DOCUMENT RESUME ED 071 262 EC 050 870
AUTHOR Rhodes, William C.; Gibbins, Spencer TITLE Environmental Forces Impinging Upon Normal and Disturbed Children in a Regular Classroom. Final Report. INSTITiTION Michigan Univ., Ann Arbor. Inst. for the Study of Mental Retardation. SPONS AGENCY National Center for Educational Research and Development (DHEW /OE), Washington, D.C. Regional Research Program. BUREAU NO BR-1-E-098 PUB DATE May 71 GRANT OEG -5 -71 -0073 (607) NOTE 130p.
EDRS PRICE MF-#0.65 HC-46.58 DESCRIPTORS Childhood; *Classroom Environment; *Emotionally Disturbed; *Environmental Influences; *Exceptional Child Research; *Interaction Process Analysis; Regular Class Placement
ABSTRACT Aspects of the interactive environments of 10 emotionally disturbed children (7 to 11 years old) were compared with those experienced simultaneously by 10 normal children (matched for age, sex and IQ) in the same setting. Two types of setting were examined: a choice setting in which the child was able to choose his activity from among several alternatives anda nonchoice setting in which the activity was planned and directed by the teacher. Two trained observers were used to describe the behavior ofa matched pair of children (one normal, one emotionally disturbed) ina regular elementary school classroom. The narrative accounts of the observers were delineated into Environmental Force Units (EFU) based on the following six variables: the frequency of total EFU'sper observation; the frequency of EFU's per observation initiated by the S; the frequency of EFU's per observation in which the Swas the sole target of the EFU; the frequency of EFU's per observation in which conflict was evident between the goals of the agent and theS; the frequency of EFU's per observation in which methods judged to be coercive were employed by the agent; and the frequency of EFU'sper observation in which the agent wasan adult. Data indicated no .significant differences in environmental forces existing between the disturbed and the nondisturbed children and only two differences between the two types of settings. (Author/GW) FILMED FROM BEST AVAILABLE COPY
/4- L-=_--- 09S ENVIRONMENTAL FORCES IMPINGING UPON NORMAL AND DISTURBED CHILDREN IN A REGULAR CLASSROOM
Spencer Gibbins Final Report
U
-4100 MI Project No. 1-E-098 r-4 Grant NO. OEG -5-71 -0073 (607) CD
William C. Rhodes Spencer Gibbins Institute for the Study of Mental Institute for the Study of Mental Retardation and Related Disabilities Retardation and Related Disabilities 611 Church Street 611 Church Street Ann Arbor, Michigan 48104 Ann Arbor, Michigan 48104
ENVIRONMENTAL FORCES IMPINGING UPON NORMAL AND DISTURBED CHILDRENIN A REGULAR CLASSROOM
The research reported herein was performed pursuant toa Grant with the Office of Education, U. S. Department of Health, Education,end Welfare. May, 1971 Contractors undertaking such projects order Government sponsorshipare encouraged to express freely their professional judgment in theconduct of the project. Points of view or opinions stated do not, therefore, neceswily represent official Office of Education positionor policy. O 41.00.41.1.1
00 U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE
0 Office of Education
Regional Research Program U $ DEPARTMENT OF HEALTH. EDUCATION WELFARE k) OFFICE OF EDUCATION THIS DOCUMENT HAS BEEN REPRO OUCED EXACTLY AS RECEIVED FROM THE PERSON OR ORGANIZATION ORIG INATING IT POINTS OF VIEW OR OPIN IONS STATED DO NOT NECESSARILY REPRESENT OFFICIAL OFFICE OF EDU CATION POSITION OR POLICY ABSTRACT
ENVIRONMENTAL FORCES IMPINGING UPON NORMAL AND
DISTURBED CHILDREN IN A REGULAR CLASSROOM
by
Spencer Gibb.tns
Chairman: William C. Rhodes
This study examines the interactive environments of children labeled as "emotionally disturbed" and compares aspects of these environments with those experienced simultaneously by "normal" children in the same setting.
Two types of settings were examined--a choice setting in which the child was able to choose his activity fromamong several alternatives, and a non-choice setting in which the activity was planned and directed by the teacher. The study assumes an ecological stance in formulating the hypotheses which revolve about differences occurring in the environmental forces impinging upon "disturbed" and "normal" children who are in the same regular classroom setting at the same time.
The experimental method used by the study, exploratory in nature combines the sophisticated techniques of observeion developed by "ecological psychology" foruse in everyday settings unmediated by artificial experimental conditions, an instrument for delineating and quantifying environmental forces, and a matched-pair analysis of variance design. In
1 the study two trained observers were used Lo describe the
on-going behavior of a matched pair of children in a regular
elementary classroom. One of the children had been labeled
as "disturbed" by the school diagnostic aeparatus and the
other, matched for age, sex and IQ had never been referred
for special services. Ten such pairs of children were observed
in both a choice and non-choice setting. These narrative
accounts were then delineated into Environmental Force Units
which were, in turn, categorized and frequency counts recorded.
The categories used for these Environmental Force Units
contributed such information as whether conflict occurred with
the environmental agent, whether the environmental agentwas an adult or peer and whether the environmental interaction
was initiated by the child or by the agent. The frequency scores derived from the classification of the Environmental
Force Units were then subjected toan analysis of variance to determine to what degree, if any, environmental forceswere dif-
ferent for "disturbed" and "non-disturbed" childrenand whether
or not these differences were affected by the nature of the
setting.
The results of the study revealedno significant dif-
ferences in environmental forces existing betweenthe "disturbed" and "non-disturbed" children and buttwo differences between the two types of settings. The study ,,oncludes that Inter- setting rather than inter-subject differencesare perhaps better studied through this experimental method andsuggests other uses for ecological techniques in psychological researchand training. ENVIRONMENTAL FORCESIMPINGING UPON NORMAL AND
DISTURBED CHILDREN IN AREGULAR CLASSROOM
by
Spencer Gibbins
A dissertation submitted inpartial fulfillment of the requirements for thedegree of Doctor of Philosophy (Education and Psychology) in The University of Michigan 1971
Doctoral Committee:
Professor William C.Rhodes, Chairman Professor Albert C.Cain Professor Rue L. Cromwell Profeplor William M.Cruickshar:: Professor William C.Morse ACKNOWLEDGMENTS
The project presentedor reported herein was performed
pursuant to a grant from the UnitedStates Office of Education,
Department of Health, Educationand Welfare. The opinions
expressed herein, however, donot necessarily reflect the
position or policy of the UnitedStates Office of Education,
and no official endorsementby the United States Office of Education should be inferred.
The author wishes to extendhis sincere appreciation
to all those whose helpmade this effort possible. I
especially want to thankthose administrators, teachers and
children who opened theirclassrooms to this study.
,The author acknowledgesthe help of his colleague,
Dr. M. L. Tracy withoutwhose support and assistance this
study could not have beendone; his chairman, Dr. W. C.
Rhodes whose own work andideas were a great influence;
Dr. Phil Schoggen and MaxineSchoggen whose encouragement
and help with methodology wereinvaluable; the observers,
Lucien Zahendra and SandraHoltz who maintained enthusiasm
and cheer against overwhelmingodds; the data coders, Joseph
Babcock, Carnelyn andMary Glinsky and Tom Lovelandwho quickly outstripped theirinstructor in coding efficiency;
Edith Klein and PattieBarr for their typing thenine-tenths of the glacier notseen in this manuscript and Pattie Tracy
ii for all her help in data collection.
The preparation of this manuscriptwas the able work of Betty Southworth and NormaFeldkamp.
iii TABLE OF CONTENTS
ACKNOWLEDGMENTS ii LIST OF TABLES v LIST OF ILLUSTRATIONS vii Chapter I. STATEMENT OF THE PROBLEM 1 Introduction 1 Emotional Disturbanceas an Interactive Phenomenon The Problem 4 5 Definitions for thisStudy The Design 7 Hypotheses 9 Study Limitations 12 14 II. LITERATURE REVIEW 16 Ecological Research in Psychology 16 Ecological Studies ofChildren 24 , Ecological Studies of"Disturbed" Children 32 Methodological Concerns Synthesis and 39 Application 44 III. PROCEDURE AND ANALYSIS 48 Population of theStudy Data Collection 48 52 Environmental ForceUnit Coding and Categorization of theData Statistical Analysis 55 63 IV. RESULTS OF THESTUDY 66 V. CONCLUSIONS 84 Examination of the Results 84 The Sample of"Disturbed" Children Methodology 88 Implications of theStudy for 92 Ecological Psychology 10 APPENDIX A 99 REFERENCES 103 115 iv 4 LIST OF TABLES Table Page 1. Total Adjusted Number of EFU's (Variable 1) 57 2. Adjusted Number of EFU Initiated by Subject (Variable 2) 58
3. Adjusted Number of EFU in which Subject is Sole Target (Variable 3) 59 4. Adjusted Number of EFU in which Conflict Occurred (Variable 4) 60
5. Adjusted Number of EFU in which a Coercive
Method was Used by Agent (Variable 5). . . . 61
6. Adjusted Number of EFU Containing Adult Agent (Variable 6) 62
7. Variable 1 Cell Means for Total Adjusted EFU's 67
8. Analysis of Variance for Variable 1, Total Adjusted Number of EFU's 68
9. Variable 2 Cell Means for Subject-Initiated EFU's 70 10. Analysis of Variance for Variable 2, Adjusted Number of EFU's which were Initiated by the Subject 71 11. Variable 3 Cell Means for EFU's in which the Subject was Sole Target 72
12. Analysis of Variance for Variable 3, Adjusted Number of EFU's in which Subjectwas Sole Target 73 13. Variable 4 Cell Means for EFU's in which Conflict Occurred 0 14. Analysis of Variance for Variable 4, Adjusted
Number cf EFU's in which Conflict Occurred . 76
15. Variable 5 Cell Means for EFU's in which Coercion by the Agent is Evident 78 Table Page 16. Analysis cf Variance for Variable 5, Adjusted Number of EFU's Containing Coercion 79 17. Variable 6 Cell Means for EFU's Containing Adult Agents 80 18. Analysis of Variance for Variable 6, Adjusted Number of EFU's with Adult Agent 81
,
vi LIST OF ILLUSTRATIONS
Figure Page 1. Analysis of Variance Design 64
2. Distribution of Adjusted Total EFU'sfor "Disturbed" Sample 89 3. 'Distribution of AdjustedTotal EFU's for "Normal" Sample 89 CHAPTER I
STATEMENT OF THE PROBLEM
Introduction
Although the classroom settingis second only to the
family setting in terms of its'Impact upon the behavior of
children, very little is known ofthe environment-behavior
relationships existing there (Gump, 1969). Most studies of
disturbed children, being "skin-bound,"do not speak to the unique contribution of theenvironment in "disturbing
situations." The study of emotionaldisturbance in children
has largely ignored the investigationof the classroom, even though ft is in thisarea where most "disturbance" becomes
apparent or identified. In order to study theenvironment-
behavior relationships ofthe classroom, especiallyas they
affect disturbed children,it would seem most profitableto utilize a naturalistic approachin psychology, to study the
child in his natural setting. This viewpoint is a newly emerging one, the assumptionsand methodology of whichare different from those of thetraditional or laboratoryper- spectives of psychology.
Applied psychology has longbeen identified asa
"clinical" science,deriving its viewpoint andmethodology from its roots in theprofession of medicine. The medical
1 2
concept of pathology and its diagnosis and treatment,within
the disturbed individual, was simply couched innew psycho- logical terms. This conception of psychology still dominates
the i,:e,ion. The rise of the learning theorist,or behav-
ioral school of thought, though still not challengingthe
concept of intrapsychic "pathology," did begin to emphasize
the importance of the externalenvironment or contingencies and their influence upon behavior.
Environmental variables are receivingan increasing
amount of attention. Sells (1969) states that thereare two
contextual prescriptions concerning the contentof psychology:
(a) there is a need for a master planto place psychological research in a phylogenetic perspective; (b)for every species and class there should be a detailed descriptionand quantitative analysis of the environment and the matchingbehavior response repertoire. It is amazing how littlewe know about species and classes of beings outside of "rats,chicks, monkeys, college students, military basic trainees, babiesand hospital patients. . . ." Taking psychology and psychological research out of the laboratory and utilizingnew observational method- ologies in "natural settings" has been theunique contribution of several psychologists who may be looselyorganized under the rubric of "ecological psychology."
Ecological psychology may t.:,ace itsdevelopment back to
Lewin and his concepts of the force field (Lewin,1951), and to
Brunswik's critiques of carefullyarranged experimental research in which contrived experiments distortthe relationships of a natural environment by artificially "tying"or "untying" 3
variables (Brunswik, 1947). Such contemporary ecological
psychologists as Barker and Willems borrow a great dealfrom
the fields of biology and ethology in their attemptsto study
human behavior in an ecologicallly respectablemanner. Their
emphasis upon observation and manseen always in the context
of an ecosystem of some type distinguish themfrom traditional
schools of psychology. This distinction is most sharply
apparent in the ecologist's repudiation of traditional,
experimental methodologies. Barker's statement that "scien-
tific psychology knows nothing, andcan know nothing, about
real-life settings in which people live . . ." (Barker, 1969)
is representative of his objections to traditionalpsychology. Methodology is not, however, the only point of
divergence between ecologically-oriented psychologistsand
those in traditional psychology. The nature of behavior as
seen through an ecological perspective takeson new dimensions.
The concept of emotional disturbance isa case in point. Most
ecological psychologists do notuse a medical model of "mental
illness," disturbanceas a pathology within the individual,
but rather see disturbed ecosystems. Sells (1966) speaks of
a mismatch between circumstance and species. Disturbance is
not seen as bound to the individual, but ratherin the inter-
action of the individual and his specificenvironment. To
understand "disturbance," therefore,one must look at the
critical interface of the individual interactingwith his environment.
Barker (1955) has repeatedly pointedout the need for more information to be gathered concerning the everyday 4
activities and influences upon behaviorexperienced by normal
children. Since the school classroom isa major environment
for children (both in time spent thereand in the myriad of
developmental influences found there),a need arises to observe
very carefully the resultant behavior in thiscritical inter-
action between child and classroomenvironment, and to try to
identify the forces of variablesinfluencing behavior in that
setting. Though such study has many implicationsin and of itself, it would seem evenmore valuable to compare such
data with those data collected fora child labelled "disturbed."
By understanding the influences ofbehavior for both children,
one may be able to shape an environmentmore conducive to child
development and one less likely to producelabels for deviance.
Before such intervention into theenvironment may occur, how-
ever, one must accurately observe, describe,and identify the
environmental forces operatingwithin the environment and the
behavior resulting from this interactionbetween child and environment.
Emotional Disturbance As AnInteractive Phenomenon
The newly emerging sector ofpsychological study based
upon "ecological" or "naturalistic" viewpointscontributes more to the study of behavior than merelyan emphasis upon environmental study. In addition t- offeringa new perspeT.t:ve
through which to view cehaviorand a set of methods for naturalistic study, ecologicalpsychology provides very dif- ferent assumptions regarding causalityof behavior. No longer is an individual separatedfrom his environment for thepurpose 5
of diagnosis. All behavio2 is assumed to bea product of the 10 interaction of the individual and hisenvironment. Such
constructs as "mental illness," "emotional disturbance,"or "schizophrenia" are of littleuse to the "ecological" psycho-
logists since they assume individualpathology--a disease
entity contained within the afflictedindividual. Not only do
these terms dictate "clinical" diagnosticprocedures but point
to the affected individualas the focus for change effects.
Disturbing behavior is viewedas an interactive
phenomenon in ecological psychology. Such a viewpoint demands
the, the disturbing phenomenonbe observed or "diagnosed" as
it occurs in its naturalsetting. Barker (1968), Rhodes
(1967, 1970) though attackingthe problem at different levels,
agree that the phenomenon of disturbance cannotbe seen, under-
stood or interpreted divorced fromits setting.
Implications for intervention generatedby such a
viewpoint place the environment ina much more critical
position. A great deal of "disturbance"can be eliminated by
manipulation of the environment--notmerely to condition or
de-condition the person labelled "disturbed"but to permit a
wider range of behavior which isnot adversely reacted to by
the setting. "Emotional disturbance" is observedand "treated,"
then, as a disruptive interactionrather than as an "illness."
The Prob.,em
The phenomenon of emotional disturbancein children
has been approached primarilyfrom a clinical stance. Though the school forms the settingin which most disturbed children 6
are identified, diagnosed and treated, little is known of the
environmental contingencies operating upon "disturbed"or even
"normal" children in the classroom. If one views disturbance
as interactive, it becomes critical to carefully observe and
compare the interaction of both "normal" and "disturbed"
children in the classroom. This study attempted to identify,
quantify, and compare the environmental forcesoperating upon children in regular elementary classrooms.
The study focused upon the environmentalforces
impinging upon children in the form ofbehavioral demands made
by other persons. Identification and quantification of these environmental forces was achieved through theuse of Schoggen's
Environmental Force Unit or EFU (Schoggen, 1963). These EFU's were delineated from lengthy and detailed narrativeaccounts of observations made by trained observers. Two populations of children were sampled--"normal"children and those children who had been labelled as "disturbed"and recommended for special education. Matched pairs of these childrenwere observed in their ordinary classrooms intwo naturally occurring situations--one in which the childrenwere engaged in an activity which they had individually chosento do, and one in which the activity was planned and administered bythe teacher. These two types of situationswere chosen to investigate the effect of "struct.re" upon behavior. This phenomenon is cited as El important source of variance bymany writers including Gump
(1969), Kelly (1966; 1969) and Gump,Schoggen and Redl (1963).
It was the purpose of the studyto utilize the instru- ment of the EnvironmentalForce Unit to obtain a description 7
of the environmental forces simultaneously impinging upon
:'normal" and "disturbed" children in thesame classroom during
"choice" and "non-choice" situations. These descriptions were then used to ascertain if significant differences in the amount
and types of environmental forces exist between the two types
of children or the two types of settings.
One may then use the categorized Environmental Force
Units as descriptors of the forces impinging upon a subject
from the environment. This quantified description of the
environment may then be used in testing hypotheses concerning
the differences found between the environments of children
labeled "disturbed" and those who are "non-disturbed."
