Front Cover
 Back Cover

Title: Analysis of the social contacts of preschool children with the aid of motion pictures
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00001445/00001
 Material Information
Title: Analysis of the social contacts of preschool children with the aid of motion pictures
Series Title: Child development series - University of Toronto ; 10
Physical Description: Book
Language: English
Creator: Bernhardt, Karl S.
Millichamp, Dorothy A.
Charles, Marion W.
McFarland, Mary P.
St. George's School for Child Study, University of Toronto
Publisher: The University of Toronto Press
Place of Publication: Toronto
Publication Date: 1937
 Record Information
Bibliographic ID: UF00001445
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: ltqf - AAA0509
ltuf - ADJ0829
alephbibnum - 000650853

Table of Contents
    Front Cover
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
    Back Cover
        Page 28
        Page 29
Full Text


v M,

m% a

,jM%- M N M.

v M 14-

M ...
e A
:: A R,



% %

% M M I I M ,
M, % %




flnivereitp of toronto 5tubiee
blb ]Development Berfes, 1Ro. 10





_ -I




The investigator in any field of inquiry is confronted with two
problems, (1) the discovery of a technique or method, and (2) the
interpretation of results.
The two studies which are included in this pamphlet are designed
to investigate these two aspects of the early social development of
preschool children.
In some fields it is possible to refine the recording device so that
changes in the phenomenon under scrutiny may be automatically
registered. In psychological study of human subjects in a social
setting the human recorder is still employed. The "personal equation"
still complicates the picture. By the use of movies a "permanent"
record may be obtained and compared with the record made at the
same time by an observer. The cinematograph data must, however,
in turn be tabulated by an observer. The first paper discusses the
contribution made by each technique to more reliable recording and
also some of the shortcomings.
The second paper discusses the results obtained by the two tech-
niques and is of especial interest since it was arranged as a control
study coincident with a similar investigation made on the Dionne

St. George's School for Child Study,
November, 1937.



PREFACE............................................... 3

INTRODUCTION ........... ............................ 7

Historical introduction............................. 7
Procedure ........................................ 10
Analysis of the data ........................... 13
Conclusions ..................................... 20
Summary ............................. ........... 23

Historical introduction ............................ 24
Procedure....................................... 26
Analysis of the data ............................... 27
Conclusions ....................................... 44
Summary ........................................ 45

BIBLIOGRAPHY ....................................... 46

APPENDIX I-Sample record form...................... 47

APPENDIX II-Comparison of sample observer's and motion
picture records. .................................... 48

APPENDIX III-Supplementary experiment ................ 52

I _

_E_ E



This report contains two studies using the same data. The first
is an attempt to evaluate the contribution of the motion picture
technique in the study of social behaviour of preschool children.
The second is an analysis of the data collected by both direct obser-
vation and from motion pictures in order to determine if there are
any age differences in social behaviour of preschool children and to
attempt to discover any characteristic patterns of social behaviour.


1. Historical Introduction
In a study of the social behaviour of preschool children, the
problem of method has presented one of the greatest difficulties.
This section is concerned with an investigation of the use and pos-
sibilities of motion pictures as an observational technique in the
study of social behaviour of the preschool child.
Dorothy Swaine Thomas's (13) discussion of techniques for
studying social behaviour marks the beginning of a concentrated
effort to devise more scientific and reliable methods for the study of
social behaviour of young children. The problem was one of finding
a method that would provide a means of accurately recording the
behaviour under observation. To this end the techniques brought
together by Thomas emphasize two main points: (1) the selection
and definition of units of behaviour for observation which in them-
selves will be capable of combination into a larger aspect of be-
haviour; (2) the control of the observer. In all, an attempt is made
to assure greater reliability by making the units of observation as
quantitative and independent of the observer as possible. The more
clearly defined and quantitative the unit of observation, the greater




