National Society for the Study of Education. Commi.

Report of the Society's committee on silent reading online

. (page 1 of 17)
Online LibraryNational Society for the Study of Education. CommiReport of the Society's committee on silent reading → online text (page 1 of 17)
Font size
QR-code for this ebook


■llli



iiiiiiiii^




iiiifin



ii



Hi



iiiiiiiiiiliill

iiilii









N



The

Twentieth Yearbook



OF THE



NATIONAL SOCIETY FOR THE STUDY
OF EDUCATION



PART II .

REPORT OF THE SOCIETY'S COMMITTEE ON
SILENT READING



THIS YEARBOOK WILL BE DISCUSSED AT THE ATLANTIC

CITY MEETING OF THE NATIONAL SOCIETY,

SATURDAY, FEBRUARY 26, 1921, 8:00 P.M.



PUBLIC SCHOOL PUBLISHING COMPANY
BLOOMINGTON, ILLINOIS

1921



Agents

PUBLIC SCHOOL PUBLISHING COMPANY

Bloomington, Illinois

PUBLISHERS OF ALL OF THE YEARBOOKS OF THE SOCIETY



COPTEIGHT 1921 Bt

Guy Montrose Whipple
Secretary of the Society



All Rights Reserved



Published February, 1921
First Printing, 3,000 copies



Composed and Printed by

PxTBLic School Publishing Company

Bloomington, Illinois



THE

Twentieth Yearbook

OF THE

NATIONAL SOCIETY FOR THE STUDY
OF EDUCATION

PART II

REPORT OF THE SOCIETY'S COMMITTEE ON
SILENT READING



Prepared by the Committee from Material Submitted

BY

J. A. O'Bkien, Mat Aykes Bukgess, S. A. Courtis, C. E. Germane,

W. S. Gray, H. A. Greene, Eeginia E. Heller, J, H. Hoover,

J. L. Packer, D. Starch, W. W. Theisen, G. A. Yoakum,

AND

-"zassfstai
Representatives of the School Systems of Cedar Rapids,
Denver, Iowa City, and Racine



Edited by Guy Montrose Whipple



THIS YEARBOOK WILL BE DISCUSSED AT THE ATLANTIC CITY

MEETING OF THE NATIONAL SOCIETY, SATURDAY

FEBRUARY 26, 1921, 8:00 P. M.

PUBLIC SCHOOL PUBLISHING COMPANY

BLOOMINGTON, ILLINOIS

1921



OFFICERS OF THE SOCIETY

for the Atlantic City Meeting



President

Harry B. Wilson
Superintendent of Schools, Berkeley, California

Vice-President

David Felmley
Illinois State Normal University, Normal, Illinois

Secretary-Treasurer

Guy Montrose Whipple
University of Michigan, Ann Arbor, Michigan

Executive Committee

(The Year Indicates Date of Expiration of Term)

Ernest Horn (1921)
State University of Iowa, Iowa City, Iowa

Frederick James Kelly (1922)
University of Kansas, Lawrence, Kansas

Paul Whitfield Horn (1923)
Superintendent of Schools, Houston, Texas

Stephen S. Colvin (1924)
Brown University, Providence, Rhode Island

Board of Trustees

Daniel Starch (1921)
Harvard University, Cambridge, Massachusetts

W. W. Kemp (1922)
University of California, Berkeley, California

Frank W. Ballou (1923)
Superintendent of Schools, Washington, D. C.



TABLE OF CONTENTS



PAGE

Introduction vn

Ernest Horn, Chairman of the Committee

SECTION 1

CHAP.