Definitions For This Study
Specimen record.--A detailed, sequential, narrative account of behavior and its immediate environmental context
as seen by skilled observers. The account describes in
concrete particulars the stream of ar individual's behavior
and habitat (Wright, 1967).
Environmental Force Unit (EFU).--An action byan
environmental agent which (1) occurs vis-a-vis the child,
(2) is directed by the agent towarda recognizable end-state with respect to the child, and (3) is recognizedas such by
the child (Schoggen, 1963). The environmental agent must be successful in getting the subject's attention at least long enough to communicate his intention. Unless this recognition by the subject of the agent and his intent is adjudgedto have occurred by the investigator, no EFV is recorded. The 8
investigator uses observercomments and subsequent recorded
behavior in making this judgment.
"Disturbed Child."--A child who forreason of behavior
problems rather than physical anomalies has been referred to
a school-recognized diagnostician and has been found by that
diagnostician as needing special education services. The
child has not, however, beenremoved from the regular class-
room and is considered to be within the "normal"ranize of intelligence.
"Non-disturbed Child."--A child considered tobe with- in the "normal" range of intelligence, whohas no handicapping physical anomalies and who hasnever been referred for special education services.
Matched pair.--Two children who, in their normal and routine placement, are participating in thesame setting simultaneously. One of the children in each pair has been defined as "disturbed," and the otheras "non-disturbed."
The children are matched as toage, sex, and IQ (within 10 points). The children were matchedon these variables since these three characteristicswere most accessible for manipu- latim in the classroom setting and sincethey are considered very potent developmental factors in determining behavior.
Choice setting.--A school-located activity period which is nit pre-plannedor struct-Lred for the child by anoner person. Though the specific behaviors engaged inmay be structured, the child has several alternativeactivities in which he may choose to participate (e.g.,library period, a "free period"). 9
Non-choice setting.--A school-locatedactivity period
which is planned and structuredfor the child. The child has
no alternative but to participate in theimposed activity (e.g., math lesson, reading drill).
The Design
Population sample
Observations were arranged within tenregular elementary
classrooms located in three separateschool districts diverse
in location and populationcharacteristics. The classrooms
consisted of one second-grade,two third-grade, four fourth-
grade, and three fifth-gradeclasses, each of which contained
a matched pair of subjects selected by theprincipal of the
school. The individuals comprising theten matched pairs of
children were never identified byname to assure confidentiality. Code names were used.
Settings within the classroom
Each pair of childrenwas observed for a period of 20
minutes on two separate occasions. One observation was made in a non-choice setting,an activity of the classroom which
is planned and executed bythe teacher; this was usuallyan
arithmetic lesson. The second observationwas done in a
choice setting--an activity whichis not pre-planned or
structured and in which there isno direct intervention by the
teacher. These settings ranged from "freeperiods" to library periods. During both observations, each of thepair was observed in the same settingsimultaneously. The settings 10
observed were suggested by theclassroom teacher.
Data collection
Data were collected in theform of specimen records
as developA by Barker and Wright(itright, 1967) by a team of trained observers taking notesby means of a pocket-sized
dictation device. Preparatory to each observation,the entire classroom was informed of the generalnature of the observation
and familiarized with the equipmentand procedures to be used to make the data collectionas unobtrusive as possible
(Schoggen, 1964). A pair of observers were utilizedin each setting, one to record thebehavior of each of the subjects
in the matched pair. The observers switched subjectshalf-way
through the activity in orderto provide a furthermeasure of observer reliability. The observer team recordedthe behavior of each matched pair ofchildren in both the choiceand non- choice settings.
The notes taken by theobserver teams for each setting were then used in constructing thenarrative typescript which was then edited and divided intoEnvironmental Force Units by two trained investigatorsworking independently. Comparisons of these independentlyderived EFU's were made and disputed identifications discussed andresolved.
Once the specimen recordswere delineated into EFU's, the number of EFU'sper child per setting was determinedand standardized as to lengthof observation (Schoggen, 1963):
Adjusted N of Duration of Shortest Record Actual EFU in Record X Duration of Record X Xnumber of EFU (Variable 1) in record X 11
Two trained investigators. working independently,then categorized each EFU in five ways:
1. Initiation of EFU by Subject or initiation by
Environmental Agent (Variable 2).
2. Subject sole target of EFU or Subjectone of group target (Variable 3).
3. Conflict between goals of Agent and Subject in EFU or no conflict (Variable 4).
4. Method of Agent in EFU adjudged coersiveor method of Agent not coersive (Variable 5).
5. Agent in EFU is an adult or Agent is a peer of the
Subject (Variable 6).
The definitions of most of the above categories (here- after referred to as variables) were taken froma previous study of Schoggen (1963). Variable 5 was defined by this writer.
The data consisted of a total of 40 specimenrecords- - one record per child per setting (10 pairs of childrenx two settings per pair). Each record was delineated into EFU's and each EFU categorized five ways. It was the number of EFU's per child per setting which became the basic unitor score for that child. Each child had a frequency score for each variable in each setting. For example, Subject A hasa score represent- ing the frequency of EFU's initiated by theSubject rather than the Agent (Variable 1) for both the non-choicesetting and the choice setting. 12
Hypotheses
Given an ecological or interactiveviewpoint and a
recognition of the contribution ofthe environment to
"emotional disturbance,"one must pustulate that the environ-
ment of a child labeledas "disturbed" must have character-
istics different from thoseenvironments of children riotso labeled.
Since children labeledas "disturbed" show less
appropriate school behaviorthan non-disturbed children
(Kounin, Friesen & Norton, 1966)one would expect the teacher
to exert much more authorityin terms of both quantity and
style in her interactions withthe "disturbed." This would
be especially true in classroomsituations which she had
structured and over which shehad complete control. In general,
ore would expect more interactions betweena "disturbed" child
and teacher and more interactionswith inherent conflict than one would find betweena teacher and a "non-disturbed" child.
By specifically isolating,describing and comparing the
environment in terms of theseinteractions, one may then be
able to better program interventionefforts.
Using the EFU's categorizedas to the aforementioned schema, the following hypotheseswere tested:
Hypothesis 1.--There will bea significant difference in the number of EnvironmentalForce Units per settingbetween "disturbed" and "non-disturbed"match-mates in both non-choice and choice settings. The greatest differencebetween the
"disturbed" and "non-disturbed"children in each pair will 13
occur in non-choice settings. (Variable 1.)
Hypothesis 2.--There willbe a significant difference
in the number of Force Unitsper setting which were initiated
by the Subject as opposedto the EnvironmentalAgent between "disturbed" and "non-disturbed"match-mates in both non-choice
and choice settings. The greatest differencebetween the
"disturbed" and non-disturbed"children in each pair will
occur in non-choice settings. (Variable 2.)
Hypothesis 3.--There willbe a significant difference in the number of Environmental Force Units per setting inwhich the Subject is the sole target of the EnvironmentalAgent
between "disturbed" and "non-disturbed"match-mates. The
greatest difference betweenthe match-mates willbe in non- choice settings. (Variable 3.)
Hypothesis 4.--Therewill be a significantdifference in the number of EnvironmentalForce Units per setting in which there is a conflict between the goal of theSubject and of the Agent between "disturbed" and their"non-disturbed" match-mates. The greatest differencewill be in non-choice settings. (Variable 4.)
Hypothesis 5.--Therewill be a significantdifference in the number of Environmental Force Unitsper setting in which the method of the Agent is adjudged coercivebetween
"disturbed" and "non-disturbed"mat h-mates. The greatest difference will be in non-choicesettings. (Variable 5.)
Hypothesis 6.--There willbe a significant difference in the number of Environmental Force Unitsper setting in which the Agent is an adult between "disturbed"and "non- 14
disturbed" match-mates. The greatest difference between the
match-mates will be in non-choice settings. (Variable 6.) These hypotheses were tested using a three factor
analysis of variance design wherein Factor A ("disturbed"-
"non- disturbed), Factor B ("choice" setting- "non-choice"
setting) and Factor S (matched pair)were each crossed with
the other two factors. Factors A and B were analyzed as
"fixed effects" and Factor S asa "random effect." Interaction
effects between Factors A and B were also analyzed todetermine
possible multiplicative relationships betweenthe two factors and their relationship to the dependent variable.
Study Limitations
Interpretation of the results of this study shouldbe
made with the sampling technique borne in mind. The size of
the sample, ten pairs of subjects observed onlytwice, seems a
formidable handicap. However, logistical considerations
dictated such size. In order to compensate for this,an
attempt was made to utilize schools representative ofa wide
cross-section of the total area school population. Each of
the schools participating might be said to representdifferent
populations: one an affluent suburban area, anotheran upwardly
mobile working class area, and the thirda small rural district.
The two half-hour periods of observation usedwere
executed to secure a sampling of the child's schoolday.
Though the class was introduced to theobservers, the observa-
tions did not begin until thepresence of the visitor seemed irs to lose its novelty. The teacher was not told who was being 15 observed so as to assure her unbiased participation. The observers themselves were not told which childwas "distLr'bed" and which was not.
The "disturbed" childrenin the sample were chosenby the participating principals. Though this seems tointroduce a confounding variable, descriptions of the referralbehaviors of the subjects were obtained to describe thesample. It was felt that "disturbed" in this study was to bedefined by the school and itsofficial labelingprocess, and therefore the sample was selectedby the school. The range of behaviors labeled as "disturbed" remains a source of possiblemis- interpretation in all studies utilizing more thanone subject. CHAPTER II
LITERATURE REVIEW
Ecological Research In Psychology
The description and analysis ofenvironmental variables in natural settings has beena major thrust of a relatively
new branch of psychology to whichthe rubric "ecological" has
been 3fixed. Though not always easily differentiatedfrom other tjoes of psychologicalresearch, its unique contribution
has been to emphasize theinteractive nature of all human
behavior with its setting. Its impact upon the field,though just now beginning to be felt,may be traced historically and
logically to writers in the fieldsof anthropology (Benedict,
1134; Mead, 1930; Frazer, 1913),ethology (Ardrey, 1966;
Lorenz, 1966), biology (Dubos, 1965)and sociology (Park &
Burgess, 1925; Caven, 1928;Hawley, 1950; Odum, 1953). This
review is, however, limitedto those writers speaking froma psychological perspective.
Many "classic" psychologicalstudies have directed
their efforts at ferretingout environmental influencesupon
behavior in the age-oldnature-nurture controversy (Skeels &
Dye, 1939; Spitz, 1945, 1947;Jenkins & Brown, 1935). The above mentioned studies of childdevelopment were primarily concerned with the effect ofthe immediate environmentupon
16 17
learning or development. Stimulus deprivation was the key
variable studied. Other studies, however, took a more molar
or epidemiological approach to the problem of environmental
influentle. The early studies of Jenkins and Brown (1935),
Faris (1938), and Faris and Dunham (1939) revealed correlational
linkages between such psychological phenomena as intelligence
and mental health with such social variables as socio-economic
level, type of neighborhood and family stability. These
epidemiological studies are representative ofa great deal of
research showing a relationship between social system organiza-
tion and such psychological "pathologies"as mental illness,
alcoholism, delinquency and marital discord (Anderson, 1923;
Zorbaugh, 1929; Shaw, 1929: 'Thaw & McKay, 1931; Sonequist, 1937;
Hollingshead & Redlich, 1958; Fried, 1965).
The Midtown Manhattan Study was undertaken in 1952to
investigate the relationship between mental disorder andthe
sociocultural environment (Langer & Michael, 1963). In its two volume report (Srole, et. al., 1962; Langer & Michael,
1963) a vast amount of datawas assembled showing positive correlations between sociocultural setting andstress situations with mental disorder. The study is also significant in that it was performed jointly by psychiatrists, Rennie and Michael,and sociologists, Langer and Srole. This synthesis of social psychology, sociology and biology J.J often called "Human
Ecology" and is well summarized by Theodorson(1961).
A psychologist dealing with social systems butusing infra-human subjects (rats) isan interesting briuge between these epidemiological studies and manipulatibleresearch. 18
Calhoun (1948; 1960; 1962; 1967) in hisstudies of population
density and its behavioral effects hasshown strong linkages
between crowding or population densityand such individual
deviance as withdrawal, homosexualityand the devouring of
offspring. Though the leap from rat toman in the generaliza-
tion of results is susceptibleto all the barbs generally given
to "rat psychology," the importof the results is most inter-
esting and nit lightly dismissedwhen viewed in light of the
above mentioned epidemiologicalstudies.
These studies in social psychologyprovide a base for
further exploration of theinteraction of man and settings.
Though they are too globalin nature to deal with individual differences or to suggest psychologicalinterventions, a sociologist has utilized epidemiologicalmethods in describing individual children and theirinteractions with school and community. Mercer (1965) describes therelationship of the environment and its labelingdevices to children called
"retarded." It contrasts the clinicalor medical model of retardation with a social systemperspective and reveals school labeled retardatesas role-players in a system ratherthan as afflicted individuals.
Still a step further along intheir concern for individuals rather than ina "systems approach" is the work of
Proshansky at al., at CityCollege of New York (Proshansky,
Ittelson & Rivlin, 1970). While these experiments deal primarily with environmentalhalf of the interaction,they do. study the interactive effectrather than limiting themselves to behavioral contingencies.'Their work, Environmental Psychology 19
(1970) reports these experiments in measuringbehavioral change as the result of the manipulation of the physical environment.
It also, however, provides a liaison between thiswork and the earlier work of the epidemiologically-oriented studies which deal with social conditions on a molar rather than molecular level. It provides a link between the epidemiologists and those primarily concerned with the unique individual inhis natural setting--the "ecological psychologists."
Contemporary studies of "ecological psychology"trace their lineage to the work of Brunswik and Lewin (Barker&
Wright, 1955). Brunswik (1947) defined ecological psychology as the study of the relationship of behavior and itsnon- psychological context. It is the lord "relationship" which sets these studies apart. The interactive nature of behavior is consistently stressed. Behavior is seen as a result of the interaction of a psychological entity,man, and his non- psychological setting, environment. Lewin (1951) spoke of the interaction of psychological and non-psychologicalrealms in his "force field analysis" methodology. It was from Lewin and his emphasis upon social influenceson behavior that Roger
Barker, the most prolific proponent of ecologicalpsychology, took the underpinnings for his theories and methodologies.
A major contribution of Barker to the field liesin his concepf of observing behavior lethin its naturalsettinc..
This has become a cornerstone of ecologicalpsychology.
Barker's thrust away from the laboratory inan attempt to learn how people behave in their everyday liveswas exemplified in his early writing in which he designated hisconcern as one 20
of psychological ecology. Barker states that:
Psychologists know little more than laymenabout the distribution and degree ofoccurrence of their banic phenomena: of punishment, of hostility,of friendliness, of social pressure, of reward,of fear, of frustration. Although we have daily records of the oxygen contentof river water, of the ground temperatures ofcorn fields, of the activity of volcan'es, of the behavior ofnesting robins, of the rate of sodium iodide absorptionby crabs, there have been few scientificrecords of how human mothers care for theiryoung, how teachers behave in the classroom (and how thechildren respond), what families actuallydo and say during mealtime, or how children live their livesfrom the time they wake in the morning untilthey go to sleep at night. (Barker, 1968, p. 2.)
The emphasis upon the study ofman within his natural
setting is also the subject of otherwriters (Gump & Kounin,
1960; Gump & Sutton-Smith,1955). Sells (1963, 1969) discusses
his theory of the "ecological niche"which represents the
adaptive match between circumstancesand species schema. He
cites the need for a master planto place psychological research
in phylegenetic perspective. Such a plan would contain "a detailed description and quantitativeanalysis of both the features of the environment definingthe ecological niche and the matching behaviorresponse repertoire." (Sells, 1969, p. 20.)
Considering the aspirations ofpsychology as a systematic science, it iE soberingto consider (a) how little we know of species outsideof rats, chicks, monkeys, college students,military basic trainees, babies, and hospital patients,and (b) the limited ana ecologicallydistorted strategies and conditions we typicallyuse to study them. (Sells, 1969, p. 19.)
Edwin Willems (1969) speaks tothis need for "natural:- istic research" ina description of the definition, considera- tions, issues, andpurposes of such research. Willems sees 21
naturalistic research as tendingtoward a low manipulation level of the experimentor and a low level in theimposition of units for measurementpurposes.
Wil.lems (1965; 1967; 1971)believes that "ecological
psychology" is not, in fact,a distinct discipline but rather
a set of "orienting attitudes"which may be useful to all
psychologists. Some of these ecologicalattitudes might be
summarized as an appreciationfor the complexity ofbehavior and its interrelationship withsetting, the value of independ- ent observation and in"openmindedness" in investigative perspective.
A major step in accomplishingthe goals of naturalistic research was accomplished when Roger Barker establishedthe Midwest Psychological FieldStation in Oskaloosa, Kansas
(Barker, 1969). Here was provideda natural setting into which as, little intrusionas possible was made by Field
Station staff. After 20 years of quietassimilation into the community, Barker, Wright and their studentsobserved and recorded the everyday livesof the population of thisrural community. Here also Barker developedhis ecologically- oriented construct ofa "behavior setting" (Barker, 1968).
In perhaps an oversimplifieddefinition, a behavior setting cons4sts of a geographicallydefinable and easily delineated milieu or setting whichencompasses a standing pattern of behavior similar instructure to the setting. For example, a football game is a behaviorsetting since it doeshave distinguishable boundariesand encompasses a recurringpattern of behavior congruent with its physical limits. By the same 22
rationale, a drugstore, diningroom, PTA meeting or class- room are behavior settings.
The "mapping" of behavior settingsin Midwest became
a major portion of Barker's work and the theoryand methodology
involved in the process comprise the mainpart of his book,
Ecological Psychology (Barker, 196q). Even in retirement,
Barker is pursuing his development ofuse of the "behavior
setting" as an ecological unit. He is presently examining the behavior settings of several smallcommunities to be flooded out by the creation ofa man-made lake. The study will compare these settings withthose established or modified in the communities in which thepopulation is relocated (Barker & Gump, 1970).
