
signalling
ber of basic sound units of English.
in t.o. there are at least
For example,
of spelling
different
ways
eighteen
common
to
such
words as
the sound
zoo, shoe, grew, through, do, blue.
In i.t.a. this variety is reduced to one
single symbol
for example:
for
the
single
sound,
Control
Group
2Yr.
1 Yr.
18Mo.
2Yr.
53.6
7.85
14.4
9.8
9.8
3.6
0.3
0.65
0.7
1.2
1.7
6.4
13.0
45.4
23.9
7.7
2.6
2.2
7.85
14.7
27.35
33.35
10.5
1.45
15.6
4.9
15.25
22.45
17.85
19.7
3.45
0.8
306
687
610
687
Beyond
Book 5
zea,
jha,
tfirco), dea,
groa,
the
Thus
beginner
down by a heavy
too many
too
characters,
bloa.
t.o.
with
weighed
redundancies:
is
burden
of
unnecessary
unnecessary
many
too many
For
print-signals.
i.t.a. the load is
visual patterns,
unnecessary
phonic
the beginner with
for he has only
lightened,
greatly
one form to learn for each
letter,
one
each
for
visual
pattern
only
in
and
word
and sentence,
whole
cases only one letter to learn
most
for
Dog
at
Reached
Percentage,
Group
whole-word
shape.
whole-word
many
Primer
Years
Reading
and Two
Showing
1 Yr.
TEACHER
1
TABLE
Percentage
READING
each
of
sound units
or
the forty
of English.
so basic
and More
i.t.a.?Simpler
Reliable
for the English
Unfortunately
speaking child, his traditional printed
and
code is extraordinarily
complex
The
inconsistent.
i.t.a.
code,
how
ever, has:
1. Consistent
In t.o.
let
ters cannot
to keep
the
same
sound
spelling.
be relied upon
for
value,
the
example,
use of the letter o in do, go, women,
gone,
more
one.
In
consistent
contrast,
is much
i.t.a.
as a code
for spoken
This content downloaded from 192.38.33.14 on Mon, 20 Apr 2015 11:15:06 UTC
All use subject to JSTOR Terms and Conditions
NOVEMBER
107
1964
English. Each of the different
in these words
dco),
, wimen,
g
gon,
the letters ch are used in t.o. to
sound in chat
represent a different
sounds
pie,
wun
from either of the sounds indicated
c and A
cat and hat.
by
separately in
printed
by a different
code-breaker
the young
would anticipate. He, therefore, finds
that he can rely on the code, and is
not led to doubt his rational approach
is signalled
symbol, as
to such problems.
2. Consistency
direction.
English we
In
of
read
traditionally
printed
words
from left to right, but within
many words the letters are not to be
left to right
to left to read
right
the final
sound
three.
signalled by letter number
to
In i.t.a. the left
right rule of
reading is never broken, for example :
maed
3. Reduced
complexity
of phonic
The
traditional
symbols.
alphabet
does not have enough letters to pro
vide one letter for each of the forty
or so sound units of English;
instead,
letters have to be used ayer again in
a variety of combinations.
For exam
for
the
Schonell
Test,
Fifth
First
Group
N
(fiat
i.t.a. a more
attempt to make
reliable
code
for young
beginners
seems to have produced
dramatic re
sults in our research.
The
pupils
to
read
and
write
with i.t.a.
learning
demonstrated
great superiority
on
in word building.
For instance,
the Schonell
word
read
(6) graded
ing test at the end of the first year
the average
i.t.a. learner could read
19 test words or more
on the i.t.a.
version
of the test, whereas
the aver
t.o. pupil
conventional
using
print could read only 5 test words on
the same test in conventional
print.
At the beginning
of the fifth term
(six months
later) the average scores
were 37 test words read
correctly in
the i.t.a. group and 11 in the t.o.