the reliability and objectivity with which the observers will record
it. The aim of these studies is the measurement of reliability.
The techniques discussed by Thomas take three general forms:
(1) those in which each child is followed for a specific length of time
and the particular aspect of social behaviour being observed is re-
corded each time it occurs, e.g. social-material activity; (2) those
in which, within the whole social situation, a definite social response
is recorded each time it occurs, e.g. laughter; (3) those using the
psychological test situation which provides for considerably less
material and social stimuli than does the nursery school setting, to
record the same type of data as in (1) and (2), e.g. use of the test
situation for studying personality differences. In all, the emphasis
is on obtaining reliable quantitative data which will allow for sta-
tistical analysis.
This approach to the study of social behaviour in young children
by Thomas and her associates gave an impetus to further research
along similar lines. As a result, considerable work has been done
on this problem of method. To mention two of these, we will turn
to the studies of Arrington and Loomis.
Arrington's (1) research was carried on at Columbia with twenty
preschool children ranging in age from sixteen to thirty-two months.
The objective of this study was the development of a reliable method
for observing social, material, and self activities of children in free
play which would yield comprehensive, significant, and comparable
data. The emphasis in this study is on reliability of method which
she sought to ensure by constant refinement and definition of the
units of behaviour under observation and by the accurate recording
of the observers. She states:

"Absolute accuracy implies a genuine and complete record of
actual events. Such a record could be approximated with the
aid of automatic recorders, such as the motion picture and the
dictograph, but, unfortunately, mechanical methods have not
yet been perfected for real life recording. Since we have no
standard record of true occurrences by which to evaluate the
records of individual observers, we are forced to assume that
agreement between different observers who are equally well
trained and are recording simultaneously the same units of


behaviour in the activities of the same individual, confirms the
probability that the events occurred" ibidd., p. 29].

Three general trends characterize Arrington's technique: (1) re-
finement of observational units to simple, objective, and compre-
hensive categories of behaviour which would facilitate observation
and allow for more accurate recording by the observer; (2) the in-
crease in reliability by the continued definition of units of behaviour
for observation to minimize subjective interpretation on the part of
the observer; (3) simplification in the actual method of recording
the data.
Loomis (9) in the development of a technique for studying social
behaviour, chose to study physical contacts as indicative of that
behaviour. The study was conducted at Columbia and thirty-seven
nursery school children were observed during the free play period.
She points out the difficulty of devising sufficiently objective cate-
gories of behaviour where the material is continuous. This difficulty
increases in the actual collection of data due to the rapid change in
situations making immediate classification imperative. She states:

"The latter difficulty will be overcome in the study of behaviour
only when details of reaction are preserved by mechanical
means, such as the moving picture or the dictograph. Then the
material can be reproduced, analysed, and classified at will"
ibidd., p. 20).

Still further she concludes:
"Mechanical means of taking and preserving records, as, for
example motion pictures, would ensure recording a larger per-
centage of the total interaction, though even by this means a
100 per cent record could not always be expected. This per-
manent record would, however, do away with the necessity for
snap judgments in the classification of contacts and would fur-
nish excellent material for training observers" ibidd., p. 87).

This latter point has been illustrated by Arrington (2) who used the
"talkies" as a means of studying some technical aspects of observer

__ __ _~__ ~_~ I ~U



The possibilities of the motion picture as an observation tech-
nique have been realized, but little has been done to develop it as
such, which is in a large measure due to the expense such a method
involves. Gesell (7) was the first to employ what he terms "cinema-
tography" in the analysis of behaviour patterns and more speci-
fically in infant behaviour. Gesell's motion picture apparatus,
known as the photographic dome or observatory, was devised as a
means of observing human behaviour, which is in its nature so com-
plex as to render description exceedingly difficult. With the aid of
motion pictures, human behaviour can be studied more thoroughly.
The motion picture records completely and impartially whatever,
so to speak, it sees. The data obtained by the use of motion pictures
are made available for study from three important aspects: (1) the
specific behaviour; (2) the specific behaviour in relation to the
episode in which it occurs; (3) the developmental significance of the
behaviour. In this way, the motion pictures become an objective
observational technique, allowing the observer to analyse the be-
haviour scientifically, reproduced in exactly the same form as it
It can be seen that within the field of method in the study of
social behaviour of young children the main difficulty has been in
devising reliable techniques that would give a reliable body of data.
The main trend has been toward objectivity in observation with a
view to securing quantitative results capable of statistical treatment.
In other words, these observational techniques have been developed
primarily with a view to obtaining reliable measures rather than for
the contribution they would make to the analysis of the behaviour
under observation. In view of this, it was decided to investigate
the possibilities and use of the motion pictures as an observational
technique in the study of social behaviour of preschool children.
The contribution of the motion picture technique was evaluated by
a comparison with the observer's record.