Another common device used to learn aboutbehavior in its natural settingwas developed by Barker and Wright--the
"specimen record" (Barker, 1963; 1969;Barker & Wright, 1951,
1955; Barker, Wright, Barker &Schoggen, 1961; Barker &
Wright, 1949, Schoggen, 1964; Wright, 1967). The "specimen record" is simplya detailed narrative account of an observation taken over a clearly delineatedperiod of time. "Specimen records give a multivariate pictureof the molar and molecular aspects of behavior and situation. They preserve the continuity of behavior andsave for study interrelations between simul-
;a'aeouscl;.,.:. successive conditions. They present undissectec. specimens of behavior and psychologicalsituations" (Wright,
1967,P. 33). One Boy's Day (Barker & Wright, 1951)is an example of a specimen recordcovering one day in the life of a boy in the Midwest. In it one reads of every activity and 23
every verbalization directed to or from the subject. The
observer is allowed to interpret mood and intent within
certain guidelines and the polished recordbecomes a readable
and image-provoking recapitulation of thesubject interactinI with setting.
The specimen record may then be analyzedby various
methods. Students of Barker devised a multitudeof criteria
for dividing records into segmentsor units for analysis and
comparison. Dickman (1963) used specimen records inan experiment to determine whether independentreaders of such records would demonstrate significant agreementon the general patterning or identification of "units" of behavior. Though the results showed poor agreementon identical incidence of units, agreement was reached on the meaningof the sequence.
Dyck (1963) divided specimen recordsinto "social contracts " - -units of social interactioncontaining one subject, one agent, one raison d'etre and one continuoustopic. Through the identification, tabulation andcomparison of such units,
Dyck was able to make quantitativestatements about naturally occurring social phenomena. He described the interaction of children with their parents at homein terms of frequency of contact, initiation of contact andritual-relatedness of contacts. His finding supported the not-surprisinghypothesis that mothers interact more frequentlywith children than do fathers but also shed lightupon the contacts by comparing frequencies in different types ofsituations and in different settings. 24
Other methods of analyzingspecimen records include
"disturbances" (Fawl, 1963) and"episodes" (Jordan, 1963;
Gump, Schoggen & Redl, 1963). Schoggen (-1963) developed the
Environmental Force Unit (EFU) whichwill be discussedmore
fully subsequently. All of these units developedto implement
analysis and use of specimen recordswere defined by the
individual investigators andindependent raters checks made to ascertain reliability.
Not all studies flowingfrom the work of Barkeruse
the specimen record, however. Using Barker's rationale for
naturalistic observation, BettyeM. Caldwell (1969) and
Caldwell, Honig & Wynn (1967)have devised a method for
translating data from recordsof behavior in situ intoa
numerical code suitable forcomputer summarization and
analysis. The APPROACH method (AProcedure for Patterning
Responses of Adults and Children)has developed a coded list
of 1) behaviors likelyto be emitted byyoung children and
adults who interact withthem, and 2) the settings inwhich these behaviors mightoccur. Five digit numerical statements
representing these factorsare recorded by the observer. The
data can then be sortedand tallied through theuse of key words. This technique is still inexploratory stages andno definitive studies haveyet been reported.
Ecological Studies Of Children
Much of the thrust of ecologicalresearch has been directed toward childrenand youth in their naturalsettings. The earliest and mostcomprehensive of the ecological studies 25
done by Barker and his colleagues was centeredaround children.
Midwest and Its Children: The PsychologicalEcology of an
American Town (Barker & Wright, 1954)reports both a detailed
description of ecological methodologY and the results of
observations made through the Midwest PsychologicalField
Station. Extensive use was made of behaviorsetting mapping
and specimen record compilation. This study has become a model or classic prototype for effortsin ecological psychology ranging from the establishment of fieldstations to the observation of children.
The goal of the studywas to describe the psychological living conditions and behavior of the childrenin the town of
Midwest and of some children with physicalhandicaps living in neighboring communities. Midwest was a small rural community in Eastern Kansas of 721 peoplein 1951/52 and study of the results should be interpreted withthose parameters. While it is impossible to summarizeall the study's extensive results, the following are representativeof their range and nature.
It was found that 585 communitysettings existed and that infants participated in 60%of them. This range of activity increased to nearly 79% ofthe settings by adolescence.
The behavior objects used by childrenduring a day numbered from 1,892 to 2,490 while the peak'ate of objection trans- actions occurred in the middle of theday. The number of things a child did duringa day, according to a criterion of behavior episodes, varied from 500to 1,300, with the duration of episodes ranging froma few seconds to 121 minutes. Seventy 26 percent of them were shorter than 2 minutes,however. Age of the child was determined a criteriafactor in amount, duration and nature ofthese episodes.
Examination of the socialinteraction of children revealed that "children lived their days withadults in a benevolent autocracy"(Barker & Wright, 1954,p. 467). Relation- ships with other children, however,were not authoritarian in nature but rather were dominance patterns metby counter- dominance. The children "generallygave what they got from all others in affection, mood, andevaluation." The most frequent single kind of social interplaywas a friendly inter- action short of cooperation which occurredin 60% of 100 episodes. Discord short of conflictoccurred in 25 of 100 episodes and in only 5 of 100 episodesdid children get into outright conflict with associates. On the other handfull cooperation was recorded in only 1 episodeper 100. Age again was found to be a criticalvariable.
With increase inage of children there occurred: gain in the complexity a of social partnersas shown by an increase in groupings of 2 ormore individuals as associates; a decrease in the frequency ofadults and especially ofparents as associates; in child associates an increase . . .; an increase in the of power and in amount fluctuation of power . . .; a definite power gain in interactions with other children . . .; an increase in harmoniousrelations and conflict with other a decrease in children, and a decreasein con- flict with adults. (Barker & Wright, 1954,p. 470.) In summary the data yielded by theMidwest study ib probably the greatest amount of informationconcerning the everyday lives ofchildren available. Its rural Midwestern origin is, however, a great limitation ingeneralizing about contemporary children. 27
"Streams of behavior,"as specimen recordsare some-
times called, were alsoused in much thesame manner as in Midwest in England (Barker & Barker, 1963; Schoggen,Barker, & Barker, 1963). Specimen records werecollected in 51
behavior settings ofa small rural community inEngland and compared with those taken in 51 settings in Midwest. A method utilizing a standard biological field procedurewas used in delineating a large behavior unit, a "social action." The
presence of social actions inbehavior settingswas tallied as well as qualitative categorizations of them made. The results were summarized by the statement that ". . . most of the social actions were common to the behaviorstreams of the American and English children, and that theydid not differ in their breadth of distribution across theAmerican and
English settings. The social life ofAmerican and English children is, according to our data, basicallythe same. The differences are differences of emphasis" (Barker &Barker, 1963, p. 159).
The structural characteristics of thesebehavioral streams in the cross-culturalstudy were then examined
(Schoggen, Barker &Barker, 1963) bymeans of the "behavior
episode" unit. Episodes were tabulatedand then categorized as to nature of initiation and termination. Results of the study rein:creed the conclusions ol Barker andBarker (1963; that the differences in the behavior streams 1)were few, 2) were more related to dynamics than to structure and 3)their locus resided relatively more often in the American childand relatively more oftenin the English environment. 28
In an approach somewhat dissimilar to Barker, James
Kelly (1966; 1969) has concentrated his effortsupon the study of adolescents in "fluid" as opposed to "constant" high schools.
The former type of high school is characterized by 42 percent of the entrants leaving during the school year and the latter by less than 10 percent of the entrants leaving during the year. He selected observation sites within the schools with the chief criteria being a high probability of representing effects of population change. Each site was classified along an "Environment-Personal" dimension and a "Public-Private" dimension. Utilizing data gathered from the observations
(slightly different methods were used for each site) anddata gathered through interviews, Kelly was able tocompare student class structure, status criteria, variability in behaviorand qualitative differences in behavior between thetwo schools.
His hypothesis that the fluid environment wouldcontain much more variability in behavior, different style of communication, less rigidity in social structure and evena different adminis- trative style were supported.
Kelly's study is similar to an earlier one done by
Barker and Gump (1964) entitled Big School, Small School. In reality a series of studies, Big School, Small School reports the findings of the investigators in high schoolsranging from an enrollment of 35 to one of 2,287. The most crucial studies were made comparing high school juniors in four schools of 83 to 151 students with those in a high school ofover 2,000 students. The schools were homogeneous in regard to economic cultural and political region. Both specimen records and 29 behavior setting mapping techniques were used inaddition to an activity questionnaire. The large and smallschools were then compared acrossmultiple dimensions.
One of the majorfindings of the reportwas concerned with student participation in school activities. It was reported that the proportion of students participatingin
extracurricular activitiesin small schoolswas 3 to 20 times
greater than in the largeschool. The small school students participated in greater numbers and more varietiesof activities than did large school students even though thelarge school provided a larger number and greater variety ofnon-class behavior settings. The satisfactionsexpressed by the two
groups of students were differentwith small school students reporting rewards from the development ofcompetence and being involved in activities while large schoolstudents gained
satisfaction from vicariousactivities and large entity
affiliation. Students of small schoolsalso reported more attractions and pressures toward participationin school non- class behavior settings than did large schoolstudents.
Because of technicalreasons, most of the study's
findings referredto non-class activities,but a study of school structure revealed that formal educationalbehavior settings constituted about the same proportion ofthe settings of schools of all sizes. Barker all' lump (1964)summarize their findings by stating that their data eid theirown educational values dictate that a school be sufficientlysmall so that all of its studentsare needed for its enterprises. "A school should be small enough thatits students are not 30
redundant" (Barker & Gump, 1964,p. 202).
Paul Gump (1967, 1969) again utilizedthe construct
of the behavior setting in hisstudy of a third grade class-
room. Specimen records of the teacherwere also taken and
divided into "segments"or naturally occurring constellations
or patterns (i.e., small readinggroup activity, class lecture,
etc.). Every segment was seen to havea site, an object and an
activity pattern consisting of (a)the nature of the teacher's
participation, (b) the groupinga. pupils, (c) the prescribed
action relationships between pupils,(d) the kinds of actions
taken by the pupils and (e) thepace of pupil action (Gump, 1969).
Six third grade classroomsserved as target populations
and observations made for twodays in each. The obtained
"segments" were then comparedsystematically across the above
dimensions. Gump found that pupil involvementreached 92% in
externally-paced small group segments(reading circle) as
opposed to 72% involvementfor self-paced groups (seatwork). The phase of the segment also seemed to be a criticalvariable, pupils becomingmore involved as the activity progressed.
Teacher "acts" were delineatedin the records and usedas units of teacher behavior. An "act" is the shortest meaningfulbit
of behavior directed towardchildren, usuallyone sentence
statements such as "All right,order. your books!" Teachers
averaged 1300 acts per classroomday. One of the more revealing
conclusions of the studyrevolved about "acts" classifiedas "Dealing with DeviatingBehavior." Such acts seemed heavily concentrated in "transition"periods or those activities which 31
close down one segment and setup another. It is during these
"transition activities" that "pupilbehavior becomes more
individual and some of the individualisminvolved behavior
divergent from that desired bythe teacher. She responds by increasing pressure, countering and rejecting moves . . ."
(Gump, 1969, p. 218). The transition activities then would
seem to constitute a fertile area forfurther study.
Preschool children have been thefocus of a series of
ecological studies conducted attne Demonstration and Research
Center for Early Education atGeorge Peabody College for
Teachers. Maxine Schoggen (1969) has collecteddata to provide
a library of specimen recordson the behavior of 24 three-year-
old children from threedifferent socio-economic populations:
low-income urban, low-income ruraland middle-income urban. The specimen recordswere divided into Environmental Force
Units and the rate of EFU'scomputed for each child. The
percent of EFU's in whichthe mother acted as agentwas
determined but nocross-group statistical measured used. These records, however, have provided thedata for several other
studies. Some of these data have alsobeen used in a "resource"
book, Children Learning (Shaw& Schoggen, 1969) to be used in
training adults to work withchildren. Brown (1969) used specimen records taken inthe homes of three-year-old children
from thres distinct socio-economic:ackgrounds: low-income
urban, low-income rural,and middle-income homes. "Behavior objects" were identifiedfrom specimen records covering 90 minutes of the child's homelife. "Behavior objects" were defined as a commonlydiscriminated part of the physical- 32
social environment which is perceived asnecessary or
appropriate for a particular behavior (Brown, 1969). Both
a frequency count and a classification schemewere used. The
study concluded that children from lowerincome homes as
compared to a middle income home dealt withfewer different
objects, were engaged in more behavior withtheir mother, were provided with fewer nutritional objects, exhibited less
verbal behavior and more motor behavior,and used objects
according to their intendedpurpose less often.
Ecological Studies Of "Disturbed" Children
Though studies involving populations ofchildren who
have been labeled as "disturbed" have beenperformed using
ecological methodologies, few of the settingsin which the
children were observed included the school. Gump, Schoggen &
Redl (1963) analyzed the behavior ofa "disturbed" boy in a
camp setting and then at home using the specimenrecord
technique of observation. Two days c-f observation,one day
at camp and one day at home, comprisethe data. The specimen records when divided into episodes (Barker, 1963)were then compared with the following results:
1) A much wider variety ofbehavior settings existed
at camp, many designed for children. The subject
engaged in active, exploratory, constructiveplay
at camp and in dallying and formallycompetitive
play at home. More strongly positive and
ambivalent emotionality occurred atcamp.
2) Camp associateswhen adults extended to the subject 33
more interest-centered, less aggressive and
less resistant social behaviors. Peer
associates at camp wereon a more equalitarian
social relationship than at home.
Many similarities in home andcamp were also found in a some-
what similar play pattern, the dominanceof adults and the
discovery that thesubject dominated weaker youngsters in both
settings. This study, limited as it is witha population of
one, is an attempt to begin to explore scientificmethods of
selecting and designing settingsrather than to test a hypothesis.
Jordan (1963) used the behavior episodetaken from
specimen records to investigate thebehavior of two eleven-year- old boys in a ward ofa residential hospital for disturbed
children. Observations were made for halfhour intervals over
a period of four months during tutoring sessionswith a special
teacher. Some of the observationswere done under a policy
that the subjects must stay inthe tutoring session but could do what they pleased, andsome were done under the condition that they need not attendthe session but if they did they must do work relevant to the tutoringobjectives. The two conditions did not show significantdifferences in the number of episodes for each childper session nor did it have effect upon episcle initiation, terminati-nor maintenance. When a comparison of the episodeswas made with data from the Midwest children of Barker and Wright (1955),striking similarities were found. These were attributed to the effectivenessof the hygenic milieu approach (Redl& Wineman, 1951) used by 34
the hospital. The effect of the change of educational policy seemed only to limit variation in the spontaneity of episode
initiation--more were initiated by the teacher.
In a different type of study, Raush, Dittmanand
Taylor (1959a; 1959b) described the social behavior ofa small group of hyperaggressive boys and analyzed changes in such behavior over a period of one anda half years in residential treatment. An interaction analysis code modified from Freeman et. al. (1951) and Leary (1957)was used in the analysis. The results on the study simply describedthe patterns of behavior discerned. Raush, Farbman and Llewellyn
(1960) replicated the study butintroduced a control group of boys matched as to age and race but drawn fromthe regular public schools and only temporarily housed inthe residential setting. The same observations and coding procedureswere utilized, for each group of six andan information transmission analysis used as a statisticalmeasure of significance. The results were summarized as follows:
1. Although the disturbed boys, at the later treatment
stage, differed from normal boys insome aspects of
behavior toward adults, theycame to act much more like
the normal children than they had early in treatment.
The changes in the patientswere not paralleled by dif-
ferenues between older and younger normal children;changes
in the patients; relations with adultswere, therefore,
judged attributable to treatment, ratherthan age, per se.
Similarly, in interactions withpeers, increase in ap-
propriateness of socialresponses seemed related to 35
treatment rather than toage changes. In several aspects of behavior towardpeers, however, the disturbed children
had failed to changeappreciably; control data showed
that the lack of changecould not be attributedto lack of disturbance.
2. As compared to theten-year-old normal boy,the normal boy approachingtwelve seemed to be progressing
toward relations ofgreater status equality with
adults. In contrast, thehyperaggressive children, who in the early stage had shown relatively littleovert dependent behavior, hadcome to exhibit increased
dependency toward adults,approaching normal age-mates in this respect. The results impliedthe dissolution
of a defensivecounter-dependency toward adults,
approaching normal age-matesin this respect. The results implied the dissolution of a defensivecounter-dependency through treatment ofthe disturbed children.
3. The behavior that thechildren received frompeers
closely paralleled thebehavior they directedtoward peers. Although the adults'behavior toward the children was related tothe behavior they received from children, the adultsproduced a far lower proportion of hostileactions than they received, particularly with thedisturbed children. The im- plications were thatthe forces for change insocial behavior. derived fromadults. A more general implication questionsthe presumption of a constant 36
social environment through which theindividual acts
while the world around him remainsthe same. The
data suggests rather, a reciprocalprocess in which
the person's effects on others and other'seffects on him are in continuous dynamic interchange.
4. Information transmission analyses showedthat the
normal children varied their behavior inaccordance with
the specific social setting. The normal children tended
to differentiate among socialsettings more than did the
disturbed children. In their actions normal children
also distinguished between theirpeers and adults to a
greater degree than did the patients. The results suggest
the relevance of concepts relatingdifferentiation to psychopathology.
5. The social setting seemeda more potent determinant r of affectional relations than ofstatus relations, where
as the status roles of the participants seemedmore
relevant to status relations thanto affectional relations.
In all groups individualdifferences tended to be greater
in relations withpeers than in relations with adults;
this suggests thata group may be more stereotyped toward
other-status recipients of actionthan it is within its own menbers.
6. Normal and disturbed childrenshowed, in general, the
same behavioral tendencies toward specificsettings.
Food settings were associatedwith relatively friendly I
I 37
actions for all groups, andcompetitive games were
associated with unfriendlybehaviors.
7. As had been true for thedisturbed children,so for the normal children, too, the iLteractive effectsbetwee.i child and setting contributedfar more information about
tae behavior than did thesum of the independent components. Thus, it would seem that the kind of behaviora setting
evokes would be toa considerable extent relatedto the
dimensions of the particularsituations that are salient to him, and that this is true for both normal anddisturbed children.
8. In general, the twonormal groups resembledeach other in their response to different settings. There was a suggestion that, inan informal setting, older
children, as compared withyounger children, may be freer to express hostility toward adults andmore controlled in relations withpeers.
