A study of interaction and selected purposes in advisory groups

MISSING IMAGE

Material Information

Title:
A study of interaction and selected purposes in advisory groups
Physical Description:
viii, 104 leaves. : ill. ; 28 cm.
Language:
English
Creator:
Springfield, Worley Thaxton, 1942-
Publication Date:

Subjects

Subjects / Keywords:
Interaction analysis in education   ( lcsh )
Teacher-student relationships   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ed. D.)--University of Florida.
Bibliography:
Bibliography: leaves 101-103.
Statement of Responsibility:
By Worley Thaxton Springfield, Jr..
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000580509
notis - ADA8614
oclc - 14037262
System ID:
AA00002228:00001

Full Text











A STUDY


INTERACTION
ADVISOR


AND SELECTED
,Y GROUPS


PURPOSES


Worley


A DI
THE


Thaxton


SSERTATION PRESENTED T
UNIVERSITY OF FLORIDA
OF THE REQUIREMENTS
DOCTOR OF


Springfield,


O THE GRADUATE COUNCIL
IN PARTIAL FULFILLMENT
FOR THE DEGREE OF
PHILOSOPHY













ACKNOWLEDGMENTS


The


author


wishes


to acknowledge


the


general


specific


contributions


thi


study


made


chairman,


Vynce

and B


Hines,


benjamin


and


Gorman.


committee


Without


members

their su


Robert


Curran


ggestions


patience


this


work


would


not


have


been


completed.


In addition,


author


wishes


to acknowl


edge


Charles


Bridges


, Jr.


suggestions


provided


valuable


stati


stical


direction


The


faculty


and


students


at P.K


. Yonge


Laboratory


School


were


quite


generous


with


eir


time


cooperation.


The


author


gratefully


acknowledges


their


help.


And,


finally,


encouragement


of hi


author


wife,


wishes


Charlotte,


to acknowl


whose


edge


patience


understanding


contributed


much


completion


this


dissertation.















TABLE


OF CONTENTS


Pag


ACKNOWLEDGMENTS


S . . . .. ii


LIST

LIST


OF TABLES

OF FIGURES


S . .I . .

. . . . .I .


ABSTRACT


CHAPTER


INTRODUCTION


. . S . 1


II.

III.


REVIEW


OF THE


COLLECTION


LITERATURE


OF THE


DATA


. . .

. . S


DATA


ANALYSIS


SUMMARY,


LIMITATIONS


, AND


CONCLUSION


. 88


APPENDIX:


RAW


DATA


FROM


OBSERVATIONAL


RECORDS


S. 96


BIBLIOGRAPHY


BIOGRAPHICAL


SKETCH













LIST


OF TABLES


Table


Page


Correlation


Original


Coeffici


and


ents


Replicated


for a Sample
Observational


Data


Change


Rate


Advisor


Interaction


Change


Rate


Group


Member


Interaction


. 64


t Values


the


Interaction


Variability
Initiated by


Number


Group


of Units


Member


Change


Rate


Advi


sor


Interaction


tion


S . 68


Change


Rate


of Advi


sor


Intera


action


section


S. . . . 70


Change i
Section


Rate


Group


Member


Inte


reaction


S . . . . 72


Change


in Rate


of Group


Member


eraction


Section


. . . . 75


Change


in Rate


Section


B Following


Sec


tion


Interaction


Made


One Group


Member


. 77


Change


in Rate


of S


section


B Following


Section


Interac


tion


. . . . 80


Change


in Rate


tion


D Following


Section


Interaction


. . . . . 82


Change


Rate


Section


D Following


tion


Interaction













LIST OF FIGURES


Figure


Page


Advisory group schedule


S. 47


Recording


sheet


S . 50


Mean values


for that


portion of


all


interac-


tion


which was made by


advisors


Mean


values


that


portion of


all


interac-


tion


which was made by


group members


Mean


values


that


portion


interac-


tion made


advisors categorized


in Section A


Mean


values


for that


portion


interac-


tion made


advisors categorized


in Section


Mean


values


that


portion of


interac-


tion made
Section A


group members


categorized


S. 73


Mean


values


that


portion of


interac-


tion made
Section D


group members


categorized


S. . . . 76


Mean


values


that


portion of


Section A


interaction
tions made


followed


Section B


one group member


interac-


. . 78


Mean


values


that


portion of


Section D


interaction


followed by


Section B


interactions


Mean


values


that


interaction followed


portion of Se
by Section D


action C
interactions


Mean


values


interaction


for that
followed


portion of
by Section


Section B
D interactions







Abstract


the


the


Diss


ertation


University


Requirements


A STUDY


Presented


Florida


Degree


INTERACTION


AND


to the
Partial
f Doctor


SELECTED


Graduate


Council


Fulfillment


Philosophy


PURPOSES


IN ADVISORY


GROUPS


Worley


Thaxton


Springfield,


December,


1973


Chairman


Vynce


. Hines


Major


Department


Foundations


Education


The


purpose


study


was


to d


determine


success


the


advi


sory


group


program


achieving


certain


their


goals


as refl


ected


changes


verbal


interaction.


an attempt


coincidental


to counteract


an innovation


negative

curriculu


potential


P.K.


Yonge


Laboratory


School,


faculty


proposed


that


advisory


groups


be establi


shed.


Every


pupil


high


school


was


placed


one


advi


sory


groups


in his


grade


level.


Instructional


staff


members


served


as advi


sors


Six


purposes


were


included


the


original


proposal


advisory


groups.


They


are


to provide


eac


h student


with


stable


group


to which he


can


relate


"home


base,


to provide


the


student


with


a place


dialogue


about


school


-related


problems,


to build


a feeling


community


among


students,


to provide


each


student


with


an adult








experiences


increase


involvement


total


school


program,


and


to provide


early


identification of


students


with problems


and


assist


them in


the


solution of


these


problems.


unit and


the


groups


also


served as


representative


unit


administrative


student government.


Purposes


and


were


those


selected


as criteria


for this


study.


The


faculty


felt


that


achievement


these goals


would


establish


the merit of


this


program


as new


valuable.


evidence could


be gathered which


indicated


that


the groups were


dialogue


take place,


providing


and


"home


sense


base,


of community,


a place


then


advisory


group model


could become a


valuable adjunct


total


curriculum.


To form one


basis


for measuring whether


or not


these


purposes


were


achieved


in any part,


eleven


hypotheses were


proposed -


All


eleven hypotheses deal


with


aspects


verbal


interaction.


suggested


The


that


review of


sufficient basis


literature


existed


this


to suppose


study


that


verbal


interaction reflected


understanding,


changes


and unity


in verbal


levels


the group


interaction,


tolerance,

situation.


insight would


trust,

By measur-


be obtained


into changes


the


levels of


tolerance,


trust,


understand-


ing,


unity.


Interaction Process Anal


ysis,


a category


observation








occasions


three


times


the


first


school


year,


three


times


at the


middle


school


year,


and


three


times


end


the


school


year.


Each


SIX


groups


the


sample


was


observed


and


verbal


interaction


recorded


twenty


minutes


nine


data-gathering


sessions


Due


to circumstances


beyond


the


control


the


res


earcher,


one


group


was


observed


only


five


times


and


one


other


group


only


times.


The


raw


data,


numbers


interactions


specific


category


, were


converted


to proportions


and


tested


ten


hypotheses


via


the analyst


variance


method.


Signifi-


cancer


was


eve


recognized


For


event


values


of F at


hypothesis


or beyond


proportions


were


tested


via


test.


Significance


was


recognize


d for


values


beyond


eve


on a one-tailed


test.


In addition


ese


observational


values,


period


were


mean


analyze


proportional


graphically


values


each


hypothesis.


the


basi


stati


stical


and


graphic


analysis


servational


data


it is not


possible


to conclude


that


advisory


group


program


made


measurable


movement


toward


achievement


goals


, and













CHAPTER


INTRODUCTION


the


Spring


1971


faculty


P.K.


Yonge


Laboratory


School


recognized


that


their


forthcoming


modular


schedule


student


might


would


encourage


perceive


an environment


an increase


in whi


ch the


distance


between


himself


school.


Unless


otherwise


established,


intercourse


among


students


and


teacher s


could


become


limited


to classroom


busin


ess,


generating


impersonal


atmosphere


school.


This


possibility


prompted


faculty


propose


system


groups


whose


purpose


was


increase


institutionalize


teacher-pupil


interaction


ultimate


educational


goal


greater


learning.


These


groups


would


would


be called


be divided


advisory


into


groups.


groups,


Each


each


grade


under


students


guidance


a member


secondary


faculty,


guidance


staff,


administrative


staff.


These


persons


would


function,


not


teachers,


but


rather


as discussion


leaders


reference


sources,


and


activity


coordinators.


Formally,


purp


ose


these


groups


was


to be


six-


fold


to provide


each


student


with


a stable


group


ys 4 fl)' 'I


-I- Anh 1^t; J" /


lI rl\


T.k: ^ /*l


-w*-^ 4-r-


11 U^/-.


fc^ -i








student with a


place


for dialogue about


school-related


problems,


to build a


feeling


of community


among


students,


to provide each student with an


adult


who


is responsible


the coordination of


provide


faculty members with


total


school


program,


common experiences


to increase


involvement


the


total


school


program,


and


to provide


early


identification of


students with problems and


to assist


them


solution


these problems.


In addition,


these


groups would


serve as


the administrative


unit and


the unit


representation


student government.


One measurable dimension of


achievement


toward


group


maintenance,


moral,


identity,


supportive


purposes


interaction.

reflected in


measurements of


Since growth and


i alteration of


rate


change


in group processes are


rate and pattern of


pattern


interaction,


should provide an indica-


tion of


group development.


The


purpose of


this


study was


to quantify


selected changes


in rate and pattern of


interac-


tion


to determine


achieving


Change


the effectiveness of


formal


in rate was


purposes numbered


identified by


advisory


and


increases or


groups


above.


decreases


in amount of


interaction


as measured by the


systematic obser-


vation


system described


Robert F


. Bales


in Interaction


Process Analys


Change


pattern was


identified by









the


increasing


or decreasing


use


certain


categories


measured


instrument.


Significance


Traditional


practices


in education


which


separate


school


-learning


environment


from


rest


learner


life


are


being


replaced


ctices


which


not


only


deny


this


duality


but


insist


upon


interdependence


two


areas.


The


practical


duality


that


still


maintains


that


classroom


exists


virtual


social


and


psychological


isolation


from


remainder


individual's


environment.


Under


these


circumstances


, the


critical


fluences


that


bear


on learning


are


directly


accessible


educator


The


ernative


position,


that


cla


ssroom


is not


isolated


and,


therefore


, some


factors


not


readily


accessible


teacher


influence


the


ability


student


time.


learn,


Only


held


recently


theoretical


alternative


importance


gaining


some


wid


spread


signifi


chance.


advisory


group


program


to be


prop


ose


as an


alternative,


it must


demonstrated


, empiri


cally,


work.


extent


that


interaction


anal


measures


movement


toward


purposes


, and


advisory


groups,


study


will


measure


some


criteria


success


lack


r


~I r ~nrt M fl- ~ -C J- --- r~ 'V v~ -


C- -1


1)


II ~\ hnV f-V n- V"^


*I v


-"i ^i -r -^-"


iZ~ulr~


~*I rn


f~









other


methods


should


tried.


But


groups


succeed,


one


model,


with


some


demonstrable


evidence,


will


avail


able


teachers


to help


them


deal


with


student


total


environment


improvement


their


education.














CHAPTER


REVIEW


OF THE


LITERATURE


This


review


literature


may


viewed


three


parts

group


Part

system


describes


scusses


programs


similar


significance


the advisory

these pro-


grams


as educational


measurable


effects,


methods,


evident


Part


when


examines


individual


three


participate


groups


, which


are


shared


programs


in part


one


and


discu


sses


these


effects


in relation


purposes


numbered


advisory


groups


Part


three


scus ses


eraction


as one


means


which


these


effect


s are


reali


zed and


degree


to which


changes


interaction


among


individual


in groups


refl


ects


movement


toward


away


from


those


ects.


idea


that


schools


are


potentially


Imper


sonal


that


steps


should


taken


to avoid


or assuage


this


potential


is not


new


Home


rooms


efined


"means


providing


pupil


-teacher


contacts


in a large


school


were


establi


shed


as early


1875


Jones


reports


that


growth


had


plateaued


at a high


eve


Harry


. McKown


off


ers


Tamn c


tIt T i' Tn


f a Cnhhh ^


n-*I


, 2,


p"d TN^


D r^T i ar IT-


T~f-n.^-


M-


*


J








more


complete


description


home


room


as a


"regular


school


period


, usually


weekly,


which


the


teacher


sponsor


meets


with


an organized


group


students


purpose

and, th


of becoming


rough


intimate


individual


acquain


contacts,


and


ted with

programs


the

and


members

activi-


ties,


promotes


development


certain


personal


ideals


knowledge


, and


habits


not


now


regularly


provide


ed for


teaching


the


traditional


school


subjects


Furthermore,


says


"The


home


room


places


main


emphasis


upon


education


body


the


Yarnell


organize


subj


work


They


action


student


matter


Joseph


begin


which


rather


than


even


Roemer,


briefly;


pupils


[the]


fuller


Charles


"that


are


pass


along


definition


. Allen,


unit


grouped


found


Dorothy


school


purposes


carrying


on school


routine


and


stimulating


and


develop-


pupil


initiative


through


various


types


of home-room


activities


Thi


framework


fille


out


these


words


. the


home


room


is not


merely


administrative


func-


tion.


intere


, at


sts,


same


problems,


time,


initiative


ace


find


where


solution


individual


outlet,


and


where


group


ideal


s are


foster


ed and


lived


Here


guidance


carried


; disc


ipline


administered;


self-consciousness


overcome;


individual


problems


are


solved;


pupil


are


inspired


to greater


effort;


sympathies


are


broadened;









leadership


intelligent


obedience


are


developed.