2 gives more detailed
group. Table
results of this test.
age
This superiority of the i.t.a.
pupils
is not confined to phonic word-build
in com
ing. They are also advanced
and
Test
of
speed
First
Year
Second
N
sentence
in
accuracy
prehension,
Standard
Deviation
Mean
be
thus :
This
at End of the
Given
Entry
Term
After
Schonell
can
character
special
used for this sound,
reading,
TABLE 2
Results
a
i.t.a.
have
in the early
stage of learning.
primitive decoding
in the word made
the
For example,
first sound is signalled by the first
letter on the left, but the second
sound is signalled by letters two and
four and the child must reverse from
from
read
In
Ac
reading.
and
Schonell
in the
Test
Standard
Deviation
Mean
Experimental
345
18.8
20.7
345
37.1
27^0
Control
623
5.2
6.6
623
10.9
10.6
significantly
t test).
at
N.B.
significance
The
means
for both
between
Schonell
the
tests
two
differ
groups
student's
(applying
the
.0001
This content downloaded from 192.38.33.14 on Mon, 20 Apr 2015 11:15:06 UTC
All use subject to JSTOR Terms and Conditions
level
of
recent report
cording to Southgate's
their
creative
too, is
(7)
writing,
in
both
and
quality
quantity.
superior
Children
Standard
In
the
schools
to
Transfer
Readily
Print
using
transfer
each
i.t.a.
to reading
he
when
individual
print
our ex
ly is ready for this step. In
a very few children have
periments
from i.t.a. to stand
been transferred
ard print after only two or three
to reach
but most
months,
appear
the
child makes
traditional
the necessary
level of fluency in i.t.a.
the
second
year of schooling.
during
in
reading i.t.a. is desira
Fluency
be
transfer
is executed,
ble before
cause Sir James Pitman
in his design
in i.t.a.
of the alphabet has preserved
extent compatible
the greatest
with the purpose of easier teaching?
those same cues, generally situated in
the top part of the line of print,
which we use when we have achieved
fluent reading of the conventionally
of words do
printed page. A minority
in appear
drastically
change more
can guess
these
ance, but children
and comprehension
than
accuracy
children who have been learning with
t.o. from the beginning.
For instance,
on the Neale
of Reading
Analysis
at
after
(4),
eighteen months
Ability
school, the average i.t.a. pupil scored
23 for accuracy
and 8 for compre
a high level of
has
in
i.t.a.
been achieved,
fluency
the
the pupil will have developed
minimal
the
skills
of
necessary
using
context.
cues and
a
ensure
Once
clues which
contextual
smooth
achieved
i.t.a. pupils
very
on
scores
tests
in
the
printed
superior
and
traditional
spelling.
alphabet
i.t.a.
The
children who began with
to t.o. have
later transferred
and
read
the
latter with
much
greater
as
com
as compared with
19 words
for the average child in
t.o. had been
in which
from the start. See Tables
3,
minute
per minute
the classes
taught
4, and 5 for details
of these results.
TABLE 3
Reading
Neale
by the
as Measured
of Reading
Ability
in t.o.)
Tested
Groups
Accuracy
Analysis
(Both
Group
N
Mean
Standard
Deviation
Experimental
146
22.9
17.4
Control
190
9.3
11.9
N.B.
Significant
two
at
groups
Kolmogorov-Smirnov
the
between
difference
cent
.1 per
level
Test
(D=21.46).
using
TABLE 4
by
as Measured
Reading
Comprehension
of Reading
the Neale
Ability
Analysis
in t.o.)
Tested
(Both Groups
Group
will
the
t.o.,
reading
pared with scores of 9 and 4 respec
tively for the child who had been on
t.o. from the outset. The
i.t.a. pupil
reads an average of 34 t.o. words per
N
Mean
Standard
Deviation
Experimental
146
7.9
6.1
Control
190
3.5
4.4
transfer.
after beginning
months
Eighteen
to learn to read with the new alpha
bet
when
hension
?to
from
TEACHER
READING
108_THE
N.B.
Significant
two
at
groups
Kolmogorov-Smirnov
between
difference
cent
level
.1 per
Test
(D=21.46)
the
using
.
complete report on the first two
years of the i.t.a. reading experiment
in Britain has just been published?
we have subse
Downing
(3)?but
A
This content downloaded from 192.38.33.14 on Mon, 20 Apr 2015 11:15:06 UTC
All use subject to JSTOR Terms and Conditions
NOVEMBER
109
1964
TABLE 5
by the
as Measured
of Reading
Ability
in t.o.)
Tested
Groups
Rate
Reading
Neale
Analysis
(Both
Experimental
146
33.?
Control
190
18.7
N.B.
Difference
not
significant
Test
(D=17.14).
between
using
Standard
Deviation
Mean
N
Group
24.2
Spelling
Test
1961
(t.o.) , September
Schonell's
Graded Word
Entrants,
A
Test
Spelling
groups
Kolmogorov-Smirnov
(reading age 7.4 years) for the
t.o.
started out with
children who
a conservative
estimate
of
Thus
24.1
for the average
effectiveness
saves
one
that
it
is
year in learn
pupil
to
t.o.
read
ing
i.t.a.'s
i.t.a.