Kounin, Friesen & Norton (1966)used videotapes inan attempt to delineate some dimensions of teaching stylesthat
affect emotionallydisturbed children ingrades 1-2 and 3-5 in regular classrooms. Data were drawn from 30classroom for 1/2 day each. Scores for deviancy andwork involvementwere
obtained fir both "disturbed"and "Lon-disturbed" children. The results indicated:
I. Pupils' scoresvary between seat work and i recitation settings. 38
2. Disturbed children exhibited lessschool-appropriate behavior than non-disturbed children.
3. Teachers who are successful inmanaging the behavior
of non-disturbed childrenare also relatively successful with disturbed children.
4. Techniques of handling misbehaviordo not always
correlate with the children'sbehavior.
5. Teacher attitude, technique inhandling group move-
ment, and programming forvariety change in learning
activities do correlate withchildren's behavior.
Other discussions ofecologically-oriented interventionshave appeared (Kelly, 1970), buthave been theoretical rather than functional in nature.
Though ecological psychologistshave labeled their
studies purely descriptiveand usually avoided the formulation
of theory, am ecologicaltheory concerning disturbanceand
"disturbed children" isproposed by Rhodes (1967). In this
view "disturbance" isseen as a reciprocal exchange betweena
child perceived as deviantin thought, word or deedand the
reacting environment. Such a viewpoint would precludea
physiological or psychodynamic"gremlin" approach and would
tend to place more emphasisupon bringing pressures to bear
upon the environment ratherthan upon the "offending" child.
Rhodes further developsthis thesis in describing therole of
the "community" indisturbance (Rhodes, 1970). Citing ecological studies ofethologists in which creatures deviant
I 39 in odor or appearance incite mayhem in the colony(Lorenz, 1966; Calhoun, 1962), Rhodes compares thiscommunity reaction to its human counterparts. Again the conclusionis reached that ecologically oriented approaches to the problemsof children must be undertaken in order tosuccessfully overcome them, and the intervention must deal with boththe child and the environment.
The time has come to begin to concentrateattention upon changing the ecologicalconditions under which children have to live andgrow, and thus reduce the number of occasions ofdisturbance and the number of children who are extruded or alienated fromtheir living units. (Rhodes, 1970,p. 313.)
Methodological Concerns The study of children in natural environmentsrequires methodologies and techniquesquite unlike those appropriate for the laboratoryor traditional experimental-manipulation-of- variable models of research. New methods of datacollection and analysis are still in the process ofevaluation by some ecological psychologists.
The chief data gatheringtechnique in ecological psychology is that ofthe specimen record. The basic procedures for learning this technique are enumeratedin Recording and
Analyzing Child Behavior(Wright, 1967). These include instructions as to observational period,note-taking, timing, dictation and revision. The observe., is giventhe following rules of reporting. (These will be usedin this study.) 1. Focus upon the behavior and the situation ofthe subject.
2. Observe and report as fully as possible thesituation 40
of the subject.
3. Never make interpretations carry the burden of
description.
4. Give the "how" of everything the subjectdoes.
5. Give the "how" of everything done byany person who interacts with the subject.
6. Report in order, all the main steps throughthe course of every action by the subject.
7. Wherever possible, state description of behavior positively.
8. Describe in some detail thescene as it is when
each behavior setting is entered.
9. Put no more than one unit of molarbehavior in one sentence.
10. Put in one sentence no more thanone thing done by a person in the situation of the subject.
11. Do not report observations interms of time intervals.
The use of mechanical aids for makingspecimen records is described by Schoggenas developed for his studies (Schoggen,
1964). The use of a mask-like deviceand tape-recorder as described would seem to expediteboth the utility and efficacy of the specimen recordas a data gathering device. Barker
(1969) als) considersthe field sta'Aon as an importanttool to be used in naturalisticmethod of psychological research.
The division of the "stream of behavior"as captured by the specimen record intosmaller units for analysis has been enumerated ina previous section. The "Environmental
- - . 41
Force Unit" (EFU) developed by Schoggen would appear to bea
very viable tool in assessing the influence of setting upon
environment. An EFU is defined as an occurrence of action
initiated upon the child, directed towardsome end, and of
which the child was aware (Schoggen, 1963). Utilizing these units he was able to show how the EFU's were distributed in
each setting, how long they lasted, who the agents involved
were, how they conflicted with the child's wishes and to what end the actions were directed.
Simons & Schoggen (1963) utilized the EFU in studying mothers and fathers as sources of environmentalpressure on
children. The EFU containing conflict and involving either mother or father were utilized from the specimenrecords of eleven children. The numbers of EFU and conflict EFUwere
tallied as well as the method and the number of methodsused
(as defined by the investigator). No statistical methods were employed to determine significant differences. Only the percentages were listed.
Schoggen's initial study (Schoggen, 1963) in whichthe
EFU was developed utilized 18 specimenrecords, the following criteria were examined: duration of EFU;rate of EFU; age and role of agent; conflictual EFU; methods ofagents (27 were identified) and the goals ofenvironmental agents (29 were identifiei,. Interrater reliabili.y was utilized to provic...1 some measure of objectivity in the assignment of EFU's into the various categories created bySchoggen.
After the design and methodology for this studyhad been formulated, an unpublishedstudy of Schoggen (1968) was 42
found which may be considered a parallel. In it, specimen
records were collected for matched pairsof children in class-
rooms and at home. One child in each pair hada clearly
visible peripheral motor impairmentand one did not. The same observer, however, watched both childrenin the pair on successive days. Environmental Force Units were delineated and categorized. The following, rather surprisingresults were indicated:
1. Other persons reached outor responded to children
with disabilities neithermore nor less frequently than to non-disabled children.
2. In general, children withdisabilities were not
singled out for special individualattention more frequently than non-disabled children.
3. Intervention by others in the ongoingactivities
of the child, "butting in," tendedto occur more
frequently in the associates of disabledchildren,
but the trend was not statisticallysignificant.
4. Attempts to benefit the subjectin some way did
not, in general,occur more frequently with the disabled.
5. Con"lict between subjectan associates occurred
neither more nor less frequently forthe disabled.
6. There was no evidence thatother persons gave more
freely of time andenergy to the disabled. 43
7. Demands for the subject to carry cut an activity
occurred as frequently for the disabled as for the
other subjects.
This study of the di::_led by Schoggen is one of the few ecological studies which approach an experimental design.
Most studies in "ecological psychology" are of a purely descriptive nature.
In summary, the area of "ecological psychology" is still in the process of "becoming." Whether it develops into a full-fledged interactional theory with its own vocabulary, constructs and methodology a la Barker or becomes a perspective or viewnoint re-shaping traditional psychological research as with Willems, has yet to be determined. This process of tin - folding, however, has to date, limited research efforts to purely descriptivc: studies. More experimental models of research may develop in the future but not before specific facility-operation problems are resolved. Barker (1969) enumerates three such problems: (1) an archival problem, or the accumulation of raw observational data; (2) a field station problem, the paucity of vantage points from which to view natural human behavior in an unbiased manner; and (3) an analytical problem, the development of measures and techniques to analyze without destroying the value of the phenomena observed. He co-oludes, "These prelem9 are great, but the possibility of arrivIng at scientifically meaningful and socially useful results are also great, and they justify the commitment of intellect, effort, and money to the task" (Barker,
1969, p. 41). 44
Synthesis and Application
The review of the literature reveals a collection of disparate studies whose relationship one to another often seems limited to the word " ecological" found somewhere in thi title. Psychologists enter their "ecological" studies with diverse backgrounds and theoretical viewpoints--from clinical psychology (Brunswik, 1947; Raush, Dittmann & Taylor, 1959a;
1959b); sociology (Fried, 1965; Langer & Michael, 1963), experimental psychology (Calhoun, 1967; Proshansky, et al.,
1970), community psychology (Kelly, 1966; 1969) andmore esoteric specialities. The studies of "emotional disturbance" in particular indicate very different assumptions underlying the concept of "disturbance" or its definitions. One en- counters unstated but evident psycho-dynamic theory in the work of Raush, Dittmann and Taylor (1959a; 1959b) while the
"disturbance" seen by Kounin, Friesen & Norton (1964)or
Rhodes (1967; 1970) has a much more distinctive sociological aura as found in labeling theory. The studies, themselies, involve widely divergent populations, from rats ina laboratory
(Calhoun, 1967) to the population of entirecommunities (Langer
& Michael, 1963; Barker, 1954). While some authors attempt to lay a foundation for a new body of theory with which to determine the parameters of "ecological psychology" (Barker,
1968; Rhodes, 1967; 1970), others declare ecologicalpsychology to be a point of view or perspective whichcan be utilized by orthodox behaviorists, clinicians or by other disciplines
(Willems, 1967).
1 45
While the diversity and seemingunrelatedness of much
of the literature may bogglethe mind, at least fourcommon
themes or issues may be gleanedfrom this survey. These
themes serve as the justification forthis study and the view- point from which itmay be interpreted:
1. The importance of observinghuman behavior in its natural setting;
2. the focus of the psychologistupon the interactive nature of behavior;
3. the influence of settingupon behavior;
4. the attempt to developa methodology which will
objectify and quantifyobservational data.
Nearly all of the writers,somewhere within their study, emphasize the importanceof understanding human behavior in the context of its naturalenvironment. With the exception of Calhoun, all studiesconcentrated upon gathering datain as unobtrusive a methodas possible in the normal habitat of the subject. The assumption that humanbehavior can best be under- stood in its naturallyoccurring context is basic to this study. Flowing from this is the assumptionthat effective intervention techniquescan best be developed using natural environments rather than laboratorysettings or artificial behavior (as in testing).
Ur the studies reviewed, virtuallyall were concerned with observing, recordingand analyzing the interactive components of behavior. This interaction may be between subject and physical settingor subject and the complex described as "behavior setting"(Blrker, 1968). In either 46
case, the examination cf intra-psychicdynamics or of the
nature of the environmental contingenciesis made subordinate
to the dynamics of interactionbetween subject and all aspects
of setting. The interpretation of thisinterae:tion may to
-lade in behavioral, psychodynamic,sociological or bicgenic
terms but an attempt is madeto capture the nature and efforLz of the interactionas completely as possible.
Though an interactive focusis evident in the studies
cited, all emphasize theheretofore underestimatedinfluence
of setting upon behavior. This powerful influenceis seen as not limited to the rewardingor punishing environmental
contingencies, but is viewed ina broader context. Barker
(1968) seems tosum it up well when he states thatone may
more effectively predict a person'sbehavior at a football
game by understanding the setting thanby garnering vast
amounts of test-derivedor case-history related information.
Even if somewhat oversimplified,the point of the statement,
that setting is too oftenoverlooked in examiningbehavior, is another primary assumptionfor this study.
Though nearly all ofthe studies cited viewedthem- selves as purely descriptive in nature, none followedthe
traditional psychologicaldata-gathering devices of testing,
interview or case history. Observation is the primarymethod in these studies. All, however, makean attempt to depart
from traditional subjectiveand intuitive "skills" inrecording and analyzing data. The use of inter-ratereliability tests and the development ofcategorical schema for quantifying observational data representan attempt to study behavior
t 47
holistically while describing results inas objective and
replicable a manner as possible. Though many hazards are
found in the attempt to integrate this holisticobservational method with "scientific analysis," this writerfeels that the direction being taken is a productiveone and that more effective methodologies will evolve.
It is from these assumptions thatthis writer formulated and executed the present study. CHAPTER III
PROCEDURE AND ANALYSIS
Population Of The Study
The schools
School districts participatingin the study were selected with a criterion ofheterogeneity. An attempt was made to observe in schools whichrepresented a wide cross section of school districts interms of socioeconomic status, urban characteristics and sizeof district. Observations were arranged in the following locations:
1. A large suburban districtserving over 30,000 children whose population consists primarilyof white-collar and professional workers. The fine reputation ofits schools was bolstered by its total operatingrevenue of $902 per pupil.
Six pairs of children fromthree different schoolswere selected as subjects.
2. A moderately large suburbandistrict serving 14,000 children whose population iscomprised of upwardly mobile blue and whtte-eollar workers. The school budget allotaa total operating revenue of $731per pupil. Two pairs of children in one school buildingwere selected as subjects.
3. A small rural district serving2,131 pupils located approximately 25 miles from thecity. Its blue-collar
48 49
population lives on farms but works in nearby auto factories.
Its total operating revenue per pupil is $706. Two pairs of children in one school building were observed.
The specific schools cooperating in the study within
each district were selected on the basis of the interest of
,the building principal in the study. The building principal acted as liaison between the investigator and the teachers.
He helped select those teachers in his buildingwho would be least threatened or conscious of observers and whose class-
rooms would be appropriate to the study in terms of their
inclusion of "disturbed" children. Because of this method of
selection, there were no "turn downs." The principals were
given full information about the study while theteachers were told only that the study was concerned with total class-
room interactions. The children were told that observers
were wanting to learn more about exactly what goes on ina classroom.
The selection procedure used has several implications
for interpretation of the results of thestudy. Since
cooperation on the part of the principalwas so important and
since he selected the teachers, his personalbiases regarding
"favorable teaching styles" must enterinto the classroom
interactions observed. The criteria,that the teachers not be
threatened by observers also may mare systematicallyintroduced
a bias toward a distinctive teaching style. Such considerations
must be kept in mind as the resultsare reviewed. 50
The subjects
The ten "disturbed" children who were subjects of the
study ranged from seven years of age to eleven years. One
of the subjects was in the second grade, two in the third
grade, four in fourth grade and three in the fifth grade.
While a preference for primary grades was stated, grades selecl,ed reflected the availability of "matched pairs." All
subjects had been referred for special education services for the "disturbed" and were in regular classrooms but receiv- ing some special help from an itinerant or "resource-type" special education teacher on a weekly basis. Reasons for referral were as follows:
Subject 1. Ten years of age in the fourth grade; referred
for special services one year ago for "school phobia"
and an obsession with death; IQ in normal range.
'(Female.)
Subject 2. Eleven years of age In the fifth grade;
referred for special services one year ago for
"acting out" aggressive behavior; IQ in low
normal range. (Male.)
Subject 3. Nine years of age in the third grade;
referred for special services one year ago for
"acting out" and aggressive, impulsive behavior;
IQ in low normal range. (Male.)
Subject 4. Ten years of age in fourth grade; referred
for special services two years ago for "bizarre"
behavior and poor peer relationships; IQ in normal
range. (Male.) 51
Subject 5. Eight years of age in i.hird grade;referred for special services one yearago for aggressive
"acting out" behavior; IQ in normalrange. (Male.)
Subject 6. Ten years of age in fifth grade;referred for special services fiveyears ago for "acting
out" behavior; IQ in low normalrange. (Male.)
Subject 7. Nine years of age in fourth grade;referred for special services fouryears ago for "acting out"
and "sneaky" behavior; IQ in lownormal range. (Male.)
Subject 8. Nine years of age in fourth grade;referred
for special services two yearsago for "acting out"
and temper tantrums; IQ in normalrange. (Male.)
Subject 9. Seven years of age in second grade;referred
for special services six monthsago for "acting out"
and lack of school achievement; IQin low normal
range. (Male.)
Subject 10. Eleven years of age in fifth grade;referred for special services fiveyears ago for "acting out"
and lack of school achievement;IQ in low normal
range. (Male.)
All of the subjects were consideredto be of normal intelligence and all achieved poorlyin school. Nine of the ten were males. Again the nature of the samplewas determined by availability of subjects. In the case of sex, however,tne sample accurately reflects the totalpopulation of the study since of all "disturbed" childrenin public schools, the great majority are males. 52
Subjects 11-20 were the "normal" matched matesof the
"disturbed" subject 1-10. In these pairs, subjects 1 and 11,
2 and 12, etc. were matchedas to sex, IQ (within 10 points),
and age (within 6 months). The chief difference between the
subjects in each pair was thatone subject had been labeled as
"disturbed" by tI'e school while the otherhad not.
All "disturbed" subjects were selected bythe principal
of the school using the criteria ofthe investigator that they be of "normal" intelligence, ina regular classroom and diagnosed as needing special services dueto "emotional disturbance." The "normal" subjects were selectedby the principal by his matching themon the basis of room membership, age, sex and IQ. The subjects were pointed out bythe principal to the investigator withno names being used and no school records being seen by theinvestigators to ensure confidentiality. The subjects were not identifiedas "normal" or "disturbed" to the observer to avoid observerbias. The teacher and the pupils in the classroomsknew only that the class activity was being observedand not that specific individuals were of primeconcern.
Data Collection
Observer training
The collection of data in the form ofspecimen records is necessarily done by skilledobservers familiar with the technique. Since no such skilled personnelwere available, the task of training at leasttwo observers was an initial concern of the study. 53
Personnel were recruited using the services ofstudent
placement offices in several local universities. Criteria
used in the selection of the observers includedmaturity;
experience with and concern for children,and where possible, teaching experience. The team selected was comprised of two women, both of whom had B.A.'s, one of whom hadan M.S. and had taught in special education fora year.
The design and scope of the studywas explained to the trainees and a copy of the proceduredetailing the process of taking a specimen record presentedto them (Wright, 1967).
The trainees were then familiarizedwith the equipment to be used in the data collection:a pocket sized dictation device and a portable tape recorder. After they felt comfortable with the equipment they were encouragedto begin "taking notes" for a specimen record usingthe dictaphone while observing routine office activity. They then progressed to taking notes while watching videotapes of children in interview situations. At this point the trainees then used the notes for dictating the observationalnarrative into the portable tape recorder for the roughdraft of the specimen record. These tapes were typed and read byan investigator who had not participated in that observationwho assisted the trainee in the editing of the tapescript. The edited tapescript., were then compared ast the congruency of the observation. The typescripts were then evaluated usingthe criteria for specimen records developedby Wright (1967).
After the specimen recordswere found to be in agreement regarding what was observed and when they fulfilledthe 54
criteria established by Wright, the actual observationswere scheduled.
The investigator andtwo observers met withthe teachers whose classes were to be observed to explainthe
study and to introducethemselves. To guard against teacher bias, the teachers were told only that the studywas concerned
with pupil-teacher andpupil-peer interactions. If the teacher asked if specific children were to be watchedshe was told that the observerswere to observe total class inter- action. None of the teachers involved appeared anxiousabout the observations. They were told thatthey could introduce
the observers to theirclasses if they liked andthat the observers could introduce themselves and their equipmentif so desired. All of the teachers,however, informed their
classes of the observersahead of time andno formal intro- ductions were made.