2. Procedure
The subjects used in this study were fifteen children, eight boys
and seven girls. These children were selected from a day nursery
where mothers who are employed in daily labour leave their children


--L 4-


from 8.30 a.m. until 6.30 p.m. Among the fifteen children seven
different nationalities were represented. The subjects were divided
into three groups of five children each, as follows: Group I-average
age 25.6 months, range 22-27 months; Group II--average age 29.4
months, range 29-30 months; Group III-average age 39.6 months,
range 37-42 months.
The observations were made of free play activity where adult
interference and direction are at a minimum. The junior playroom
of the St. George's School for Child Study, with the accompanying
play material which was constant for each group, was used for the
observational setting. The children played with material, in groups,
or alone at will.
Each of the three groups was observed once a month for four
months. Each subject was observed in his group for two five-
minute periods each month. The observations were divided into
half-minute intervals.
A social contact was the unit of observation. A social contact
is defined as any relationship with another child or overt behaviour
involving other children. The main objective was to record, with as
little interpretation as possible, the actual overt social interaction
which occurred. Two observers recorded the social behaviour in
terms of the social contacts of the child under observation. A third
observer operated the motion picture camera.
Record forms were prepared showing at the top the group to
which the subject belonged, his identification in that particular
group, and the date of observation. The balance of the form was
marked off into divisions representing the half-minute intervals.
The social contacts observed were recorded by means of arrows, the
direction of which illustrated the direction of the social contact.
The content of the behaviour observed was recorded descriptively
rather than in symbols. At the end of each complete observation,
the observers examined their records, checking one with the other
for agreement and disagreement. (See Appendix I for sample
Motion pictures were taken at intervals throughout each obser-
vation with a Cine-Kodak 16 mm. camera, using panchromatic and
supersensitive films. On the average, two hundred feet of film were

_ __I X __ ;__~_

-L 1


used for one complete observation of five children for ten minutes,
which averaged approximately forty feet for each child.
All five children were together for the complete observation
period and were observed one at a time. The subjects were pro-
vided with white shirts lettered in black, back and front, A, B, C, D,
and E, respectively, in order to facilitate identification in the motion
pictures. The subjects were taken into the observation room where
they were permitted to play a few minutes before the observation
began in order to become accustomed to the setting and bright lights
necessary for taking motion pictures inside. This time was de-
creased at each successive observation as the children became more
familiar with the situation. The room was divided in half by a row
of low toy cupboards. This was done to keep the group in a smaller
field to facilitate the taking of the motion pictures.
The observers and motion picture operator took up their posi-
tion in the unused half of the observation room and were free to
move around while observing. At the beginning of the observa-
tional period, one observer gave a signal to the motion picture opera-
tor. This was noted from a stop watch. The observers then
proceeded with the observation and recording. The motion picture
operator took a picture of the subject at the beginning of each
observation period, irrespective of the activity of the child, for
identification. Thereafter, motion pictures were taken throughout
the observation whenever the unit of observation appeared most
frequently; in other words, whenever the social situation became
complex and difficult to record. An attempt was made by the ob-
servers to record at what points during the observation motion pic-
tures were taken, depending for this on the sound of the machine.
However, this proved to be unsuccessfully recorded. One of the
observers gave a signal again at the end of the observation period
to the motion picture operator indicating that no more pictures
were to be taken at that time. This procedure was repeated for
every child in the group under observation.
From the use of this procedure two sets of data were obtained:
(1) the observers' records; (2) the motion pictures. For a com-
parison of these records it was necessary to reduce the motion pic-
tures to a form comparable to the observers' records. As a first step
in this procedure a measure of agreement was found between the


lk --


records of the two observers, the average of which, for all records,
was 82%. These records were then combined to form one
record including every item on both records. Hereafter this com-
bined record will be referred to as the observers' record.
The second step was to take a record of behaviour from the
motion pictures. The motion pictures were made available for
observation by use of a projector operated by one of the observers.
A description of the behaviour observed on the screen was dictated
by the second observer to a stenographer recording in shorthand.
The same observers were used here as in the original observation.
While only one observer gave an account of the activity observed
on the screen, the observer operating the projector watched the
screen closely and, where differences were noted, the pictures would
be shown again. The films for each complete observation of one
group were projected at one time and shown as often as was thought
necessary by the observers. This record was then typed and checked
on a second day when the films were projected again, additions and
deletions being made wherever necessary.
This procedure provided two sets of comparable records. The
last step was to combine these. The problem presented here was to
fit the motion picture records, which were continuous without refer-
ence to time, into the proper time intervals. This was done by
checking the motion picture records against the observers' record
and using descriptions of behaviour as landmarks. This gave the
time interval in which the motion pictures were taken, but not the
length of time over which they were taken. These two sets of data
were then combined in one record showing: (1) the social contacts
common to both the observers' record and the motion picture re-
cords; (2) the additional contacts from the motion picture records;
(3) the additional contacts from the observers' record.