I. Factors Affecting Results in Primary Reading .... 1 f^

W. W. Theisen, Director of the Division of Keference
and Eesearch, Cleveland Public Schools

II. Controlling Factors in the Measurement op Silent

Reading 25

May Ayres Burgess, Department of Education, Russell
Sage Foundation

III. Individual Difficulties in Silent Reading in the

Fourth, Fifth, and Sixth Grades 39

William S. Gray, The University of Chicago

IV. The Development of Speed in Silent Reading 54

John A. O'Brien

V, Motivated Drill Work in Third-Grade Silent

Reading 77

J. H. Hoover, Cape Girardeau, Missouri

VI. The Effect of a Single Reading 90

G. A. Yoakum, Director of Teacher Training, Nebraska
State Normal School, Kearney, Nebraska

VII. Outlining and Summarizing Compared w^ith Re-
reading AS Methods of Studying 103

Charles E. Germane, Department of Education, Des
Moines University

VIII. Measuring Comprehension of Content Material... 114
Harry A. Greene, State University of Iowa

IX. The Vocabularies of Ten First Readers 127 t-

J. L. Packer

X. The Contents of Readers 145 iX

Dandel Starch, Harvard University

SECTION 2

I. Exercises Developed at Detroit for Making Reading

Function 153

Reginia R. Heller and S. A. Courtis, Detroit Public
Schools

II. Silent Reading Exercises Developed at Denver,

Cedar Rapids, Racine, and Iowa City 162

Information Concerning the National Society for the Study
of Education 173



INTRODUCTION

As stated in the introduction to the EigJiteenth Yearbook,
Part II, of this Society, it was hoped by those who contributed
that the book would serve to stimulate further investigation in the
various fields of work which were reported. This hope has been so
well realized for the subject of reading, that at the Cleveland meet-
ing the Executive Committee of the Society decided that it was very
important to collect and publish the results of such studies as have
been completed since Dean Gray made his report two years ago.
This work was assigned to the committee whose names are signed to
this report. These members were asked to suggest others who had
something to contribute. Dr. Thorndike and Dean Haggerty found
it impossible to finish their manuscripts in time to have them in-
cluded in the Yearbook. Dr. O'Brien's study was submitted
through Dr. Buckingham, under whose direction the investigation
was made. Other contributors are indicated in the table of contents.

After considerable correspondence among members of the com-
mittee it seemed best to arrange the Yearbook in two sections, the
first part dealing with investigations which presented data bearing
on the problem of reading, the second part containing examples of
concrete exercises which have been actually tried in the classroom.
It was hoped to give a large proportion of the space to these class-
room exercises, but the difficulty of gathering and editing them was
so great that it has seemed necessary to include at this time only a
few samples of lessons which were submitted, and to suggest that
an entire Yearbook be later given to this work.

It is very essential that such studies as are described in the first
part of this report be made. Teachers are particularly ready just
now to undertake any new method which goes under the name of
' ' silent reading. ' ' No doubt, the teaching which results from this
interest will, in general, be superior to that which we have had in
the past. On the other hand, many mistakes will be made, some of
them perhaps quite serious. This is particularly likely to be true
in the case of certain types of speed exercises. In a way, it is un-

VII



VIII

fortunate that changes in methods cannot be delayed until we have
more assurance as to the efficiency of the methods which are being
recommended.

The problems which need investigation are almost without limit.
Most of them, however, may be grouped under five heads : first, the
thorough-going analysis of the various types of reading abilities
required in life outside the school ; second, the construction of a
course of study which would show the proper relation, on the one
hand, between oral and silent reading, and, on the other hand, be-
tween reading and literature; third, a study of the problems of
reading in the area where reading overlaps study ; fourth, the dis-
covery of exercises for the development of each of the major types
of reading abilities; fifth, an investigation of the diagnosis and
treatment of individual cases. These problems naturally overlap ;
each is a center of focus rather than an isolated problem.

In attacking any of these groups of problems it is important
to distinguish among four qualities and to study the relation
existing among them. These are speed (including skimming), com-
prehension, organization, and remembrance. There are, in addition
to these, certain technical skills, such as the use of reference ma-
terial in libraries, the use of encyclopedias, dictionaries, etc. There
are also the various abilities involved in the proper use of indexes
and tables of contents. Each of these abilities needs to be studied,
moreover, in relation to the various types of materials which are
commonly read and in relation to the various purposes for which
these types of material are read.

The exercises which are given in Section II represent but a
small sampling of a great number which were submitted. Since it
was impossible to print all of the really excellent lessons which
were reported, it seemed wise to include only exercises for the
first three grades. Even with these limitations there was space but
for a small part of the lessons which were submitted. The effort of
the Chairman has been to select lessons which represent a wide
range of types of exercises.

These lessons embody attempts to work out methods of teaching
under the guidance of the data which have been disclosed by such
investigations as were summarized by Dr. Gray in the Eighteenth



IX

Yearbook, Part II, and as are also reported in the first section of
this Yearbook. In this sense they represent experimental work.
They are for the most part in the stage where the technique of
practical method is being worked out. So far, little has been done
to isolate and test the effect of any one exercise, but we have evi-
dence that satisfactory results can be obtained from certain com-
binations of exercises. Studies like those of 'Brien and of Hoover
in this Yearbook, for instance, lead to just such conclusions. The
conclusions seem to be substantiated also by the superior scores made
on the standard tests by schools which have featured such work, and
by the rapid improvement which has resulted when such exercises
have been introduced.