The teachers were asked to select twotwenty-minute periods during the day for the observationsin their room.
They were asked toindicate one period inwhich the children had a choice in their activity and one periodin which the children had no choice but were told what to do. The order of the observations was not a variable in the study. These periods were then scheduled for observation and theobservers reported to the rooms one-half hour priorto each activity period to provide the class a "period of adjustment"to the observer. In all instances, the "non-choice"or structured setting was an arithmetic or reading lesson. The "choice setting" observed was usually a "free" period inwhich 55
children could do seat work, playgames or talk quietly,
though one unstructured art lessonwas observed and one period in a "resource materials center."
The observers took verbalnotes with dictaphone during
the scheduled period, delineatingone-minute intervals as time
references. Afterward these notes were utilizedin dictating
a narrative account of the observationinto a portable tape-
recorder. Typescripts were then made, editedand final copies
used for the process ofdelineating Environmental ForceUnits.
Approximately 600 pages oftypescript were produced. (See
Appendix A for sample specimenrecord typescript.)
Environmental Force Unit CodingAnd Categorization of the Data
The investigator had beentaught the skills involved
in delineating EnvironmentalForce Units using the definitions,
study materials, and "rulesof thumb" of Schoggen. Sample
specimen records which hadbeen EFU'd by Schoggenwere obtained and used in training.
Four individuals were trainedto identify Environmental
Force Units (EFU's) in thespecimen records of behavior compiled by the observers. The purpose of the studywas explained and didactic material concerningthe identificat4ln of EFU's presented (3choggen, 1963). The coders practiced the identifica- tion of EI..'s on plainspecimen rect.rds whichwere then checited against the EFU's found inthat specific record by Mrs. Schoggen. After some proficiency was developed the coders would EFUa sample record and thencompare them among themselves. When the reliability quotientreached 80% agreementon the location 56
and label of EFU's, the coders began workon the data of the
study. Two coders divided each specimen record into EFU's
independently. The EFU'd records would then be compared and
any disagreements worked out by the coders. Only 12% of the
EFU's revealed such disagreements. The end product was then a specimen record in which Environmental Force Units had been identified by two independent coders.
The coders then, working independently,categorized and tallied the frequencies of Environmental ForceUnits according to the six variables represented in the study hypotheses:1
1. The frequency of total EFU'sper observation.
2. The frequency of EFU'sper observation initiated
by the subject as opposed to those initiatedby the agent.
3. The frequency of EFU'sper observation in which
the subject was the sole target ofthe EFU as
opposed tc those in which agroup target existed.
4. The frequency of EFU'sper observation in which
conflict was evident between the goals ofagent
and subject as opposed to those in whichno cOnflict was evident.
5. The frequency of EFU'sper observation in which
methods adjudged to be coercivewere employed by
the agent as oppc3ed to those inwhich non-
1 Frequency counts were adjusted to compensatefor length of observation by usingthe formula: Adjusted N = Duration of shortest PeriodX Actual number of Duration of Record X EFU in Record X 57
coercive methods were employed.
6. The frequency of EFU'sper observation in which
the agent was an adultas opposed to a peer of the subject.
These data are summarized byTables 1 through 6.
TABLE 1
TOTAL ADJUSTED NUMBER OF EFU'S(VARIABLE 1)
Subject Choice Setting Non-choice Setting 1
Disturbed 1 6.67 5.83 2 15.65 7.00 3 13.57 14.78 4 1.90 10.91 5 12.00 28.18 6 26.00 4.17
7 , 9.00 6.00 8 13.55 7.62 9 17.14 7.83 10 4.35 3.81 Non-disturbed
11 8.00 10.91 12 15.86 15.20 13 11.54 9.33 14 25.71 4.57 01 15 20.00 22.86 16 6.40 8.00 17 16.92 8.8o 18 15.24 14.29 19 17.27 :1.82 2, 9.57 10.91 58
TABLE 2
ADJUSTED NUMBER OF EFU INITIATED BY SUBJECT (VARIABLE 2)
Subject Choice Setting Non-choice Setting
Disturbed
1 2.67 3.33 2 5.22 0.00 3 3.57 3.48 4 0.95 6.36 5 5.00 11.82 6 12.00 2.50 7 2.00 0.00 8 5.81 0.95 9 8.57 0.00 10 0.00 0.00 Non-disturbed
11 5.33 2.73 1? 6.21 1.60 13 1.54 2.67 14 11.43 2.86 15 (...67 5.71 16 0.00 1.00 17 5.38 3.20 18 4,76 0.95 19 5.45 5.45 20 1.74 0.91
{ 59
TABLE 3
ADJUSTED NUMBER OF EFU IN WHICH SUBJECT IS SOLE TARGET (VARIABLE 3)
Subject Choice Setting Non-choice Setting
Disturbed
1 3.33 5.83 2 13.91 1.00 3 12.14 9.57
/I 0.95 8.18 5 9.00 20.00 6 13.00 4.17 7 5.00 0.00 8 10.32 1.90 9 16.19 1.74 10 1.74 0.00
Non-disturbed
11 8.00 9.09 12 13.10 4.00 13 6.92 6.00 14 21.90 4.00 15 20.00 15.24 16 4.80 8.00 17 12.31 4.00 18 10.48 3.81 19 12.73 8.18 20 3.48 1.82 60
TABLE 4
ADJUSTED NUMBER OF EFU INWHICH CONFLICT OCCURRED (VARIABLE 4)
Subject Choice of Setting Non-choice setting Disturbed
1 0.0 0.0 2 6.96 1.00 3 7.14 7.83 4 0.0 2.73 5 3.00 9.09 6 5.00 0.0 7 4.00 0.0 8 4.52 0.95 9 4.76 0.0 lo 0.0 0.0 Non-disturbed
11 1.33 0.91 12 3.45 0.0 13 3.08 4.00 14 4.76 0.57 15 4.17 4.76 16 0.80 2.00 17 3.08 1.60 18 4.76 0.95 19 3.64 0.91 20 0.87 0.0 61
TABLE 5
ADJUSTED NUMBER OF EFU IN WHICH A COEnCIVE METHOD WAS USED BY AGENT (VARIABLE 5)
Subject Choice Setting Non-choice Setting Disturbed
1 0.00 0.00 2 1.74 0.00 3 3.57 3.48 r 4 0.00 0.00 5 1.00 1.82 6 2.00 0.00 7 1.00 0.00 8 0.65 0.00 9 0.95 0.00 10 0.00 0.00 Non-disturbed
11 0.00 0.00 12 2.76 0.00 13 0.77 2.00 14 3.81 0.57 15 0.83 2.86 16 0.00 0.00 17 0.77 0.00 18 0.00 0.00 19 0.00 0.00 20 0.00 0.00 62
TABLE 6
ADJUSTED NUMBER OF EFU CONTAINING ADULT AGENT (VARIABLE 6)
Subject Choice Setting Non-choice Setting
Disturbed
1 0.00 2.50 2 6.09 7.00 3 11.43 7.83 11 1.90 6.36 5 4.00 15.45 6 12.00 2.50 7 4.00 6.00 8 4.52 6.67 9 3.81 7.83 10 4.35 3.81 Non-disturbed
11 0.00 2.73 12 6.90 14.40 13 6.15 4.67 14 10.48 1.71 15 4.17 11.43 16 1.60 3.00 17 6.92 6.40 18 10.48 12.38 19 2.73 8.18 20 7.83 10.00 63
Statistical Analysis
Hypotheses tested
The hypotheses found in Chapter Imay be summarized
for statistical purposes in null formsas follows:
For each Variable X, "disturbed"children will
not have significantly differentnumbers of
EFU's than their "non-disturbed"matchmates (Factor A).
For each Variable X, there will beno significant
difference in the numbers of EFU'srecorded in
choice as opposed to non-choicesettings (Factor B).
For each Variable X, theinteraction of sample
'(Factor A) and setting(Factor B) will produce
no significantly different numbers ofEFU's.
Analysis of Variance
The following three-factoranalysis of variance model, an A x B x S design, was used where FactorA( "Disturbed"-
"Non- disturbed "), Factor B ("Choicesetting"- "Non-choice setting"), and Factor S (Matched Pair)were each crossed with the other two factors(Lindquist, 1953). Factors A ariu
B were said to be "fixed effects"while Factor S was "random."
Because of this interactionof fixed and random effects, it was statistically possible to test F ratiosonly for Factors 64
A and B and their interaction. Factor S was used to eliminate
variance due to differences between classroomsand schools.
A("Disturbed") A S (Pair) I 2("Non-disturbed")
(choice) (non-choice) B(choice) B 1 1 2(non-choice)
1 Subject 1 Subject 11 2 Subject 2 Subject 12 3 Subject 3 Subject 13 4 Subject 4 Subject 14 5 Subject 5 Subject 15 6 Subject 6 Subject 16 7 Subject 7 Subject 17 8 Subject 8 Subject 18 9 Subject 9 Subject 19 10 Subject 10 Subject 20
Fig. 1.--Analysis of variancedesign.
This analysis of variance comparedthe mean scores of the two samples of subject (Factor A)in the two settings
(Factor B) to determine ifsignificantly higher means exist for either sample within eithersetting. The interaction of
Factors A and B was computed todetermine if there exists a significant interactional effect betweensample and setting.
Factor S is used to eliminate variancedue to inter-classroom differences.
In using this analysis ofvariance model one assumes that the relatedness of the dataderived from each of the two subjects in a matched pair iscomparable to the relatedness 65 of data derived from observing the same subject intwo separate observations. Given the impossibility of manip- ulating many variables in psychological research (onemay not expose a s4ngle subject to the condttions of disturbance and non-disturbance) this assumption regardingmatching is a common one. CHAPTER IV
RESULTS OF THE STUDY
The analysis ofvariance model used comparedthe samples of "disturbed" and "non-disturbed" childrenacross choice and non-choice settings to determine ifsignificant
differences occurred inthe frequency counts of eachof the six variables. These differences could bedetermined to be between the two samples of children, between thetwo types of settings or to be a result of the interaction ofsample and setting. The results of thisanalysis were used totest the six hypotheses of the study. The F-ratioswere obtained as folloWs:
For Factor A, F =Mean Square A Mean Square AS
For Factor B, F=Mean Square B Mean Square BS
For Factor C, F =Mean Square AS Mean Square ABS
F-ratios for Factor Swere statistically impossible to obtain since S was a random effects factor. Factor S in this design was used to eliminate varianceoccurring between classrooms and schools.
66 67
The hypotheses of the study are presented here in null form and presented in their component parts for purposes of clarity.
Hypothesis 1
A. "Disturbed" children will not have significantly
different numbers of EFU's per setting than their
"non- disturbed" matchmates. (Factor A.)
B. There will be no significant difference in the
numbers of EFU's recorded in choice as opposed
to non-choice settings. (Factor B.)
C. The interaction of Factors A and B will produce
no significant difference in the numbers of
EFU's.
The cell means for Factors A and B are found in Table 7.
TABLE 7 , VARIABLE 1 CELL MEANS FOR TOTAL ADJUSTED EFU'S
Al A 2
B 1 11.98 14.65
B2 9.61 11.66
The analysis of variance results are seen in Table 8. TABLE 8 Source ANALYSIS OF VARIANCE FOR VARIABLE 1, TOTAL ADJUSTED NUMBER OF EFU'S Sum of the Squares Degrees of Freedom Square Mean F-Ratio Significance Level Factor(Choice-(Disturbed BA non- - disturbed 141:5.7904 1 55.7904 2.3374 0.16 Factor Snon-choice)(Pair) 511.1865 71.6097 91 56.798571.6097 2.3143 --- 0.16 - -- ABAS 214.8114 0.9363 19 23.8679 ------ABSBS 447.6780278.4799 9 49.742030.9422 0.9363 0.0188 ______- -- 69
Inspection of the means revealsthat the "disturbed"
sample (A1) were recorded with fewerEFU's than the "non-
disturbed" (A2) and the choice situation(B1) resulted in
more EFU's than the non-choice (B2) The F-ratio for the
disturbed-non-disturbed factor was foundto reach the .16
level of significance. The F-ratio for interaction of choice-
non-choice factor was also foundat the .16 level of
significance. The F-ratio for interaction ofthese two
factors was not found to reachsignificance.
Statistically, all components ofthe null hypothesis
would be supported--thereare no differences in the numbers
of EFU's when comparedacross sample or setting, nor is there
an interaction effect. The absolute differences noted,how- ever, should be kept in mindas further results are revealed.
Hypothesis 2
A. "Disturbed" children will nothave signifi-
cantly different numbers ofsubject-initiated
EFU's per setting thantheir "non-disturbed"
matchmates. (Factor A.)
B. There will be no significantdifference in the
numbers of subject-initiated EFU'srecorded in
choice as opposed to nnn-choicesettings. (Factor B.)
C. The interaction of FactorsA and B will produce
no significant difference in the numbersof 70
subject-initiated EFU's per setting.
The cell means for Factors Aand B are found in Table 9.
TABLE 9
VARIABLE 2 CELL MEANS FGR SUBJECT- INITIATED EFU'S
Al A2
31 14.57 14.85
B 2 2.84 2.70 1
The results of the analysis ofvariance are found in Table 10.
Inspection of the means revealsthat the "disturbed" sample (Al) was recordedas having very similar frequencies of subject-initiated EFU'sas the "non-disturbed" sample (A2), but the choice situations (B) resulted in 1 more subject- initiatea EFU's than thenon-choice situation (B2). The F-ratio for the disturbed-non-disturbedfactor was not found to reach significance. The F-ratio for the choice-non-choicefactor was found to be at the .04 level ofsignificance, one of the most significant differencesfound by the study. The F-ratio for the interaction of these twofactors was not found to reach significance.
Statistically, the component of thenull hypothesis regarding differences between thetwo types of children would be supported, but thehypothesis chat thereare no difference' between the two types ofsituations would be rejected. The null hypothesis regardingan interaction effect would also be ANAWSIS OF VARIANCE FOR WHICH WERE INITIATED BY THE , VARIABLETABLE 10 2, ADJUSTED NUMBER OF SUBJECT EFU'S Factor A (Disturbed-, Source Sum of the Squares Degrees of Freedom Square Mean F-Ratio Significance Level Factor 3 (Choice-non-choice)non-disturbed) 37.5972 0.0462 1 37.5972 0.0462 5.77350.005 0.04 ___ ASFactor S (Pair) 115.659881.5916 9 12.8510 9.0657 ______- -- ABSBSAB 113.077758.6088 0.4161 91 12.5641 6.51200.4161 0.0331 ______------72
supported. One may conclude that choicesituations result in higher frequencies of subject-initiatedEFU's than non- choice situations for the two typesof children.
Hypothesis 3
A. "Disturbed" children will not havesignificantly different numbers of EFU's pEr settingin which
the subject isthe sole targetthan their "non-
disturbed" matchmates. (Factor A.) B. There will be no significant diference in the numbers of EFU's in which the subjectis the sole target recorded in choice as opposedto
non-choice settings. (Factor B.)
C. The interaction ofFactors A and B will produce
no significant difference inthe numbers of EFU's in which thesubject is the sole target. The cell means for Factors A and B are foundin Table 11.
TABLE 11
VARIABLE 3 CELL MEANSFOR EFU'S IN WHICH THE SUBJECT WAS SOLETARGET
The results of the ( analysis of varianleare found in Table 12. ANALYSIS OF VARIANCE FOR VARIABLE TABLE 12 Source Sum of the Squares IN WHICH SUBJECT WAS SOLE Degrees of 3, ADJUSTED NUMBER OF EFU'S MeanTARGET Significance Factor(Disturbed- Anon-disturbed) 39.7803 Freedom 1 Square F-Ratio Level FactorFactor(Choice- S Bnon-choice) 171.2718 1 171.2718 39.7803 7.84122.9907 0.020.12 AS4, (Pair) 119.7115476.2908 9 13.301252.9212 --- - ___-- ABSBSAB 282.2534196.5821 6.7158 91 31.361421.8424 6.7158 0.2141 --- ___------74
Inspection of the means reveals that the "disturbed"
sample (A1) was recorded as having fewersole target EFU's
per setting tan the "non-disturbed" sample (A2). The choice
setting (B ) resulted in higher frequencies ofsole target
EFU's than the non-choice setting (B2). The F-ratio for the
disturbed-non-disturbed factorwas found at the .12 level of significance. The F-ratio for the choice-non-choice factor
was determined significant at the .02 level--thehighest level
of significance found by the study. The F-ratio for the AB interaction was not significant.
Statistically, the null hypothesis regarding dif-
ferences between the two types of children would besupported, but the hypothesis that thereare no differences between the two types of settings would be rejected. The null hypothesis regarding the interaction effect would also besupported. One may conclude that though no differences exist between the matchmates regarding the number of sole target EFU'sper setting, choice settings do produce for bothgroups a higher frequency of such EFU's than non-choice settings.
Hypothesis 4
A. "Disturbed" children will not havesignificantly
different numbers of EFU'sper setting in which
conflict occurred between the goals ofthe
subject and environmental agent than willtheir
"non- disturbed" matchmates. (Factor A.) B. There will be no significant difference in the
number of EFU's in which conflict occurred 75
recorded in choice as opposedto non-
choice settings. (Factor B.)
C. The interaction ofFactors A and B will
produce no significant differencein the
numbers of EFU's in which conflictoccurred. The cell means for FactorsA and B are found in Table 13.
TABLE 13
VARIABLE 4 CELL MEANS FOREFU'S IN WHICH CONFLICT OCCURRED
A 2 A2
B1 3.53 2.99
B 2 2.16 1.57
The result of-theanalysis of varianceare found in Table 14.
'Inspection of themeans reveals that the "disturbed"
sample (A1) was recordedin both settingsas having greater numbers of EFU's in whichconflict occurred than didthe
"non-disturbed" sample (A2). The choice setting (B1) for both samples resulted inhigher frequencies of EFU's containing conflict than didthe non-choice setting (B2).