3. Analysis of the Data

Since this is primarily an investigation of method in the study
of social behaviour, the analysis will be made from that point of
view. In this analysis three major questions will be discussed:
(1) to what extent do motion pictures used as an observational
technique increase the data; (2) does this increase in data change

I ____ --NEER



the picture we have of the behaviour or merely expand it; (3) do
motion pictures contribute to the further analysis of social behav-
iour? In the discussion of these three questions, the data will be
dealt with from both their quantitative and qualitative aspects.
From a comparison of the motion picture records with the ob-
servers' record, from the point of view of actual number of contacts
recorded, it is possible to ascertain to what extent the data were
increased by the use of motion pictures as an observational tech-
nique. However, if this comparison is to be valid, it can only be
made at those points in each record where the situations recorded
are parallel. All those situations were chosen for comparison from
each record that had common starting points and common finishing
points. The contacts in between these two points in all such situa-
tions in each record were then counted and compared. In all, 90%
of the motion picture records were made available on this basis
of comparison. In each situation the social contacts recorded were
noted in three respects: (1) social contacts common to observers'
record and the motion picture records; (2) additional contacts from
the motion picture records; (3) additional contacts from the ob-
servers' record. Hereafter when either the observers' record or the
motion picture records are referred to, they will be taken to include
only those comparable sections of the records. All further analysis
will be based only in those comparable sections.
common to Additions Additions Motion Picture
Observers' from from Additions
Groups Totals and Motion Motion Observers' Total
Picture Picture Records Straight
Records Records Addi- Corrections
Group Total 301 248 41 590 235 13
I % 51 42 7 100 40 2
Group Total 231 176 32 439 155 21
II % 53 40 7 100 35 5
Group Total 197 162 38 397 145 17
III % 50 41 9 100 37 4

7, _

mi I


Table I demonstrates in totals and percentages the proportion
that: (1) the social contacts common to both the observers' record
and the motion picture records; (2) the additional contacts from
the motion picture records; (3) the additional contacts from the
observers' record; each is of the whole body of comparable data in
the three groups. It is seen here that the additional contacts from
the motion picture records comprised 42%, 40%, and 41% of the
total in each of the three groups respectively. These figures repre-
sent an actual increase in data of 72% in Group I, 67% in Group II,
and 69% in Group III.
Table I also demonstrates that the observers' record adds to the
data, material that, within the same limits, was not included in the
motion picture records. A further analysis was made to discover
what social contacts were recorded by the observers and not ob-
served in the motion pictures. In the main, these were found to be
of two classes: watching of two specific types and talking. The
watching not included in the motion pictures was found to be either
that directed to or from the child under observation, beyond the
field of the camera, or that so closely allied with some other social
activity as not to be separated out as such. These two types of
watching accounted for 2.5%, 1.5%, and .8% of the additions from
the observers' record in each of the three groups respectively.
The observers found it very difficult to observe when a child was
talking in the motion pictures, unless the subject under observation
was facing the observers and in the immediate forefront of the
screen. Further, when verbalization was observed, it was exceed-
ingly difficult if not impossible, except in the most obvious cases,
to determine whether or not it was talking directed to another child,
talking to self, or, as in some cases, singing. These additions of
verbalization from the observers' record accounted for 2%, 3.5%,
and 6% more of this class in each of the three groups respectively.
The remaining 2.5%, 2%, and 2.2% of the total 7%, and 9%c of
additions from the observers' record in each group, comprised social
contacts which could conceivably be verified by still further repe-
tition and observation of the motion pictures. It may be assumed
that, up to a point, the oftener the motion pictures were observed
the more complete would be the record therefrom.
Table I demonstrates further that the motion picture records