It is the opinion of the Committee that this or some similarly
constituted committee should continue the study of these problems.
As rapidly as possible the efficiency of each type of exercise should
be scientifically determined. Investigation must of necessity be
slow. Meanwhile, an exhaustive search should be made for all types
of silent reading exercises which seem to give good results. These
may be subjected by the Committee to a critical examination made
in the light of present knowledge, and printed with explanatory
notes as a future Yearbook.

Committee on Silent Reading,
H. A. Brown,
B. R. Buckingham,
S. A. Courtis,
W. S. Gray,
M. E. Haggerty,

D. Starch,

E. L. Thorndike,
G. M. Whipple,
Ernest Horn, Chairman.



SECTION 1

CHAPTER I

FACTORS AFFECTING RESULTS IN PRIMARY READING



W. W. Theisen

Director of the Division of Reference and Research

Cleveland Public Schools



INTRODUCTION

Reading instruction may be said to be in a process of transition
as regards aims, methods and content. If one examines the older
literature, one frequently encounters such terms as "expression,"
"enunciation," "articulation," "pitch," "inflection," and "em-
phasis," while today "silent reading," "thought," "content
value, " " rate, ' ' and ' ' individual differences ' ' are terms which chal-
lenge the attention of the student. It is not difficult to locate the
cause. The development of a more scientific attitude toward edu-
cation has tended to make educators more critical. Studies of
failures (16)\ the development of standard educational and intelli-
gence tests and the use of methods of classroom experimentation
have served to point out some of the shortcomings of the old sys-
tem and to indicate some of the possibilities of the new. Even so,
there has been far too little actual testing of results and too little
experimentation to discover possible achievements. When we con-
sider the time that is ordinarily given to primary reading and the
bearing of reading achievement upon the future success of the
child, it is important that we bring together such evidence as we
have, concerning the effect of various factors in producing results.
This article is offered as a brief summary of experimental evidences
and current thought concerning factors affecting results in primary
reading.

FACTORS

1. Attendance. For a group of children just learning to read,
attendance is a factor that may be important. It is a matter of



^ Numbers in parentheses refer to the list of references at the end of this
chapter.

1



2 THE TWENTIETH YEARBOOK

observation that normal children sometimes fail to acquire satis-
factory proficiency in primary reading, because of excessive absence.
However, absences that aggregate less than two months in the course
of a year probably have little effect upon attainment in primary
reading as compared with such other factors as intelligence, quality
of teaching and amount of reading done. The writer found the
correlation between attendance and attainment on the Haggerty
Achievement Reading Tests to be negligible for the children who
had had kindergarten training, and who had attended school for
130 days or more at the time of the test in May/ The Peai*son
correlation between attendance and score for 210 first-grade chil-
dren, selected at random, was .017 for Test I, — .003 for Test II, and
.008 for the two tests combined. For 190 second-grade children,
similarly selected, the corresponding figure in the case of Test I
was .073 (31).

Long periods of interruption would probably interfere materi-
ally with progress. Packer and Anderson found in the case of rate
of reading that summer vacation lowered it materially. Children
in 1 B, who read 50 words per minute in May, read but 44 in Sep-
tember. Corresponding figures were : for 1 A, 84 and 49 ; for 2 B,
125 and 68; and for 2 A, 145 and 125, respectively (21).

2. Time Devoted to Reading. At first thought, many persons
would probably say that results in primary reading vary directly
with the time given to reading on the daily program. If all other
factors were equal, this would probably be true within limits.
Under present ways of teaching, other factors apparently over-
shadow it. The Pearson correlations obtained in our own study
between total time (including recitation, study and phonics) and
score on the Haggerty tests for 200 first-grade children, selected at
random from a group of nearly 600, were slightly negative, being
—.035 for Test I, —.153 for Test II, and —.102 for both tests com-



'^ Figures recently given us by Miss Engel, of the Psychological Clinic at
Detroit, indicate that a percentage of absence greater than 15, or a lesser
amount of continuous absence, is causally related in a definite way to first-
grade failure. The operation of this factor might not be revealed by tie method
of correlation, but it is revealed when differences in intelligence are first allowed
for and the pupils then classified into those who are promoted and those who are
not promoted. — Editor.