The F-ratio for thedisturbed-non-disturbed factorwas found at the .30 level ofsignificance. The F-ratio forthe choice-non-choice factorwas determined to be at the .10 level. of significance. The F-ratio for ths. AB interaction was also found not to reach significance. ANALYSIS OF VARIANCE FOR VARIABLE IN WHICH CONFLICT OCCURRED TABLE 14 4, ADJUSTED NUMBER OF EFU'S Factor A (Disturbed- Source Sum of the Squares Degrees of Freedom Square Mean F-Ratio Significance Level Factor B (Choice-non-choice)non-disturbed) 19.6280 3.2148 11 3.2148 1.1851 0.30 ASFactor S (Pair) 105.933324.4142 99 11.770319.62802.7126 3.2812 ___ 0.10 ___ ABSBSAB 53.837533.5926 0.0052 91 3.73255.98190.0052 0.0013 ______- -- 77
Statistically, the null hypothesisregarding differ-
ences in frequencies of EFU's containingconflict would be
supported for both sample andsetting. The null hypothesis
regarding Interaction effect wouldalso be supported. One concludes that differencesoccurring between samples and
settings are likely to havebeen due to chance, but the
proximity to significance forFactor B is a continuation of a trend throughout all hypotheses.
Hypothesis 5
A. "Disturbed" children willnot have significantly
different numbers of EFU'sper setting in which
coercion by the agent is evidentthan their
"non- disturbed" matchmates. (Factor A.)
B. There will be no significantdifference in the
number of EFU's in whichcoercion by the agent
is evident in choiceas opposed to non-choice
settings. (Factor B.)
C. The interaction of FactorsA and B will produce
no significant difference in thenumbers of
EFU's in which coercionby the agent is evident. The cell means for Facto: A and B are found in Table15.
The results of the analysisof variance are found inTable 16.
Inspection of themeans reveals little difference between the frequencies recorded for the "disturbed"sample
(A1) and the "non-disturbed"sample (A2) The nhoice setting
(B1) for both samplesresulted in higher frenUencieSof EFU's 78
containing coercion bythe agent than didthe non-choice setting (B2). Neither the F-ratiofor the disturbed-non- disturbed factor nor the F -ratio for the ABinteraction was found to reach significance. The F -ratio for thechoice -no,- choice factor reached the .20 level ofsignificance.
TABLE 15
VARIABLE 5 CELLMEANS FOR EFU'S IN WHICH COERCION BY THE AGENTIS EVIDENT
A A 2 2
B 2 1.09 0.89
B2 0.53 0.54
Statistically, thenull hypothesis forall three components was supported.
Hypothesis 6
A. "Disturbed" childrenwill not have significantly
different numbers ofEFU's per setting inwhIch the agent isan adult than their "non-listurbed"
matchmates. (Factor A.)
B. There will beno significant differencein the number of EFU's containing an adult agentin choice as opposed to non-choice setting. (Factor B.) C. The interaction ofFactors A and B willproduce no significant differencein the numbers of EFU's
containing adultagents. The cell means for Factors A andB are found in Table17. ANALYSIS OF VARIANCE FOR CONTAINING COERCION VARIABLE 5,TABLE ADJUSTED 16 NUMBER OF EFU'S Factor A (Disturbed- Source Sum of the Squares Degrees of Freedom Square Mean F-Ratio Significance Level Factor B (Choice-non-choice)non-disturbed) 2.07930.0846 11 2.07930.0846 1.85700.0686 0.20 - -- ASFactor S (Pair) 11.084823.2977 99 2.58861.2316 ___ ------0 ABSBSAB 10.0775 4.92030.1102 991 0.54671.11970.1102 0.2015 ______------80
TABLE 17
VARIABLE 6 CELL MEANS FOR EFU'S CONTAINING ADULT AGENTS
Ai A 2
B 1 5.21 5.72
B 2 6.59 7.49
The results of theanalysis of varianceare found in Table 18.
Inspection of themeans reveals that the "disturbed"
sample (A1) were recordedin both settingsas having slightly
fewer numbers of EFU'scontaining an adult agentthan the
"non-disturbed" sample (A2). The choice setting (B1)for both
samples contained fewer EFU'swith adult agents thandid the non-choice setting (B2). Neither the F-ratio for the
disturbed-non-disturbed factornor the F-ratic for the AB
interaction was found toreach significance. The F-ratio for the choice-non-choice factor was found to beat the .24 level of significance.
Statistically, the nullhypothesis would be supported for all threecomponents.
Summary
A total of 18 testsof significancewere administered. The frequency scores for each of the sixvariables were analyzed for differences between types of children,between settings and for the interaction effect of type ofchild and ANALYSIS OF VARIANCE FOR VARIABLE WITH ADULT AGENT TABLE 18 6, ADJUSTED NUMBER OF EFU'S Factor A (Disturbed- So urce Sum of the Squares Degrees of Freedom Square Mean F-Ratio Significance Level Factor B (Choice-non-choic3)non-disturbed) 24.7905 4.9773 11 24.7905 4.9773 1.56590.3669 0.24 - -- ASFactor S (Pair) 122.0779188.2623 9 13.564220.9180 ___ - -- ABSBSAB 142.476593.4349 0.3591 919 10.381615.8307 0.3591 0.0345 ______- -- 82
setting. Of these 18 tests, onlytwo were found to be
significant, and bothof these were concernedwith differences
between settings. The fact that sucha small number of significant results were found in proportion toto the namb-r of tests administered, greatly increaseE, theprobability that such results may bedue to chance.
Differences found in thesix EFU variablesbetween the "disturbed" and"non-disturbed" samplesdid not reach statistical significance at an .05 level ofconfidence. In the case of three of the variables, (total EFU,Sole Target EFU, and Conflict EFU) the chance that thesedifferences between the two samples was due to random factorswas 30% or less. Such differences,while not remotelyapproaching statistical significance, may suggest that there isa very slight tendency for"disturbed" children:
,a) to experience fewertotal EFU's per setting
than "non-disturbed"children, b) to experience fewerEFU's in which theyare
the sole target of theenvironmental agent, and c) to experience more EFU's in which there iscon-
flict between the goalof the agent and thatof the child.
This latter trend may be more significant thanit appears since it is, by necessity, related to the first variable,total number of EFU's, but indicates results in the oppositedirection. Even though "disturbed" children generally hadfewer EFU's, thf tended to havemore EFU's containing conflict. 83
Differences found in the EFU variable between choice and non-choice settings were found to be significant in two of the six variables. The differences revealed in the other four variables, though far from statistically significant, were consistently in the same direction. These data reveal the choice situations when compared with non-choice settings for the two samples tended to result .in:
a) greater numbers of Environmental Force Units
which were initiated by the subject, and
b) greater numbers of Environmental Force Units in
which the subject was the sole target ofan
environmental agent.
Furthermore, while not statistically significant, choice situations tended to result in:
a) greater numbers of EFU's,
b) greater numbers of EFU's in which a conflict
existed between the goal of the subject and
that of the agent,
c) greater numbers of EFU's in which a coercive
method was used by an agent, and
d) fewer numbers of EFU's in which the environmental
agent was an adult.
There seems to be a consistency in the trend toward greater numbers of EFU's across al: variables in cilice situations. That these were with peers also is of interest. CHAPTER V
CONCLUSIONS
Examination Of The Results Applying the commonly used levels of confidence,.01 or .05, the statistically substantiated results ofthe study are limited to the difference found for two variablesbetween choice and non-choice situations. (These are but twofindings of significance out ofa total of 18 tests of significance.) The frequencies of Environmental Force Unitsinitiated by the subject and the frequencies of Environmental ForceUnits in which the subject was the sole targetappear significantly higher in choice c.ettings. These results are notunexpected given the nature of a choice situation. One would expect children to be able, if not encouraged, to initiatemore
conversations or other formsof interaction ina setting in which they have the freedom to choose froma variety of activities. The increased numberof EFU's in which theywere the sole target would seem also to reflect thisincreased individual i7teractionas children engaged in individual conversations. These results would alsoseem to give validity to the teacher's designation of these periodsas truly less structured and as presenting "choice" to thechild. The other trends, albeit based on consistentlynon-
84 85 significant findings, identified in the data results regarding choice and non-choice situations support these two significant findings. The tendency for choice situations to contain greater total EFU's and more EFU's "In which the agent was a peer again coincides with what one would expect in a less structured classroom situation. At first glance, the tendency for greater numbers of conflict or coercive EFU's to occur in choice situations may seem contradictory to "choice."
Inspection of the specimen records themselves, however, reveals that many of these frequencies were recorded when the environmental agent was a peer having some mild disagreement with the subject.
The fact that neither difference nor a consistent trend toward difference in numbers of Environmental Force
Units was found between the "disturbed" and "non-disturbed" samples,must be the most significant result of the study.
The results of this study dictate the conclusion that the environments of "disturbed" and "non-disturbed" children in the same setting as measured by EFU's are essentially the same. These results are in agreement with the results obtained by Schoggen (1968) using much the same methodology. The
Schoggen study, using Environmental Force Units as the measurement instrument, revealed no differences in the quantity or types of interactions experienced by ort:Aopedically handicapped youngsters and a matched control group.
The agreement of the results of these two studies, both of which observed handicapped children and a matched control group, leads to a conclusion that, despite preconceptions 86
regarding the reaction of the environment tohandicapped
youngsters, no environmental differences did, infact, exist.
The handicapped child recordedas many verbal interactions as
the normal youngster in the setting andthe type of inter-
actions as measured by this studywere no different than those of his peers.
Since "emotionally disturbed" childrenare identified by the school in terms of theirovert behavior and the
"problems" they cause in theirclassroom interactions, the finding of no difference in environmentalinteraction is rather startling. One may conclude that the label "emotionally disturbed" as applied by the schoolis based upon criteria unrelated to environmental interactionassessed, that the phenomenon of emotional disturbance isindependent of environ- mental interaction, or that theinstrument used by this study was not effective in measuring the environmentalphenomena.
Any of these conclusionsmay be valid. Many writers (Fried,
1964; Kounin, Friesin &Norton, 1966; Mercer, 1965) have postulated that such labelsas "emotional disturbance" anC
"mental retardation"are used as social system devices for reducing problems caused by behaviorin some way deviant from the behavior of the majority. The behavior exhibited by tnose so singled out need not correspondto the label applied.
The devidm.e instigating the labeliligprocedure may be the result of cultural background,physical appearance or socio- economic phenomena. Though the school is a prime offender in the expeditious categorizationof children for purposes of maintaining uniformity withthe classroom, in this study 87 1 one of the other two alternativesappears more plausible to this author.
The assumption that emotionaldisturbance is unrelated to environmental interactionwould also be extremely difficult
to substantiate. Even if one were toassume a ps_nodynamic
viewpoint concerning the etiologyand nature of the "disturb- ance," the manifestation of this intrapsychic phenomenonis
usually seen in overtinterpersonal behavior and inthe
quality and quantity of verbalinteractions. For example,
one of the chief behavioral characteristicsof such conditions as autism is a lack of verbal communication. The existence of "emotional disturbance"without its manifestation in environmental interactionsseems untenable.
This author is led to the third alternativeas a result of his own theoretical biasand because of some less formal data obtained from the studyin the form of observer reactions. Theoretically, emotional disturbance viewedfrom an ecological perspective must be evident inthe interchange between individual and setting. Disturbance from this view- point is seen as an incongruencebetween individual and setting or milieu. This incongruence must be made manifest in an observablemanner or the child would not be labeledas deviant. Intuitively, the transcripts from whichthe data were gathz.ed reveal distinctive di:ferencesin the two samples. Although the observers werenever told Lhich subject was "disturbed" and which was not "disturbed"and though they had no background in psychology,they could identify the
"disturbed" child withoutexception after each observation. 88
(This fact also has implications regarding possiblecontamina- tion of the study results.) The reasons given for this
identification were alwayscouched in terms of that child's
interaction with the otherin the room. Often that child was said to be a barometer of the class climate. In several
instances in which actingout or anxious behaviorwere rampant throughout theroom, it was the child labeled
"disturbed" who seemedmost sensitive to andreflective of this atmosphere.
The identification ofthe reasons why theseinter-
actional differenceswere not detected may bea valuable
contribution of the study. This investigator offerstwo primary causes forconsideration:
1. a "hidden" sub-sample ofchildren;
2. gaps in the developing methodologyof
r "ecological psychology."
The Sample Of "Disturbed"Children
For purposes of thisstudy the sample of "disturbed" children were those children who had beendiagnosed and labeled as "disturbed" by the school and itsspecial education component. It was the intentof this study to deal withthe population of children classified as disturbed inschools whether or not these children may be seenas "disturbed" by other types ofinstitutions
Inspection of the datafor the sample of "disturbed" children did, however, reveal the possibilitythat there were in reality two populations of childrenrepresented in the [ 89
sample. Because of the nature of the interactionsof two populations, the data within the "disturbed"sample tended to "cancel itselfout" by increasing the variancebetween pairs. Fi_Nres 2 and 3 demonstrate the bi-modalnature of the distributionof adjusted total EFU's for the"disturbed" sample as contrasted with the normalcurve of the "non- disturbed" sample.
,00 N alS1;J:7 Subj.9v 1 / (15.0) % I (24.97) / Subj.4 / Subj.2 , # (12.84) % (22.65) / %. %Wm. e Subj.10 ...... Subj.l Subj.8 Subj.3 (8.6) (12.50) Subj.6 Subj.5 \ (21.17) (28.35)(30.17) (40.18) 5 10 15 20 25 k 30 35 40 Adjusted totalEFU's .
Fig. 2.--Distribution of adjustedtotal EFU's for "disturbed" sample.
roar agargift riSUbj 18 /(29.53) \ do".SUbjeljli %11 ... isSubj.19 Subj.12 40 (20.87) (29.09)(31.06) N ,... Subj.16 Subj.11 Subj.20 ,oe Subj.17 Subj.14 Subj.15%.. owl. (14.40) (18.91) (20.48) or Ilo, (25.72) (30.28) (42.86) 5 10 15 20 25 30 35 40 Adjusted totalEFU's
Fig. 3.--Distribution of adjusted totalEFU's for "normal" sample. 90 Four of the subjects within the "disturbed"sample were recorded with fewerthan 15 EFU's per setting. There wasa range of approximately 7 in thisgroup of four. The remaining six were recordedas having at least 21 EFU's per settingw4th a range of 19. The "non-disturbed"sample of children recorded totalEFU scores which somewhat more closelyresemble a normal curvecentering about a mean of 26.30 EFU'sper setting. All but one of the "non-disturbed"sample recorded at least 15 EFU'sper setting. These data would indicate that twogroups of children were represented bythe 'disturbed" sample--children whoseem isolated or withdrawn and have few EFU'sas a result, and those whosesocial contacts as measured by EFU'scorrespond more closely to the "non-disturbed"distribution. The latter group also contains several individualswhose EFU frequencies are exceedingly large. By including inthe overall "disturbed" group these two groups, whose frequencieson the very basic measure of total EFU are dissimilar,one would appear to confound theresults. The high andlow frequenciesfound within the samplewould tend to reduce the potencyof dif- ferences betweenthat sample and the "non-disturbed"sample. The exceedinglyhigh frequencies of EFU's foundfor several of the "disturbed"sample were "washed out" in theresults as they were combinedwith the unusually low frequenciesof several subjectsin the "disturbed" sample for themean value. It is interesting to note that the"clinical descriptions" given by theschool for each member of the "disturbed"sample include "withdrawal" as part of the descriptionin only one 91
case. All of the others are described as "acting out"or
"aggressive." Either the school records deal with tehavior
description in terms too broad to be accurateor members of
the "disturbed" sample were intimidated by theobservation and
tended to behave inan inhibited manner. Either or both of these explanations may be valid.
On the basiS of this "split" within the "disturbed"
sample, one may conclude that children labeledas "disturbed"
may be classified into two distinct groups--thosewho have
characteristics associated with socialisolation or withdrawal
and those whose frequency of interactionswith adults and peers more closely corresponds to "normal" frequencies.
Included in the lattergroup may be the "acting out" or
"aggressive child" described byschool records. This hunch
gained by inspection of the data iscorroborated by the
research efforts of others in the fieldof emotional disturb-
ance in children to "sort out" the types of childrennow
lumped under the rubric of "disturbed." Such work is being currently pursued by Rue Cromwellat Wayne State University
and promises to be a valuable addition tothe literature of disturbance. It again points out the inadequacy ofusing stereotypic categories for dealing withchildren. The category of "disturbance" wouldappear to house children exhibitin6 quite different behavivIalrepetoires and having quite different needs. The children exhibiting actingout or aggressive behavior tend to havemany social contacts in school with peers and withthe teacher, though many of these are of a conflictual nature. Other "disturbed" children, 92
however, interact very little with peers and their school
interactions seem limited to infrequent directions given by
the teacher. With two such distinct behavior patterns, it
seems very inappropriate for either research or intervention
efforts to use the term "disturbed" as a grouping criteriaor description of children.
Methodology
A second reason for the inability of thisstudy to
discern statistical differences in the interactiveenviron-
ments of "disturbed" and "non-disturbed" childrenmay lie in
several of the data gathering and analyticalmethods of
"ecological" psychology. Because of the emphasis in this area
of psychology upon data gathering withinthe natural setting, many problems arise in regard to sampling and data collection.
Gaining access to a public school classroomfor purposes of research is rapidly becomingan impossibility.
The reasons for thisare varied and both practical and bureaucratic in nature. When one finally does manage to acquire access a great many "degrees of freedom"are lost.
The composition of the sample of classroomsor teachers tends to be influenced most by accessibilityrather than by the variables to be studied. In the case of this study, the volunteering and cooperation of the building principalwas a pre-condition to selection of the sample of classrooms. The principal became ';he key figure inboth the selection of the classrooms and of the individual subjects. His choosing of the subjects wasnecessary because the school records were 93
inaccessible to the investigator. Such influence from a
variable unaccounted for by the study (the principal)can
certainly lead one to qualify the results. The fact does
remain, however, that the children labeled "disturbed"had
been so diagnosed through official school channelsand
thereby reflect the concept of "disturbance"utilized by the entire school systemrather than the principal.
This problem of accessibility to subjectsand the
necessity of compromise in experimentaldesign, is certainly
not new to psychology. The difficulties in arranging
observations in natural settings area result not only of
logistical problems but area result of a growing resistance
of many institutions, particularlyschools, to being "exploited"
as research sites. In the six months it took this investigator to get permission to observe in theschools included in this study, a general mistrust concerningthe university was repeatedly voiced by teachers, principalsand administrators.