Contacts Common to Observers' and Motion Additional Contacts from Common Plus Additions from
Picture Records Motion Picture Records Motion Picture Records
Groups and I II III IV V VI VII Totals I IIIII IV V VI VII Totals I II III IV V VI VII Totals

Group Total 60 13 12 34 21153 8 301 53 14 0 41 14113 13 248 113 27 12 75 35266 21 549
I % 20 4 4 11 7 51 3 100 21 60 16 6 46 5 100 21 5 2 14 6 48 4 100
Group Total 58 8 15 20 10 111 9 231 34 6 3 43 11 82 2 181 92 14 18 63 21 193 11 412
II % 25 4 6 9 4 48 4 100 19 3 2 24 6 45 1 100 23 3 4 15 5 47 3 100
Group Total 45 9 10 16 14 98 5 197 37 8 2 34 1 0 162 8217 1250 24168 6 359
III % 23 5 5 8 7 50 2 100 235 121 6431 100 23 5314 746 2 100

Additional Contacts from Common Plus Additions from
I. Watching Observers' Record Observers' Record
II. Gestures Group Total 15 5 14 2 0 5 0 41 75 18 26 36 21 158 8 342
III. Verbalization I % 37 12 34 5 0 12 0 100 22 5 9 11 6 44 3 100
IV. Approach and Withdrawal Group Total 7 7 16 0 0 1 2 33 65 15 31 20 10 112 11 264
V. Adjacent Use of Material II % 21 21 50 0 0 2 6 100 25 6 11 8 4 42 4 100
VI. Associative Use of Material Group Total 4 25 1 0 5 0 38 48 13 35 17 14 103 5 235
VII. Physical Contacts III % 8 11 66 2 0 13 0 100 21 6 15 7 6 42 3 100


served to amend the observers' record to some extent. The extent
to which the motion picture records did amend the observers' record
appeared to have some direct relation to the amount of social be-
haviour and its complexity as manifested by the child under
Motion pictures when used as an observational technique do
increase the amount of the data. Does this increase in data change
the total picture of the behaviour under observation as well as
expand it? To explore this question the comparable data in each
of the records were investigated in terms of categories of social
behaviour. Seven broad categories of social behaviour were chosen
on the basis of objectivity of interpretation.
Table II demonstrates in totals and percentages the frequency
of occurrence of each of the seven categories of behaviour in each of
the three groups. These are considered in respect to: (1) that part
of the data common to both the records; (2) the additional data
from each of the records; (3) a combination of each of these addi-
tions with the data common to both records. This table illustrates
that the additions from the motion picture records represent in
terms of percentage of frequency of occurrence of behaviour, ap-
proximately the same general picture as do the data common to
both records. This point is further corroborated when these are
seen in combination. It should be noted here, however, that in
that category labelled "Approach and Withdrawal" the additions
from the motion pictures change the proportion of that behaviour
more than in any other.
The additions from the observers' record are seen to be pre-
ponderantly in favour of watching, gestures, and verbalization, dis-
playing quite a different picture to that found in any other section
of the data. These differences, however, tend to level out when
combined with the common elements of the records in all categories
of behaviour except that of verbalization. On the whole, the motion
pictures would seem to confirm rather than change the group picture
of behaviour in all except verbalization.
It must be observed here that this analysis was made on a group
basis and as such there is a tendency for the group results to obscure
the individual cases. There is also a possibility that the breadth of
the categories chosen excluded the subtle variations of behaviour.




Approach Adjacent Associative
Watching Gestures Verbalization and Use of Use of Physical
Withdrawal Material Material Contacts
A' 3.5 8 3 3 8.5 10 12 7.5 1 1 4 2 10 2.5
B1 5 2.5 5 5 15 15 3 4 2.5 3 9.5 7 10 4.5

C1 11 6 10.5 7 8.5 13.5 3 7.5 8.5 8 2 4 3 6.5
DI 13 11 2 3 2.5 1 1 1.5 8.5 6 9.5 5 6 1

E' 1 1 12.5 7 12.5 10 5.5 5.5 14.5 15 3 6 10 6.5

A2 6 7 8 7 5 10 7.5 3 12.5 13.5 1 1 3 4.5
B2 11 14.5 14.5 15 11 3.5 9 11.5 14.5 5 14 14 3 9 0
C2 2 5 5 14 10 3.5 14.5 9.5 8.5 11 6 6 6 11.5