TABLE 6
two
age 7.1 years).
year
(chronological
The average t.o. score for i.t.a. pupils
are still on
(this includes some who
was
34.4 (reading age
i.t.a. readers)
as compared with
8.4 years),
only
of
turned, but Table 6 shows the analy
sis of the data so far available.
18.9
the
quently analysed the results of giving
test in t.o. to both
the Schonell
course of their third
in
the
groups
Effect
Earlier this year a t.o. spelling test
was administered
to both i.t.a. and
t.o. groups. At the time of writing not
all the answer papers have been re
on Writing
and
Spelling
Over-all
Scores
Comparison
Group
t.o. Group
(N-318)
(N-602)
Mean
28.7
24.1
S.D.
16.4
N.B.
level.
i.t.a.
t=4.34,
14.0
at
significant
.1 per
cent
It is noteworthy
that by the mid
of their third year of schooling
the i.t.a. pupils are able to spell t.o.
better
than the
words
significantly
dle
have been reading and
t.o. only.
test was
the spelling
Although
t.o.
to
both
in
of chil
groups
given
children who
writing
with
dren, 49 (15 per cent) of the 318
on the first
two years
report
of the i.t.a. experiment
(3) stated
to be
that creative writing
appears
in
i.t.a.
much
classes, and
improved
some teachers claim that "the stand
The
ard of creative writing has improved
These
almost beyond
comparison."
claims are at present under objective
at the Reading
Re
investigation
to
i.t.a. children had not transferred
t.o. at the time of testing. It should
be pointed out that these spelling re
sults
are
and that the
incomplete
of the two
information
background
terms
in
of
sex,
groups
intelligence,
and social class has not yet been de
termined.
In
conclusion,
be exercised
to date,
findings
search Unit,
but already
they have
some support from Southgate,
who
observed
in her
has
independent
must
study of i.t.a. (7) that "Free writing
in the class appeared more spontane
ous, prolific and correctly spelt than
that
experiment
a fruitful
been
found.
is usual with
such young
children."
i.t.a.
caution
although
in respect of the
the
results
in Britain
line
of
of
the
indicate
inquiry
has
is the Reading
(John A. Downing
Research
the
Officer of
Reading Re
This content downloaded from 192.38.33.14 on Mon, 20 Apr 2015 11:15:06 UTC
All use subject to JSTOR Terms and Conditions
search
Unit,
Institute
University
of London
of Education.)
6.
References
7.
Journal
Experiment,"
109 (1961),
of Arts,
F. Psychology
Schonell,
Reading.
1949.
1. Downing,
Roman
Reading
336.
John
Alphabet
Teacher
A.
"The
Augmented
to Read,"
for Learning
16 (March
325
1963),
Initial
Downing,
J. A. The
Teaching
New
and Illustrated.
Explained
Alphabet
York: Macmillan,
1964.
Ex
3. Downing,
i.t.a. Reading
J. A. The
Evans
London:
1964.
Bros.,
periment.
4. Neale,
M. Neale
of Reading
Analysis
London:
1963.
Macmillan,
Ability.
to Read:
An
5. Pitman,
I. J. "Learning
2.
Federal
Support
TEACHER
READING
110_THE
Edinburgh:
of Royal
Society
149-180.
and Teaching
of
& Boyd,
Oliver
Roman
Al
V.
Southgate,
"Augmented
An Outsider's
Re
Experiment:
phabet
16
Educational
Review,
(1963),
port,"
32-41.
(This article was originally
uled for the October
issue
which
journal,
ing instruction
editor
regrets
delayed
for Reading
had
as
beginning
its theme.
that printing
sched
of
this
read
The
problems
publication.)
Instruction
A
in the United
States is marked
instruction
significant
point in reading
by the inclusion of reading in the extension of the National Defense Education
two kinds of help
IRA supported
the passage of this bill, under which
Act.
to improve the teaching of reading. Under Title
III, the
purchase of certain books and equipment will be made possible on a matching
basis through state educational
agencies. Title XI will provide for institutes
will
be available
for the improvement
of reading.
as
for the implementation
of the new law will be circulated
soon as possible by the U. S. Office
of Education,
and further information
will be published
in IRA journals as it is available.
Guidelines
This content downloaded from 192.38.33.14 on Mon, 20 Apr 2015 11:15:06 UTC
All use subject to JSTOR Terms and Conditions