The general feeling of school personnelwas that university researchers "exploited" schools by settingup the experiment, gaining a degree and leaving the schoolwithout feedback concerning the results. Ecological psychology must develop methods of feedback regarding informationgained or its accessibility to sites will be severlyhampered. The results of this st.zdy will be presented the staff of each participating school in an in-service trainingsession.
The role of the principal in the study,though more active than one might wish in controllingall variables, was a necessary compromise. It was through this use of the 94 principal that the school's concernsregarding confidentlality of school records were assuaged. The school recordswere never seen by these investigators since theprincipal provided all necessaryinformation for the study. In leaving a laboratory setting andattempting to observe in a natural setting, ecologicalpsychology makesan implicit value judgment. The ecologicalpsychologist must decide that despite the "hassle" andmethodological compromise inherent in using "natural settings,"the resultsare still more desirable than are those gained from studyin a more controlled but artificial setting. Though this resultsin much more complexityin design and interpretation andgreatly increases the "contamination" ofthe results by extraneous variables, the results do reflect real lifebehavior with its natural motivations, conditionings and influences. Though the chances of ferreting out pure causalrelationships is much reduced, one doesobtain a much more valid and usefuldescription of human behavioras it naturally occurs. A case could alsobe made for the position that the experimentallaboratory is, in fact, introducing confounding variablesin its artifical environment. This must be more valuable to theecological psychologist as he seeks to apply thisknowledge in "real life situations" than tne applicabilitywhich is extremely confined bj thelaboratory. 1: this value judgmentwere more explicit in the experimental studies ofecological psychologists, pernaps the field would be less "paranoid"concerning the validity of itsresults. 95
Ethical dilemmas also ariseonce access is made into
a natural setting. In order to observe ina "natural
setting," one must either dothe observation undetected and
without informing the subjects (aprocedure rapidly becoming
ethically and legally disreputable)or the time frame of the
observation must be long enoughfor the initial effects of
observer intrusion to "wear off"and for the observer to
become "absorbed" into the setting. Even if the latter
methodology is adopted,one faces the problem of explaining
the purposes of the observationto those observed. If
individuals are to be observed,how much is told to them with-
out "contaminating the results"? An attempt was made by this
study to compromise in thesolution of these problems.
Subjects were told the generalnature of the observation but
not that individualswere to be observed. It was logistically
impossible to establisha time frame for the study to
completely eliminate "observereffects," though some timewas alloted to this purpose. The issue raised, however, hasyet
to be fully confronted bypsychological research. The
increasing ethical andlegal parameters being appliedto research in regard to individualrights to privacy and informed consent must beaccounted for in research efforts.
The extent to which "knowledgeablesubjects" will influence the results has yet to bedetermined. One possible method of circumventing sucheffects may lie in long termexperiments in which the observereventually becomes "imbedded"in the setting. Such an alternative againpoints to the desirability of a "field station" approach in which many studiesoperate 96
from a central locationover a long term period. The ''field station" becomes a ratherpermanent part of the community.
The problem of observer biasis also an omnipresent
one. In Net, in a specimen recordthe observer is encouraffed
to make some types of subjectivejudgments. The question of the objective validity ofdata derived from these specimen
records was found by thisinvestigator to be one never fully
answered. The judgment concerningthe validity of the
specimen records in the endrested upon the intuition of the
investigator. These biases must be weighed inconsidering
results. Such subjective elements,however, do not invalidate the results.
The sheer effort and great amountsof time and resources demanded by the specimen recordmethodology may also influence observers insubtle ways. How much data is lost through sheer observerfatigue or frustration isa matter of speculation. The difficulties encounteredin making an intensive 30 minute observation,followed by a written description and then later theediting of these resultsare not to be underestimated. It becomes a tedious taskand is plain "hard work."
None of the above methodologicalproblems and potential biases, however, wouldappear to distort the resultscon- sistently in the "no difference"direction. The Environmental
Force Unit itselfray be insufficient as a measure ofthe interaction of child andenvironment. While it can be viewed as recording the verbali-teractions rather efficiently, its detection of subtlenon-verbal interactions and its failure 97
to capture the range of affect inherentin all interactions
severely limit its uses especially ininteractions involving
"disturbance" where affect playsa major role. This
investigator knows of no single instrumentwhich successfully
captures all aspects of an interactionincluding a range of
affect. Perhaps an interaction analysissy ung a quick
and convenient numericcode for recording affective andnon-
verbal aspects of interactions mightbe used in tandum with
the EFU system in analyzinga specimen record. Though this
author knows of no other study whichattempted to find
statistical differences betweengroups using the EFU as a
measure, with the exception of Schoggen (1968)who also
found no differences, otherauthors (Schoggen, 1963; Gump,
Schoggen & Redl, 1963) have usedthe percentages and absolute
numbers of EFU's to distinguishbetween groups. If these
figureswere subjected toa statistical test, they also might
prove to be insignificant differences. The ever present
problems of observer bias, coderreliability or sample
representativeness might wellaccount for differences they
have obtained. The small samples involved inthese studies
(1 to 16 subjects) alsoleads one to interpret the results rather conservatively.
Inherent in the task ofcategorizing EFU's, as with
any measurement device, isa losb of information and it may
be this subtle affectiveaspect of interaction which suffers
most in the process of categorization. The affective dif-
ferences present between twopeers arguing over a crayon and a teacher confronting a child witha wrongdoing tend to be 98 lost as both interactions fall into the "conflict"category. While some conflictual interactions are considered"normal" and even healthy, differences in thequality and quantity of su.1- eractions often are the chief characteristicin "deviant behavior." Very few personsperceived as deviant
engage in behavior which istotally alien toa normal behavioral repetoire. It is the qualityor intensity with which the behavior is expressed, its quantity(frequency), or its patterning which identify it as "different." The differentiation of behaviors and the effect ofenvironmental interaction must depend upon the development ofinstruments sensitive to thesequalitative variables. The fact that the Environmental Force Uniton two occasions did distinguish in even an extremelylimited degree between the two types of behavioral settingsmay be suggestive. It would seem to indicate that furtherresearch be done to determine theuse of Environmental Force Units in theobjective identification and description of environmentalsettings. Instead of using the EFU to distinguishbetween individuals, it may be more suited to distinguishingbetween settings. The results of this study in this regardare corroborated by the findings of Barker (1968) andGump (1961). Too often, those responsible for providing andmaintaining settings for children (..e., schools, day care centers,playgrounds, etc.) are content to describe the types of experiencegained there by children interms of the trainingof the professional staff or in ambiguous terms such as "freeplay," classlesson, etc. The EFU may provide an opportunity foron-going 99
evaluation of settings bycapturing the specific types of 1 interactions occurring there ina quantifiable manner.
Through this instrumentone may measure the nature of inter-
actions occurring as wellas identify the environmental aents and their characteristic methods. The EFU and itsuse in this manner provides many opportunities for futurestudy.
Implications Of The StudyFor Ecological Psychology
The nature of ecologicalmethodology and its tremendous
logistical requirements wouldimply that such researchbe of a long term nature. The emphasis shouldbe upon on-going
data collection in depthin a few localities ratherthan upon
short term studies witha sampling technique. It may be that
the methods and goals ofecological psychologyare incompatible with traditional sampling procedures. This may imply theuse of the "field station" concept previously mentioned. While sampling attempts to make generalizations about populationson the basis of sample representativeness and probabilitycurves,
ecological methods attemptto formulate generalprinciples
through in-depth study ofone sample. Ecological methodology
seems more related to the methodsof anthropology and ethology and their emphasis upon the process rather than thecontent of behavior than to psychology and its contentfocus. Such an emphasib is used to justifythe lack of representativeness.
Ecological psychologycan, however, become muchmore credible if care is taken tochoose the select samplesfrom environ- ments less divergent from the mainstream of culture (asin the case of rural Kansas). 100
Long term and in-depth study in fewer localitieswould also overcomesome of the methodological concerns. Observers could become better trained, more highly skilledand less obtrusive tf such studieswere centralized and on-going. Indeed, the direction taken by Barker and hisfield station concept would seem to be correct in light of thisstudy.
The potential theEnvironmental Force Unitmay have in the description of behavior settingsmay offer a means of quantifying and analyzing differences between behaviorsettings which heretofore relied upon intuitive judgmentsor categorical systems in which much data was lost (as inan Interaction
Analysis approach). It could be usedas one of a series of
measures to describe a settingand the interactionof that setting with individuals. Other measures to beused in conjunction with it may include means of mappingthe physical environment, traditional case historyapproaches and in- novative methods for measuring the affectivecontent of interactions.
The experience ofthe investigator withthe observers who were trained to take specimen recordsleads to a conclusion that such training has a serendipitous effectapplicable for pre-service training programs for professionals whowill deal with children. The careful observationrequired by the methodology, its specific focus and its techniquefor transferring the observation experience to writtenform proved to be stimulating and learning-filledtasks for the observers. 'Their unsolicited and independentcomments were unanimously enthusiastic about their observationsas learning 101 experiences for them atout children, "emotionaldisturbance" and its relationship to teacher or peer behavior,and the general dynamics of a classroom. This technique mightbe well used in the initial training experiences ofteachers, psychologists or others who must developan understanding of and sensitivity to childbehavior. Such techniques,as a method of feedback forin-service training,might be explored but its limitation in terms of time lapsebetween observation and results are a limitation in feedbackprocedures. In light of experiences in this study, thisinvesti- gator would view the "status" of the four principlesof "ecological psychology" identified in the literaturereview as follows:
1. The importance ofobserving human behaviorin its natural settingpresents considerable
methodological and ethicalproblems but ones
which may beovercome. The development of
innovative measuringtechniques and methods of analysis may aid in freeing thisapproach from the restrictiveparameters of traditional
statistical procedures. The "field station"
concept in studyingbehavior with an intensive
and long-range focusseems to be a step in the right direction.
2. The focus upon the interactive nature ofbehavior
has not yet been fullyrealized. Though present
techniques attempt tostudy the interaction,they appear to eliminate criticalaspects of the 102
interaction such as affect.
3. The influence of settingupon behavior is
perhaps the best developedarea in ecological
psychology. The methodologies presentlravailahle such as the EFUseem to be effective descriptions
of this phenomenon. It is the task of formulating
the principles andprocesses involved in this
influence of setting whichawaits. From these efforts may come the,as yet, unborn "ecological intervention strategies."
4. The development ofa methodology which will
objectify and quantifyobservational data has
a good beginning in Barker'sobservational
techniques. Care must be taken, however,to
combine these techniqueswith other tools to
gain as comprehensivea view of individual and setting as possible.
The task of ecologicalpsychology wouldseem to lie in the further identification of all aspects of settingand its interaction withthe individual. Not only must techniques be devised for determining these qualitative aspectsof inter- active behavior but means must be found to differentiatethe various levels of intensitywith which these variables interact. It is to this task thatthis study addressed itself. APPENDIX A
June 7, 1971. We are observing in Clara Barton Elementary School. We are presently observing Subject 16 inthe third grade. The situation is a non-choiceone. The following is that observation:
Observer 1.
10:30
Subject 16 is doing his math.
The teacher is instructing a portion of the classin math, while the rest of the students have been instructed to do their own work at theirown seats.
The Subject is now leaning over hispaper, apparently working.
The teacher has given special instructions tohim that when he finishes with his math hecan work with the multiplication cards or the addition card, whichever he feels he needs.
Presently he's just working with the problemsat hand. m c One of the boys seated next to himsays, "What did you ta say?" to the Subject. o 44 The Subject ignores him, and justcontinues to work quietly at his desk, says nothing to theboy who's just asked a question of him.
The Subject is making some numbers down theside column on his paper. He looks over to the boy seated at his left to see if perhaps he'sfollowing the instructions correctly;now he looks back to his own paper and continues tonumber down the side of the paper.
103 104
The boy seated next to him leansover and says some- P M thing to him. The Subject fails to look up at him, N c he only shrugs his shoulders and continues withhis work, seemingly ignoring the boy, at least forthe ro CN.) present. 43 0
cn ocr "Did you do the problems?" theSubject says to the o boy seated next to him. The other boy nods his head, in affirmative.
Now the Subject getsup from his desk.
He looks over quickly toward theobservers, now he walks toward the back of theroom.
He walks over to the drinking foulltain,gets a drink.
He starts his return trip back tothe desk; he walks back to the desk, pulls the chair out, sitsdown, slides the chair in close to the desk, andonce again begins to lean over his work andprepare to coitinue with math problems.
The Subject finishes the numberingand now draws a line, a neat line, slightly to the rightof the numbering to create a nicemargin.
He brushes the hair out of hiseyes, looks up for a moment to the other side of theroom, watches some of the activity of some of the children,a couple of boys talking.
He watches them rather curiously,interestedly; now he returns his glanceonce again back to his own paper.
He leans over his paperonce again and becomes involved in his work.
The Subject seems to be able toeasily fall into his work; he seems quite interested init, quite involved in it, compared to some of theother students who are still fussing, still perhaps just gettingthem- selves reak.y to start their work,cx else they're flitting about the roomor talking to their neighbor.
The Subject seems able to work easilyon his own. The teacher, as Isay, is working with other students. 105
The teacher at the moment has just said something to the boy seated next to him, and he looks over toward the boy for a moment, perhaps just to observe his reaction to the teacher.
But once again he returns quickly to hisown work, leans over the paper once again and continues to write on the paper.
10:32
Once again the teacher says something to the boy seated next to the Subject.
This time he takes no notice, he just continues to hover over his paper, continuing to write, preparing for various answers he's going to give--apparently he's numbering the paper for subsequent answers from his textbook.
The Subject looks up for a moment toward the blackboard to check out the instructions, the page of the assignment, and so forth.
Now he looks down toward his paperonce again, continues to write with his pencil.
The Subject then looksup toward the blackboard in the direction of where the assignmentis written.
Again he returns his glance to hisown paper.
The Subject looksup once for a moment toward the other side of the class where the teacher isworking with other students on their math problems.
He looks over toward the blackboarddirectly behind the teacher where she's writtensome of the problems that the other studentsare doing.
He locks down toward hisown paper, and resumes his writing and his work.
The Subject looks uponce again at the blackboard, pushes the hair out of his face.
He now looks down toward histextbook; he takes his index finger and marks the firstproblem with his finger to know where he ison the page.
Now he takes the instructionsheet and slides it over the textbook so as to be able to justpull it down over each successive problem, so hecan easily find the problem to answer and he beginshis work, starting to do the individual problems. 106
10:33
Several students are walking, filing past the other side of the room, since the doors givean access to other room::, arranged in sucha way the children have to welk through the room to g- fromone room to another.
The Subject looks up for a moment, observes them, apparently distracted by the shuffle of feet.
He shifts his chair pack,away from his table.
Now he pulls it forwardonce again, rearranges him- self in the seat, leans over hispaper and starts to work once again.
He has apparently finished theanswers of one of the instruction sheets.
He picks it up, gets out of his seatnow, walks over-- Ex] to the teacher's desk. r-4z Cr* CDC-0. He says, "I finished it." 0 CD W ct F-4ct The teacher says, "Pat it o on the desk," and he looks z at her for a moment and puts it on top of the desk. o o
"3(I) He stands there for a moment, hesitating, o perhaps 1_, thinking she's to give him another instruction.
Then he whirls around and startstc walk back toward his own seat.
He walks toward hisown desk, or table; hesitates for a moment.
Now he walks around back toward another table,slightly to the left of the teacher's desk and he startsto go through some papers there.
He finds apparently another set ofinstructions, math problems, and takes the new sheet ofpaper and walks back to his own seat.
He sits there in the wrong seat, theseat next to his own, and realizes his mistake, then gets up and rearranges himself and sits himself at hisown seat.
He puts the sheet ofpaper out in front of him, picks up his pencil as though he's prepared to start his work again. 107
He looks over toward theboy seated at his left's paper, leans over it for a moment,examines it, says nothing now, and returns tohis own paper.
The Subject takes thenew instruction sheet and holds it up for a moment, examinesit, now he places it down on thI table and startsto do the work.
The Subject seems tohave considerable interest in his work; he seems to bemore productive than most of the students around him.
He seems to be able toconcentrate easily and works well on his own.
Most of the studentsare, who have been told to work at their desks, are not doingvery well, they're talking to neighborsor looking about the room, involved in other activities,most of them easily distracted.
The Subject, on theother hand, seems to be ableto handle the assignment athis desk very easily.
He looks up fora moment from his own paperover toward the boy seated nextto him once again,peers over his shoulder, looks at hispaper, now he looks back toward hisown paper.
10:36
The Subject starts torenumber his paper,as he had earlier.
He looks'up fora moment as a boy walks by his desk, now back down toward hisown paper and continues to number the margin side ofthe paper.
The boy looks over hisshoulder for a moment,says something to him.
The Subject once againlooks up and also sayssome in return.
He continue's to talk.
Now he turns around andsays something to the boy seated at his left.
The boy who is standingsits down now at the Subject's right and they continueto talk. 108 The other boy leans 0 (n17 over and cups his mouth with 0 mi his hand as though to ctC talk more confidentiallyto C 0 the Subject. (D ti Cl) CC The other boy apparentlyfinished his conversation, (D 0 he gets up andmoves away. cn
The Subject now goes back to his work andcontinues to number hispaper.
10:37
The Subject beginsto examine his instruction now. sheet
Now he looks downtoward his own paper He looks up for a moment toward the girlseated across from him; he's watchingher, watching and looking about theroom now, observing the activity going on.
10:38
The Subject continuesto examine the activity room. in the
He's gazing almost indifferently, aimlesslyabout the room, without really selectingany one activity to concentrate on; hiseyes seem to just momentarily glance about theroom. He turns to the boy seated at his left andsays something to him.
The other boy leansover closer to him to perhaps him better. hear
The Subject continues to talk to the boyalthough he's also continuing tonumber his page, workingwhile he's talking at thesame time. He's not actually at the otter boy, looking only occassiona17y looksup in the direction of the otherboy. Most of the time he's talking and hovering over the paper, lookingup only occassionally.
10:39
The Subject holds hisinstruction sheetup in the air now, glancing from hisinstruction sheet to his 109
writing paper, going from question toanswer, apparently, with his glance, writing all the while.
A boy in an orange shirt, apparently asksa question of the Subject and the Subject turns to him and says, "WELL, you better solve it yourself. Why don't you go on to the next problehi?" 0 cc The Subject continues to stare at theother boy, as the other boy turns away and apparently taking the cn advice and starting a new problem...... _ cr 0 Now the Subject once again looks back toward hisown paper.