The


home


room


in fact,


a means


contact


connecting


correlating


interests


teacher,


pupil,


school,


parent,


and


community,


and


fostering


a broader


view


privi-


leges,


duti


responsibility


an intelligent


citizenry.


In ideal


conditions


the


home


room


was


serve


as a


place


where


guidance


was


available


was


no other


place


school


environment.


Other


units


school


their


responsibility


administration,


subject


matter,


or physical


health, but


no other


one


was


concerned


with


emotional


While


attitudinal


practical


development


application


as was


these


ideas


home


room.


resulted in


abuse

tial


and


modification,


original


theoretical


concept


remained


erest


high.


Compare


poten-

the


justification


1935


with


those


given


during


twenty


years.


first


World


War,


urbanization,


industrialization,


a developing


educational


philosophy


combined


to produce


a situation


significantly


different


from


that


middle


and


late


1800


Roemer,


Allen,


and


Yarnell


described


differences


between


latter


half


nineteenth


cen-


tury


and


the


first


third


twentieth


century


this


manner


. homogeneous


personnel


changed


to hetero-


generous


student


bodies;


programs


studies


multiplied


their









offerings;


secondary


school


varied


their


curricula


to meet


the


ever


changing


and


broadening


need


their


student


en-


rollments


new


psychology


sugg


ested


new methods;


and


improved


transportation


and


communication


facility


gave


democracy


new


meaning.


The


greater


amount


of leisure


time


made


pos-


sible


inventions


gave


more


time


education.


The


World


War


had


tremendous


influence


the


social


and


civic


phases


citizenship.


Philosophical


changes


educational


objectives


suggested


Commission


Reorganization


of Secondary


Education,


Franklin


Bobbitt,


and


others


added


increasingly


complex


educational


environment.


6 Roemer


et al.


state


that


"the


stud


ent


was


being


lost


sight


mel


mass


organization


admini


provide


station.


closer


contact


had


to be


teacher


some


and


way


devised


student.


That


way


was,


course


home


room.


Twenty


years


later,


same


general


position


was


not


only


sound


own,


but


now


achieved


called


deve


a di


lopmental


guidance.


In 1953,


base


Little


and Chapman

each pupil


defined

guidance


developmental


and


services


provide


which


are


planned


and


implemented


in accordance


with


powers


develop


personally,


socially,


educationally,


vocationally


--~ ~~~ S -a a


There


between


stinct


theoretical


guidance


instructional


__ A


*A


, I


*


I.


jr


1









empha s i s


child


on providing


a distinctive


needs


facet


each


of developmental


every


guidance


a guidance


program.


Thus


Peters


and


Farwell,


a bri


historical


cancer


sketch


distinctly


developmental


modern


guidance,


American


assign


ideal


signifi-


individual


uniqueness


individual


right


maximum


development


instrumental


the


growth


of developmental


guidance.


Dinkmeyer


Caldwell


describe


with


clarity


that


"guid-


ance


that


part


educational


program


which


emphas


izes


individual.


Whil


it does


not


deny


the


remedial


and


corrective


aspects


guidance,


developmental


guidance


stead


works


toward


ass


listing


each


child


utilize


oppor-


tunities


learning


personal


development.


con-


cerned


with


average


youngster


who,


because


he presents


no special


problem,


frequently


receives


minimal


attention.


service


is not


a specialized


therapeutic


service


adjunc-


tive


school,


but


is a part


educational


process.


is concerned


with


helping


child


learner.


The


degree


to which


developmental


guidance


can


"help


child


a learner"


related


ability


to develop


positive


attitudes


child


toward


self


and


one


s self-


Herman


Peters


and


Gail


. Farwell,


Guidance


A Devel-


oomental


ADoro


ach


(Chicaao


Rand.


a


McNallv


Co.. 1967)


. D.


1M









worth.


For


as both Dinkmeyer


and CaldwellI


and Peters and


Farwell12


state,


a great deal


evidence,


empirical and


theoretical,


has


been produced


to support a relationship be-


tween self-concepts


ability to achieve.


Therefore,


a program of


developmental


guidance,


such as


the advisory


group program,


to be


successful,


it must


influence


the


growth of


positive attitudes of


self


and


self-worth.


Three effects of


individuals


interacting


in groups


are


chosen,


from many,


to demonstrate


that certain aspects of


the group,


as a social


tem,


may


be empirically related


to purposes


individual


and


interacting


the advisory groups.


in groups that has


One effect


been related


to feelings of


community


and


openness


is cohesiveness.


Hubert Bonner


defines cohesiveness as,


forces which induce members


"the totality


remain in a group.


of

In


other words,


cohesiveness might be equated


with


sense


community


in a


group.


These


forces are divided by Dorwin


Cartwright and Alvin


Zander


into


least


types of


components:


forces


that derive


from


the group'


attrac-


tiveness


and b)


forces whose


source


the attractiveness of


alternative memberships.


In a


footnote


those authors


Ibid.,


Peters and Farwell,


catio


Hubert Bonner,
ns (New York:


Group


Dynamics:


Ronald Press Co.,


Principles


1959),


and Appli-


A.-


, 2,


___









add a


third component,


"forces


against


leaving the group


that result


from costs associated


with


leaving


from other


restraints.


Cohesiveness


been operationally


defined


scores on


various


self-report


inventories,


sociometric


analyses,


and


interaction profiles.


The


been


relationship


explored


between


several authors.


cohesiveness


community


Stanley Seashore


found


support


for the hypothesis


that degree of


cohesion,


our


terms,


community,


was


positively related


to personal


at-


traction among members,


opportunities


interaction,


power


the group


to common


conformity.


Cooperation,


which


Martin Grossack


says


is a determinant of


group cohesiveness,


can be


related


to communication


patterns,


which are measur-


able by


systematic observation


instruments.


In his


study,


Grossack


found


that members


perceived


as cooperative


"received


significantly more communications,


more


instrumental


com-


munications


(opinion and


information) ,


fewer


consumma-


tory


communications


(tension and antagonism)


Using


slightly d

expressed,


differentt te

individual


of participation.


Th


rminology,

satisfacti

e results o


Robert Porter


would

Porter'


suggested


be related


study


that


to amount


indicated


Ibid.,


footnote,


Stanley Seashore,


Industrial Morale


0 II i


"Group Cohesiveness


Prnlrtl-i vi v -


a Factor


" n i sertati on Ahstracts.








that


instead


participation


general,


expressed


sati


faction


was


related


to amount


group-oriented


participa-


tion.


Another


eff


ect


we might


examine


to explore


prac-


ticality


and values.

basic goals


advisory


groups


advisory


of community


and


is similarity


group


system


communic


action


attitudes


to approach


we should


expect


that


the


attitudes


the


group


members


, particularly


those


relate


to the


value


respect


other


tolerance


would


become


increase


ingly


similar.


The


research


does


indi


cate


that


this


takes


ce.


In a general


statement,


Mary


Alice


Monk


said


that


membership


in a group


is related


perception


group


pothe


s member


that


greater


similarity


James


in situations


in attitudes


found


believed


among


support


to be constructive


individual


perceives


others


with


increased


similarity


himself


an experimental


study


communication


small


groups


whose


member


represented


various


ethnic


groups,


Lawrence


Hamilton


found


significant


support


believing


18.
Sati


Robert


sfaction


tional


. Porter


Small


search


"Relation


Group


ship
*


of Parti


scussions


(Novemb


cipation


" Journal


of Educa-


130.


19. Mary
Attitudes


station


Ali
and


Abstracts


Monk,


ome


Perception


, 14


(1954),


ects


Others


Group


Member


' Attitude


ship


" Diss


er-


1267.


I









that


attitudes


towards


self


and


other


ethnic


group


members


can


be changed.


A third


viduals

advisory


measurable


in groups


groups


effect


in relation

sociometric


interaction


purposes


ratings.


indi-


and


Sociometric


rating


may


be described


as diagrams


intra-group


interaction.


One


explanation


how


sociometrics


reflects


growth


community


that


offered


a self-supporting


George

cycle t


Homans.


ends


Homans


to develop


points


which


out

people


thrown


together


interact


more


with


group


member s


than


with


nongroup


members.


From


this


interaction


evolves


sense


community.


similarity,


among


individuals,


interest


and


attitude


that


results


when


sense


community


emerges


in a group


is reflected


sociometric


analysis.


These


techniques


show


that


these


groups


average


number


interp


personal


relationships


individual


increases.


And


to complete


cycle,


expansion


social


network


means


that


shared


activities


increase


Experimental


support


for

subj


this


hypothe


ect-matter


cla


ene


sses


that


Stanford


compares


incorporated


human


traditional

relations


training


with


subj


ect-matter


classes


that


included


no overt


effort

ing.


include


He found


that


principles


academic


human


achievement


relations


was


train-


comparable









the


two


groups


but,


more


importantly,


that


the


experimental


group


was


significantly


different


from


the


traditional


group


acceptance


others


In a study


group


dynamics


Trotzer


found


that


group-centered


groups


were


significantly


different


from


topic-centered


(structured)


groups


the


bas


empathy,


unconditional


positive


regard,


and


depth


self-exploration


Silluzio


explored


relation


ships


between


interaction


sentiments


in hi


doctoral


study.


The


experimental


groups


were


arranged


to attend


cla


sses


tog


ether,


be graded


a distinctive


system,


have


flexi


ability


their


schedule 1


, and


have


same


teachers


The


control


group


was


drawn


from


population


students


that


had


few


common


classmates,


traditional


grading


system,


inflexibl


schedules


varying


teachers


The


experimental


group


proved


to hav


significantly


more


favorable


attitudes


toward


their


ers


than


did


control


group


Expr


essed


terms


this


study,


individuals


who


interact


at high


level


they


will


advi


sory


groups


eve


a sense


sameness


or community


study


which


demonstrates


one


advantage


of a disper


period


treatment,


as are


Gene


lum,


Stanford,


" Educational


"Sens


Lead


ers


itivity
hip, 28


Educ


ation
ember,


and


Curricu-


1970),


James


Topic-C


Peter


entered


Trot


d Methods


zer


"The


on Group


Eff


ect


Process


Group


and


Out


-Centered


comes


S cP rfI-h i nn


Ah~i-tr t


A9^7A


nl G;Q


I |










advisory


groups


, versus


a concentrated


period


treatment


that


Marilyn


Bates.


examine


the


cognitive


and


affective

groups:


potential

control, t


group


traditionally


unseling,

counseled


Bates

group


(one


Three

meeting


per


group


week


(two


thirteen


meetings


weeks),


days


an accelerated


total


counsel


thirteen


hours)


The


traditionally


counseled


group


was


significantly


different


from


control


accelerated


groups


cognitive,


behavioral,


connative,


and


acculturative


outcomes.


subscore,


acceptance


others,


traditionally


counseled


both


group


control


proved


and


to be


acc


significantly


elerated


different


from


As members


group


show


greater


understanding


one


another


acceptance


of others


group,


barrio


ers


fri


endship


should


reduced


and


amount


involvement


with


oth


ers


increased.


In a study


group


counseling


in which


sub-


jects


were


below-grade-level


readers,


Schmidt


found


that


counseling


group


only


made


significantly


better


reading


improvement


than


the control


also


made


significant


change


in sociometric


rankings


particular


interest


was


that


Schmidt used

treatment was


two

the


treatment

counseled


levels

group


plus

which


a control.

included


One

no reading


Marilyn


Bates,


"A Test of Group


Counseling,


" Personnel


groups


1 I_ _


_


___


'I









improvement


materials


The second


treatment


was


based


the


SRA


Reading


Laboratories


material


The


counseled


group


was


compared


this


group


which


devoted


entire


time


reading


improvement


and


control


The


counseled


group


not


only


surpass


both


ese


groups


reading


improve-


ment


but


growth


a sense


community


as re-


elected


sociometri


analy


that


indicated


that


more


members


had


more


friends


the


counseled


group.


. Of


the


way


which


these


and


other


effects


group


participation


may


be realized,


study


is concerned


with


social


interaction


among


group


member


About


interac-


tion,


Anderson


says


, "Of


factor


s whi


impinge


on a


counseling


group


none


is more


pervasive


than


interac-


tion


among


group


members


thi


pervasive


quality


that


significant


s study


For


reali


zation


advisory


group


purpo


ses


establishing


a home


base, 2)


pro-


viding


community


a place


are


for

all


dialogue,

dependent


and

upon


building


interaction


feeling


among


member s


the


group.