FACTOBS AFFECTING RESULTS IN PRIMARY READING 3

bined (31). The Spearman correlation obtained by Woody for
Grades III, IV, and V between the number of minutes per week de-
voted to reading recitation (actual reading, phonics and word drills,
exclusive of study) and the scores attained in the Monroe reading
test were those of Table 1.

TABLE 1



COBBKLATIONS BbTWBBK TIMB DeVOTED TO READING AND SCOEKS

Test (Woody)


IN MONBOB RBADINQ


Grade


Number of
Teachers
Reporting


Correlation
between time
and comprehen-
sion scores


Correlation

between time

and rate

scores


ni

IV
V


51
52
60


.13

.06
.06


.17
.29
.06



This lack of correlation, he concludes, seems to indicate that
other factors are more influential in determining the score attained
than the time element (33). While these two studies are by no
means conclusive, they indicate how completely other factors may
submerge the time factor. It is entirely possible, e. g., that many
children may gain more from ten minutes spent in independent
silent reading than they would in a twenty-minute oral-reading
recitation period of the conventional type. Similarly, a class pos-
sessing a high average intelligence may make greater progress in
half the time taken by one of low intelligence. Again, one teacher
will use the entire period profitably while another squanders two-
thirds of it. The teacher should endeavor constantly to have each
child spend his reading time in ways that will be the most profitable
to him.

3. Kindergarten Training. Does kindergarten training influ-
ence the character of the reading work done by children in the
primary grades? No comprehensive studies of this problem have
been made. Our own results with the Haggerty tests in the first
grade showed that the group that had attended kindergarten ex-
ceeded the group that had not, even though their median ages were
the same. The median score for the kindergarten group was 7.5 in
Test I, and 4.2 in Test II, and for the non-kindergarten group 6.0
and 3.6, respectively. In the second grade, the kindergarten group



4 TEE TWENTIETH YEABBOOK

scored 13.8 in Test I, and 9.6 in Test II, while the non-kindergarten
group scored 12.6 and 7.4, respectively. In the third grade the
non-kindergarten group excelled. The number of pupils, however,
was less in this grade, as the figures in Table 2 will show. The
average difference between the kindergarten and non-kindergarten
groups is roughly equal to one-fourth of the interval between the
first and second grades (31).



TABLE 2



Mhdian Soobes ok


THE Haggeety Achievement Reading Test for Kindeboabtjin

AND NON-KlNDEEOARTBN GROUPS




Grade I


Grade II


Grade III




Test I


Test II


Test I


Test II


Test I


Test II


Kindergarten

Group
Non -kindergarten

Group


7.5 (526)
6.0 (92)


4.2 (516)
3.6 (91)


13.8 (378)
12.6 (77)


9.6 (378)

7.4 (70)


17.8 (106)

18.9 (70)


15.6 (106)
16.8 (59)



Figures in parenthesis indicate number of children.

4. Intelligence and Mental Age. Dickson had children of the
low first grade segregated on the basis of intelligence quotient and
mental age. After an experiment covering a year and a half, he
concluded that "mental age and I. Q. are important factors in re-
vealing a child 's chances for success in his school work. ' ' Children
who tested low were very slow to learn to read. They had little
initiative. What they appeared to learn one day was not retained
to the next. Much repetition was necessary. Their reading was
marked by a tendency to name words without thought of their
meaning. In a group of "borderzone" children (I. Q. 85 or below)
only 6 could read in an easy primer after nearly a year and a half
of effort under a strong primary teacher. Of 42 pupils who tested
normal or above, "all but five passed the work of the first grade
at the end of the term. The teacher attributed the failure of these
to irregular attendance, and of one to excessive timidity. ' '

In another school, thirty of the pupils, who tested below six
years mentally, and who classified in the "dull normal" group, or
below, were placed in a special first-grade division. Ten were re-
peating the work of the grade. At the end of two terms, under the
experiment, two pupils out of the thirty were promoted into the
* ' high first ' ' grade, regular class. Near the end of the third term.



FACTORS AFFECTING RESULTS IN PRIMARY READING 5

five more were ready (7). While this is but one experiment, the
results are exceedingly significant for primary education. Hag-
gerty has found that there is a significant correlation between
intelligence and ability to perform the exercises of his primary
reading tests (12) . He correlated teachers' estimates of intelligence
when weighted according to grade location, with scores on his read-
ing tests. In the case of 200 pupils in Grades I to III the Pearson
coefficients were .71 for Test I, .69 for Test II, and .76 for the two
tests combined. Similar figures for 144 eight-year-old pupils were
.67, .67, and .71, respectively. Using the scores obtained with his
intelligence test, the figures were .65, .67, and .70 for the same
group. For 200 pupils of Grades I and II the intelligence and
reading test yielded a correlation of .84. Terman reports five third
grades tested by Dickson in the results shown in Table 3.



TABLE 3
Distribution of Intelligence in Five Thied-Gbadb Rooms (After Dickson)









Percent


Percent


Boom


Median

Mental

Age


Median
I. Q.


below
5.5 men-


above
7 mental






tal Age


Age


A


6-0


87


31


10


B


5-7


76


46


5





6-0


85


20


20


D


7-2


108


14


60


E


7-8


112





71



He points out the average mental age of Room E was fully two
years above that in Room B, and the median I. Q. 36 points higher.
' ' One third of the pupils in Room A, and half of those in Room B
were incapable of doing standard first-grade work." The lack of
progress in Room B was so evident that the teacher was in despair
and the superintendent doubted her efficiency (28).

Of all the factors which make for progress in primary reading,
intelligence is probably the most significant. It has not, however,
been sufficiently recognized in dealing with children,

5. Chronological Age. Chronological age at any stage of school
life is less indicative of probable success than mental age, or mental
maturity. The younger children of a grade on the whole excel the
older in reading (e. g., see 9, 30). Their superior inteUigenee
enables them to do so. The six- and seven-year-old first-grade pupils



6 TEE TWENTIETH YEABBOOK

in St. Louis read better orally than the eight- and nine-year-olds.
Similarly, in the second grade the seven- and eight-year-olds aver-
aged better than the nine- and ten-year-olds. The average rate of
silent reading in the second grade was found to decrease with
age (9).

It is not surprising that our results showed no correlation be-
tween score on the Haggerty tests and age. For 210 first-grade
pupils, and 190 second-grade pupils, selected at random, the corre-
lations between age and score were zero, except one correlation of
0.13 between Test I and age in the second grade. These figures
are too small to be of significance. If age were one of the strong
factors in producing results in primary reading, we should expect
to find a decided positive correlation between it and reading per-
formance. Such a condition would mean, in general, that the older
the child in a given grade, the better his reading. This, however,
is not the case. Only in the event that we selected children of about
the same intelligence quotient should we expect to find a positive
correlation between chronological age and reading achievement.

6. Nationality and Home Influence. Foreign language spoken
in the home is a distinct handicap to a child 's reading development.
This agrees with common observation. It is difficult, however, to
determine the influence of nationality. Intelligence, rather than
nationality per se, probably accounts for a large part of the differ-
ences in attainments of different language groups. The Rochester
studies tend to show that Hebrew children can be expected to read
better than foreign children of other nationalities, and that Italian
children do not read as well. 'Hern concluded from the Rochester
studies that foreign children made relatively better showings in
oral than in silent reading. But even in oral reading tests "the
children seemed to labor under a distinct language handicap. This

TABLE 4

Obal Reading Scoees in the Gray Test by Pbedominating Nationality
(Afteb Jxtdd)



Cleveland Polish and

Grade Average American Italian Bohemian Jewish



I


31


37


21


21


82


II


42


44


25


40


48


ni


46


47


28


44


50



FACTOBS AFFECTING RESULTS IN PRIMARY BEADING 7

was decidedly noticeable in the case of Italian children. The differ-
ences due to nationality were more marked in the silent reading
scores than in oral reading scores" (20). In Cleveland, as Table 4
shows, the oral reading of Jewish children was distinctly above
that of the average, and that of Italian children much below. Poles
and Bohemians were reported as making slow progress during the
first year but approximating the average in the next four grades
(16). In St. Louis, English-speaking and German-speaking chil-
dren represent average achievement. ''Jewish children rank above
the average, and Italian pupils rank distinctly below the average.
With the exception of German and Jewish children, practically all
foreign-speaking children are seriously retarded by language handi-
caps. ' ' In quality of silent reading ' ' foreign children made lower
scores than did English-speaking children. Jewish children formed
an exception to this rule. There seemed to be little correlation be-
tween rate of silent reading and nationality." The exact figures
are not shown. In the case of oral reading, the report recommends
that selections ''be provided which are simple in construction and
phraseology, and that will enable the pupil to develop gradually in
the mastery of language forms as well as in the recognition of
symbols." With reference to silent reading it recommends the
extensive reading of simple selections which relate to familiar ex-


1 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Online LibraryNational Society for the Study of Education. CommiReport of the Society's committee on silent reading → online text (page 1 of 17)