10:40
The Subject continues to work with hisown problems. He spreads the paper out with his hands, touches it with his hands, readjusts its position, continuesto solve the problems.
He continues to glance continually from instruction sheet to his answer sheet and gaining the gist ofthe question and then producing ananswer on his piece of paper.
Now he's studying the questionson the instruction sheet, nervously twittering with the pencil in his hand as he reads the questions on the instruction sheet.
Tile Subject looks up for a moment,distracted, as it were, from his work, looks up at a student who is walking by.
He once again looks down toward hisown paper, resumes his work, continues working.
10:111
He looks up for a moment toward the blackboard, studies it for a moment, perhaps .;hecking outthe assignment.
Now he once again looks down towardhis paper, continues his writing, continues to look fromhis instruction sheet to his paper.
There's a bit of activity occurring inthe direction of the observers; some studentsare questioning our presence, why we're here. I 110 The Subject turns around fora moment to watch this activity.
One of the boys who's just enteredinto conversation with the observers walksaway and the Subject follows him with his eyes as he walksaway to his own seat.
The SubjecL; thenonce again resumes his work, looking down toward his paper and writingwith his pencil, apparently resolvingsome of the problems.
Observer 2.
10:42
He looks at his sheet,concentrating quite hard on what he's doing.
He alternately worksa problem, and then sits back and looks at it to see whetherhe's done it right.
He looks at the teacher,then back at his sheet.
He stands up, bracingone leg against the side of his seat and then sits again.
With his head on his left hand,writes with the right hand.
There is some evidence ofrestlessness as he slightly taps his right leg.
He doesn't seem toointerested any longer in the math work.
He is beginning to be easilydistracted by other forces in the classroom.
10:43
Children are moving around, buthe pays no attention to them, but he continuesto concentrate on his work.
He's chewing gum sort ofabsentmindedly as he glances at me.
He really isn't interestedin the math at all. 111
Tossing his head lightly to get the hairout of his eyes, he shrugs his shoulders.
By now distracted by what's goingon around him, he looks at me, looks at the kid nextto him, rattles his nencil, looks back atme again, and shrugs his shoulders.
He seems to be sort of tryingto dis'dract himself.
He says something to the boy nextto him something unintelligible.
The boy doesn't answer, heleans back in his desk, leans forward again.
He looks at me, grins, picksup his pencil and rattles it.
10:44
He's watching the teachernow, she's helping some of the kids in the class.
He continues to watch her ina very abstracted manner for a long time.
He slouches back in his chair, hisleft and right legs thrown in front of him.
Looking'ahead, he then glances atthe ceiling, puts both hands on head.
He makes a mark on the sheet,erases it, makes another mark, busy scribbling.
He gets up, begins to returnhis paper to the kit box, walks up to the table, there'sanother boy there.
He says something to the boy nextto him there.
He taps with his pencilon his own work sheet. Says something to himself quietly.
10:45
He leans forward to write, he'sgot a new sheet of paper now.
His lips move silentlyas he works.
He reaches down, touchesthe page with his pencil, 112
touches his own collar, and alternateschewing his gum and tapping his head.
He leans over, touches the desk of the boynext to him.
Pushing the hair out of hiseyes, his head in his left hand, he begins writing again.
10:46
Lifting his head, he stops writing and restshis head again on his left hand.
His "concentrating gesture" seems to bethat his left hand shades his eyes.
He's back doing that again.
He begins scribbling, leans back, looksat the sheet.
he writes something on a piece ofpaper, then some- thing on the worksheet.
He's holding the sheet in his lefthand.
He may be bored by it, itmay be too easy for him too.
He doesn't seem distracted byany of the other children in the classroom unless hereally actively goes out of his way to be distracted.
10:47
He erases, begins to write again,shading his eyes again.
His hand brushes the hair out ofeyes.
He begins rocking back and forthin his chair.
He leans back very far in his chair,stretching.
The hand that is shading hiseyes moves down to his mouth.
He's moved his left handnow down to the desk and he runs it up and down the side ofthe desk, tapping absently on the side of the desk.
He looks at.me, looko back downat the page, day- dreaming again now, looking offinto space, not at the teacher, not atanyone. 113 He seems able to go through the sheets veryquickly, one problem after another, when he'snot daydreaming. 10:48
He continues to work.
He shades hiseyes with his left hand, althoughthere isn't really any directsunlight. It seems to bea gesture of concentration.
His legs are lockedunder him in the chair.
He works steadily. He writes rather lightlybut I can't tell how largehis handwriting is. He doesn'tmove, he appears quite relaxed.
10:49
He continues to writewith his right hand.
Looking down, he mouthsthe words silent7.y to himself.
Scratching his head withhis right hand, he leans his head (.1r, his lefthand to see what he's done, leans on his elbow, checkingthe work on his sheet.
10:50
He writes new without shading his eyes, his lefthand is holding the sheet.
Raises his hand towave the teacher over.
Says something tothe teacher as shecomes up to him.
She says somethingback, it may bea correction or just simply a statementof fact. As the teacher turns and leaves, he continuesto work. She turns and comes back and says, "Ifyou want to you can take tile post test, doyou want to do that?" He nods affirmatively.
He gets another worksheet from his deskto take the post test.
He leans over to the desk of the boy nextto him, says something to him, theother boy does not respond. 11/1
He looks arouna the room, heleans back in his chair, leans forward, picksup his pencil.
His left hand on the desk, he drawsa line down the middle of his paper.
Beginning to write now, he numbersdown one side of the paper.
Folding the sheet again, he's leaningsomewhat on his left elbow.
He writes in whatseems to be a light, clear hand, with his right hand.
The noise level in the classroomis slightly higher now.
He makes another line, carefully,down the side of the paper and continues to number. r Standing up, he takes his sheetup to a shelf and places it there.
He stands over by the shelf,evidently chosing another arithmetic sheet.
10:52
Someone is with him now. They're both thumbing through the box.
The Subject picks up somethinghe's dropped andmoves back to his seat.
Returning to his seat, he beginsto write on a sheet of paper now with his righthand, holding thepaper with his left hand.
10:53
A boy comes up sort of toshow him what he's doing ct L71 on 41 his sheet, and the Subjectsays, "Let's see." The ct other boy teases, drawing the CA 0 sheet back. CD may The Boy movesaway, the Subject stands up, watching 0 some other children in the classroom,gets out of his ce seat, and moves toward the backof the room. The observation ends. REFERENCES
Anderson, N. The hobo. Chicago: University of Chicago Press, 1923.
Ardrey, R. The territorial imperative. New York: Atheneum, 1966.
Barker, R. G., ed. The stream of behavior. New York: Appleton-Century-Crofts, 1963.
Barker, R. G. "Explorations in ecological psychology." American Psycholop,ist, 1965, 20, 1-14.
Barker, R. G. Ecological psychology. Stanford, California: Stanford University Press, 1968.
Barker, R. G. "Wanted: An eco-behavioral science." In E. Williams & H. Raush, eds., Naturalistic viewpoints in psychological research. New York: Holt Rinehart & Winston, 1969. pp. 31-43.
Barker, R. G. & Barker, L. S. "Social actions in the behavior streams of American and English children." In R. Barker, ed., The stream of behavior. New York: Appleton-Century- Crofts, 19-63. pp. 127-159.
Barker, R. G. & Gump, P. V. Bigschool, small school. Stanford, California: Stanford University Press, 1964.
Barker, R. G.& Gump, P. Personal interview, March 1970.
Barker, R. G. & Wright, H. F. "Psychological ecology and the problem of psycho-social development." Child Development, 1949, 20, pp. 131-143.
Barker, R. G. & Wright, H. F. One boy's 'day. New York: Harpers Bros., 1951.
Barker, R. G. & Wright, H. F. Midwest and its children. New York: Harper & Row, 1955.
Barker, R. G., Wright, H. F., Barker, L. S., & Schoggen, M. Specimen records of American and English children. Lawrence, Kansas: University of Kansas Press, 1961.
115 116 Benedict, R. F. "Anthropology and the abnormal." The Journal of General Psychology,1934, 10, 59-80. Brown, E. "A study inecological psychology:the behavior objects used by three year-old children fromthree income groups." DARCEE Papers andReports, 1969, 3 (6). Brunswik, 7. Systematic andrepres ntative design of Rsychological experiment. 3erkeley, California: University of CaliforniaPress, 1947.
Caldwell, B. M. "A new 'Approach'to behavioral ecology." In J. P. Hill, ed., Minnesota symposiaon child psychology. Vol. 2. Minneapolis: Universityof Minnesota Press,1969. Caldwell, B. M., Honig, A. S., & Wynn, R. APPROACH--A procedure for patterningresponses of adults and children. Unpublished manuscript, 1967. Calhoun, J. B. "Mortality and movementof brown rats (Rattus Nornegicus) inartificially supersaturated Journal of Wildlife populations." rianagement, 1948, 12 (2),167-172. Calhoun, J. B. The ecology and sociology of the Norwayrat. Bethesda, Maryland:Public Health Service No. 1008, U.S. Publication Department of Health,Education and Welfare, 1960.
Calhoun, J. B. "Population densityand social pathology." Scientific American,February, 1962, 206 (2), 193-148. Calhoun,. J. B. "Ecological factorsin the developmentof behavioral anomalies. In Comparativepsychopathology. New York: Greene &Stratton, 1967.
Cavan, R. S. Suicide. Chicago: University 1928. of Chicago Press,
Dickman, H. "The perception of behavior units." In R. Barker, ed., The stream ofbehavior. New York: Appleton-Century- Crofts, 1963. pp. 23-41.
Dubos, R. Man adapting. New Haven: Yale 1965. University Press,
Dyck, A. "The social contacts of some Midwestchildren with their rlrents and teachers." In R. Barker, ed., The stream of behavior. New York: 7:963. pp. 78-98. Appleton-Century-Crofts, Faris, R. E. L. "Demography of urbanpsychotics with special reference to schizophrenia." American Sociological Review, 1938, 3,203-209. Faris, R. E. L., & Dunham, H. W. Mental disorders inurban areas. Chicago: Universityof Chicago Press, 1939. 117
Fawl, C. "Disturbances experiencedby children in their natural habitats." In R. Barker, ed., Thestream of behavior. New York:Appleton-Century-Crofts, 1963. PP. 99-126.
Frazer, S. G. "The scapegoat." In The Golden Bough. Vol. 9. 3rd edition. New York. St. MartinPress, 1913. Freedman, M., Leary, T., Ossorio, A., &Coffey, H. "The interpersonal dimensionof personality." Journal of Personality, 1951, 20,143-161. Fried, M. "Social problems andpsychopathology." In Urban America and the planning of mental Healthservices. Vol. 5, Symposium10, 1964, 403-446. Gump, P. V. The classroombehavior setting: Itsnature and relation to studentbehavior. Final report to the U.S. office of Education, Project No. 5-0334,Contract No. 0E-4-10-107, 1967. Gump, P.V. "Intra-setting analysis:The third grade class- room." In E. Williams & H. Raush, eds.,Naturalistic viewpoints in psychologicalresearch. New York: Holt, Rinehart & Winston, 1961. pp. 200-220. Gump, P. & Kounin, J. S. "Issues raised byecological and 'classical' researchefforts." 1960, 6, 145-152. Merrill-Palmer Quarterly,
Gump, P. Schoggen, P., & Redl, F. "The behavior ofthe same child in differentmilieus." In R. Barker, ed., stream of behavior. The New York: Appleton-Century- Crofts, 1963. pp. 169-202. Gump, P., & Sutton-Smith, B. "Activity-setting and interaction." social American Journal ofOrthopsychiatry, 1955, 25, 755-760. Hawley, A. H. Human ecology: Atheory of communitystructure. New York: RonaldPress, 1950. Hollingshead, A. B., & Redlich, F. C. Social class andmental illness. New York: Wiley, 1958.
Jenkins, R. L., &Brown, A. W. "The geographical of mensal deficiency distribution in the Chicago area." Proceedings of the AmericanAssociation on Mental 59, 291-308. Deficiency, 1935,
Jordan, N. "Some formalcharacteristics of the behavior two disturbed boys." of In R. Barker, ed., Thestream of behavior. New York: pp. 203-218. Appleton-Century-Crofts, 1963. 118 Kelly, J. G. "Ecological constraints on mental health services. American Psychologist, 1966, 21, 535-539.
Kelly, J. G. "Naturalistic observations in contrastingsocial environments." In E. Williams & H. Raush, eds., Naturalistic viewpoints in psychologicalresearch. New York: Holt, Rinehart & Winston, 1969. pp. 183-199.
Kelly, J. G. "Towards an ecological conception ofpreventive interventions." In D. E. Adelson & B. Kales, eds., Community psychology and mental health. San Francisco: handler PublishingCo., 1970. pp. 126-145.
Kounin, J. S., Friesin, W. V., &Norton, A. E. "Managing emotionally disturbed children in regularclassrooms." Journal of Educational Psychology, 1966, 57(1),1-13.
Langner, T. S.& Michael, S. T. Life stress and mental health: The Midtown Manhattan Stull,Vol. 2. New York: Free Press, 1963.
Leary, T. Interpersonal diagnosis of personalitL. New York: Ronald Press, 1957.
Lewin, K. Field theory in social science. New York: Harper & Row, 1951.
Lindquist, E. Design and analysis of experiments inpsychology and education. Boston: Houghton-Mifflin,1953.
Lorenz, K. On aggression. New York: Harcourt, Brace & World, 1966.
Mead, M. Growing up in New Guinea. New York: Marrow, 1930.
Mercer, J. R. "Social system perspective andclinical perspective: Frames of reference forunderstanding career patterns of persons labelledas mentally retarded. Social Problems, 1965, 13, 18-34.
Odum, E. P. Fundamentals of ecology. Philadelphia: W. B. Saunders, 1953.
Park, R. E., & Burgess, E. W. The city. Chicago: University of Chicago Press, 1925.
Proshansl7, H. M., Ittelson, W. H. Rivlin, L. G., eds. Environmentalpal1921and his physicalsetting. New York: Holt, Rinehart &Winston, 1970.
Raush, H., Dittman, A.& Taylor, T. "The interpersonal behavior of children in residential treatment." Journal of Abnormal Social Psychology, 1959a, 58,9-27. 119
Raush, H., Dittman,A.& Taylor, T. change in social "Person, setting and interaction." Human Relations, 1959b, 12(4), 361-378.
Raush, H. Farbman, I. & Llewellyn, L. "Person, setting and change in socialinteraction." 13, 305-332. Human Relations, 1960,
Redl, F. & Wineman,D. Children who hate. Macmillan, 1951. Glencoe, Illinois:
R1)odes, W. C. "The disturbingchild: A problem of management." ecological Exceptional Children, 1967,33, 449-455. Rhodes, W. C. "A communityparticipation analysis of disturbance." emotional Exceptional Children,1970, 36, 309-314. Schoggen, M. "An ecological study of three-year-oldsat home." DARCEE Papers andReports, 1969, 3(7). Schoggen, M., Barker, L. S. & Barker, R.G. "Structure of the behavior of Americanand English children." Barker, ed., The In R. stream of behavior. New York: Appleton-Century-Crofts, 1963. pp. 160-168. Schoggen, P. "Environmental forcesin the everyday lifeof children." In R. Barker,ed., The stream of New York: behavior. Appleton-Century-Crofts, 1963. pp. 42-69. Schoggen, P. "Mechanical aidsfor making specimen of behavior." records Child Development, 1964,35, 985-988. Schoggen, P. "Environmental forcesin the everyday lives children with physical of manuscript. disabilities." Unpublished Eugene, Oregon:University of Oregon, 1968. Sells, S. B. "An interactionistlooks at the environment." American Psychologist,1963, 18, 696-702. Sells, S. B. "Ecology and thescience of psychology." In E. Williams & H.Raush, eds., Naturalistic psychological research. viewpoints in New York: Holt, Rinehart& Winston, 1969. pp. 15-30. Shaw, C. R., et.al. Delinquency areas. Chicago: University of Chicago Press,1929.
Shaw, C. R., & McKay,H. D. Report on the National Commission causes of crime. on Law Observance andEnforcement. Washington, D.C.:U.S. PrintingOffice, 1931. Shaw, J. W. & Schoggen, M. Children learning: Samples everyday life of of children at home. Nashville: Demonstration and Research Center forEarly Education, George PeabodyCollege for Teachers, 1969. i 120
Simmons, H.& Schoggen, P. "Mothers and fathersas sources of environmentalpressure on children." In R. G. Barker, ed., The stream ofbehavior. New York: Appleton-Century-Crofts, 1963. pp. 70-77.
Skeels, H. M., & Dye, H.B. "A study of tne effectsof differential stimulationon mentally retarded children." Proceelin s of the AmericanAsLociation on rental De iciency, 1939,, 11 -13
Sonequist, E. The marginal man. New York: Scribners, 1937.
Spitz, R. A. "Hospitalism: An inquiryinto the genesis of psychiatric conditions in earlychildhood." Psycho- analytic Study of the Child. Vol. 1. New York: International UniversitiesPress, 1945.
Spitz, R. A. "Hospitalism: A follow-up report." Psychoanalytic Study of theChild. Vol. 2. New York: International UniversitiesPress, 1947. Srole, L., Langner, T. S., Michael, S. T., Opler,N. K. & Rennie, T. C. Mental health in the metropolis:The Midtown Manhattan study. Vol. 1. New York: McGraw -Hill, 1962.
Theodorson, G. A., ed. Studies in human ecology. New York: Harper & Row, 1961.
Willems, E. P. "An ecologicalorientation in psychology." Merrill-Palmer Quarterly, 1965,11, 317-343.
Willems, E. P. "Sense of obligationto nigh school activities as related to school size andma-ginality of student." Child Development, 1967,38, 1246-1280.
Willems, E. P. "Planning a rationalefor naturalistic research." In E. Willems & H. Raush,eds., Naturalistic viewpoints in psychologicalresearch. New York: Holt, Rinehart & Winston, 1969.pp. 44-73.
Willems, E. P. Personal communication, May1971.
Wright, H. F. Recording and analyzing childbehavior. New York: Harper & Row,1967.
Zorbaugh, :1. W. The gold coast am.. the slum. Chicago: University of ChicagoPress, 1929.