This


relations


between


interaction


and


purpo


ses


and


may


seen


study


reported


Porter


above.


that


study,


Porter


found


that


a significant


relationship


exists


between


expressed


satisfaction,


comparable










feeling of


community,


amount of


group-oriented


participation,


interaction.


The


dissertation referred


to previously


by Silluzio


also


bears on


this relationship.


His


study


compared


stu-


dents who attended all


classes


together


those who


few classmates


in common.


Measuring


their


attitudes


toward s


peers and


teachers,


another measure of


community


feeling,


Silluzio


found


significant


differences


favoring


those


students


interacting


throughout


school


day.


Further


evidence


that


interaction


related


to growth


toward


purposes


is obtained


from Borgatta and


Bales.


This


study was designed


assess


possibility


predicting


interaction rate


patterns


in groups


con-


stituted


from


individuals


previously measured


interac-


tional


rate and


pattern.


finding


that


study perti-


nent


relationship


in question


was


that


quality of


performance,


Bales


as measured


supportive


systematic observation model


is directly related


supportive categories


the rate of

in Bales's s


categories of


(categories

interaction.


system measure


The

feelings of


solidarity,


helpfulness,


acceptance,


and


satisfaction.


These


characteristics are associated,


usually,


with


such feelings


as community,


home


base,


people


you


can


talk to.


Thus,









measures


rate


interaction


should


reflect


exi


tence


these


character


stics.


In another


study


CO-


authored


Bal


, the


authors


found


that


when


individual


in a group


are


ranked


order


number


acts


they


initiate,


they


will


also


tend


to be ranked


number


acts


they


receive,


.address


specific


other


individuals,


and


addr


ess


the


group


as a whol


Thus,


measures


change


interac-


tion


rate


(acts


initiated


plus


acts


received)


measures


change


interaction


pattern


(who


said


what


to whom)


should


measure


whether


advisory


groups


show


growth


toward


being


a pi


ace


where


dialogue


can


occur.


Summary


The


members


groups


that


are


experiencing


po s


itive


growth


toward


achi


eving


solidarity


and


a sense


belonging


may


chara


cterized


as i)


working


harmoniously


with


others


having


attitudes


similar


to other


group


member


having


many


friends


within


group,


identifying


with


group,


having


eater


tolerance


others,


and


having


more


pos


itive


attitudes


toward


group.


These


characteristics


can


be measured


separately, but


they


also


can


be measured,


indirectly,


examining


an element


basic


them


all.


D/riha4+


'U L3nr" I


fla 1


C 4-mA^^ 4-1-m


'Tihn Al rrr


M411


1-1


E- NIV


^ .j









Interaction


in a group,


as behaviors


to which


some


kind


response


is made,


that


common


element.


Previous


research


shows


change


that


measurements


in attitudes


toward


interaction


peers,


reflect,


quality


example,


group


relationships,


in attitudes


toward


group.


mea-


surfing


changes


interaction,


rate


and


pattern,


advisory


groups,


growth


toward


or away


from


purposes


and


may


be demonstrated.













CHAPTER


COLLECTION


OF THE


DATA


Introduction


Chapter


study


is concerned


with


hypotheses


which


were


used


to detect


growth


toward


or away


from


the


selected


advi


sory


group


purposes


process


and


techniques


used


their


data


to evaluate


those


potheses.


Eleven


hypotheses


were


proposed


as criteria


to evaluate


movement


advi


sory


groups.


hypothe


ses


were


phrased


terms


various


aspects


interaction


and


group


were


intended


to provide


one


or more


measures


growth


three


ected advi


sory


group


purposes


The


design


study


was


selected


to maximize


degree


inte


rpretability


In nonlaboratory


situations,


such


this


study,


poss


ibility


variables


interven-


treatment


ect


is great.


When


variables


inter-


vene


treatment


are


not


recognized


such,


the


de-


gree


interpretability


decreased.


The


time


-seri


design,


as described


Campbell


and


Stanley,


offers


greatest









opportunity


conditions of


to minimize unrecognized


this


intervention under


study.


As discussed


in Chapter Two,


interaction


proposed as


a phenomenon


which reflects


change


in the group.


The Bales


model,


Interaction Process Analysis,


received


wide use


was developed


international


analysis


use in


small


have been


groups.


developed


Other models of


teacher-pupil,


labor-management,


or client-counselor


interactions.


present discussion

descriptive terms


a


will

nd in


examine

terms


Bales model


related


level


in general


interpre-


ability

The


that may b

procedures


e expected


used


when


to sample


usin

the


g


this model.


population are de-


scribed


this


chapter.


The


final


section


this


chapter


outlines


timetable


data


collection,


describes


unit


of datum


to be


recorded,


describes


recording


technique.


Hypotheses


The


hypotheses of


this


study


consist


basic


types:


those related


to change


in rate and


those related


to change


in pattern.


Change


in rate


signifies


that


individuals


interact more or


less.


Change


in pattern refers


three


processes.


One,


the distribution of


individual


s total


interaction among


categories may


change.


For


example,


fl an c. r r' 411- n^n .4- r a n C a a n a: n 4. 4- nm a n' r


aa url 4- aF


**'--r -^~ hnh4- nn /-v^ ^\


ilC


tnn









interaction


exceeds


a single


unit,


second


and


succeeding


units


may


be classified


the


same


category


first


unit


they


may


be distributed


among


other


categories.


a point


time,


is possible


to determine


proportions


succeeding


interactions


stributed


among


cate-


gories.


At a second


point


time,


the


proportions


suc-


ceeding


interactions


may


indicate


that


some


categories


have


become


more


actively


used


and


some


less.


Thi


change


the


distribution


acts,


among


categories,


which


succeeds


initial

pattern.


acts

And


second


three


when


type


first


change in interactional

individual temporarily


discontinues


his


interaction.


interaction,


one


time,


a second


possible


individual


begins


to determine


pro-


protons


stribution


responses


made


second


individual


At another


time,


distribution


responses


among


the


category


may


change.


This


change


stri


bution


responses


a second


individual


third


type


international


change


in pattern.


Hyl


total


number


interactions


, that


propor-


tion


made


crease
measure
gories


the advi


across


d by


time


sum


Interaction


sor


the


unit


group


time


will


situation


interactions


Process


Analy


in all


de-


as
cate-


S1S


Hy2


total


number


interactions


group


member


per


unit


tne group
interaction


time


situation
n in all


will


increase


as means


ured


categories


across


time


sum


Interaction


Drn'cc a


C C2


S'n a










One


measure


success


in achieving


purposes


, and


3 of


the


advisory


groups


is change


in rate


intera


action.


individuals


meet


as a group


a period


time,


friends


will


quite


be made


unusual


perhaps


no change


enemies


will


occurred


be made.


the d


It would


degree


per-


sonal


involvement.


consequence


change


in involvement


degree


to which


one


interacts


with


other


the


group


Some


individual


should


receive


more


interaction


and


some


ess.


advi


sory


group


to become


a home


base


purpose


, a place


where


dialogue


can


take


ace


a com-


munity


, then


friend


ships


must


be made,


trusts


established,


and


personal


barriers


dissolved


studi


Porter


and


Silluzio


would


indicate


that


these


character


stics


are


related


increased


rates


interaction.


Hypothesi


designed


measure


Hypothesi


designed


test


assumption


that


the


egree


interaction


dominant


ed by


one


individual


a few


individual


the


realization


advisory


group


purposes


will


be repressed


feeling


that


advisory


a place


where


discussions


sonal


evance


can


occur


will


not


eve


when


one


individual


or one


group


individual


command


a share


interaction


out


proportion


their


numbers.


advisory


group


purposes


__









Hypothesis


tests an aspect of Hypothesis


That


some


individuals must


interact less,


some more,


variance


between


individuals


to decrease.


Since


advisor,


the


initial


group meetings,


should


play


a pri-


mary role


in structuring


group activities and


then allow


this


role


to be assumed by the group members as


they gain


experience


the group


process,


proportion of


the


total


interaction made


The


unit


advisor


proportion of
time devoted


Reactions,
situation
categories


will


should decrease across


advisor's


the categories


increase across


as measured


time.


interaction


time


sum of


Interaction Process


1-3,
in


Positive
the group


interaction


Analysis.


Hy5


The


proportion of
time devoted


unit
tive


group
action


Reactions,


situation


will


advisor'


the categories


decrease a


as measured


in categories


10-12


by
of


cross


interaction per


10-12,


Nega-


time


the sum of
Interaction


inter-
Process


Analysis.


Hy6


The


unit


proporiton of
time devoted


Reactions,


will


group member


interaction


categories


increase


across


time


1-3,
in


per


Positive
the group


situation as
in categories


measured by


sum of


interaction


of Interaction Process


Analysis.


unit


proportion of
time devoted


tive Reactions,


group member


interaction


categories


will


decrease


across


10-12,
time


per
Nega-
in the


group


situation


as measured by


sum of


interac-


tion in


categories


Analysis.


10-1


Interaction Process


A second measure of


success


in achieving


advisory group


purposes


and


change


in pattern.


Hypotheses


7 were designed


test


that


change


in pattern









More


fully,


time


spent


in a group


increases


propor-


tion


an individual


total


interaction


devoted


one


four


sections


of Bales


system


will


change.


For


example,


when


a number


individuals


meet


first


time,


one


would


expect


number


of questions


(Questions),


in propor-


tion


to other


sections,


to be high.


But,


they


become


acquainted


become


more


secure


with


one


another,


pro-


portion


interaction


interaction


devoted


devoted


to questions


to questions


should


decreases,


decrease.

the pro-


protion


interaction


devoted


to other


sections


should


increase.


Hypotheses


represent


test


specific


case


measure


success


discussed


above.


critical


role


advisor


advisory


groups


is developing


climate


conducive


to movement


toward


purposes


interacting


in the


categories


of Section


, Negative


Reactions,


advisor


can


create


a situation


entirely


unsuited


friendly,


open


atmosphere


necessary


home


devot


base and


community


to Sections


to develop.


A and


should


Change


reflect


attention


affective


direction.


members


group


perform


a function


setting


group


climate


least


as critical


advisor's.


attitudes


toward


group


life,


toward


other


group


members


, 2,









negative


nature


Success


in moving


member s

toward


reactions


purposes


Sections


and


A and

should


facilitated


proportionate


increase


in positive


reac-


tions.


As positive


reactions


(supportive


comments,


help,


and


agreement)


antagonism)


increase,


should


negative


decrease.


reactions


Hypotheses


(disagreement


6 and


measured


thi


change


Hy8


The
per


which


will


proportion


unit


time


follows
increase


member


proac


in category
categories


across


intera


, Attempted


, Positive


time


group


action


Answers


Reactions,
situation


as measured


Interaction


Pro


cess


Analy


S1S


An act


proaction


is d


efined


Bal


, "Very


simply,


an act


who


which


a direct


has produced


ast


continuation

act. .,, 2


same


With


member


reference


Hypothesis


, thi


means


that,


across


time,


number


times


an individual


commits


an act


answering


(categories


whi


ch follows


of positive


reaction


categories


1-3)


will


increase.


Research


Bal


shows


that


large


part


time


when


an individual


interacts


that


action


consists of


sequence


acts


rather


than


single


act


The


acts


suc-


ceeding


initial


are


proactive


acts


as defined


above.


These


proactive


acts


are


important


indicators


success


the


advisory


groups


in movement


toward


purpose.


For,









individual


finds


advisory


group a place where


may


recei ye


support and


friendship,


need


to defend


him-


self


or withdraw


inwardly decreases.


His


proactive acts


should


reflect


this move


toward a


greater


sense of


security


in an


environment


that


offers acceptance


support.


Hypothesis


is designed


to test


case


change


pro-


active acts


of Attempted Answers


(Section B)


following


ini-


tial acts of


Positive


Reactions


(Section A).


The


proportion of member reactive


interaction


unit


time


in categories


4-6,


Attempted Answers,


which follows categories


10-12,


Negative Reactions,


will


increase


across


time


the group


situation


as measured


Interaction


Process


Analysis.


HylO The
unit


proportion of


time


member


in categories


reactive
10-12, Ne


interaction


gative


Reaction,


which


decrease
measured


follows categories


across


time


Interaction


7-9,


Questions,


group


Process


will


situation


Analysis.


HY11


unit
which


will


proportion


time


of member


in categories


follows
decrease


categories


across


reactive


10-12,
4-6,


time


interaction


Negative
Attempted


the group


per


Reactions,
Answers,
situation


as measured by


In contrast


Interaction


to proaction,


Process


Bales defines


Analysis.


reactive


act as,


"an act


which follows


immediately the


last act of


another member.


Thus,


a reactive


analysis


indicates


what


type of


statements are made by one


individual


in direct


response


second


individual.


As the


advisory


groups


become what


purposes


outline


they


should,


we may


expect


that


individual


would










attitudes,


more


tolerance


and


understanding,


more


re-


aspect,


more


worth


individual,


more


support


Hypotheses


and


are


designed


test


whether


these


effects


occur.


Specifically,


Hypothesis


designed


the


as sump-


tion


that


reacting


supportive,


tolerant


atmosphere


created

stated


as operative


purposes


group


and


practices


a member


become

will D


parallel

erceive


those


acts


categorized


destructive


as Negative


and,


hence


Reactions


, a member


constructively,


s responses


not


will


constructive


responses.


One


the constructive


sections


Attempted


Answers.


Thus,


hypothesis


will


measure


whether


individual


increasingly


perceives


supportive


moving


atmosphere


toward


group


feeling


community.


Hypothesi


process


tests


increasingly


a different


supportive


aspect


environment


same


should


reduce


need,


individual,


to assert


or d


efind


his


status


Responses


in Section


Negative


Reactions,


are


characterized


as responses


des


signed


to defend


the


self,


actively


through


hostile


behavior


or passively


through


with-


drawal


or denial


help.


When


rate


response


Section


D decr


eases


we have


some


evidence


that


group


llfnlat flfl


bv-tT Fnnmdn ~'


cil, an r aC ani r


ma nZVV


1 r.









to reduce


level


negative reaction


to Attempted


Answers.


In a


situation


in which


individual


feels


lack


support or


self-worth,


he is


likely to respond


to opin-


ions,


suggestions,


offers of


information


(the


categories


of Attempted Answers


though


they were


personally threat-


ening.


individual


other


senses


hand,


support


those


himself


situations where


others, Attempted


Answers are more


likely to


be perceived


as helpful.


The


response may


By measuring


be measured


decrease


as a diminution of


in rate of


negative reaction.


negative reaction


Attempted


Answers,


should


obtain


some


indication


of move-


ment


toward


purposes


and


Design


Data


were collected


evaluation of


applying


hypotheses


Bal


this


Interaction


study


Process


Analysis


category


system


to a random sample of


twenty-


four


advisory groups at


. Yonge


Laboratory


School.


The


sample groups were


observed


three


times on


three


occasions;


beginning


school


year


, and


03),

and


as near


o06)
6


as possible the


and at


end


school


school


year mid-point


year


(07'


The


two-week


period


immediately


after


school


year


begins


and before


ends was


recognized


as either


a period


* I *


.i B ^ *


(01,









D.T.


Campbell


and


Julian


Stanley


s notation


re-


search


design


will


represent


exposure


a group


an experimental


variable


or event,


effects


of which


are


to be measured;


0 refers


some


process


of observation


or measurement;


Xs and


a given


row


are


applied


the


same


specific


persons


The


left


-to-right


dimension


indicates


the


temporal


order,


and


the


Xs and


Os vertical


one


another


are


simultaneous.


* a


symbol


indicating


random


assignments


to separate


treatment


groups,


is necessary


Following


notation


model,


design


thi


research


appears


follows:


012


015


016


021


022


023


025


031

041


032

042


033

043


035

045


046


051


052


053


055


056


5


058


061


062


067


Comparisons


interac


tional


data


in regard


the


hypotheses


were


mad


all


pairwi


relationships


assum-


significant


F ratios


Change


was


expected


com-


parsons


data


from


and


and


and


Change


was


expected


appear


between


these


comparisons


I I









change


parameters.


more


than


ten


school


days


could


elapse


between


and


between


and


between


and


However,


at least


three


months


should


elapse


between


and


between


and


and,


six


and


one


-half


months


should


elapse


between


and


Thi


design


Stanley


similar


time-


that


series


referred


design.


6 In


to by


their


Campbell


escrip-


tion


it as a quasi-experimental


sign,


subject,


potentially,


the eight


sources


internal


eri-


mental


invalidity


The


scussion


validity


or invalidity


Campbell


and


Stani


refers


to experimental


esign


not


instrument.


this


sense


, they


distinguish


tween


two


types


validity,


inte


rnal


and


external.


design


that


perfectly


valid,


internally,


possesses


factors


. which


themsel


ves


could


produce


changes


whi


ch might


mistaken


the results


X. ."


Few


designs


achi


eve


perf


ect


internal


validity;


rather


different


signs


are


internally


valid


to differ


degr


ees.


A de-


sign


that


is perfectly


valid,


externall


contains


factor


which


. represent


a potential


cific


effe


of X


some


undepsirably


limited


con-


editions.


Thus,


extent


that


design


is externally


valid,


the results


may


be generaliz


larger,


un-









perfectly


valid,


externally.


However,


some


designs


may


shown


to be


more


valid,


externally,


than


others


The


first


the


two


sources


internal


invalidity


scuss


ed by


Campbell


Stanley


relation


"time


series


" design


story


Theoretically,


length


time


history


between

or the


observations


possibility


becomes

some e


greater,


xternal


effect


event


occurring


which


will


greater


influence


the


expe


present


rimental


study,


eight


outcome


calendar


also


becomes


months


would


elapse


in whi


subj


as cla


ssmates


same


school


would


ave


a great


deal


opportunity


interaction


outside


limits


of the


advisory group.


This


ect


would


be atten-


uated


process,


somewhat


nonbiased


as described


nature


section


selection


on sampling


procedures


The


second


these


variabi


, instrumentation,


did


not


appear


as serious


present


study


as other


In studi


designed


"time


series


" the


principal


source


inva


lidity


due


instrumentation


stems


from


the


varia-


ability


means


during


devi


ce.


When


using


obs


ervation


hniques


aware


, the


thi


observe


, the


that


author


source


minimized


variability


this


Be-


effect


developing


observer


skill


as d


escr


ibed


above.


Campbell


and


Stanl


identify


four


factors


which


fluence


he degree


external


validity


in experimental









design.


these


four potential


sources,


three


are men-


t ioned


as possibly


operative


"time- series"


design.


When


ment,


the pretest


the effect of


sensitizes


subjects


treatment is modified.


treat-


Campbell


and


Stanley


call


this


interaction of


testing


and


treatment.


this


study,


the pretest consisted


three episodes of


nonparticipatory


observation.


treatment


consisted


the activities of


the advisory


group


and


opportunities


interaction


found


there.


For


testing


treatment


interaction to b

nonparticipatory


fluence


le operative,


observation must


activities


as a source of


interfere


or opportunities


invalidity,


with or


interaction.


Since,


subjects


this


study,


students


Univer-


sity of Florida


laboratory


school,


were


to be


frequently


observed


class


and


would


not be


aware


that


author was


observing


different reasons


than


ordinary


observers,


there


was


little


reason


to suspect


that


treatment and


test-


ing would


interact


this


study.


Studies


conducted,


in the


past,


with


population of


students of


the P.K.


Yonge


Laboratory


School


have


been


limited


nonrepresentative


quality


that


population.


the past


few years,


efforts


have been made


to make


student body more representative of


populations


least


large as


state


school-age


population.


Statistics on









difference


population


exists


between


school


-age


those


students


students


the


and


the


basis


statewide


I.Q.


and


racial


composition.


S1S


socio-economic


status


some


difference


exists.


In relation


the


statewide


popu-


lation,


the


P.K.


Yonge


School


population


is overrepresented


higher


ranks


and


lower


ranks


and


underrepresented


the


middle


and


lower-middle


ranks.


Thi


mentioned


regard


to Campbell


and


Stanley


second


source


external


invalidity


the


interaction


of selection


and


treatment.


It has


been observed,


in some


studi


, that


factors


ec-


tion


lead


to differential


capac


ity


to respond


or deal


with


treatment.


Since


subj


ects


this


study


were


not


selected


on any


a prior


notions


ability


or attitude


, thi


interactive


effect


was


to be


eliminated.


Campbell


and


Stanley


propose


one


more


variabi


whi


ch is


sometimes


influential


"time


-seri


es" design


Reac-


tive


arrangements


are


those


conditions


experimental


environment


which


influence


outcomes


cause


an in-


accurate


interpretation


treatment


ect.


In a study


which


data


are


being


collected


observing


subject


behavior,


example,


observer


may


sturb


true


effect


treatment


behavior


so distractive


that


alters


Despite al


efforts,


Activity

testing


inte


reaction


conditions


will


the

have


group

some









the


observer


attends more group meetings


than


is neces-


sary to gather data


for the


purpose of desensitizing


subjects


to his


presence,


data recording


accomplished


unobtrusively,


testing phases of


sites normally used during


experiment occur


treatment


phases,


the observer,


otherwise,


makes


his presence unobtrusive.


ability


to generalize


results of


this


experl-


ment


were


to be


limited by


one more


factor--variance of


treatment.


The


process


within


each advisory


group was


determined


advisor.


Since


this was


first


year


this


plan and


first


effort


this


area


for most advi-


sors,


a great deal


of between-group difference


was


expected


treatment.


Each advisor


sample


was


asked


pre-


pare


a 100-word


statement describing


activities


in his


group


that


were


intended


to produce movement


toward


pur-


poses


, and


the advisory


groups.


The


administration of


school


foreseen


that


this


might


be problematic and


scheduled


a series


of workshops


the advisor


These were


voluntary,


one meeting per week,


under


the direction


the guidance counselor.


It was


not


intended


that


these workshops


standardize group


process and


activity


The


administration


expected


that


advisors


would gain


insight


into the relationships


between group


havior


and


purposes


set


forth


groups.









Instrumentation


A number


category


systems


have


been developed


systematic


describe these


observation of


systems as


behavior.


. .'shorthand


Simon and Boyer


' methods


collecting


observable objective data about


the way people


talk and act.


They make possible a relatively


simple


record


of what


is happening


but


they


do not record


what


being


said.


These


systems


are made up of


sets


of categories


of behaviors.


they observe,


These


number


systems vary


individuals


type of


that may


behavior


be observed,


the materials


required


to record


observations,


and


several


other ways.


The


intent


system developed


Robert F


to observe behavior


small


. Bales


groups.


distinct


Most cate-


gory


systems are


vironment.


Bales


intended


defines


for use


small


teacher-pupil


group as


en-


"any number


persons


engaged


interaction


with


each other


in a


single


face-to-face meeting


or a series


such meetings,


which


each member


receives


some


impression or perception


each


member


distinct


enough


that he can,


either


time or


in later


questioning,


give


some


reaction


to each


others


an individual


person,


even


though


be only to


recall


that


the other was


present.


Operationally,


Bales









says


groups of


two and


twenty


are possible,


and


even


larger


groups are not necessarily


ruled out.


The


twelve categories


in Bales


s Interaction Process


Analysis


are


"meant


to be completely


inclusive


sense


that


every


act


which can be observed


can be classified


one


positively


defined


category


The categories


Bales


instrument are:


Shows


Solidarity


Shows Tension Release


Positive Reactions


Agrees


Gives


Gives


Suggestion
Opinion


Attempted Answers


Gives Orientation


Asks
Asks
Asks


Orientation


for Opinion
for Suggestion


Questions


Disagrees


Shows
Shows


Tension


Negative


Reactions


Antagonism


Development of


present


set of


twelve categories


was


begun


1946.


Bales


sought


to formulate a


general


purpose categories


observation and


analysis of


small


group


for the


final


goal


"developing


a more


adequate


body


theory


relevant


analysis of


full


scale


social


systems.


The


process


selecting


cate-


gories was


as much


empirical


theoretical.


Initial


obser-


nations were made with no specific


category


system


in hand.


The observers merely attempted


to categorize


interaction as








which were obvious


repeaters


formed


the


first category


sys-


tem observers


took


into


field.


From repeated


field


testing


subsequent revisions,


twelve major revisions,


the


"final"


set of


twelve was produced.


For


this


instrument,


or any


instrument,


to provide data


concerning


change


in rate and


pattern of


interaction


that


may be accurately


interpreted,


certain


criteria must be


met.


David Fox


describes


five criteria


that are necessary.


Objectivity


is described by


Fox as


"the extent


to which


the data


obtained


are


a function


of what


being mea-


sured.


When


systematic observation


instruments


are


being


used,


objectivity


consistency.


The


is a question

process of de


inter-


veloping


and


observe


intra-observer

er consistency


described


by Bales


Interaction Process


Analysis and


essentially practice


using


instrument.


In an article,


Borgatta


and Bales


achieving


present data


consistent


observer


supporting


behavior.


possibility


The data are


presented


as Pearsonian


correlations


category


and


range


from


category


category


The overall


correlation


was


.92.


Objectivity


this


case,


is a func-


tion of practice and


understanding


the explanations of


David J


. Fox,


The


Research


Process


in Education


(New


York


Holt,


Rinehart and Winston,


Inc.


, 1969),


Ibid.,


p. 380.









behavior appropriate


each category.


The


author,


sole


observer,


tion of


expects


subject


to develop these


code numbers


skills


categorical


through memoriza-


definitions


and in group


observation


practice for no


less


than


50 hours.


Intra-observer


reliability will


be calculated by


taping


the experimental


tion on a randomly


group


sessions,


selected


performing


sample


the categoriza-


percent)


two


separate occasions,


and


computing


correlation coefficients.


iThe


second


criterion by which


one may


evaluate


the data-


gathering


plan


interpretability


is appropriateness.


Fox defines


this


factor


"the extent


to which


respon-


dent


group can meet


the demands


imposed


instrument.


Most category


systems


systemic observation


inter-


action make


no demands


upon


the observed


except


that


they


stay within

capable of


sight


recording


the observer.

nonverbal as


These


well


systems


as verbal


are

behavior.


Since


subjects


this


study were


required


meet


as a group


for the


purpose of


providing


a source of


data


this


study,


purposes


the advisory


group and


school,


instrument


should be appropriate


this


instance.


Sensitivity,


third


criterion,


first


that


measurable.


Fox


defines


sensitivity


the ability of


instrument


to make


the discrimination


required









research


ment


ability

produced


has


problem.

sensitive

is wheth

by the i


occurred


. If


Acceptable


e


enough

r or no


instrument


no change


to meet

t the r


evidence


the s

results


indicate


that


standard


analy


significant


that


occurs,


strument


an instru-


interpre-


of data

t change


may


still


sensitive


but


will


have


to be


tested


a situation


which


change


known


occur


or when


gros


differences


are


expected.


When


general


zing


from


other


studi


to evaluate


the


sensitivity


an instrument,


crucial


to be


aware


that


an instrument


sensitive


one


program


or population


may


not


sensitive


to a second


program


or population.


Complete


evaluation


sensitivity


system


use


this


study


will


come


after


data


are


analyze


Significant


different


ces


would


indicate


that


was


sufficie


ntly


sensitive.


no differ


ences


are


measured,


we may


However,


the


blame


we may


Bal


instrument


establish,


system


or the


partially


examining


experimental


, sensitivity


studi


treatment


as present


similar


present


one


which


h h


ave


found


significant


difference


es.


One


study


which


found


signifi


cant


change


as measured


Bales


instrument


was


that


John


Kirby


In hi


study


public


school


princ


ipal


, Kirby


proved,


among


other


things,


that


variance


number


acts


individual


that









observers


could


categorize


Bales


instrument


decreased


time


spent


treatment


program


increased


possible


use


this


study


as evidence


sensitivity


Bales


instrument


present


because


certain


parallels


may


be drawn


between


present


study


Kirby


study.


One,


subjects


in both


cases


were


required


meet


reasons


other


than


to supply


a source


data


the studies i

those usually


n question.


reported


Two,

the


both


groups were larger


literature.


And,


than


three,


Kirby


measured


change


across


a longer


(nine


months)


period


time


than


other


studies


using


Bales


instrument,


did


present


study.


-The


fourth


attribute


an instrument


must


have


to be


accurately


interpretable


reliability.


says,


reliability


we mean


accuracy


data


sense


their


stability,


repeatability,


or preci


sion.


Reli-


ability,


when


using


a sy


stematic


observation


instrument


con-


cerns


degree


to whi


behavior


subject


consistent


or not


subject


gross


periodic


changes


degree


to which


observer


can


agree


with


himself


and


second


observer


when


category


zing


behavior.


Bales


chose


deal


directly


with


this


problem


in an article


co-authored


with


Edgar


Borgatta.


authors


categorized


subject









interaction


across


time


and


found


that


subjects


do behave


in a manner


that


is categorically


consistent.


Since


only


operational


way


to category


behavior


this


time


through


observer


judgment,


reliability


observation


tech-


unique


finally


observer


reliability.


The


sec


ond


purpose


the


artic


question.


report


Those


ed above


results


are


was


to present


reported


above


evidence


under


cussion


objectivity


instrument.


The


procedures


that


were


used


this


study


to maximi


observer


reliability


are


reported


above.


The


final


criterion


interpretability


is validity


Validity


is defined


"the


extent


to which


pro-


cedure


actually


accomplishes


what


see


to accompli


measures


what


see


measure


The


extent


which


instrument


accomplishes


what


see


accomplish


may


evaluated at


seven


eve


confidence.


eve


of predictive


validity


the


research


the


greatest


confi-


dence


validity


a particular


instrument.


This


type


validity


depends


upon


successful


prediction


future


behavior


that


based


on res


ults


from


instrument


que


stion


Evidence


that


thi


eve


validity


may


achieved


when


using


Bal


ins


trument


may


drawn


from


a study


categorized


Borgatta


behavior


Bales


a number


author


groups


observed


to establi


U-









baseline rate and


pattern data


for each


participant.


comparing th

the authors


base data and


hypothesized


their


various


theory


interaction


interaction,

1 rate and


pattern characteristics


groups made when members of


original


groups were combined


form different groups.


test


their


hypotheses,


they


observed and


categorized


havior

predict


the new groups.


international


Results


behavior


proved


that


basis of


they


could


earlier


havior.


A second


study which supports


contention


that


Bales


instrument may


be validly


interpretable


that


George Talland.


Perceiving


that most


laboratory


groups


are


studied


the


process of


solving


problems,


Talland


sought


examine


interaction


in groups whose


task


was


identify problems


rather


than solve


them.


Because


they


seek


more


to identify than


to solve


problems,


advisory


groups


more closely parallel


Talland's


therapeutic groups--those


groups


seeking


to identify problems--than


problem-solving


groups.


dimension of


interaction


which Talland


chose


measure was


characteristic of


phase movement,


interaction


proposed by


in certain


Bales


types of


to be


small


groups.


Bales


theoretically


excluded


therapeutic groups


George A.


Talland,


"Task and


Interaction Process:


Some Characteristics of


Therapeutic Group


Discussion,









from


demonstrating


the


that


groups


the


in hi


therapeutic


phase


study.


groups


movement


Talland


not


complete


found


applicable


hypothesized


the


proved


orientation-


evaluation-control


phase


movements


Finally,


movement


as a


occurred


test of


"real"


whether


predicted


problem-solving


groups


phase


, as


opposed


to laboratory


groups


, Henry


Landsberger


used


Inter-


action


Process


Analys


to collect


interaction


data


twelve


labor-management


mediation


group


His


positive


result


affirmed


the


Bal


hypothesis


existence


phases


group


interaction.


Application


the


criter


accurate


interpret-


ability


defined


Fox


Bal


instrument


indicates


that


result


from


that


instrument


may


meaningfully


in-


terpreted.


Sampling


The


advi


sory


groups


were


formed


when


core


teachers


for


each


grade


ass


signed


students


that


grade


an ad


visory


group.


assigning


first


student,


alphabeti-


call,


advisory


group


one,


sec


ond


student


group


two,


and


so on,


teachers


fill


six


groups


per


grade


with

were


fifteen

correct


students

d to make


each.

each


Gross

group


imbalances


similar


in sex


the


or race


larger









population


of P.K.


Yonge


students.


From


a total


twenty-


four


groups,


sample


was


selected


randomly


drawing


two


groups


from


the


twelfth


grade


groups,


one


group


from


eleventh


grade


groups,


one


group


from


the


tenth


grade


groups,


and


groups


from


ninth


grade


groups.


Two


groups


were


drawn


from


twelfth


and


ninth


grade


groups


include


larger


proportion


subjects


from


population.


groups


and


advisors


produced


subj


ects


or one-fourth


population.


Randomness


was


limited


groups


grade


level


and


students


per


grade


level.


Data


Collection


P.K.


Yonge


Laboratory


School


began


a modular


schedule


fall


1971.


Each


module


30 minutes


long, and


the

one


school

cycle.


day


has


nineteen


That


modules

seventh


Six

school


school

day,


days equal

a student


begins


to repeat


the


schedule


previous


school


days.


The


school


year


broken


into


five


parts.


Each


part


lasts


five


cycles


school


Between


idea,


plan,


implementation


compromises


frequently


occur.


A number


factors


forced


such


an occurrence


upon


advisory


group


program


Adminis-


trative


detail,


need


class


mee


tings


school


assem-


blies,


and


testing


were


few


impose


itions


upon


timr


rie=di nn i-0d^


fnr


^ rl cnrv


rrrnin


rtiuni i- ii


Ac nli nn -


PSi









during


the


first


and


third


parts


the


school


year


Eleventh


and


twelfth


grade


groups


met


from


A.M.


to 1


P.M.


and


ninth


and


tenth


grade


groups


met


from


1:30


to 3


P.M.


During


the


second


and


fourth


parts,


advisory


groups


met


one-


half


hour,


twice


per


cycle,


as planned.


Eleventh


and


twelfth


grade


groups


met


from


A.M.


and


ninth


and


tenth


grade


groups


met


from


1:30


P.M


. During


fifth


part


the


school


year,


final


school


days


the


groups


were


to have


met


they


did


during


the


first


and


third


parts


year.


However,


sev


eral


cons


iderations


required


that


they


meet


one


-half


hour,


twice


per


cycle


they


did


during


Twelfth


cycle.


the


days

sixth


the second


grade


Eleventh


cycle


the


groups


grade


Tenth


cycle


days


coincide


and


met


groups


grade

Ninth


cycl


with


calendar


fourth


the

met


group

grade

Since


wee


first

the t

met t


groups


parts


and


hir

he

met


school


the


d


fift

and


second

the f

cycle


calendar


h


school


year.


days


fifth

and

ourth

does


day


days

sixth

and

not


each


grade


had


scheduled


advisory


group


meetings


changed


from


week


to week.


Thus,


one


group


did


not


always


meet


on Monday,


while


another


met


alwa


on Friday


see


Figure


The


author


collected


interactional


data


a nonpar-


ticipant


observer


in each


groups


. Ob


ser


vations


took


place


room


assigned


to each


group


school


admin-

































































































0 00
W O M


0

-H
~I
ra'






r-i
r-I






r^t
a'



rl
H-




('I
H-


T I I


WW










period


advisory


group


s one


and


one-half


hour


session,


the


first


and


second


observational


periods.


Since


the


advi


sory


groups


met


only


one


-half


hour


per


session


dur-


third


observational


per


iod,


data


were


collec


ted


dur-


middle


twenty-minute


period


the


one-half


hour


sess


ion.


the


groups


met


as per


the


projected


plan,


minutes


observation


data


per


group


per


observation


period


could


have


been


coll


ecte


d within


school


days.


men-


tioned


above


certain


factors


intervened


regular


focus


advi


sory


groups


on activities


rec


toward


achievement


their


purposes.


The


intervening


activities,


while


not


necessarily


unproductive


with


respect


advisory


group


purposes


, and


were


such


a nature


prec


lude


coll


section


of observation


data.


For


example


during


ass


meetings,


one


intervening


activity


interaction, might


community


occur


feeling, but


that

the


would co

presence


tribute


students


the

who


growth

were


not


experimental


subjects


made


that


instance


inadequate


the


observational


purposes


this


study.


Thus


, rather


than


the


ten


data


collecting,


per


observational


period,


the


first


observational


period


lasted


twenty-


SIX


school


days


The


sec


ond


period


lasted


twenty


-one


school


days


and


third


1 c4- ca


T 4 rta 4n 0 a -


4-I ,OO


rn'/ ?-t-S/^


F3 -y71


wijiv +-r-


-T T.TaV 4- C r


-- ,-


T-


C-


I


O









between


and between


and


Likewise,


only


five months


passed


between 0


and


instead


six


and


one-half months projected


under


"Design"


this


study.


In addition,


sample of


during


twelfth grade


second


advisory


observational


groups met


period


to accomplish


advisory group


purposes


infrequently that


it was decided


cease


observing


and


use what data had been


collected


the time complete data


were collected


other


grades'


advisory groups.


Thus,


for the


second


observation


period,


the two twelfth


grade groups were


represented


the data,


evaluation


change was


based


on data


from observations


and


and 0O


and


A representation of


recording


sheet


used


shown


in Figure


The


horizontal


lines


represent


time and


vertical


lines


represent


category


boundaries.


Group


inter-


action


was


scored by


assigning


each group member


and advisor


a code number


and recording


code numbers


beneath


proper


category


heading.


The


"unit"


to be scored


is defined


Bales


"the


smallest discriminable


segment of


verbal


nonverbal


behavior


to which


the observer,


using the


present


set of


categories s


after


appropriate


training,


can


assign a


classification


under


conditions of


continuous


serial


scoring.


He further


says


that


this


smallest discriminable


segment


. often


. a


single


simple


sentence,


" or that


com-















SHEET


GROUP


ABSENT


DATE


TIME


1 2 3 4 5 6 7 8 9 10 11 12







illll === = i === = = = ==== === -*''















Ril










This


formal


definition


was


operationalized


follow-


ways.


separate


unit


was


scored


individual


when


or she:


initiated


an interaction,


moved


from


one


category


vidual


interaction


entering


to another


sequence,


without


extended


another


their


indi-


interac-


tion


single


category


some


length


time.


For


example,


the


conversation--


"Where


you


next?"


"Math.


" "That


too


like


math.


--would


scored


four


separate


units


ass


signed


appropriate


individuals.


The


response--


"That


was


a good


guess,


a better


answer


would


have


been


'ovoviviparous


What


you


think?"


would


have


been


scored


three


categories


, VI,


and


VIII).


All


ree


units


would


have


been


attributed


same


individual


and


scored


as directed


same


second


indi-


vidual.


An example


third


type


scoring


decision


might


look


like


this


think


mini-courses


are


not


very


good


because


are


so short.


We have


little


time


get


into


subj


ect


that


just


as our


interest


built


, it


time


to get


back to


regular


program.


Also,


interests


instructor


subject


are


much


more


developed


than


ours


consequently


, does


deal


with


aspects


that


U_| S 1 1r


nnvt


f i nre c


t-rnnhl es-nmn


This


i interaction









separate


scores


within a


single


interaction of


some


duration


is determined by


Bales's


observation


that


achieve maximum validity,


a new


score


should be entered


every


four


seconds.


Thus,


in an


interaction of


twelve


seconds duration,


three


scores would


be entered.


Opera-


tionally,


time


To collect


is estimated.


proactive and reactive data,


was necessary


to record,


beneath a


category


heading,


the code


number


speaker


and


code number


receiver.


Bales


has


adopted


notation


to represent


the group as a


whole,


speaker


or receiver,


and


to represent some general


specific


other


outside


the group,


and


we used


the


notation.


Except


the one change,


Bales


s category


system was used


as defined


appendix


to Interaction


Process


Analysis.


Conclusion


Eleven


hypotheses were proposed


to measure


the direction


growth


toward


selected


purposes


the advisory


groups


the P.K.


Yonge Laboratory School.


A random sample of


these


groups


were


sele


cted


to provide data


hypothesis


evaluation.


For


this


study,


unit of


datum was


that


defined by


Robert F.


Bales


Interaction Process Analysis.


The data


were


gathered by


systematically


observing


sample over an


eight-month


period


and categorizing the units
















CHAPTER


DATA


ANALYSIS


Chapter


IV of


this


study


concerned


with


statistical


treatment

hypotheses


presentation


proposed


of data


Chapter


III.


gathered

The re


to evaluate


suiting


values


were


used


to detect


growth


toward


away


from


selected


advisory


group


purposes.


data


ana


lyzed


this


study


consisted


units


interaction


as defined


Bales


Interaction


Process


Analysis.


All


hypotheses,


except


Hypothesis


were


evaluated


using


anal


is of


variance


method


Hypothesis


concerning


variance

statistic


change


among


reported


individuals,


in Walker


and


was


Lev


evaluated


This


using


statistic


used


special


case


when


one


wishes


compare


variance


beginning


experiment


with


vari-


ance,


the


same


group,


later


experiment.


The


model


selected


use


this


experiment


termed


randomized


block


design.


The


randomized


block


design


WaIker


Helen


nn .Tnc nh


c:f-1


st-i al


Tnfsrpnncp


f


.


T.a\,









was


selected because


it allows


statistical


control


dif-


ferences between


groups.


Four


general


conditions must be


met


when


using


analysis of


variance.


In addition,


three


criteria


specific to the randomized block design must be


met.


The general


analysis of


conditions


variance are:


applicable


all models of


observations are drawn


from


normally


distributed


populations,


observations


represent


random samples


from populations,


variances of


populations


are equal,


and


numerator


and denominator


of F ratio are


independent.


Criteria


one


specific


treatment with k


randomized


= two or more


block design are:


treatment level


assignment


subjects


to blocks


that


the variability


among


subjects within any


block


less


than


the variability


among


the


blocks.


The


number


subjects and observations


within


eac


h block must be equal,


and


random assignment of


treatment


levels


the experimental


units


within


each


block.


An exception


is made


this


randomization proce-


dure when


a block consists


one


subject


who


receives


treatment


levels and when


the nature of


treatment


pre-


eludes


the


randomization of


order.


2. Roger
i- _- i__ -


Kirk,


Experimental 1


Design:


Procedures


for the


. n S -


n ?


-^ __ i __ joIi_ _


/~ 1^ -









The general


criteria may


be considered


following


ways.


The


population s


from which observations were drawn


were


total


units of


interaction made by


the members and


advisors of


advisory


groups


P.K.


Yonge Labora-


tory


School


during


any


one observational


period.


For


lack


of evidence


contrary,


it may


assumed


that


these


units of


interaction


were


distributed normally with regard


to categories of


Bales's


instrument.


The observations


that


were made


to produce data may


be perceived as


random samples


populations


respects.


One,


advisory


groups


were


selected


on a


random basis.


And,


two,


"a priori"


schedule or


criteria


were


established


which determined


particular


regards


grou

the


session


third


to be used


general


as a data


criterion,


base.


populations are


normally


distributed,


then


population


variances will


equal.


Finally,


when observations are


randomly


sampled


from


normal


populations,


numerator


and denominator


F ratio are


independent.


The


randomized


block design


is a common model


used


analysis of variance.


Kirk,


one,


describes


it at


length


in Experimental


Design:


Procedures


the Behavioral


Sciences.


In particular,


Kirk


says


this design


is quite


useful


when


subject


variability may partially


obscure


treatment effect.


This design controls


that


source of









The


criteria


specific


randomized


block


design


may


considered


these


ways


The


present


experiment


had


one


treatment,


advisory


group


ambience,


and


nine


treat-


ment


leve


based


the


passage


time


a factor


re-


lated


to change


groups.


Kirk


proposes


that


an accep-


table


alternative


used


to reduce


within-block


variability


below


that


among


-block


variability


expose


every


sub-


ject


to all


levels


treatment.


4 Thi
Thi


was


done


the


present


study.


Finally,


this


study


was


excepted


from


third


order


specific


criterion


treatment


since


eve


this


is not


possible


study,


that


to randomly


specifi-


cally


noted


as an exception


criterion.


The


results


derived


from


analyst


variance


no more


than


indicate


presence


absence


change


that


statistically


different


from


chance.


Thus,


designs


where


more


than


two


blocks


are


used


and/or


more


than


two


treatment


level


are


used,


a significant


value


F will


indicate


that


east


one


pair


blocks


treatment


levels


refl


ects


sig-


nificant


change


, but


will


not


indicate


which


pair


or pairs


are


significant


comparisons


present


between


treatment


study,


eve


and


there


are


pairwise


pairwise


com-


parsons


between


block


, groups


level


SThis


study


was


designed


Therefore


examine


we were


change


only


across


time,


interested


not


between


pairwise


groups.

treatment


j


=









significance


was


found


analysis


variance


for


a particular


hypothesis


, Tukey


s Honestly


Significant


Difference


procedure


was


used


to determine


whi


pair


pairs


were


significantly


differ


ent


no significance


was


obtained


a particular


hypothesis


further


analy


was


made nor was


neces


sary


Nonsignificant


F ratios


derived


from


analy


SiS


variance


indicate


that


no one


pos-


sible


pairwise


comparisons


invol


ves


significant


change.


According


to Kirk,


Tukey


s Honestly


Significant


Differ-


ence


procedure


designed


test


all


pairwise


comparisons


among


means


Signifi


chance


rmined


comparing


difference


between


means


being


consider


ed and


value


derive


per


Honestly


Significant


Diff


erence


formula


difference


excee


honestly


signifi


cant


difference,


then


change


significant


between


those


comparison


eve


Thi


cedure


simplified


building


a matri


composed


differences


among


means


as described


Kirk


necessary


prerequisite


consideration


of obser-


national


data


as an obj


active


basi


s upon


which


judge


hypothesis


det


ermination


intra-observer


reli-


ability.


As d


escr


ed in


earlier


section,


Instrumenta-


tion,


intra-observer


reliability


was


determined


repeating









the categorization


process on a randomly


selected


sample /


percent)


advisory groups.


Audio


tapes of


the


verbal


interaction


were made on


occasions.


Each of


these


sessions


was


assigned


a number


(01-47)


drawn from a


table of


random numbers.


Entering the


table at a different


point,


the


first


twelve of


these numbers


01-47)


were


selec-


ted as


the


sample


to be


used


to determine


intra-observer


reliability.


To measure


this


reliability,


the categoriza-


tion process was


repeated


these


twelve


sessions.


Data


from that repetition were


correlated


with data


same


sessions gathered during


original


categorization


interaction.


Correlations


coe


fficients


were calculated


interactions


category


interactions


per participant.


The data


are


found


in Table


Table


Correlation


Coefficient


Sample of


Original


and


Replicated


servational Data


Data for Group Members Data for Advisors


Total interaction .959 .906

Section A .731 .810

Section B .870 .633

Section C .883 .938

Section D .754 .473









correlations above


indicative of


sufficient relation-


ship


between


two


sets


of data.


The correlation of


original


Section B


In the context


duplicate data


is quite near


from advisor


interaction


Bales's minimum acceptable


of eight high correlations,


this one


level.

rela-


tively


value of


low value


should not


correlation


cause much concern.


advisory


interaction


low


in Section D


based on a


very


small


number


interactions.


It is


actually not


numbers.


valid


In general,


to compute a


it must


correlation


be said


that


based


on so few


correlations


between


the original


and duplicate data


are of


sufficient


size as


indicate


that


samples came


from


same


population.


In Hypothesis


was


proposed


that


proportion


interaction


in all


categories of


interactions made by


advisors would


the group,


decrease


including


time


the advisor,


passed.


the members of


become more


intimate


need


for the


advisor


to enter


interaction


should


decrease.


To evaluate


this


hypothesis,


that


proportion of


all


interactions


which was made


each advisor was


computed.


These


proportions were


the data


used


to determine


the


sums


squares


necessary


to calculate F


ratios


from


analysis


of variance method.


change


interaction rate was


to have


occurred,









03- 07


Between


these


three


pairs of


observational


periods


the greatest amount of


time


passed and,


theoretically,


the


greatest opportunity


the


treatment,


advisory


group


inter-


action,


to have


its effect.


A portion of


power of


the


time


series-design derives


from the repeated


inter-treatment observations.


It is


assumed


that no


significant differences


will


occur


between


pairwise observations drawn


from 01,


or 04'


and 06


or 07,


and


In Table


and


following tables,


significant


F value will


indicated


level.


Thus,


an F


value obtained


that


is equal


to or


greater


than


will


judged


significant.


The


following

represents


proportions,


calculations may


the


which served as


found


summary presentation of


data


these


the Appendix.

the analysis of


Table

vari-


ance


treatment of


advisor


rate.


Table


Change


in Rate of


Advisor


Interaction


Source of Sum
Variance of Squares df Mean Squares F

Treatment .094506 8 .011813 .760

Blocks .622010 5 .124402 7.999*

Residual .622075 40 .015552









The F ratio


for that


portion of


variance attributable


to change between


treatment levels


is not


significant.


Non-


significance


indicates


that no one of


pairwise com-


parsons


possible


between


treatment


levels


involves


a sig-


nificant


change.


There


is no


need


to look at


separate


pairwise comparisons


Tukey's method.


The F rat

between blocks


io derived


for variance due


significant.


Significan


to differences

ce between groups


indicates


that


real


differences


exist


between advisors with


respect


designed


their rate of


to examine


interaction.


differences


present


between


study was


groups or


advisors,


only


between


treatment


levels


(i.e.,


across


time).


The


fact


that


differences


between


groups


advisors may


influence


ment


treatment


block design.


effect


reason


Effects due


to blocks


the develop-


(i.e.,


groups or


advisors)


are


partitioned out


of effects due to


columns


(i.e.,


treatment)--the


result


which


is a more


accurate assessment of


effect due


treatment.


To facilitate


advisor rate of


reader


interaction and


s perceptions


change


of change


succeeding


hypotheses


display


the data are


the data may


presented

contribute,


graphically.


times,


Graphic


information


not contained


in summary


statistical


tables.









The graph of


the mean


proportions of


advisor


interaction


appears


in Figure


.32 -L
T


Observational


Periods


Figure


Mean


values


that


portion of


all


interaction


which was made


advisors.


A line graph of


the mean


proportions of


advisor


inter-


actions


per


observational


period does


not


offer


clear-


cut pattern.


possible,


statistically,


examine


this


line


trend, but only where


the F value


treatment


variance


significant.


Statistical


and


graphic data


presented


Hypothesis


do not offer


any


evidence


that change


has occurred.


measured


the Bales


Interaction Process Analysis model,









Hypothesis


change occurs


interaction.


This


was proposed


rate at


to determine


which


hypothesis


whether


the group members


the complement


or not


enter


of Hypothesis


advisor


rate goes down,


group-member


rate must


go up.


In addition,


advisory


groups


are


approaching


their


avowed purposes of


providing


a home


base,


a place where dia-


logue occurs


and


a sense of


community,


then member


inter-


action


should


increase.


And


these goals


are


neared,


interactional


increases will


follow.


For


computational


purposes,


all member


interaction


per


group was


tion made


observation


lumped


together.


all members


period


proportion of


a group


equaled X.


present


sums


interac-


one


in Hypothesis


2 were


computed


from


proportions of


interaction


attri-


butable


group members


in each


group


each


obser-


vation


tion


period.

all ir


The values


iteraction


X which represent


initiated


members of


propor-

a group


may


found


the Appendix.


The


summary table


analysis of


variance


Hypothesis


found


in Table


Based


on analysis of


variance any


change


that may


have


occurred


between


treatment


levels


was


not


significant.


in Hypothesis


differences


between blocks or


groups were


significant


but not of


hypo-


thetical concern


this


study.


Graphic


representation









Table


Change


in Rate


of Group


Member


Interaction


Source of Sum
Variance of Squares df Mean Squares F


Treatment


Blocks


.096965


.012120


.591807


.760


.118361


7.421*


Residual


.637987


.015949


Total


1.326759


.05;8,40
p < .05.


.69

.67

S.65

(63


.18.


Observational


Periods


Figure


Mean


values


that


portion


interaction


which


was


made


group


member


-- I __









As expected,

graph of mean prc


this graph is


portions


a reflection


from Hypothesis


Again,


line

little


can be


said


with regard


to general


trends


the data.


Trend analysis,


a posteriori,


is not appropriate


because


treatment F


is not


significant.


Analytical results


Hypothesis


2 must be construed


indicate


that


no change


taken


place


rate at


which group members


participate


in interaction.


The


third hypothesis


was


proposed


to examine


possi-


ability that


change might


occur


the extent


which one


group member


small


group of members dominated


con-


versation.

entered in


situation


Lteraction


out


existed


proportion


in which


their


few


number,


move-


ment


toward


advisory


group goals would


impeded.


Social


give


take among


all members


the group


is necessary


the group membership and


total


school


population


are


to find


achievement

action could


the groups


purpose.


occur


source of


A more


regardless of


security


crucial


even distribution


change


inter-


the group member


proportion of


total


interaction.


The


raw data


used


to compute


t for


Hypothesis


were


obtained by


group member


computing


interaction


as a proportion


initiated by


total


units of


each


interac-


tion


initiated


all members of


a group.


proportions









formula


cited


earlier.


The


results


this


computation


are


found


Table


Negative


variability


values

less.


for

Whe


t reflect

n this si


change


tuation


from g

obtain


greater


we expect


to find


more


group


members


parti


cipating


interaction.


Or,

the


every


total


group


amount


member


interact


participating,

on to be spread


we would


more


expect


evenly


among


the


members.


this


were


two-sided


test,


signifi-


cant


vals.


change


Since


would


change


indicated


in a particular


seven


comparison


direction


was


inter-


antic


pated,


a one-sided


test


was


considered


appropriate.


this


basis,


only


comparison


which


significant


change


occurred


was


at 04


Group


Change


was


expe


cted


-04


-07,


and


-07.


The


decrease


variability


amount


interaction


which


occurred


this


point


obviously


does


not


warrant


conclusion


that


a general


decrease


variability


occurred.


There


was


no change


number


units


interaction


stributed


among


the


group


members


over


course


this


exp


eriment.


Growth


a result


statements.


toward


and


The


purpose


cause


climate


groups


of a greater


a group


should


number


in which


part,


of positive


members


find


support


and


opportunity


discussion


must


one


character i ze


- -


warm,


supportive


comments.


The


purpose


w m


r













>1

pa

a,
43
*r
4-)


C
(0


H

C
O
-1
4-)
u
C(
U
14-

C



0

Q)

0

IM
0
5.4


C1

43




r4I



O1

'C
.94
k








5.4
43


4(0






a)
rd



'-






>0

) 14
44W
+> 0


N r
X


* *

.3 <
U
J oU

r-
J 0
SU









Proportions were calculated


each advisor which


represented


that part of


the advisor


total


interaction


categorized


in Section A.


These


proportions were used as values of X


when cal-


culating


the


sums of


squares


needed


to compute


F ratios


from analysis of


variance.


The


summary


table


analysis of variance of


Hypothesis


found


Table


Table


Change


in Rate of


Advisor


Interaction


in Section A


Source of Sum of
Variance Squares df Mean Squares F


Treatment

Blocks


.013008

.009020


.001626

.001804


1.053

1.168


Residual


.061790


.001544


Total


.083818


.05;8,40
< .05.


.18.


Neither


treatment F nor the


blocks


significant


this


hypothesis.


36 possible comparisons,


none,


par-


ticularly 03-04,


06-07'


or 03


significant.


-07,


Figure


5 displays


the mean proportions of


advisor


inter-


action categorized


in Section A.









on analysis


variance


graphic


display


of observational


data


no change


taken


place


rate


at which


advisors


made


Positive


Statements,


Section


advisory


groups.


Observational


Periods


Figure


Mean
made


values


advisors


that
cate


portion
gorized


interaction


Section


A second


reflection


growth


toward


advisory


group


purpose


tion


found


of rate


advisor


of Negative


interactions


Statements,


Section


the


diminu-


initiated.


the


forces


action


at work

Positive


producing

Statement


growth


prompt


section,


they


more

also


inter-


tend


rn E CY : n


* t I *
in~~~~* 4i- rh njt 4-k -w n4? t^ a" Wnl4 n W


* U S J
fi 4^^-* ^ ^*-;^^


*---


*t __n C I


h l /-T-a


^ v /-


I


f^ ^









requirement


to respond


in a negative manner.


Hypothesis


was


proposed


to detect change


the rate of


initiation in


Section D


by advisors.


Computation


similar to


portions were established


action in Section D as


that in Hypothesis 4.

each advisor's rate of


a part of his total


Pro-

inter-


initiated inter-


action


These


determining


proportions were


sums of


used as values


squares


X when


the analysis of variance.


Table


is a


summary


table of the results


analysis of


variance


Hypothe


Table


Change


in Rate of Advisor


Interaction


in Section D


Source of
Variance


Sum of
Squares


Mean Squares


Treatment


.000531


.00006638


.985


Blocks


.000036


.000007


.104


Residual


.002681


.00006703


Total


.003248


.05;8,40


.18.


Nonsignificant F


ratios


indicate that


there


was


statistical


difference


between


treatment


levels or


between


groups.


Further,


this means


that no pairwise


comparisons







.013

.011

.009

.007

.005

.003

.001
.000


Observational


Figure


Mean
made


values


that


Periods
portion


advisors categorized


interaction


in Section D.


There


is an apparent


trend


toward


greater


use of


verbal


behavior


advisors


that could


categorized


as Negative


Statements,


Section D.


This was


not


anticipated


trend.


During


observational


periods


and


higher percentage


an advisor'


s total


interactions


were


type


that


re-


flected


tension,


disagreement,


or defensiveness.


reader,


however,


must


bear


in mind


that


increase


from a mean


pro-


portion


.001216


at 07


to a mean


proportion


.011316


involves an


increase of


actual Negative


Statements of


one


to five.


This may


represent a


trend, but


the data


not


conclusively


support


that assumption.


Indeed,


analysis


of variance


indicated


that


this


increase


is not significant.


On available evidence,


it must


be concluded


that no change








For


same


reasons


that


advisors


were


expected


increase


their


proportion


interaction


Section


Posi


tive


Statements,


was


also


expected


that


group


members


would


increase


their


participation


Section


members


group,


including


advisor,


became


more


familiar


with


one


another


and


anxiety


expected


innovative


situations


was


replaced


with


security,


pro-


portion


Positive


Statements


initiated


group


members


should


increase.


Thi


assumption


reflected


Hypothesis


The


proportions


used


as data


thi


hypoth


esis


were


calculated


acitons


from


initiated


ratio


Section


number


A to


group


group


member


member


inter-


s'total


number


interactions


initiated.


These


proportions


were


use


as values


X when


com-


putting


sums


squares


analy


variance


Hypothesis


The


analy


ss1


variance


summary


table


Hypothesi


appears


in Tabl


Table


Change


in Rate


of Group


Member


Interaction


Sec


tion


Source of Sum of
Variance Squares df Mean Squares F


Treatment


Blocks


.007633


.000954

.002702


.013512


1.040

2.947*









The hypothetical


group members would


expectation at


increase


this point


proportion of


was


their


that


total


interaction


that was categorized


in Section A,


Positive


Statements.


This


treatment F


increase was


indicates,


expected


no significant


across t

t change


ime.


occurred


the rate of Positive Statements.


The


significant F


ratio


block


effect


implies


that a


real difference existed be-


tween


the groups.


The mean


values per


group


proportions


total


interaction attributable


to Section A


were


signifi-


cantly different.


The


randomized


block design controlled


this


source of


variation


to minimize


influence


upon


analysis of


treatment


effect.


Graphically,


data appear


in Figure


.100

.090

.080

.070

.060

.050


*P
*


Observational Periods


Figure


Mean
made


values


for that


by group members


portion


all


categorized


interaction


in Section A.









discernib

Positive


reflects


Statements


a net


group


decrease


members


proportion


important


of

to


remember


that


graphed


data


presentation


does


not


con-


trol


the


for


effect


statistical


apparent

The


trend

data,


between-group


presentation


may


data.


attributable


especially


differences


A portion


that


statistical,


as does


the


effect.


present


no founda-


tion


from


which


may


be assumed


that


change


has occurred.


The


rate


at which


group


members


interact


Section


Positive


Statements,


remains


unchanged


across


time.


Hypothesis


was


propo


to determine


change


occurs


the


rate


interaction


initiated


Section


Negative


Statements


, by


group


members.


This


hypothesis


complement


Hypothesis


the


group


matures


and


posi


tive


statements


are


more


in evidence


, proportionally,


the


rate


at which


interaction


initiated


Section


should


decrease.


For


Hypothe s


the


proportions


necessary


to calculate


were


determined


comparing


number


units


Negative


Statements


made


group


members


the


total


number


interactions


initiated


group


member


These


propor-


tions


were


used


as X


in the ANOVA


computations.


The


res


obtained


to evaluate


Hypothesis


are


found


Table


The


effect


time


spent


advisory


group


environ-









occurred on F test by analysis of


variance.


Differences


between


group mean rates of


interaction


in Section D were


significant.


Some groups


interacted at


significantly


higher


rates of Negative Statements


than did others.


Again,


this


effect is controlled by


statistical


design.


Figure 8


con-


tains


line graph representation


the data.


Table


Change


in Rate of Group Member


Interaction


Section D


Source of Sum of
Variance Squares df Mean Squares F

Treatment .027253 8 .003406 1.889

Blocks .085156 5 .017031 9.446*

Residual .072139 40 .001803

Total .184548 53


.05;8,40


.18.


The


trend


in rate of


Negative Statements


apparently


reflects a net


increase.


A net decrease was expected.


But


should be noted


that a


part of


apparent


increase


rate


is due


to significant differences


between


the groups.


That


not all


increase


in rate


attributable


time


the advisory


groups.


Statistically,


there


no evidence


that change


rate


nF KTan4- 4 tin 04- -, 4- /nnrnn- e, I : 7 ,. i i n n r


?n^*4"^ 1 1 nlnl y **^ /++ ^ i"


V^n rr f^ ^^f^ ^ ^^li /t


LIJT









occurred.


This


is especially


true


when


the


graphs


the


hypotheses


are


viewed


together.


.130

.120


.110

.100

.090

.080

.070

.060

.050


\/


,--s...;


servational


Periods


Figure


Mean
made


values


group


that


members


portion of
categoriz


all


eraction


Section


pre


ceding


hypothe


ses


are


concerned


with


change


amount


interaction


The following


four


hypotheses


are


concerned


with


change


sequence


categories


interaction.


Hypothesis


prop


oses


that


groups


move


toward


reali


change.


nation


purpose,


Specifically,


when


sequence


Section


of categories


Positive


will


Statements,


WIIIIIIII








To compute


proportions


necessary


to calculate


values


vidual


F in Hypothesis


s verbal


act,


number


categorized


time s


in Section A,


was


an indi-

followed


act,


determined.


uninterrupted,


This


categorized


number was divided


in Section B,


total


was


number


times an individual


changed


from Section A


to any other


section of


categories without


interruption.


This


value was


computed


were


per group per


then utilized as


observations


the


period.


formulae


These values


to determine F.


The results of


those calculations


are


found


in Table


Table


Change


in Rate of


Section B


Following


Section A


Interaction Made


by One Group Member


Source of Sum of
Variance Squares df Mean Squares F

Treatment .146993 8 .018374 .652

Blocks .084183 5 .016836 .597

Residual 1.127292 40 .028182

Total 1.358468 53


.05;8,40
< .05.


=2.18.


Neither


treatment nor the


blocks


effects


showed


significance.


There


were no


statistical


differences


between


time


periods represented by treatment


variance and between


groups


represented by


blocks variance .


statistically









Graphical


representation


mean


proportions


from


the


data


Hypothesis


appear


Figure


.270

.250

.230

.210


.190

.170

.150

.130

.110

.090

.070


Observational


Periods


Figure


Mean


values


actions


that


followed


portion
Section


Section


interactions


A inter-


made


one


group


member.


The


expected


trend


in graphed


data


Hypothesis


was


a net


increase.


relatively


high


level


interac-


tion


at 08


resulted


from


one


group


in whi


ch all


Section


1 nf+ltorn* i n na


0 n4 a 4 'I r r\ c


FnlPwr~l1r\*Q


CQ^4- i rar


T.TQ V*


nrr









categorized


in Section B,


resulting


in an


inflated mean


proportion


observation


Considering


this when


look-


for trends


the graph,


there


apparent change


rate of


interaction.


sum of


statistical


graphic


data


indicates


that no change occurred


the rate at which


group member


interaction


in Section A


was


followed by


group


member


interaction


in Section B.


A second

sequence of


type of


interaction


change occu

ns between


when


individu


the categorical

als is altered.


Bales


terms


this


sequence


between


individuals


"reactive


interaction.


characterized by a


example,


pattern


casual


acquaintances may


interaction


quite different


from that of


close


friends.


the acquaintanceship becomes


closer,


complex.


simple


Hypotheses


pattern


through


interaction becomes


11 explore


rich and


possibility


that


some of


these alterations


in pattern


will


occur


advisory groups


will


reflect movement


toward


purposes


and


Hypothesis


9 proposes


Attempted Answers,


that


response


number


follows


times a Section


a Section D,


Negative


Statements,


interaction


will


increase.


the groups move


toward a


sense


of community,


we expect


that


the members will


become more


the extent


tolerant of, and


this


less defensive


Negative Statements


toward, one another.


should be








perceived


increasingly


as honest expressions of


attitude and


feeling


and


less


as personal,


derogatory remarks.


To compute F, proportions were determined from


number


times an individual


responded


to an


interaction


categorized


in Section D with an act categorized


in Section


B and


the


total


number


responses made to


acts categorized


in Section


The results of


the calculations are


found


Table


Table


Change


in Rate


Section B


Following Section D


Interaction


Source of Sum of
Variance Squares df Mean Squares F


Treatment

Blocks


Residual


.105017

.120786


.30465


.013127


.024157


.402


.741


.032616


Total


.530458


.05;8,40


.18.


F ratios derived


from analysis of variance of


the data


are not


significant.


There was


no statistical


difference


between


treatment


levels or


between groups.


The


number


times Section B


interactions


follow Section D


interactions,


reactively,


did not


increase


anticiapted.








.250

.230

.210

.190

.170

.150

.130

.110

.090

.070


Observational Periods


Figure


Mean


values


that


portion


Section D


inter-


action


followed by Section B


interactions.


The anticipated


trend


the data


Hypothesis


9 was


a net


increase


in proportion.


The


steady


increase


from 03


to 06

later


is negated


the decrease


experiment.


in proportion


values at


that occurred


and 07


are


in-


flated,


in part,


by proportions


from single groups


which


greatly


exceed


the mean


values


those observational


periods.


data


The Appendix


evaluation


contains


Hypothesis


those raw values.


indicate


Available


that









From


the


expectations


that


the


members


group


will


find


group,


increase


ingly,


source


security


and


reinforcement,


Hypothesi


proposes


that


Negative


State-


ments


will


follow


Questions


at a decreasing


rate.


supportive


emotional


climate


expected


that


Questions


members


and


advisors


will


contain


little


or no


threaten-


ing a

sive,


toward


aspects.


emotional


achievement


such, Questions

response. Thus,


purpose


should


not


require


groups


, Negative


are


Statements


a defen-

moving


should


result


from


Questions


lesser


degree


time


advisory


groups


passes.


The


proportions


prepared


computational


purposes


Hypothesis


were


obtained


from


number


times


Negative


Statement,


a group


member,


was


made


in response


to a Question


made


a first


group


member


relation


the


total


number


time s


any


response


was


made


to a Question.


These


proportions


were


raw


data


formulae


The


results


those


calculations


are


found


Tabl


Tabl


Change


in Rate


of Section


D Following


Section


Interaction


Source of Sum of
Variance Squares df Mean Squares F

Treatment .013289 8 .001661 1.196

Blocks .039771 5 .007954 5.722*









Inspection of


summary


table


Hypothesis


indi-


cates


that


there was


no significant


change


between


treatment


levels.


anticipated decrease


rate at which


Negative Statements


follow Questions


was


not evident


data.


There was,


however,


real


difference


between groups.


As before,


the randomized block design


partitioned


that


influence out


total


effect.


The group differential


did not mask or


inflate any


treatment


effect.


Graphically,


these data appear


.070-

.060

.050

.040

.030

.020

.010


in Figure


0


Observational


Periods


Figure


Mean values
interaction


that


followed


portion


of Section


Section D


interactions.


The


slight observed


trend


net


decrease


the










any


evidence


available


which


would


support


the


hypotheti


cal


contention


that


Section


interactions,


Negative


State-


ments,


would


follow


Section


interactions,


Questions,


a decreasing


rate.


same


respect


that


Negative


Statements


should


used


less


frequently


in response


to Questions,


they


should


used


ess


frequently


as a response


to Attempted


Answers


The


proposed


development


a group


atmos


phere


trust,


tolerance


, and


under


standing


should


result


the


percep-


tion


that


Attempted


Answers


, correct


incorrect,


are


posed


sincere


efforts


to solve


problems


offered


advisor


or group


member.


And,


such,


they


should


be empty


threat.


These


thoughts


are


refl


ected


Hypothesi


Negative


Statements


should


be used


less


frequently


response


to Attempted


Answers


advisory


group


situa-


tion


time


passes.


Proportions


used


to compute


values


were


deter-


mined


as before.


Ratios


were


established


between


the


number


times


an Attempted


Answer


from


a group


member


was


followed


a Negative


Statement


from


a second


group


member


and


total


number


times


any


response


was


made


Attempted


Answer


The


result


the


calculations


are


found


Table


ans lci .c


t- y i a n r1 o


4-n


f rnm


Wu7nn(-h


i nd i -


* iI


I l l


-- wh








Attempted


answers


did


not


decrease.


Difference


between


groups


was


significant.


This


effect


doe s


not


alter


statistical


validity


treatment


variance.


Table


Change


in Rate


of Section


D Following


Section


Interaction


Source of Sum of
Variance Squares df Mean Squares F

Treatment .216073 8 .027009 1.012

Blocks .347232 5 .069446 2.602*

Resudual 1.067574 40 .026689

Total 1.630879 53


.05;8,40


.18.


< .05


The


mean


proportions


per


observation


are


graphically


illustrated


Figure


The


obvious


trend


toward


a net


increase


propor-


tion


interaction.


The


expected


trend


was


a net


de-


crease.


Two


factors


contribute


apparent


incre


ase


proportion.


First,


proportion


interaction


during


one


tion


group


period.


was


That


quite


single


above


value


mean


value


inflated


that


mean


observa-


value.


The


same


may


be said


relatively


high


mean


proportion


dur-


Secondly,


significant


inter-group


differences


-~ a -1 *1 a n


.


~










are


used


ess


frequently


as a response


to interactions


initiated


Section


Attempted


Answers.


None


analy


ses


data


proved


to be


significant.


Thu


there


little


or no evidence


to support


thi


hypothesis.


.230

.210

.190


.170

.150


.130

.110

.090


.070

.050

.030


0--^'


Observational


Periods


Figure


Mean


values


interaction


that


followed


portion of
by Section


Sec


tion


inte


actions.


Summary


Tin1 a----


1.. -- a A- a_ --_.


_A.a--- -


J-l- -


r*L


**^^


-r --~- --~


*a -c-


lIA--A









those were comparisons of


03-04'


06-07'


and


03-07


in all


hypotheses.


There were


362 other pairwise comparisons


which change was


possible.


Analysis of


variance was made


the data


these hypotheses.


Mean


values of


each


observational


period


were


plotted


graphically,


also.


None of


possible


pairwise comparisons


treatment


effect


proved


to be


significant.


treatment,


time


spent


advisory


group,


had no


statistically


significant


effect


the hypothesized directions.


Graphs of


sistent aid


data


various


establishing


hypotheses offer


success


advisory


no con-


group


in meeting


its goals


, and


Trends


in graphed data


Hypotheses


and


are not


anticipated direc-


tions.


Trends


in graphs


Hypotheses


and


are


anticipated


direction.


However,


these


trends are


qualified


deviant group effects


which bring


their


mean-


into


serious


question.


The


lack


significant


results obtained


this


experiment


clearly


does


support


hypothesized


changes


in interaction.


Based on a hypothetical


correlation between


change


in interaction and growth


toward


acceptance of


group as


a home


base,


a place where dialogue


takes


place,


and as

sumed


the center


that


of the


the advisory


school

group p


community,


program did


it must be

not make si


as-


gnifi-














CHAPTER


SUMMARY,


LIMITATIONS,


AND


CONCLUSION


Summary


The


purpose


this


study


was


to determine


success


the advisory


group


program


in achieving


certain


their


goals


as reflected


changes


verbal


interaction.


an attempt


coincidental


to counteract


an innovation


the

the


negative

curriculu


potential


Yonge


Laboratory


School,


faculty


proposed


that


advisory


groups


establi


shed.


Every


pupil


the


high


school


was


placed


in one


of six


advisory


groups


in his


grade


level.


Instructional


and


staff


members


served


as advi


sors.


Six

advisory


purposes

groups.


were

They


include


are


to provide


original

e each


proposal


student


with


a stable


group


to which


can


relate


home


base,


to provide


student


with


a place


dialogue


about


school


-re


lated


problems,


to build


feeling


community


among


students,


to provide


each


student


with


an adult


who


response


ible


for


the


coordination


total


school


experience,


periences


to provide


increase


faculty


involvement


members


the


with

total


common

school


ex-

pro-









problems.


The


groups


also


served


admini


strative


unit


and


representative


unit


student


government.


Purposes


, and


were


those


selected


as criteria


this


study.


The


faculty


felt


that


the achievement


these


goals


would


establi


merit


program


new


and


valuable


If evidence


could


be gathered


which


dicated


that


grou p


were


providing


a home


base,


place

then


where


dialogue


advisory


could


group


take


model


place


could


and


become


sense


a valuable


community,


ad-


junct


total


curric


ulum.


To form


one


basi


measuring


whether


or not


these


purp


oses


were


achieved


any


part,


eleven


hypotheses


were


proposed


All


even


hypoth


eses


deal


with


aspects


of verbal


interaction.


The


review


literature


this


study


suggest


that


a sufficient


basi


s existed


suppose


that


verbal


fraction


refl


ected


level


tolerance


, trust,


understanding,


unity


group


measuring


changes


verbal


interaction,


insight


would


be obtained


into


changes


eve


erance,


trust,


understanding,


and


unity.


maximize


validity


in a situation


in which


no control


groups


were


available


times


-series


design,


described


Stanley


Camb


ell1


was


utili


data-gathering


model.


To meet


conditions


this


des


ign,


data


were


gathered


on nine


separate


occasions;


three


times









first


the


school


year,


three


times


middle


school


year,


and


three


times


at the


end


school


year.


The


observation


instrument


stem


used


to obtain


developed


R.F.


data


was


Bales.


category


2 Thi
Thi


stem


was


specifically


verbal


interaction


developed


in one


to characterize


twelve


types


category


Each


groups


sample


was


observed


and


verbal


interaction


recorded


twenty


minutes


nine


data-gathering


occasions


to circumstances


beyond


control


researcher,


one


group


was


observed


only


five


times


and


one


other


group


only


times.


The


raw


data,


numbers


interactions


in specific


category


were


converted


to proportions


and


tested


ten


hypotheses


particular


via


mod


analyst


analy


variance


method.


variance


The

s study


was


random


zed block


esign


Thi


model


was


used


adaptability


to repeated


measures


same


subject


and


s intended


control


effects


due


to variation


be-


tween


groups


values


of subjects


F at,or


hypothesis


Significance


beyond, the


proportions


was


level.


were


teste


recognized


the


d via


eleventh


test


variance


reported


Walker


and


Lev


Significance


was


Tfl v1 4an T i-a nn .- -%.t


Ols


"ft -1


aDrnnnn ni^-<"


rfEl









recognized


t values


level


on a one


-tailed


test.


The


randomized


block


design


model


produces


values


One


these


refl


ects


diff


erences


between


treatment


levels.


study,


treatment


eve


were


different


lengths


time


that


the advi


sory


groups


been


operational


second


F value


reflected


diff


erences


between


blocks


The


blocks


were


different


advisory


groups.


Blocks


were


formed


this


level


to control


various


different


ces


between


groups.


The


results


analy


S1S


treatment


difference


were


not


significant


in any


case


results


analyst


of block


differences


were


significant


Hypotheses


, 10,


and


Thi


study,


however,


was


designed


examine


inter-group


diff


erences,


only


inter-treatment


differences


That


s study


was


des


signed


to evaluate


concept


advi


sory


group,


not


success


single


group


advi


sor


in using


concept.


use


fulness


technique


would


limited


success


was


determined


criteria


external


technique.


In addition,


mean


proportional


values


per


observational


period


were


graphed


eac


h hypothesis.


These


graphed


means


did


not


offer


conclusi


evidence


with


regard


achievement


purpose


advisory


groups


basi


of stati


stical


and


graphic


analysis









Limitations


Every


experimental


result


very


nature


limited.


The


present


experimental


conclusions


must


not


be construed


implying


closure


this


case


or as generalizable


other


situations


with


some


character


stics


common


advisory


group


program.


These


results


apply


this


one


instance


existed


when


tested.


The


principal


consideration


which


must


made


with


regard


these


res


ults


that


this


study


dependent


upon


a hypothes


zed correlation


between


interaction


and


personal


growth


and


group


growth


toward


certain


purposes.


correlation


does


not


obtain,


then


no basi


exists


to expect


that


measured


interactional


changes


are


reflective


group


growth


toward


purposes.


Signifi


cant


interactional


change


could


obtained


and


group


growth


inferred,


when,


in reality,


no growth


occurred.


Or growth


might


occur


when


interactional


change


immeasurable


, leading


con-


clusion


that


no growth


occurred


However,


research


reviewed


Chapter


this


study


straightforward


with


regard


relation


ship


between


eraction


change


and


change


the


quality


the


group


life.


A second


consideration


which


may


have


been


influential


thi


study


was


instrument


used


to gather


data.


The


Bales


mod


may


not


have


been


responsive


enough