D2 7 4 1 3 1 2 3 5.5 8.5 11 7.5 3 10 9

E2 9 12.5 14.5 11.5 12.5 10 12 13 4 4 15 15 6 9
A3 14.5 14.5 8 1 5 6 10 11.5 5 8 7.5 11 14 14 Z
BE 8 10 5 11.5 14 10 12 14 8.5 11 11 9 10 11.5
C3 14.5 12.5 12.5 11.5 5 13.5 14.5 15 2.5 2 13 13 14 14
D3 11 9 10.5 11.5 7 6 7.5 9.5 12.5 13.5 12 12 1 2.5
E3 3.5 2.5 8 9 2.5 6 5.5 1.5 8.5 8 5 10 14 14
R .80 .58 .52 .77 .78 .80 .61

*C & O-Contacts common to Observers' Record and the Motion Picture Records plus Additions from the Observers'
tC & MP-Contacts common to Observers' Record and the Motion Picture Records plus Additions from the Motion
Picture Records.



Because of this, an individual analysis was made in respect to
the same categories of behaviour. Each of the fifteen children was
ranked in each of the seven categories. The ranking was done, first,
in respect to those contacts common to both the observers' and
motion picture records plus the additional contacts from the ob-
servers' record; secondly, in respect to those contacts common to
both records plus the additional contacts from the motion pictures.
In other words, the subjects were ranked first, in respect to the
observers' record and, second, in respect to the motion picture
records. The differences between ranks were noted and a rank
order coefficient of correlation computed for each category.
Table III demonstrates the degree of similarity there is between
ranks in each of the categories of behaviour when examined in rela-
tion to both records. In the case of each category, the obtained
coefficient indicates the degree of similarity there is between that
picture we have of the child's behaviour when directly observed,
and when the additional data made available by the use of the
motion pictures are considered. In all but 15% the ranks were
changed by the inclusion of the additional data from the motion
picture records. Using the observers' record as a basis, any
differences in rank of behaviour in respect to the motion pictures
indicate that there is a difference in the picture of the child's be-
haviour as obtained by direct observation and as obtained by
motion pictures.
The low coefficient obtained in the category designated verbal-
ization is further proof of the discrepancy there is between the
observers' record and the motion picture records in this particular.
Motion pictures limit such a study when talking is an integral part
of social behaviour.
This analysis demonstrates that motion pictures do make con-
siderable additional data available. It demonstrates also that the
additional data confirm in general the picture obtained by direct
observation when the subjects were considered in groups. However,
when the subjects were considered individually rather than in
groups, the additional data not only expanded the picture of the
child's behaviour but also changed it.
If the motion pictures are to be of real value in the study of
social behaviour, they should contribute to the analysis of that

_ ___ ,

M - -



behaviour. To demonstrate this, representative situations from
each group were lifted from the main body of the records and appear
as parallel situations, one from the observers' record and one from
the motion picture records. (See Appendix II.) These situations
are shown actually as recorded by the observers and as recorded
from the motion pictures. The description was added in each case
to give a comprehensive and concise picture of the behaviour as it
occurred. From a comparison of these situations, the observers'
record with the motion picture records, it can be seen in each case
that the observations made from the motion pictures not only in-
creased the data but also changed the picture of the child's behaviour
in varying degrees. Hence, in this study of social behaviour, the
additional data made available by use of the motion picture con-
tribute largely to the analysis of this behaviour and to our further
understanding of the young child in consideration of the data.

4. Conclusions
From an analysis of the data it is apparent that motion pictures
represent a development of observational technique in the study of
social behaviour. The importance of this development depends
upon the contribution the method makes to the study of the be-
haviour under investigation. An analysis of the data has shown
three specific contributions which motion pictures used as an obser-
vational technique make to the study of social behaviour.
First, in the body of the comparable records, the data were
increased by the use of motion pictures as an observational technique
72%, 67%, and 69% in each of three groups respectively. From
this it is apparent that two observers recording simultaneously the
same activity of the same individual with a high percentage of
agreement, do miss a considerable amount of the available data.
This is due to the impossibility of recording all that is seen and a
failure to see all that occurs. The discrepancy between the ob-
servers' records and the motion picture records would, however,
decrease as the amount of film taken increased. If the pictures were
taken when the manifestations of the unit of observation were
simple and less frequent as well as when more frequent and complex,
the similarity would have been greater between the two records.
Second, a qualitative analysis was made of this additional data,

-L .. -Y


in terms of certain specific categories of social behaviour, to deter-
mine whether or not the additional material presented the same
picture of behaviour as did the data common to both the observers'
and motion picture records. This additional material was consid-
ered in relation to each group observed and in relation to each indi-
vidual observed. It was found by analysis that the inclusion of the
additional data made available by use of the motion pictures did
not change the picture of social behaviour when that behaviour was
considered in relation to each of the three groups observed. How-
ever, from an examination of differences in rank for each individual,
it was found that the inclusion of the additional data from the
motion pictures did change the picture of the social behaviour of the
child when considered individually rather than in groups.
Third, from a comparison of the records, in a qualitative analy-
sis, the additional data from the motion pictures showed many
variations in behaviour not found in the observers' record. As such
they contributed largely to a further qualitative analysis of social
Since this investigation has been exploratory in nature from the
point of view of method, some discussion of the use, limitations, and
possibilities of motion pictures is warranted. This investigation
illustrates the use of motion pictures as an observational technique.
The observations could have been made entirely by trained ob-
servers. Had this method been followed solely, a great deal of
available data would have been missed as has been illustrated. On
the other hand, the expense that the motion picture technique in-
volves made the use of this method alone, prohibitive. A combina-
tion of these two methods would seem the adequate solution. A
further question is raised here: should the motion pictures be taken
using the behaviour-sampling technique as was done in this study,
or using a time-sampling technique? If the motion pictures are to
be used as an additional technique in the study of social behaviour
to augment the data and increase its reliability, they will do so by
recording the more frequent and complex manifestations of that
behaviour rather than the simpler, less frequent, and more easily
recorded data. If the time-sampling method is used, many such
situations would assuredly be missed. If a system of signals or some
mechanical means were devised whereby the observer could record,



or would know at what times and over what intervals the motion
pictures were taken, the combination and comparison of the two
records would be greatly facilitated and more reliable.
It is apparent that the expense which such a method as that
requiring motion pictures involves will limit its development con-
siderably. In view of this, it should be pointed out here that the
cost of an observation such as this investigation represents, could
have been reduced by half, had the observations been made of out-
door play activity in a similarly restricted area. When this can be
done, standard films may be used rather than the more expensive
supersensitive films necessitated by indoor observation.
Undoubtedly the most important aspect of motion pictures as
an observational technique is that of permanency. The motion
picture provides a means of permanently recording the behaviour
as it occurred and makes repetition of the observation possible.
Because of the complexity of social behaviour, this is exceedingly
important. The observations can ,be repeated at will, permitting
the observer to make the best possible judgment of that behaviour
and to revise and change his more immediate classifications. In
view of this, motion pictures should prove of value as an observa-
tional technique of longitudinal studies. The permanency of the
motion picture record makes preservation of a record of behaviour
possible, in exactly the same form as it occurred at different stages
of development.
A further contribution which comes as a result of this perma-
nency of record became apparent to the observers through the con-
tinued repetition of the motion pictures. As the motion pictures
were repeatedly observed, the observers became increasingly aware
of that child as a social individual. His behaviour was gradually
identified as having a definite individual pattern, assuming signi-
ficance as it differed from that of every other individual observed.
In other words, the social behaviour of the child under observation
came to have a significance, the meaning of which became apparent
from continued observation of that behaviour as it actually oc-
curred. The significance of such behaviour could never have been
realized from analysis of the observers' record alone, where each
contact is seen and recorded separately, becoming significant as
such, rather than in consideration of the individual behaviour pat-

I II .



tern as a whole. If motion pictures are to become an integral part
of methods of observation in the study of social behaviour, their
greatest development and contribution will be in the investigation
of individual differences.

5. Summary
Three groups of five preschool children were observed once a
month over a period of four months. In all, each child was observed
for forty minutes. Social behaviour was observed during free play
activity. Two observers observed and recorded social behaviour in
terms of social contacts. The more complex manifestations of social
behaviour during the observation period were recorded by motion
pictures operated by a third observer. Records were taken from the
motion pictures and compared with the observers' records.

An analysis of these data shows:
(a) The extent to which the comparable data were increased
by use of the motion pictures as an observational technique.
(b) That the motion pictures tend in general to confirm the group
picture of social behaviour obtained by direct observation, in all
except verbalization. However, when an analysis of behaviour was
made for individuals rather than for groups, the additional data
changed the picture of individual behaviour from that obtained by
direct observation.
(c) This increase in data contributed largely to a further analy-
sis of social behaviour and an understanding of it in relation to the
individual child.
(d) Conclusions: From a comparison with direct observation,
motion pictures represent a development in observational techniques
contributing as such to the study of social behaviour of preschool
children. The expense such a method involves will limit its develop-
ment. However, the permanency and accuracy of observation
which the method makes possible render it invaluable as an obser-
vational technique, especially in the study of individual differences.




1. Historical Introduction

The purpose of this investigation is to study in terms of the
frequency of social contacts directed to other children and received
from other children:
(1) Age differences both in frequency and types of social be-
(2) Individual patterns of social behaviour and their interrela-
tionship within each group.
The available research relating to the social behaviour of young
children is extensive. In an effort to simplify a complex problem,
attempts have been made to isolate certain social responses from
the total situation. By means of the time sampling technique, rela-
tively simple aspects of social behaviour have been analysed in
quantitative terms.
The quantitative data from a large number of studies indicate
that we may anticipate certain general trends in the behaviour of
nursery school children as follows: an increasing number of social
contacts with age. Arrington (1) found that the frequency of social
contacts was about twice as great at the three-year level as at the
two-year level. Correlation between age and social contacts was
.66. Beaver (3), isolating the factors involved in the initiation of
social contacts, found a distinct tendency for number of contacts to
increase with age, the correlation closely approximating that of
Arrington, .69. Robinson and Conrad (12) found an even higher
correlation between age and social contacts, .71.
Certain aspects of behaviour are, on the whole, much more easily
rated than others. As the command of language increases, there is
a general tendency for verbal contacts to replace physical contacts.
According to Arrington (1), physical contact contributed a less fre-
quent mode of social interaction than language. Beaver (3) found a
similar trend in that neither group uses physical contact in large
proportion but that it is more frequently used by the younger group.
In fact, there appears to be a gradual progression from "solitary"
behaviour, when a young child deals with other children only in an
exploratory manner, to active co-operation where we often discover




complex group interrelations. Bott (6), in a study of personality
differences, found that both imaginative and constructive play are
positively correlated with chronological age. The coefficient be-
tween chronological age and constructive play was found to be .56.
Most studies indicate that watching follows a trend similar to phy-
sical contacts, resulting in a negative correlation between this cate-
gory and age. Bott's results confirm this trend with a correlation
of -.70.
"Talking", "physical contacts", "gestures", "use of material",
and other behaviour items are probably influenced by certain types
of variations in the situations, yet they are not so dependent on the
situation as a less specific trait such as "sympathy". All that has
been discovered is that there are reliable measures of behaviour
trends in a given situation. A vigorous, convincing attack on the
question of the influence of the situation upon such a fundamental
aspect of behaviour as "aggressiveness", has been undertaken by
Jack (8). Five children found to have low ascendance scores from
a previous experiment were familiarized with certain situations and
supplied with the required skill and knowledge. These five were
then placed in the same situation with other children to whom the
situation was unfamiliar. The five subjects showed a greater in-
crease in ascendance scores than did the remainder of the same pre-
school group over the same period of time. As Murphy (11) sug-
"We should expect consequently that both the range of social
behaviour and the frequency of specific patterns of behaviour
would be affected by specific training and by the range of
situations to which the child has been exposed."
The kind of variation which deals with factors in the situation
beyond the experimental set up, and those inherent in the child, are
extremely hard to express in qualitative terms. Berne (4) measur-
ing patterns of social behaviour found that individual differences
were more significant than age differences in certain traits. Arring-
ton (1) throws more light on her quantitative results through a
qualitative analysis of the behaviour of individual children. Among
twelve profiles charting changes from one year to another, eight
instances were found of children whose behaviour was predomi-

I _

An analysIs ol the soIal cont main
136 7T686c no 10

3 1262 03155 2390

1 r "



I I-,


IV r I I.e

IV 11 1 VI h I I

I I A.,
Inm I

IN III P,: I .:A ... V. K I I I, II., W H VV

IN ... IV I I
I Vs *! V .11




1, 14 L
4V I

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs