AN. ANALYSISOF THE RELATION. 07F .
VERTICAL INTEGRATION AND SELECTED ~ _: x E
”Wm “‘0 9mm ‘
RELATEONSHIPS m commune, 1; ‘

  CHANNEL SYSTEMS

 
 
 

: .VThesismormeDegreeofpno r

7* MICHIGAN STATE‘UN'IvERSITE-ii?2:*>9?2-;ggégggéigaiéii=j”:5:321:23;2:1?
T ’ ~ JAMES" L. WIEK. *‘ i

 

J

MICHIGAN STATE

HI H NW

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3 1293 00

Emma“

7

 

 

LIERARY
Michigan State

 

 

U ntversity
< ‘I A ’1 fl,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I
PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
DATE DUE DATE DUE DATE DUE
5
I i
I
I
I
I
I
I
MSU Is An Affirmative Action/Equal Opportunity Institution I I
c:\c|rc\datedue.pm3-p.1 I
i L
I ~ 1“" "“

 

 

 

 

 

 

 

nTHESi—S

 

 

 

 

 

 

ABSTRACT
AN ANALYSIS OF THE RELATION OF

VERTICAL INTEGRATION AND SELECTED ATTITUDES AND
BEHAVIORAL RELATIONSHIPS IN COMPETING CHANNEL SYSTEMS

By

JmmsL.ka

The purposes of this study are: (l) to study the
relationships between vertical integration and selected
attitudinal and behavioral variables in competing channel
systems; and (2) to develop operational measures for each
of the channel system variables under study.

The study was undertaken in order to answer the
following questions:

a.Is there a relationship between the amount of vertical
integration in a channel system and the:

(1) amount of cooperation and conflict displayed by its
members?

(2) speed with which its members adopt new ideas and
practices?

(3) size of the business units in the channel?

(4) age of its members?

(5) education of its members?

b. Do the attitudes and behavior of channel members seem
to vary depending upon geographic region?

c. Is there a relationship between the amount of vertical
integration in a channel system and the attitudes held
by the members of that system?

d. Is it possible to predict the cohesiveness of an individ-

ual with the others in his system, before he joins that
system?

e. PO the members of similar channels have certain attitudes
ln common?

 

 

 

 

James L. Wiek
f.Do channels which are alike in terms of vertical integra—
tion tend to be alike in terms of other variables?

g.Is it possible to operationalize and measure several com-
plex behavioral and attitudinal variables?

Hypotheses were phrased for these questions and data
collected to test these hypotheses. The study employed per-
sonal interviews with 176 persons who were involved in table
egg production—marketing systems. Respondents operated at
one or more of five different channel levels; some were lo—
cated in four southern states while others were in four
northern states.

A research instrument for each channel level was
developed to elicit responses for each of the dependent vari—
ables (vertical conflict, vertical control, cooperation, in—
rmvativeness, operational size, education and age), as well
as to gather information on attitudes with the use of Q—
methodology.

The main unit of analysis in this study is the chan—
nel, which is defined as a group of business firms which
supply inputs to, or market eggs which come from a particu—
lar producer. The producer was the determinant of a chan—
nel; once a producer was selected into the study, the people
with whom he dealt were automatically included.

Data analysis procedures included: factor analysis
Of 'tYPeS' (P analysis), stepwise multiple regression, simple
correlations, and t tests of the difference between two means.

Statistically significant relations were discovered

 

 

 

 

 

wee: 'ncegratnr. a .

r~ r
9 ' "'5"O:: '3 SE'.&.:. .I
“C L0 ”.1“
' .— ~- A o-r L:
EliciLJf. LS oSSC--:-€- n-
r--‘ .

"nastier cf sys:e:r. pa. ----,

:fizat'. differences if. 2:»: :2'
..w.::.1; waters 0 .--.. aria:
‘ .1-" esp-ed :r.a:.r.e- “'5'...
r =_'- characzerize: 2;; a .--
n a .::< o: :3::’--::, :
“era: a tecisior. :* 3: r
If, at: a W---;r.;'.€5: :.
. mi :23: a::-:-:»;—s t: t".
3* l“:e*:a:ec “r.ar.r.e;s :e
l5: described.
Ire “4613' o-ffer-r.; -

T- ‘3‘ ' ‘
~ Lite Al‘F" . ‘ I I
thanueis lI‘LC-UCC‘G

may {Hated to SYS‘CEmic a C‘:

ESI'S ' '
Item is accessible and reC

nt. IL. appeared that ‘

C

 

III“ Chan
“9‘ s .
« ystems consxster

 

 

 

James L. Wiek

between integration and several behavioral variables and
13nded to support the general hypothesis that vertical in—
tegration is associated with efficiencies which result from
Um behavior of system participants. The study found sig-
rfificant differences in the attitudes held by members of
competing systems which varied in terms of integration.
ifighly integrated channel systems are composed of persons
Mm>are characterized by a willingness to cooperate with
oflmrs, a lack of conflict, a willingness to consider the
effect of a decision on the whole system, a marketing ori—
entation, and a willingness to innovate. In general, it
was found that attitudes of the members of non—integrated
and low integrated channels tended to be the reverse of those
just described.

The widely differing levels of competitive effective-
ness of the channels included in this study seem to be di-
rectly related to systemic openness—~the degree to which
a system is accessible and responsive to its external en—
vironment° It appeared that a small number of relatively
OPEH channel systems consistently out—performed the remain—
hm group of less open channels because of a willingness
On the part of these open system participants to: adapt
t0_a rapidly changing market situation, maintain and foster
an attitude of intra—system feedback and communication, and
to operate as a team with a minimum of divisiveness and a

maximum of intra—channel assistance.

 

 

 

AN ANALYSIS OF T
'IERTICAL INTEGRATION ALL

I

EEI-LI'IIORAL RELATIONS 3‘11? 3 1..

u:

James 1

. Submit
Mlchigan Sta
1 fulfillmen

f0r the

inputia

DOCTOR OF

Departm
arketin
g and

l?

AN ANALYSIS OF THE RELATION OF
VERTICAL INTEGRATION AND SELECTED ATTITUDES AND

BEHAVIORAL RELATIONSHIPS IN COMPETING CHANNEL SYSTEMS

By

James L° Wiek

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Marketing and Transportation Administration

1969

 

 

)h

To my best friend, Michelle

 

 

 

Is'nould like to than
sslvemen in this projec: fr:
spatience, deliberation, an:
ratio: while reviewing draft:
szsges of research. and manuscr

Ikeciated. Dr. Henry La."

=.;;es:1ons. The Department 0

"on, under the Chai

:w‘v’lGEQ the author with finan

team from graduate assista

2m " '
So: oata collection, as

we
Icomputer programmino and
Several members of the

235' '
lon Serv1ce at Michigan St

Iall Iio
It, John Wolford, and l

m

”ledge about the poultry ir
951%“ 0f the research in‘

d . .

e519” 0f the quota s

with ample. i
t .

he asmstance of the f ‘
I I o
tillstsl I

. Jerry C

. 0 I .
lsfippi; x, Universa.

T
homas B. Morris, No

I———

ACKNOWLEDGMENTS

I should like to thank Dr. E. J. McCarthy for his
hwolvement in this project from its inception to completion.
lfls patience, deliberation, and constructively critical sug—
gestions while reviewing drafts and talking through critical
stages of research and manuscript formulation are deeply
appreciated. Dr. Henry Larzelere and Dr. Donald Bowersox,
acting as members of the dissertation committee, spent many
hmms reading drafts of this work and offering advice and
mggestions. The Department of Marketing and Transportation
Amfimistration, under the Chairmanship of Dr. Donald Taylor,
provided the author with financial support in the form of
fimm off from graduate assistantship duties during the three
months of data collection, as well as by giving a grant to
cover computer programming and data analysis expenses.

Several members of the staff of the Cooperative Ex—
tension Service at Michigan State University——Charles Sheppard,
Carl Hoyt, John wolford, and Tony Rapes-—provided technical
knowledge about the poultry industry which was necessary for
the design of the research instruments, as well as for the
I design of the quota sample. Data were gathered in the South

with the assistance of the following Extension Poultry Spec—

ialists; Jerry Cox, University of Georgia; Paul Yount, Miss—

issippi; Thomas B. Morris, North Carolina State University;

 

   

 

:3. Lime; Barton, '.n;vers;t§
"3:? Chairman of the Polar;

Sire Cniver s icy i r. t: 2 cu : e<

 

izn Pontry Spec;a-;s:s .

Clyde Springer, 3". Cr:

""53 ' " ‘ ' 5.. .- '
(7,-..1, an “(ZYIESC Fw x F
:‘or' . .~ 5 ' : .
:ss-s.eo o; p:cv-c-r.;
fl 2:

 
   
 

' tens of mews}-.. -
-- e .

 

 

mm.T. Lionel Barton, University of Arkansas. Dr. Howard
Zindel, Chairman of the Poultry Science Department at Mich—
iganState University introduced the author to these four
mdfinsion Poultry Specialists.

Clyde Springer, Dan Crowle, Wayne Playford, Gene
quwrden, and Wayland Fox of Farm Bureau Services, Incor—
mmated, assisted by providing the author with a background
oftflm egg industry in Michigan-~particularly with the opera—

Upnal details of contractually integrated channels.

iv

 

 

 

 

 

TABLE OF

I. INTRODUCTION . . . . .
PURPOSE OF THE STUD! .
RESEARCH 335:3); AND ME

The Product: Eggs
Variables Included .
Development of Hypo:
lnscruments . .
Selection of Responc

CONTRIBUTIONS OF THE 5

THEORY AND PRACTlCE

LIMITATIONS . . .
ORGANIZATION OF THE RE

-3. Run»; or LITERATURE .
GEAERAL CONCEPT 0? CH)

THE CHANNEL CONCEPT DI

ALLY .

o o
. o o

A SYSTI
BEHAVIORAL ELEMENTS ll
Contr

THE CHANNEL AS

01 in the Cham
The locus of cham
Methods of channe
The measurement 0
Control and chann‘

Conflict in the Cha
Causes of channel

Conflict re solu ti

TABLE OF CONTENTS

Chapter

I. INTRODUCTION . . . . . . . . . . . . . . . . . 1

PURPOSE OF THE STUDY . . . . . . . . . . . . 2
RESEARCH DESIGN AND METHODOLOGY. . . . . . . 3
The Product: Eggs . . . . . . . . . . . . 4
Variables Included in the Study. . . . . . 5
Development of Hypotheses and Research
Instruments . . . . . . . . . . . . . . . 5
Selection of Respondents . . . . . . . . .

CONTRIBUTIONS OF THE STUDY TO MARKETING
THEORY AND PRACTICE . . . . . . . . . . . . 7

LIMITATIONS O 0 0 0 0 O 9 0 0 0 ° 0 O . D O O 8
ORGANIZATION OF THE REMAINDER OF THE THESIS. lO ,

II. REVIEW OF LITERATURE . . . . . . . . . . . . . 13
GENERAL CONCEPT OF CHANNEL . . . . . . . . . 13
THE CHANNEL CONCEPT DEFINED MORE SPECIFIC—

ALLY. . . . . . . . . . . . . . . . . . . . 14
THE CHANNEL AS A SYSTEM. . . . . . . . . . . l6
BEHAVIORAL ELEMENTS IN CHANNEL SYSTEMS . . . 19

Control in the Channel . . . . . . . . . . 19
The locus of channel control . . . . . . 20
Methods of channel control . . . . . . . 21
The measurement of channel control . . . 21
Control and channel efficiency . . . . . 22

Conflict in the Channel. . . . . . . . . . 23
Causes of channel conflict . . . . . . . 23
Conflict resolution. . . . . . . . . . . 25

Cooperation in Marketing Channels. . . . . 28

Innovativeness in Channel Systems. . . . . 3o

VERTICAL INTEGRATION IN CHANNEL SYSTEMS. . . 33

   

3m
MAJOR STUDIES OF STRUC
AGRICULTURAL. CF ‘iliELE
REPORTS ON VERTICAL I)
INDUSTRY. . . . . ..
REPORTS ON THE ATTII‘UI
EGG CHANNELS. . . .
CONCLUSIONS ABOUT THE
NRMJCHANNEL LIIERAT
:1. RESEARCH DESIGN AND ANA}
INTRODUCTION . .
}KPOTHESES . . .
Eiypothesis L . . .
tatistical HypOt
\mneral Hypotnes
Sample Null Hypo:
Statistical Hypo
HYPOt'nesis 2 ,
Statistical Hypot
Hypothesis 2.

Samplg Null HYPOt
Statistical Hypo

HYPOthesis 3 .
Statistical Hypo:

 

Hypothesis 3.
Sample Null
. H ot
Statistical §§pc
Hypothesis 4 .

Statistical H
HYPOthesis 4

Sample Nu
Statist-ll “Ypo1

lcal HYP‘

Hypothesis 5
Hypothesis 6

DEVELOPMENT OF RESP.

constructi
Cons

YPOI

A]

tr on of 01.14
UCtiOn
EmplOYQd in, :fieAé

7—7 .I
i 1

Chapter Page
MAJOR STUDIES OF STRUCTURAL ORGANIZATION OF
AGRICULTURAL CHANNELS . o . . . . . . . . . 4O
REPORTS ON VERTICAL INTEGRATION IN THE EGG
INDUSTRY. 0 o O 0 o I a 0 O O I O O 0 O 44
REPORTS ON THE ATTITUDES HELD BY MEMBERS OF
EGG CHANNELS. . . . . . . . . . . . . . . 45

CONCLUSIONS ABOUT THE MARKETING AND AGRICUL—
TURAL CHANNEL LITERATURE SURVEYED . . . . . 47

III. RESEARCH DESIGN AND ANALYSIS OF DATA . . . . . 49

INTRODUCTION 0 ° 0 D 0 O 0 ° 0 0 9 O 9 6 I I 49
MPOTHESES o o 0 0 0 o 0 0 0 o 0 O 0 O 0 O O 49
Hypothesis 1 . . . . . . . o . . . . . . . 49

Statistical Hypotheses Which Relate to
General Hypothesis lo . . . . . . . . . 50

Sample Null Hypothesis Which Relates to
Statistical Hypotheses 1.1 to 1.8 . . . 52
Hypothesis 2 . . o o o . . . . . . . . . . 52

Statistical Hypotheses Which Relate to
Hypothesis 2. . . . . . . . . . . . . . 52

Sample Null Hypothesis Which Relates to

: Statistical Hypotheses 2.1 to 2.8 . . . 53
i Hypothesis 3 . . . . . . . . . . . . . . . 53
Statistical Hypotheses Which Relate to
Hypothesis 3. . . a . . . . . . . . . . 54
Sample Null Hypothesis Which Relates to
Statistical Hypotheses 3.1 and 3.2. . . 54
Hypothesis 4 . . . . o o o . . . . . . . . 54
Statistical Hypotheses Which Relate to
Hypothesis 4° . a . . o . . . . . . . . 54

Sample Null Hypothesis Which Relates to
Statistical Hypotheses 4.1 to 4.4 . . . 55

Hypothesis 5 0 0 o O 0 0 O 0 0 I 0 O U o o 55
Hypothesis 6 . . . o . o . . . . . . . . . 55
DEVELOPMENT OF RESEARCH INSTRUMENTS° . . . . 55

Construction of Questionsa . o . . . . . . 56

Construction of Attitudinal Statements
Employed in the Q—Methodology . . . . . . 57

 

 

 

 

 

7v
J.

Field Testing the R‘
FIELD PROCEDURE.

Sample Size .

’h

Selection 0 Produc‘
Interview Procedure

Procedure for Inter
of the Channel.

Procedure Employed

Respondents .
ANALYSIS OF DATA .

Explanation of Step

Analysis of Hypothe

Hypotheses 1.1 to

Hypotheses 2.; to

Hypotheses 3.1 ar.
4

Hypotheses

-l tc
Hypothesis 5.1 .
HYPothesis 6.1 .

Summary Of HypOthes

FINDINGS OF THE STUDY

THE RELATIONSHIP BET'U
SELECTED BEHAVIORAL
Hypothesis 1.1:

H C<
I'Il/Dothesis 1.3: Q
HYPothesis 1.4; Q
HYPothesis 1.5: I]
HYPOthesis 1.6. A
HYpothesis l . 7 . E
HYDOthesis 1.8, O
A I

13:?“ka of the R

ween Integrat

Chapter Page
Field Testing the Research Instrument. . . 58
FIELD PROCEDURE. . . . . . . . . . . . . . . 59
Sample Size. . . . . . . . . . . . . . . . 6O
Selection of Producers . . . . . . . . . . 61
Interview Procedure. . . . . . . . . . . . 62
Procedure for Interviewing Other Members
of the Channel. . . . . . . . . . . . . . 63
Procedure Employed in Interviewing Southern
Respondents . . . . . . . . . . . . . . . 64
ANALYSIS OF DATA . . . . . . . . . . . . . . 64
Explanation of Steps in Data Processing. . 65

Analysis of Hypotheses . . . . . . . . . . 68
Hypotheses 1.1 to 1.8. . . . . . . . . . 68

Hypotheses 2.1 to 2.8. . . . . . . . . . 69

Hypotheses 3.1 and 3.2 . . . . . . . . . 69

Hypotheses 4.1 to 4.4. . . . . . . . . . 69

Hypothesis 5.1 . . . . . . . . . . . . . 70 ‘f

Hypothesis 6.1 . . . . . . . . . . . . . 71 ‘
Summary of Hypothesis Testing Procedure. . 72

IV. FINDINGS OF THE STUDY. . . . . . . . . . . . . 73
THE RELATIONSHIP BETWEEN INTEGRATION AND

SELECTED BEHAVIORAL VARIABLES . . . . . . . 73
Hypothesis 1.1: Conflict Received . . . . 73
Hypothesis 1.2: Conflict Given. . . . . . 76
Hypothesis 1.3: Cooperation Received. . . 77
Hypothesis 1.4: Cooperation Given . . . . 78
Hypothesis 1.5: Innovativeness. . . . . . 78

Hypothesis 1.6: Age . . . . . . . . . . . 79
Hypothesis 1.7: Education . . . . . . . . 80
Hypothesis 1.8: Operational Size. . . . . 80
A Summary of the Relationships Which Exist

Between Integration and Selected Depend-

ent Variables . . . . . . . . . . . . . . 82
THE RELATIONSHIP BETWEEN NORTHERN VERSUS
SOUTHERN LOCATION AND SELECTED CHANNEL
BEHAVIORAL VARIABLES. . . . . . . . . . . . 82

 

 

Hypothesis 2.1 CO:
Hypothesis 2.2. Co.
Hypothesis 2.3: CO
Hypothesis 2.4: CO
Hypothesis 2.5: Co
Hypothesis 2.6: In
Hypothesis 2.7: A;

hypothesis 2.8: LO
Summary of :‘ne Diff
Between Northern a
T'hT-J RELATIONSHIP SET»
TION AND THE ATTI CD

Discussion of the R
Hypotheses 3.1 and
PREDICTION OF CHANEEL
Hypothesis 4.1: Pr
Received. .
Hypothesis 4.2: Pr
Given .

 

Hypothesis 4 . 3: P:
Received. .

Hypothesis 4.4: Pr
Given . . . . . .
A Summary of the Re
Between the Critex
Predictors of Suc<
THE DIFFERING ATTITUI
Explanation of Tab:
Description of Eacl
Factors l and 2.
Factors 3 and 4.

Factors 5 and 6.
FaCtor 7 .

Persons . . . . .
TYPES . . RT BEHAVI

Chapter

Hypothesis 2.1: Control . . . . . . . . .
Hypothesis 2.2: Conflict Received . . . .
Hypothesis 2.3: Conflict Given. . . . . .
Hypothesis 2.4: Cooperation Received. . .
Hypothesis 2.5: Cooperation Given . . . .
Hypothesis 2.6: Innovativeness. . . . . .
Hypothesis 2.7: Age . . . . . . . . . . .
Hypothesis 2.8: Education . . . . . . . .

Summary of the Differences Which Exist
Between Northern and Southern Channels.

THE RELATIONSHIP BETWEEN VERTICAL INTEGRA—
TION AND THE ATTITUDES OF PERSONS . . . . .

Discussion of the Results of Tests of
Hypotheses 3.1 and 3.2. . . . . . . . . .

PREDICTION OF CHANNEL MEMBER COHESIVENESS. .
Hypothesis 4.1: Prediction of Conflict

Received. . . . . . . . . . . . . . . . .
Hypothesis 4.2: Prediction of Conflict

Given 0 0 a I O I O O O O O 0 O O i O O O
Hypothesis 4.3: Prediction of Cooperation
Received. . . . . . . . . . . . . . . . .
Hypothesis 4.4: Prediction of Cooperation
Give n O O a o O 0 0 (7 O O 0 9 0 D I O D I

A Summary of the Relationships Which Exist
Between the Criteria of Success and the
Predictors of Success . . . . . . . . . .

THE DIFFERING ATTITUDES OF TYPES OF PERSONS.
Explanation of Table 4.5 . . . . . . . . .

Description of Each Factor Type. . . . . .
Factors 1 and 2. . . . . . . . . . . . . .
Factors 3 and 4. . . . . . . . . . . . . .
Factors 5 and 6. . . . . . . . . . . . . .
Factor 7 . . . . . . . . . . . . . . . . .
A Summary of the Differing Attitudes of
Persons . . . . . . . . . . . . . . . . .

THE DIFFERENT BEHAVIOR PATTERNS OF CHANNEL
TYPES C 0 O Q 0 0 ° 0 0 O 0 O 0 0 0 0 9 D C

viii

Page

83
83
85
85
86
87
88
88

89
9O

9O
95

95
97
97
98
98
99
100
100
104
105

106
107

107

108

 

 

 

Behavioral Descripti
Factors 1 and 2. .
Pactor3. . . . .
Factor 4 . . .

Types of Channels ‘n‘r.
Bach Variable . . .

Explanation of Tat
A Summary of the Diz‘
Descriptions of Cna

. SWEAR! OF RESEARCH APPF
EVALUATION OF COMPEIIHG
CONCLUSIONS. .

c o o 0

INTRODUCTION . . . -
SUHMARY OF RESEARCH AI
Purposes of the Stuc
Literature Reviewed.
Research Design and
Operationalizatior
Internal Consister
Operationalizatiox
Pilot Study.
Selection of Resp<
SWTMRY 0F FINDINGS.
EVALUATION OF COMPETII
Physical Characteri.
Systems . . . . .
Operational Size .
sYstem Structure
other Characteristi
COmpeting C
PhYSical a
Relativel
EXPLANATION O

hannels
nd Other

Y Open 3

_'__—_7

Chapter
Behavioral Description of Each Factor Type
Factors 1 and 2. . . . . . . . . . . . .
Factor 3 . . . . . . . . . . . . . . . .
Factor 4 . . . . . . . . . . . . . . . .
Types of Channels Which Are Described by
Each Variable . . . . . . . . . . . . . .

Explanation of Table 4.7 . . . . . . . .
A Summary of the Differing Behavioral
Descriptions of Channel Types . . . . . .

V. SUMMARY OF RESEARCH APPROACH AND FINDINGS,
EVALUATION OF COMPETING CHANNEL SYSTEMS, AND
CONCLUSIONS. . . . . . . . . . . . . . . . . .

INTRODUCTION . . . . . . . . . . . . . . . .
SUMMARY OF RESEARCH APPROACH . . . . . . . .
Purposes of the Study. . . . . . . . . . .
Literature Reviewed. . . . . . . . . . . .
Research Design and Methodology. . . . . .
Operationalization of Variables. . . . .
Internal Consistency of Questions. . . .
Operationalization of Attitudes. . . . .
Pilot Study. . . . . . . . . . . . . . .
Selection of Respondents . . . . . . . .
SUMMARY OF FINDINGS. . . . . . . . . . . . .
EVALUATION OF COMPETING CHANNEL SYSTEMS. . .

Physical Characteristics of Egg Channel
Systems fi 0 O D 0 0 o o o 0 O O 9 Q o 0 0

Operational Size of System Members . . .
System Structure . . . . . . . . . . . .
Other Characteristics of Systems . . . . .

Competing Channels Described in Terms of
Physical and Other System Characteristics

Relatively Open Systems. . . . . . . . .
EXPLANATION OF TABLE 5.4 . . . . . . . . . .
Moderately Closed Systems. . . . . . . .
Relatively Closed Systems. . . . . . . .

 

Page

111
111
112
113

113
113

115

117

117
117
117
117
120
122
123
123
124
125
128
133

134
134
134
134

138
138
147
147
151

 

 

 

CONCLUSIONS AND REFLL
Meaning of This Res

Meaning of “his Res
tems Opera-tin: -r.

.YANAGERIAL IMPLIC 0"! S
FUTURE RESEARC

am A GLOSSARY.

12353.32“ RESEARCH INSTRCHE
Rana: ITS: ANALYSIS RESL
2: n or VARLAELEE
Rana Foqu'tAr: or or Th:
INDEX. .
Rama :xsmrcnozs FOR .
RPSRDEX e PROCEDURES eraprozz
REIDZQX ~ PROCEDURE FOR CHO<
SOLUTIONS. .
Nit»
"MATH DESCRIPTION or THE
“MULTIPLE REGRES:
ROUTINE. . . . .

ESLIOGRAPHY. .

T__7

Chapter

CONCLUSIONS AND REFLECTIONS. . . . . . . . .
Meaning of This Research for Egg Channels.

Meaning of This Research for Channel Sys-
tems Operating in Other Industrial Settings 156

VI. MANAGERIAL IMPLICATIONS AND OPPORTUNITIES FOR
FUTURE RESEARCH. . D O O 0 D D O O I C O I C O

Managerial Implications of This Research

Approach. . . . . . . . . . . . . . . .
Opportunities for Future Research. . . . .
APPENDIX A GLOSSARY O o 0 D 9 0 0 0 0 0 0 I) a 9 O I 3
APPENDIX B RESEARCH INSTRUMENTS. . . . . . . . . . .
APPENDIX C ITEM ANALYSIS RESULTS AND OPERATIONALIZA—
TION OF VARIABLES. . o . . . . . . . .
APPENDIX D FORMULATION OF THE VERTICAL INTEGRATION
INDEX 0 0 O O O . D I O O \O O 0 O O O O 0
APPENDIX E INSTRUCTIONS FOR INTERVIEWERS . . . . . .
APPENDIX F PROCEDURES EMPLOYED IN THE TYPE ANALYSES.
APPENDIX G PROCEDURE FOR CHOOSING AMONG FACTOR
SOLUTIONS. . . . . . . . . . . . .
? APPENDIX H DESCRIPTION OF THE OPERATION OF THE
"MULTIPLE REGRESSION DELETE" COMPUTER
ROUTINEO O o 0 0 0 0 o O a o o D 0
BIBLIOGRAPHY o o o n o b o 0 o o o o o o a a o o o n o

 

Page

155
155

158

158
161

163

167

187

196
205

210

217

226

227

 

 

LIST (a?

5.4.5

3.1 Types of Respond :ts. .

R.

Types of Egg Producers i
Classified by Gecgrapf

Type......--°

Simary of Hypotheses Te

=». analysis of the Results

tions'nip 3e twee: In 2e;

Behavioral Variables &
Results Based on Ch

- .
an.

1 Results of Tests of vac

on t Tests of Relatior
Location and Selected
35 Channels

(1)

The Relationship Betwee:
and the Attitudes of

 

is

Prediction of the Futurt

Potential Channe l Mem'

3
Representative Types of
INoe, and Attitudinal
MOSt Strongly Charact

.‘I

HYPOtheSeS.

Table

3.2

3.3

LIST OF TABLES

Types of Respondents. . . . . . . . . . . . . .

Types of Egg Producers Included in the Study,
Classified by Geographic Region and Channel

Type o o o o o o o o o o o c o o o o o I o o 0

Summary of Hypotheses Testing Procedure . . . .

Analysis of the Results of Testing the Rela—
tionship Between Integration and Selected
Behavioral Variables and Comparing the

Results Based on Channel and Individual Data.

Results of Tests of Hypotheses 2.1 to 2.8 Based
on t Tests of Relationship Between Geographic
Location and Selected Dependent Variables for

35 Channels o o o o o o o I: o o o o a o a c o

The Relationship Between Vertical Integration
and the Attitudes of 148 Persons° . . . . . .

Prediction of the Future Cohesiveness of a
Potential Channel Membero . . . . . o . . . .

Representative Types of Persons in Each Factor
TYpe, and Attitudinal Response Patterns Which
Most Strongly Characterize Each . . . . . . .

Behavioral Description of Channel Types v ° ° '

Median Integration Scores of Channels the
Extreme Ends of Each Behavioral Continuum . .

Statistical Procedures Used in the AnalYSis Of

a o 9 9 o o n o

HypotheSeSo O 0 9 0 0 0 0 0 0

xi

Page
60

62

72

74

84

91

96

101

109

114

127

 

 

 

 

file

5.2 Three Groups of Channel
Criteria of System 0

3.3 Cohesiveness in Competi

Systems.......

3.4 Fean Size of Business U
of Channel Systems

Exhibit A. . .

Exhibit B. . .

Exhibit C. . .

C.l Validity of mestions U
Each Research Instrum
Factor Analysis . .

3-1 Values Employed in Comp
ration Index Measure

Ff; An Example of Perfectly

5'1 Heasures of Factor Homog
Solution, Analysis 5 . l

3-2 Measures of Factor HOMO!
Solution, Analysis 6.;

 

 

 

Table Page

5.2 Three Groups of Channels Rated on Each of Four

Criteria of System Openness . . . . . . . . . 137

5.3 Cohesiveness in Competing Types of Channel
Systems 0 O O O O O O O O 0 O 0 0 O O O O I O 141
5.4 Mean Size of Business Units in Competing Types

of Channel Systems

Exhibit A. I O 0 0 0 0 O 0 O D C C O I 144

Exhibit B. O 0 0 I 9 O o o o C 5 O O I 145

Exhibit C. O O O 0 0 0 O 0 I U B I I O 146
CJ. Validity of Questions Used in Each Index for
Each Research Instrument, as Determined by

. . 193

Factor Analysis . . . . . . . . . . . . .

DJ. Values Employed in Computing Individual Inte—
gration Index Measures for Each Channel . . . 200 I

F.l An Example of Perfectly Homogeneous Factors . . 213

GJ. Measures of Factor Homogeneity, Seven Factor
Solution,Analysis 5.1 . . . . . . . . . . . . 221
G.2 Measures of Factor Homogeneity, Four Factor

Solution, Analysis 6.1. . a . . . . . . . . . 224

xii

 

 

 

 

 

 

C

INTRO

Traditionally, the ec
leendiscussed in the marketi
isbeen directed toward the
Ifficiency through the minimi
More recent approach to cha
We view that channels are
Nultinate success of which
flies and costs, but also u
Mutation, conflict, contr

innovativeness . 2

\_—.

., 1See, for example: I.
llIllcture of Channels of Dist
‘9“, Harketin : A Maturinc
Ba{listing Association, 19605 ,
inner, "The Division of Lat
ofthe Market," Journal of Pc
99- 185-193. The University
Ad Leon 3. Richartz, Vertica
“Ill and Bacon, Inc., 15675

256e, for example: I
“inflict in Marketing Channe:
Reflections on Pro ress in Me
HAlerting Association, 1964 5 .
"humus of Channel Centre.
All. keting and the New Sciencc
“a“ MEmketing Association, .
Palmountain, The Politics 0:
lard University Press, 1955"
underStanding the Retailer,
ll(Nay-J1me, 1962), pp- 147‘
IIAlternative Explanations 2f
3‘Million,“ in Stephen Ac, 17‘
"it. keting (Chicago: Amer l
I’D» 477-490.

  

CHAPTER I

INTRODUCTION

Traditionally, the economic aspects of channels have
been discussed in the marketing literature1 and much effort
. has been directed toward the goal of increasing marketing
efficiency through the minimization of certain channel costs.
A more recent approach to channel analysis is characterized
by the View that channels are social systems, the performance
and ultimate success of which is based not solely on eco—
nomics and costs, but also upon such behavioral factors as
cooperation, conflict, control, power, communication, and

. . 2
innovativeness .

1See, for example: Louis P. Bucklin, "The Economic
Structure of Channels of Distribution," in Martin L. Bell
(ed.), Marketing: A Maturing Discipline (Chicago: American
Marketing Association, 1960), pp° 379-385; and George J.
Stigler, "The Division of Labor Is Limited by the Extent
of the Market," Journal of Political Economy (June, 1951),
pp. 185-193. The University of Chicago; and Helmy H. Baligh
and Leon E. Richartz, Vertical Market Structures (Boston:
Allyn and Bacon, Inc., 1967).

2See, for example: Bruce Mallen, "Cooperation and
Conflict in Marketing Channels," in L. George Smith (ed.),
Bejlections on Progress in Marketing (Chicago: American
Marketing Association, 1964), pp. 65-85; and Louis P. Bucklin,
"The Locus of Channel Control," in Robert L. King (ed.),
{lirketing and the New Science of Planning (Chicago: Amer—
ican Marketing Association, 1968), pp. 142-147; and Joseph c.
Palamountain, The Politics of Distribution (Cambridge: Har-
vard University Press, 1955); and Warren J. Wittreich, "D1115-
understanding the Retailer," Harvard Business Review, Vol.
40 (May-June, 1962), pp. 147-159; and Bert C. McCammon, Jr.,
"Alternative Explanations of Institutional Change and Channel
Evolution," in Stephen A. Greyser (ooh), Toward Scientific
Egrkzgin‘lggéChicago: American Marketing Assoc1ation, 19637,

 

 

 

 

 

 

l'uch of the channel 11

52, abstract level , however;
title has been verified by e
sites have been reported in
giered to support or reject
:-:behavior.3 This study ex
amoral variables which ma‘

:vard integration in competi

PURPOSE 0
ifnile the present tre
pied production-nuke ting s
:economies of scale which p
:ease in the efficiency of s
ES'Jlt of certain attitudinal
Played by members of these sy.
The purposes of this
relationships between vertica
‘iiudinal and behavioral vari
MS; and (2) develop operati

M.

.. 31m exception is: Do
fillet, “Patterns of Conflict
‘ IS," in George L. Smith.
W (Chicago: Amer:

. 4The relation betweer

mtEgration is examined by E.

mums, “Simulation of Proc

“Raymond M. Bass (ed.), _S_<_:_fi

gihicago: American Marketing
SI

 

i - 2

Much of the channel literature is written at a gen-
erah,abstract level, however; hypotheses are advanced but
Initle has been verified by empirical research. Only a few
shxfies have been reported in which 'real-world' data were
gaflmmed to support or reject hypotheses about channel sys-
UMlbehavior.3 This study examines several attitudinal and
Mfimvioral variables which may help to explain the trend

Unmrd integration in competing production-marketing systems.

PURPOSE OF THE STUDY

While the present trend toward increasingly inte—
gnated production—marketing systems has been due in part
UJeconomies of scale which produce lower costs,4 the in-
cxease in the efficiency of such systems may also be the
result of certain attitudinal and behavioral patterns dis-
Iflayed by members of these systems.

The purposes of this study are to: (1) study the

relationships between vertical integration and selected at—

 

titudinal and behavioral variables in competing channel sys—

tems; and (2) develop operational measures for each of the

3An exception is: Donald H. Granbois and Ronald P.
Vfillet, "Patterns of Conflicting Perceptions Among Channel
thbers," in George L. Smith (ed.), Reflections on Progress
in.Marketing (Chicago: American Marketing Association, 1964),
pp. 86‘100 o

4The relation between channel costs and vertical

integration is examined by E. Jerome McCarthy and Robert J.
Vfilliams, "Simulation of Production—Marketing Channels,"
in.Raymond M. Hass (ed.), Sciencei Technology, and Marketing
gigicago: American Marketing Association, 1966), pp. 3354“—

   

 

 

 

amel system variables unde

The study has been un
zfolloving questions:

a. Is there a relationsh
gration in a channel

(1) amount of coopera
its members?

(2) speed with which
and practices?

(3) size of the busin

(4) age of its member

(5) education of its

'2. Do the attitudes and
seem to vary dependin

('y

Is there a relationsh
tical integration in
tudes held by the mem

5:.

Is it possible to pre
individual with the 0
he joins that system?

(I:

Do the members of sin
attitudes in common?

0-0.

Do channels which are
tend to be alike in t

«3

Is it possible to ope
complex behavioral an

RESEARCH DES
This is a study invol
keting channel systems for eg
if the attitudinal and behavi
acterize the members of these
kfinitions of previously 115‘
We collected in several Ste

“tough the use of personal .

 

- 3

dwnnel system variables under study.

The study has been undertaken in order to answer

Hm following questions:

a.

Is there a relationship between the amount of inte-
gration in a channel system and the:

(1) amount of cooperation and conflict displayed by
its members?

(2) speed with which its members adopt new ideas
and practices?
size of the business units in the channel?

(4) age of its members?
education of its members?

Do the attitudes and behavior of channel members
seem to vary depending upon geographic region?

Is there a relationship between the amount of ver—
tical integration in a channel system and the atti—
tudes held by the members of that system?

Is it possible to predict the cohesiveness of an
individual with the others in his system, before

he joins that system?

Do the members of similar channels have certain
attitudes in common?

Do channels which are alike in terms of integration
tend to be alike in terms of other variables?

Is it possible to operationalize and measure several
complex behavioral and attitudinal variables?

RESEARCH DESIGN AND METHODOLOGY

This is a study involving a set of production—mar—

keting channel systems for eggs. The study quantified some

of the attitudinal and behavioral relationships which char-

acterize the members of these systems. Rigorous operational

definitions of previously used concepts were developed; data

were collected in several states in the North and the South

through the use of personal interviews; and a number of

 

 

 

 

 

potheses regarding the relat

retested for statistical si

:eProduct: Eggs
The table egg is the p

zdzannel systems included i
:amelsuhich exist to aid t;
:‘fer several attractions to r
mt channel systems. Egg
tesfor analysis because:

4 Table eggs are physically
resorted according to Unit
:iture grade and size, they
iiconsequently they have re
insof consumers. Thus the
:tribute variability to the
57ers of the channel systems
zilarket them.

3. Auide range of structural
ailing from those which are c
muons; to those which cor
lI‘étied to feed suppliers by
lamels which contain larger
$55 firm has assumed the peri
3m or marketing functions
3915 in the channel.

3 The structure of egg Chan
faithe evolutionary pattern
ff“?! and turkey industries
{Mini change in these indu
hung the same sorts of cha

 

5; Traditionally, several 1e
:55 members of each of thes
JedObJectives which were in

 
 

Luther channel members. Th
h to sub-optimization, dupl
fictions which have served
011°f these channels.

 

(—7.

hypotheses regarding the relationships among the variables

were tested for statistical significance.

The Product: Eggs
The table egg is the product which links together

the channel systems included in the study. Eggs, and the
channels which exist to aid their production and marketing,
offer several attractions to researchers who wish to analyze
extant channel systems. Egg channel systems are good candi—
dates for analysis because:

1. Table eggs are physically homogeneous. While they have
been sorted according to United States Department of Agri—
culture grade and size, they typically have not been branded
and consequently they have remained undifferentiated in the
minds of consumers. Thus their homogeneous nature does not
contribute variability to the attitudes and behavior of the
members of the channel systems which have evolved to produce
and market them.

2. A wide range of structural types of egg channels exist,
ranging from those which are composed of small, independent
operations; to those which contain larger operations which
are tied to feed suppliers by production contracts; to those
channels which contain larger operations in which one busi—
ness firm has assumed the performance of two or more produc—
tion or marketing functions traditionally performed at other
levels in the channel.

3. The structure of egg channel systems seems to be follow—
ing the evolutionary pattern previously established in the
broiler and turkey industries. This evolution produced sub—
stantial change in these industries and may presently be
Causing the same sorts of changes within the egg industry.

4. Traditionally, several levels have existed in egg Chan-
nels; members of each of these levels frequently have pur—
sued objectives which were incompatible with the interests
Of Other channel members. This divergence of interests has
led to sub—optimization, duplication of effort, and possibly
to actions which have served to weaken the competitive posi_
tion of these channels.

 

 

 

 

 

5, Data on the relation bet
iteration in egg channels w
fl!ians.5 Hence a study of
:u'ty to examine how behavio
enforce the economic reasons
:egration in channel systems

tables Included in the Stuc

To fulfill the objectl
miables were defined,6 opera
bindependent variables are:
:operation, vertical control,

be, age, and education.

heloEnt of Hmtheses and

Six general hypothese!
ntistical hypotheses were d1
fthe data. The study 8111910}
'3’: individuals who operated ‘
£57111 channel systems.8 A
\--——--,—-

SEQ.

6Each of these variab
L)llendix A).

| 7The vertical positio
“Inferred to as 'levels'--
WY level, the producer 1e
1"“: and the retailer level

i 8In this study, a 'ch
J“’“t‘supply firms (feed mill
Wit-receiving firms (proce
the“) and the egg producer

 

5. Data on the relation between channel costs and vertical
hungration in egg channels were collected by McCarthy and
wuliams.5 Hence a study of egg channels provides an oppor-
huuty to examine how behavioral variables complement and
reinforce the economic reasons for the trend toward vertical
hmegration in channel systemsn

Vmfiables Included in the Study

 

To fulfill the objectives of this study, several
variables were defined,6 operationalized, and analyzed.
flm independent variables are; vertical conflict, vertical
cooperation, vertical control, innovativeness, operational

size, age, and education.

Ibvelopeeet of Hypotheees and Research Instruments

Six general hypotheses and twenty—two null and
statistical hypotheses were developed to guide the analysis
of the data. The study employed personal interviews with
176 individuals who operated at each of five different lev—
els7 in channel systems.8 A separate research instrument
‘—————-——.——

Sieg-

6Each of these variables is defined in the Glossary
(Appendix A) .

 

7The vertical positions of the members of a channel
are referred to as 'levels'-—the feed supply level, the bird
supply level, the producer level, the processor-distributor
level, and the retailer level are included in this study.

8In this study, a 'channel system' includes the

input-supply firms (feed mills and hatcheries) and the
output-receiving firms (processors, distributors, and re—
tailers) and the egg producer with which they are associ-
ated.

 

 

 

 

redeveloped for each channe

ionsincluded in each instr
:25 and concerns of persons
thresearch instrument was
zsection designed to measure
25:10:15, the behavior of ea
meat variables; and (b) a
ataents designed to elicit
éav'nich relate to each by
idof simple correlations, t

:Eactor analysis .

hiection of Reswndents

The producers include
:the basis of a representat
ire: was asked who supplied
.tccessed and distributed his
IUIuas asked who retailed t
f‘nswho made up each entire
i-“terviewed.

All of the producers
located in the state of Mich
Ere conducted with channel
located in the states of Ohi
Wide a regional perspecti
3°rthern channel members, in

luthem states of Mississip

  

was developed for each channel level in order that the ques—
‘Uons included in each instrument be relevant to the activ—
ifles and concerns of persons operating at that channel level.
Emfllresearch instrument was composed of two sections: (a)
asection designed to measure, through the use of direct
mmstions, the behavior of each respondent on several de—
pendent variables; and (b) a section containing twenty-seven
statements designed to elicit attitudinal responses° The

data which relate to each hypothesis were analyzed with the
ahlof simple correlations, t tests, multiple regression,

m:factor analysis.

Selection of Respondents

The producers included in this study were selected
m1the basis of a representative quota sampleo Each pro—
mmer was asked who supplied his birds and his feed and who
{recessed and distributed his eggs° Each processor-distrib—
utor was asked who retailed the eggs he handledo Thus, the
firms who made up each entire channel were named and later
interviewed.

All of the producers included in the North were
located in the state of Michigan; some interviews, however,
were conducted with channel members other than producers
located in the states of Ohio, Indiana, and New Yorko To
Provide a regional perspective for the responses of these
northern channel members, interviews were conducted in the

southern states of Mississippi, Alabama, North Carolina,

 

 

 

uhkansas. These four size
they produce a large sham
:u‘JeUnited States; (b) the
widespread in these states
sent suggestions in the eg
may have a competitive ad‘
:Jgressive attitudes and dis

separate with one another an

CONTRIBUTIONS OF T
THEORY A

The present study has
mined as significant cont
iépractice: First, it was
'L'attitudinal and behavioral
Eketing systems operate 'be
Hing systems. Whereas prev
fined, economics-oriented rea
EClef these systems, this 5
Mining the non—economic r
M systems.

Second, in order to
in, a pioneering methodolo
lbles which had been treated
Metal terms were defined a

”It and quantitative analys

Third, whereas previ

 

 

mm Arkansas. These four states were included because:

(a) they produce a large share of the eggs that are produced
tithe United States; (b) the use of production contracts

is widespread in these states; and (c) there have been per—
sistent suggestions in the egg industry that southern egg
maimay have a competitive advantage because they hold more
mmgressive attitudes and display a greater willingness to

cooperate with one another and to innovate.

CONTRIBUTIONS OF THE STUDY TO MARKETING
THEORY AND PRACTICE

The present study has several strengths which may
be viewed as significant contributions to marketing theory
and practice: First, it was undertaken in order to explain
hiattitudinal and behavioral terms why vertically controlled
nmrketing systems operate 'better' than less controlled com—
PMflng systems. Whereas previous studies have provided cost—
based, economics—oriented reasons for the superior perform—
mme of these systems, this study has been directed at de—
termining the non—economic reasons for the popularity of
such systems.

Second, in order to provide this behavioral explana_
Hon, a pioneering methodology was developedo Complex vari—
ables which had been treated in the literature in vague,
general terms were defined and operationalized for measure-
ment and quantitative analysis.

Third, whereas previous research efforts have

 

 

 

mutated on the analysis
the present study entire c
335, the present study analy
:‘hsh’ps as they exist at di
pies.

Fourth, this study is
armchmel systems whi
is. The inclusion of diffe
inctures provides an unusua
hiysis.

Fifth, the analysis c
hate in different geograph
fibehavior of northern vers
.35: why the egg industry see

Ethe South than in the Nort

LIM

Several aspects of t

539“ limit the generalizabil‘
Wthe study. First, the
astribution management are
“his. While much analytic
mhe areas, a great deal mo
70" physical distribution sy

uWily. Instead of concent

‘Jd economics-or iented f ac to

*1 systems, this study is c

 

concentrated on the analysis of one or a few channel levels,
hithe present study entire channel systems were analyzed.
flms, the present study analyzes several behavioral rela-
tionships as they exist at different points within each
system.

Fourth, this study is unique in that it analyzes
competing channel systems which differ in vertical integra-
timn The inclusion of different types of competing channel
structures provides an unusual opportunity for behavioral
analysis.

Fifth, the analysis compared channel systems which
operate in different geographic regions. Differences in
Um behavior of northern versus southern channels may sug—
gest why the egg industry seems to be advancing more rapidly

hithe South than in the North.

LIMITATIONS

Several aspects of the methodology and research de-
Sign limit the generalizability of the conclusions derived
from the study. First, the areas of logistics and physical
distribution management are excluded from the present an—.
a1YSis. While much analytical work has been completed in-
flmse areas, a great deal more remains to be learned about
hOW physical distribution systems may be managed more effi-
ciently. Instead of concentrating on the impact of cost
and economics—oriented factors on the performance of chan—

le systems, this study is concerned with the analysis of

 

 

 

 

 

and attitudinal and behav
:influence the competitive
Second, respondents a
tied on the basis of a rep:
ssesult, not all persons at

..re of being included in t

 
 

Third, several interv
rough the cooperation of Ag
may have introduced some
:o‘sone latitude in selectin
itniewed. Each interviewe
nertain number of contract
Eisner-integrated producers
lgpersons who occupied the
3215, the interviewers may ha
fitticularly easy to intervie
3&1 others might have been.

Fourth, this analysi
i’dbehavior of persons enga
inthe southeastern, north c
Slites. While these regions
htional egg industry, it sh
liews were conducted in othe
Wehas the Southwest and th

Fifth, 176 persons w

“Y appear to be a relativel

 

 

several attitudinal and behavioral variables which appear
tojnfluence the competitive effectiveness of such systems.

Second, respondents at the producer level were se-
lected on the basis of a representative quota sample. As
aresult, not all persons at the producer level had an equal
chance of being included in the study.

Third, several interviews were obtained in the South
umough the cooperation of Agricultural Extension personnel
wmamay have introduced some interview bias because they
hmisome latitude in selecting the respondents which they
hmerviewed. Each interviewer was requested to interview
acertain number of contract producers and a certain number
drowner—integrated producers and to follow up by interview-
ing persons who occupied the other levels in the channels.
flms, the interviewers may have selected producers who were
gmrticularly easy to interview or who were more cooperative
flmn others might have been.

Fourth, this analysis concentrated on the attitudes
and behavior of persons engaged in the egg industry located
tithe southeastern, north central, and northeastern United
States. While these regions are extremely important in the
national egg industry, it should be recognized that no inter—
views were conducted in other major egg producing regions
such as the Southwest and the Northwest.

Fifth, 176 persons were interviewed. While this

may appear to be a relatively small sample, at some channel

 

 

 

 

 

mls—e.g., the bird suppl

hals~nost of the major fi
:dchigan were interviewed
Sixth, the data rele
nswere factor analyzed,
:separate a large set of h
siler groups, each of whic‘
tens of attitudes or beh
man's correlation, or fac
cue group to which he was
Sinus a descriptive proced
situation more simple and un
frential technique, and doe
listical significance under ‘
‘w‘currence of the results by
talysis was thus employed t

Which channels were most lik

ORGANIZATION OF THE

Chapter II is titled
Brketing literature is summ
Iritten about each of the V
my, Chapter III is titl
ӣ33. In this chapter t
”theses are described; the

s“linens is explained; the

mtlthe data analysis metho

 

T_________________________________________ii
lO

levels—-e.g., the bird supply, feed supply, and processing
levels—-most of the major firms performing these functions
hmMichigan were interviewed.

Sixth, the data relevant to general hypotheses 5
mm.6 were factor analyzed, using a mathematical procedure
to separate a large set of heterogeneous people into several
smaller groups, each of which was relatively homogeneous
filterms of attitudes or behavior. On the basis of each'
person's correlation, or factor loading, he was assigned
to the group to which he was most alike° Factor analysis
is thus a descriptive procedure used to render a complex
situation more simple and understandable° It is not an in—

ferential technique, and does not employ the notion of sta—

 

tistical significance under which the probability of the
occurrence of the results by chance is estimated. Factor
analysis was thus employed to determine which persons and l

which channels were most like one anothero

ORGANIZATION OF THE REMAINDER OF THE THESIS

Chapter II is titled Review of Literatureo Selected
marketing literature is summarized to show what has been
written about each of the variables included in the present
StUdy. Chapter III is titled Research Design and Analysis
of Data. In this chapter the general and statistical hy—
Potheses are described; the development of the research in-

struments is explained; the field procedure is described;

and the data analysis methods are discussedo Chapter IV

 

—

ititled Findings of the Stt
2:31 hypothesis testing proc

:‘sunary tables. Following
:ficsneaning. Chapter V i:
Izbriefly reviews the reseax
nedures, and discusses the
in the study. Chapter VI :2
rdSuggestions for Future Re
Following Chapter VI
L'pendix A, titled Glossar ,
tel technical terms employec

flied Research Instruments ,

 

Bats designed for each of tr

fixCis titled Item Analysis

If Variables. This appendix

15the questions used to inde

   
  
  
   
   
  
  
     

ift‘ne other questions used
hdix C also indicates the
lintitative measure on each
bundix D, titled Formulati
35$, explains the means by
aligned to the members of e
‘itled instructions for Inte
StNotions sent to the Agric

thinted interviews in ”155

 

“Arkansas. Appendix F, t

 

y

11

is titled Findings of the Studyo The results of the statis-
tical hypothesis testing procedure are presented in a series
of summary tables. Following each table is an explanation
ofits meaning. Chapter V is titled Summary and Conclusions.
szriefly reviews the research design and data analysis
pmmedures, and discusses the conclusions which resulted

fnmlthe study. Chapter VI is titled Managerial Implications

 

mm.5uggestions for Future Research.

 

Following Chapter VI several appendices are presented.
Appendix A, titled Glossar , presents definitions for sev-
eral technical terms employed in the study. Appendix B,
titled Research Instruments, contains the research instru—

ments designed for each of the five channel levels. Appen—

 

dix C is titled Item Analysis Results and Operationalization

of Variables. This appendix shows the degree to which each

 

of the questions used to index each variable is like each

of the other questions used to measure that variable. Ap-
pendix C also indicates the methods used to determine a
(mantitative measure on each person for each variable.
Appendix D, titled Formulation of the Vertical Integration
E§§§§, explains the means by which an integration score was
assigned to the members of each channel. Appendix E is
titled Instructions for Interviewers, and contains the in-
structions sent to the Agricultural Extension personnel who
conducted interviews in Mississippi, Georgia, North Carolina,

and Arkansas. Appendix F, titled Procedures Employed in

 

—

:1 yes Analysis, explains

ziteria for selecting a fac
is, and the procedures for

elyses which relates to ge
hpendjx G, titled Procedure
535M, explains the proc
frtor solution for the anal
phases 5 and 6. Appendi

Caeration of the 'Hultiple R

amine, explains the analys

:3 hypothesis 4.

   

1-——_,

' 12
the Eype Analysis, explains the factor analysis procedures,
criteria for selecting a factor solution for further analy-
sis, and the procedures for describing each type in the
analyses which relates to general hypotheses 5 and 6.

Appendix G, titled Procedure for Choosing Among Factor

 

Solutions, explains the procedure for choosing a particular
factor solution for the analyses which relate to general

hypotheses 5 and 6. Appendix H, titled Description of the

 

Operation of the 'Multiple Regression Delete‘ Computer
Routine, explains the analysis of the data which relate

to hypothesis 4.

 

 

 

 

 

C'

REVIEW 0

The purpose of this
aiming marketing literatur
tiables of this study. Th
in of the channel concept
as; it then moves to a co
filing with each behavioral
azeview of some studies of
:itural channels, vertical

flattitudes of egg channel

GENERAL CON

Introductory marketi
aIarketing mix as being com
“Product, price, promotion
I'll term 'channel of distrir
will these basic marketing
aMiller say that a chanm
3°“ of institutions throng

tociucts to the user or Ult

\______
lThomas A. Staudt 5

,gélntroduction to Fluke“cj
5), p. ao'we.

CHAPTER II

REVIEW OF LITERATURE

The purpose of this chapter is to summarize the
existing marketing literature which relates to each of the
variables of this study. The chapter begins with a discus—
flpn of the channel concept in both general and specific
farms; it then moves to a consideration of the literature
dealing with each behavioral variable, and it concludes with
areview of some studies of structural organization in agri—
mfltural channels, vertical integration in the egg industry,

and attitudes of egg channel members.

GENERAL CONCEPT OF CHANNEL

Introductory marketing textbooks frequently picture
a marketing mix as being composed of several elements, such
as product, price, promotion, and channels of distribution.
The term 'channel of distribution' is used in a general man—
ner in these basic marketing textbooks. For instance, Staudt
and Taylor say that a channel of distribution is "a combina—
tion of institutions through which a seller markets his

PrOducts to the user or ultimate consumer."l

“—-

1Thomas A. Staudt and Donald A° Taylor, A Manager-

ial Introduction to Marketin (Englewood Cliffs, New Jersey,
19655, p. 306.

13

 

 

 

 

 

McCarthy defines a c

hang institutions, from pr
finding any number of (or
antes that "every producer
iaarketing intermediaries
i'jectives. This set of mar‘
Marketing channel (also
:i‘mtion).“3 Zober also 5
Quad of intermediaries; he
inanimate, impersonal or
tii'v'iduals who compose the
::stitute the multiple Chan
hicto these four definiti
i'econposed of institutions

ziiivities which result in t!

iproducer to a final consum

THE CHANNEL CONCEPT
More specific defini
flStribution' have been give

511th. As Bucklin traces 1

\______

j 23. Jerome McCarthy
W (3rd Ed., Homewood
“1968), p. 312.

E-

, 3Philip Kotler, 5?},
”f5: New Jersey: Pren -
4 .
Martin Zober, 313.3

in
Us” and Sons, Inc., 196

 

l4

McCarthy defines a channel as “any sequence of mar—
keting institutions, from producer to final user or consumer,
hmluding any number of (or perhaps no) middlemen."2 Kotler
states that "every producer seeks to link together the set
ofnmrketing intermediaries that best fulfills the firm's
objectives. This set of marketing intermediaries is called
um marketing channel (also trade channel, channel of dis—
tribution).“3 Zober also stresses that channels are com—
posed of intermediaries; he says, "these intermediaries,

Hm inanimate, impersonal organizations and the chains of
individuals who compose the wholesale and retail trades,
constitute the multiple channels of distribution. . . ."4
Basic to these four definitions is the notion that channels
are composed of institutions or intermediaries which perform
activities which result in the movement of a product from

a producer to a final consumer.

THE CHANNEL CONCEPT DEFINED MORE SPECIFICALLY

More specific definitions for the term "channel of
distribution' have been given by Bucklin and McCammon and
Little. As Bucklin traces the origin and meaning of the

“a

23. Jerome McCarthy, Basic Marketing: A Managerial
Approach (3rd Ed., Homewood, Illinois: Richard D. Irwin,
Inc., 1968), p. 312.

3Philip Kotler, Marketing Management (Englewood
Cliffs, New Jersey: Prentice—Hall, Inc., 1967), p. 387°

' 4Martin Zober, Marketin Mana ement (New York: John
Wiley and Sons, Inc., 19645, p. 1‘09." "‘-

 

 

 

 

  

 

11 'channels of distributi

:mere too restrictive in
gsical flow of the good, (
ments, information, and r‘
men of title and thus f
nature of the institutio
Zwercome the limitations
:;gests the following defin‘
:zs'nall be considered to
an performs all the acciv
:eaproduct and its title
HcCammon and Little
1': 'channel' through time.
:channel definition are m
.. reseller groups classifi
1'?! conclude that descripti<
L’mld include: (1) the type
Eteach stage in the product
Wies, such as financing :

lhproduct; and (3) the fun

 

Metween all channel part

\_______-

5Louis P. Bucklin,
Cture (Institute of Busi
Ilium-e School of Business
‘thmia, Berkeley, 1966),

r.

f.

it 6Bert C. McCammon, J
“1th Channels: Analytical
mm (ed.), Science in
Sons, Inc., 1965 , pp.

   

 

15

tenn'channels of distribution,‘ he notes that early defini—
fibns were too restrictive because they: (1) omitted the
musical flow of the good, (2) excluded the flows of orders,
payments, information, and risk, (3) concentrated on the
nwvement of title and thus failed to satisfactorily explain
Um nature of the institutional linkages which developed.
fl>overcome the limitations of previous definitions, Bucklin
suggests the following definition: "a channel of distribu—
timishall be considered to comprise a set of institutions
much performs all the activities (functions) utilized to
move a product and its title from production to consumption."5

McCammon and Little also trace the meaning of the
tenn'channel' through time. They state that most approaches
u>channel definition are manufacturer—oriented and assume
Umt reseller groups classified by type are homogeneous.
'Hmy conclude that descriptions of the marketing channel
fimuld include: (1) the types of decision makers involved
at each stage in the product flow; (2) the facilitating
agencies, such as financing firms and firms which transport
Hm product; and (3) the functional relationships which oc~
uu'between all channel participants.6

“~—

5Louis P. Bucklin, A Theory of Distribution Channel
Structure (Institute of Business and Economic Research,
Graduate School of Business Administration, University of
California, Berkeley, 1966), p. 5.

. 6Bert C. McCammon, Jr., and Robert W. Little, "Mar—
keting Channels: Analytical Systems and Approaches," George

Schwartz (ed.), Science in Marketin (New York: John Wiley
and Sons, Inc., 1965), pp. 321-385.

 

 

 

 

 

THE CHANN

Such authors as Alde
'zayer, Ridgeuay, Stern, and
king channel is an operati
rdistinctive pattern of b
35'6 set of objects togethe
:eobjects and between thei

Alderson, for instan

The functional approach

with the study of organi
keting functions are dis
or by individuals acting
0f system of interest he
logical system because 0
the bond among the compo
integrated to permit the
but the bond is loose en
sent or addition of comp

Various writers have
1‘! said to characterize sys
finance, state that marketi

wilting systems, and have

" The channel consists of
Structured to produce pr
2- Members of the channel 5
Ceptexble objectives. . .

\M

r _7Ralph F. Breyer, "5
i:Flnation and Growth of Mark
n‘i‘fflderson, and Stanley
M (second series) (
N
1“) Ines, 1964), p0 1630
8

Wroe Alderson, Mar

2‘:—

tion

We. (Homewood, Illinois:

 

 

16

THE CHANNEL AS A SYSTEM
Such authors as Alderson, McCammon and Little, Fisk,
Breyer, Ridgeway, Stern, and others, have argued that a mar-
keting channel is an operating system with an identifiable
am.distinctive pattern of behavior. A system is defined

as"a set of objects together with the relationships between

Um objects and between their attributes."7

Alderson, for instance, notes that:

The functional approach to marketing theory begins
with the study of organized behavior systems. Mar—
keting functions are discharged by behavior systems

or by individuals acting within systems. The type

of system of interest here is classified as an eco—
logical system because of the peculiar nature of

the bond among the components. They are sufficiently
integrated to permit the system to operate as a whole,
but the bond is loose enough to allow for the replace-
ment or addition of components.

Various writers have listed sets of attributes which
are said to characterize systems. McCammon and Little, for
instance, state that marketing channels may be viewed as
Operating systems, and have the following characteristics:

1. The channel consists of interrelated components that are
structured to produce predetermined results. . . .

2. Members of the channel strive to achieve mutually ac—
ceptable objectives. . . .

“——

7Ralph F. Breyer, "Some Observations on 'Structural'
Formation and Growth of Marketing Channels," in Reavis Cox,
Wroe Alderson, and Stanley J. Shapiro (eds.), Theory in
Marketing (second series) (Homewood, Illinois: Richard D.
Irw1n, Inc., 1964), p. 163.

8Wroe Alderson, Marketing Behavior and Executive
ggtion (Homewood, Illinois: Richard D. Irwin, 1957), p.

 

 

 

 

  
 
   
 

  

3. Activities performed by
sequentially and thus it
activities as “marketing

:. Marketing channel is a
participation in it is v
freely, subject to insti
and withdraw with compar

f. Asingle enterprise usua
It initiates, coordinate
the activities undertake

5. The behavior of channel

established channel, is
ifies types of acceptabl
occupational code consis
group norms, and a subtl
is used in most channels
participants . 9

 
 
 
   
   
 

   
 

 

 
    

 

 
 
    
  

 
 

Stern and Brown, on

:mels as social systems w
(a) Structural Variables

l. Roles--a subset a
behavior for a pa
collectivity in t

(b) Functional Variables

1. Pattern maintena
to maintain its
ticipants and th

2. Goal attainment-
system needs thr
vironment. . . -

3- Adaptation-42he
sources for the
the allocation 0
ten components.

4 Integration-45119
components relat

effective furlc ti

\I.

t, 9th tt McCanWOn’
“‘l 329-330.

 

I 7 l7

3. Activities performed by channel members are undertaken
sequentially and thus it is logical to think of such
activities as “marketing flows". . . .

4. A marketing channel is an open system in the sense that
participation in it is voluntary. Firms enter a channel
freely, subject to institutional and market constraints,
and withdraw with comparable ease. . .

5. A single enterprise usually “administers" the channel.
It initiates, coordinates, and controls most or all of
the activities undertaken. . . .

6. The behavior of channel members, particularly in a well
established channel, is “regulated“ by a code that spec—
ifies types of acceptable competitive behavior. The
occupational code consists of informally established
group norms, and a subtle but clear array of sanctions
is used in most channels to control the behavior of
participants.

Stern and Brown, on the other hand, choose to describe
channels as social systems which are characterized by:

(a) Structural Variables, such as:

 

l. Roles--a subset of norms that define expected
behavior for a particular unit——a person or a
collectivity in the social system. . . . and

 

(b) Functional Variables, such as:

 

 

1. Pattern maintenance—-the tendency of the system
to maintain its structure both in terms of par—
ticipants and their behavioral prescriptions. . . .

2. Goal attainment-—activity directed to satisfy
system needs through interaction with task en—
vironment. . . . ,

3. Adaptation-—the activity needed to provide re—
sources for the attainment of system goals and
the allocation of resources among different sys—
tem components. . .

4. Integration--the mutual adjustment of the system
components relative to their contribution to the
effective functioning of the system as a whole.10

W

. 9Bert C. McCammon, Jr., and Robert W. Little, 92.
E$_., pp. 329—330.

10Louis W. Stern and Jay W. Brown, "Distribution Chan-
nels: A Social Systems Approach, " in Louis W. Stern (ed. ),
Distribution Channels: Behavioral Dimensions (New York:
Houghton Mifflin Company, 1969), p. 16.

 

 

  

Among those behavior

  
 

fussed by Miller are the

  
 

rally relevant to the ope

  
 

1. Cohesiveness~the te
sufficient physical
components. . .to enab
forces that would di

 
 
 
 

    
 
 

  

L"

Feedback-~(when feed
channels which carry
8 loops back from th
nel A and transmits :
emitted by Channel A.

. Integration versus Se
integrated). ..they a:
decider of the systet
If they are segregatt
ers each controlling

Conflict--conflict a1
hitters send incompa‘

  
  

u..-

.r-

u-

Relative Openness ve:

(a) Closed system--a
aries through uh,
nation transmiss
No real or concr
so concrete syst
open or relative

(b) Open system-~a S
are at least par
sizeable magnitu
of matter energy
to cross them

when production-4ndrk
pm which have the P

W attributes, the a“

lie .
interlotions and Ward

MWMthS

18

Among those behavioral attributes of real systems
(fiscussed by Miller are the following which seem to be es—
pmflally relevant to the operation of marketing channels:

1. Cohesiveness—-the tendency of systems to maintain
sufficient physical contact among subsystems and
components...to enable them to interact resisting
forces that would disrupt such relationships.

2. Feedback—-(when feedback exists)...there are two
channels which carry information such that Channel
B loops back from the output to the input of Chan—
nel A and transmits some portion of the signals
emitted by Channel A.

3. Integration versus Segregation--(If systems are
integrated)...they are centralized, the single
decider of the system exercises primary control.
If they are segregated, there are multiple decid—
ers each controlling a subsystem or component.

4. Conflict-—conflict arises when two or more trans-
mitters send incompatible commands to a receiver.
5. Relative Openness versus Relative Closeness—-

(a) Closed system——a system with impermeable bound—
aries through which no matter——energy or infor-
mation transmissions of any sort can occur.

No real or concrete system is completely closed,
so concrete systems are therefore relatively
open or relatively closed. ‘ .

(b) Open system-~a system with boundaries which
are at least partially permeable permitting
sizeable magnitudes of at least certain sorts
of matter energy or information transmiSSions

to cross them.

When production—marketing channels are conceived of
as Systems which have the previously mentioned characteris-
tics and attributes, the analyst may explicitly consider
Um interactions and interdependencies of the elements of
Hm sYstem, and may be less naive about the complex nature

M

11James G. Miller, "Living Systems;

BShavioral Science, Vol. 10, NOo 3: JUlY9 19659 pp.

Basic Concepts,"
193—237.

 

 

 

  

:‘cause and effect which fol

  
 

film of the system is dist

  
 

.. r’uingchamels as systems

  
 
 

 
 

1. This approach recogni
is a purposive and re
rather than a random

     

.' "

   

The systems concept
cooperative, as well
within the channel.

    
  

w

. The channel is percei
ganism that reflects
tions of its particip

h

. The marketing channel
is recognized as a be
a concept that broade

valry. . . .

The notion that a che
provides a basis for
are system generated.

on

BEHAVIORAL ELEMENTS

As systems, channels
fitted to one another by th
1:.er by certain element
38 studied such intrachann

h, Conflict, cooperation:

Whom from the liter

Mr'lobles is discussed

it .
\hl 1‘1 the Channel

Although several Per
who

. .. . -.mS

l9

dfcause and effect which follows when the state of one
element of the system is disturbed. Other advantages of
viewing channels as systems are:

1. This approach recognizes the fact that a channel
is a purposive and rational assemblage of firms
rather than a random collection of enterprises.

2. The systems concept emphasizes the existence of
cooperative, as well as antagonistic, behavior
within the channel.

3. The channel is perceived as a unique social or—
ganism that reflects the hopes, goals, and aspira—
tions of its participants.

4. The marketing channel, from a systems point of View,
is recognized as a basic "unit of competition"——
a concept that broadens the study of economic ri—
valry. . . .

5. The notion that a channel is an operating system
provides a basis for identifying dysfunctions that
are system generated. . .
BEHAVIORAL ELEMENTS IN CHANNEL SYSTEMS
As systems, channels contain individuals who are
related to one another by the performance of channel activ—
ififls and by certain elements of behavior. Various writers
kwye studied such intrachannel behavioral variables as con-
trol, Conflict, cooperation, and innovativeness. Selected

cmuributions from the literature which deals with each of

Umse variables is discussed below.

Control in the Channel
Although several persons have discussed control in

CIlannels, much more remains to be done in the way of

________________
12McCammon and Little, op. cit., Po 330°

 

 

  

 

     
  
  
  
  
    
  
  

 

:a'erstanding: (1) who cont
sated; (3) how it may be m
7:_se, contributes to charm
:‘persons who have written
srevieved below.
The locus of channel
:Ze of economic power in de
fizzle concludes that in 0rd
tithe efficiency of channe
secentral direction.l3
Hallen, another writ‘

a. and control in the chem

The usual pattern in the
relationships is that th
ator who puts the struct
and who holds it togethe
whether through command
through an autocratic or

Hallen discusses the

 
  
  
   
  
     
  
  
  

39! time by manufacturers,
‘lscusses the present-day s

hinted and retailer-domi

M

”Robert w. Little,
:3 Channels.“ A mimeograp
ss Seminar sponsored by t
h Force on Marketing Met
‘an, Illinois.

14

W Bruce Mallen, “Co
Messémelsg" in L. George
‘3 lgpMagketin (Chicago
1 . . .
em h5 85. Reprmte

   

20

Lmderstanding: (1) who controls whom; (2) how control is
amrted; (3) how it may be measured; and (4) whether control,
parse, contributes to channel efficiency. The contributions
ofpmrsons who have written in one or more of these areas

is reviewed below.

The locus of channel control. After reviewing the
role of economic power in determining channel leadership,
little concludes that in order to enhance the productivity
mm the efficiency of channel members, there is a need for
some central direction.l3

Mallen, another writer who reviews the role of leader-
flup and control in the channel, states that:

The usual pattern in the establishment of channel
relationships is that there is a leader, an initi—

ator who puts the structure into this relationship

and who holds it together. This leader controls,

whether through command or cooperation, i.e.,
through an autocratic or democratic system.l4

Mallen discusses the struggle for control in channels
over time by manufacturers, wholesalers, and retailers, and
discusses the present-day struggle between manufacturer—
dominated and retailer—dominated channelSo
~————__________

13Robert W. Little, "Power and Leadership in Market—
ing Channels." A mimeographed paper delivered at the Chan—
nels Seminar sponsored by the American Marketing Assoggation
Task Force on Marketing Methodology, November 6—8, 19 ,
EVanston, Illinois.

Bl: " . alld cooperation in. I'Iarket—
uce Mallen, COnfllCt ‘

. . ) . f‘j
18," in L. George Sml tll (ed0 9 ItEfle'CthIls 311 EIO

less in Marketin (Chicago: American M

1964 , pp. 65-85. Reprinted in Bruce E, Mallen éediz, ggfih
fl¥3§ting Channel: A Conceptual Vieprint (New or .

Wiley and Sons, Inc., 1967), pp. 124-134.

 

 

    
  
  
  
  
  
    
 
 

 

islet, channel control may b
ads, coercion, power gaine
Zegitinacy, and power throug
iscuss efforts of channel
epover in the channel is
rver is shared equally amo
The measurement of
on the general discussion
21th channel control. n

Him in one level of t
control another firm in
makes, participates in,
of that firm. Simple me
each channel does not ir
relationship.

lhe extent to which one
determined by the propor
decisions which the forn
of influence on these de
influence obviously can
of probability that cho
Of this choice, or an a
mented by the other, an
change in this choice w

   
   
  
 
 
 
 
   
   

In'general, elements of
said to exist if the pr
batween two firms in di
nel involves some enfor
Present and/or future 111

\__

lathanISLouis W. Stern an
id.) Dnel of Distribution
in" istrlbution Channel
' Houghton Mifflin Comp

     

21

Methods of channel control. According to Stern and
Beier, channel control may be gained through the use of re—
wards, coercion, power gained by expertness, power through
legitimacy, and power through identification. These authors
discuss efforts of channel members to balance power when:

(a) power in the channel is asymmetrically shared, (b) when

power is shared equally among channel members.15

The measurement of channel control. Baligh improves

upon the general discussions of other writers by operation—
alizing channel control. He says:

A firm in one level of the channel . . . is said to
control another firm in another level, . . . if it
makes, participates in, or influences some decisions
of that firm. Simple membership of two firms in
each channel does not involve, as such, any control
relationship.

The extent to which one firm controls another is
determined by the proportion of the latter firm's
decisions which the former makes, and by the amount
Of influence on these decisions which it has. Such
influence obviously can be measured along the scale
Of probability that choice, and the communication
of this choice, or an alternative by one is imple—
mented by the other, and the probability that a
change in this choice will also be implemented...

In general, elements of managerial control can be
said to exist if the product exchange transaction
between two firms in different levels of the chan—
nel involves some enforceable limitations on the
present and/or future marketing behaVior of either

\——_

15Louis W. Stern and Frederick J. Beier, "Power in
the Channel of Distribution," Chapter 6 in Louis w° Stgrn
(ed.), Distribution Channels: Behaviora;_2£E92§$2E§ os-
ton: Houghton Mifflin Company. 1959M PP° 92‘116“

 

 

 

 

or both of these firms.

fligh's suggested means of

hemloyed in this thesis,
The concept of cnann
:ylouis Bucklin, who adopts
motes the difficulties e
in to measure control. Bu
Instead of attempting to
control directly, one al
channel authority center

various symptoms of con
be regarded as the inci‘

brands within the chann

measures of advice seeki
Embers. O O 01

Control and channe l

 

ithe notion that some firn
lunative for organizing ar

L“ states that:

  
 
    
   
     
      
   
 
 
  

- - . a more cautious a
the conclusion that aut
within the channel will
ence. For control to b
it must be shown that c
Channel activities will

\‘—‘-—-————

l
6Helmy H. Baligh,

ihlflnel Choice, “ in P. 0-
”lat tition and World Mark
”‘9 missociation,
alien, Th
“'1‘ John Wiley and Sons,
17LouisP. B kl
. uc in
EighertLF King (ed. ), M,
ark F,all 1968. Con
lion at eting Association
ion, 1968), p. 143.

22

or both of these firmso . . 016

BaUgh's suggested means of Operationalization of control
is employed in this thesis, as explained in Chapter 3.

The concept of channel control was recently reviewed
by Louis Bucklin, who adopts Baligh's definition of control
and notes the difficulties encountered by researchers who
wish to measure controlo Bucklin concludes:

Instead of attempting to measure the presence of
control directly, one alternative for locating
channel authority centers would be to search for
various symptoms of controlo Such symptoms might

be regarded as the incidence of producer or reseller
brands within the channel, and/or the sociometric
measures of advice seeking and giving among channel
members. . . .

Control and channel efficiencya Bucklin is critical
of the notion that some firm in the channel should take the
initiative for organizing an efficient channel. Instead,
he states that:

. . . a more cautious attitude is warranted toward
the conclusion that authoritarian relationships
within the channel will invariably improve perform—
ance. For control to better channel cooperation,

it must be shown that close coordination of all
channel activities will enhance profit opportunities

l6Helmy H. Baligh, "A Theoretical Framework for
Channel Choice,“ in P. Do Bennet (edo), Economic Growth,
Sflmgtition, and World Markets (Chicago: American Market—
lng Association, 1965), ppo 631—654o Reprinted in Bruce E.

Mallen, The Marketin Channel; A Conce tual View oint (New
York: John Wiley and Sons, Inca, 1967), p0 205°

17Louis P: Bucklin, "The Locus of Channel Control,"
in Robert L. King (ed.), Marketing and the New Science_g§
Plannin . Fall, 1968. Conference proceedings Of the Amer—
Efifirfiggketing Association (Chicago: American Marketing
Association, 1968), p0 143°

 

 

 

‘

by either expanding cons
channel costs, that some
'thenecessary level of c
resulting system ri idit
otherwise secured.
The literature which
raprimitive stage of soph
easexist to operationaliz

ifnriters have not univers

geater channel efficiency w

inflict in the Channel
Several writers have
:gof (l) the causes of con

:ccedures by which channel

Eire channel conflict. Th

Causes of channel cc

 

i'ee forms of channel confl

1. Horizontal competiti
tween middlemen of t
discount store versr

Ix:
.

Intertype competitic
tween middlemen of c
channel sector; for
versus department 51

Vertical conflict-~1
channel members of (
discount store versr

 

3...;

\~_____'

181mm, p. 145 .

‘ ' 19Joseph C. Palamou
.gibutlon (Cambridge, Massa
€53,1955), pp. 24-58.

 

23

by either expanding consumer demand or reducing
channel costs, that some form of control can achieve
the necessary level of coordination, and that the
resulting system rigidities do not offset the gains
otherwise secured.l

The literature which treats channel control remains
ata.primitive stage of sophistication. Relatively crude
nmans exist to operationalize control in a given channel
and writers have not universally accepted the notion that

greater channel efficiency will result from greater control.

Conflict in Ehe Change;

Several writers have contributed to our understand-
hg of (l) the causes of conflict in channels; and (2) the
procedures by which channel participants have attempted to

reduce channel conflict. These topics are discussed below.

Causes of channel conflict. Palamountain specified
Hmee forms of channel conflict relationships:

1. Horizontal competition-~this is competition be—
tween middlemen of the same type; for example,
discount store versus discount store.

2. Intertype competition-—this is competition be—
tween middlemen of different types in the same
channel sector; for example, discount store
versus department store.

3. Vertical conflict—~this is conflict between
channel members of different levelsi for example,
discount store versus manufacturer. 9

“—

18Ibid., p° 145.

. ngoseph C. Palamountain, Jr., The Politics of Dis-
gribution (Cambridge, Massachusetts: Harvard University
ress, 1955), pp. 24-58.

 

 

 

 

Palanountain's work
field of channel conflict.
sit occurs in the channels
nities. For instance, Pal
:‘zoccurred in the distribut
‘nact and growth of chains,
hypouer, the uses and aims
tithe reactions of differe
athis bargaining power. P
111'autonobile channels, for
Iodto legislative efforts 1

Mallen isolated two
Pitt: price and non-pr ice
ntof the exchange act; it
Mparties in the exchange
RSpect to price-—one wishe:
hisible while the other wi
Sible- Non-price conflict
ness in business philosoph
“11 and the tendency for b
towant to eliminate anothe
is wholesalers .

Another major contr
nil conflict has been made

hive conflict to be a pro

20

Bruce Mal len ,

 

l 24
Palamountain's work is a pioneering effort in the
field of channel conflict. It is an analysis of conflict
as it occurs in the channels of distribution of several com—
modities. For instance, Palamountain studied conflict as
it occurred in the distribution of groceries and noted the
impact and growth of chains, the measures of chain bargain-
ing power, the uses and abuses of chain bargaining power,
and the reactions of different parties in grocery channels
tO‘Ufij bargaining power. He also studied pharmaceutical
and automobile channels, focusing on the conflicts which
led to legislative efforts to remove the sources of conflict.
Mallen isolated two general sources of channel con—

flict: price and non-price factors.20 Price conflict arises

 

out of the exchange act; it is the result of the fact that
two parties in the exchange act have opposite goals with
respect to price--one wishes to lower the price as much as
possible while the other wishes to raise it as much as pos-
sible. Non—price conflict arises out of the basic differ-
ences in business philosophy which may exist within a chan-
nel, and the tendency for both manufacturers and retailers
to want to eliminate another middleman in the channel, such
as wholesalers.

Another major contribution to the literature of Chan-
{ml conflict has been made by Stern and German. They per—

ceive conflict to be a process of two broad classes of

M
20

 

Bruce Mallen, o . cit., pp. 125-126.

 

 

 

 

mes: (l) a change that

nonbip and (2) a change in

fitter leads to resolution 0

Stern and Goman ana

arses of conflict; (2) the

for conflict; and (3) the p
mess.

According to Stern a

hthe action of one or more

1. Role Deviance. Role

to rights or obligat
as one or more of tr

M

Issues. An issue is
which is desired by
when two parties wis

4....-

Incongruence of char
tations, decisions,

Stun and German note while
“Ch can result from confli
hconflict may make an ef
mother components of the
“be! may alter himself so

an be accommodated .

Conflict resolution

 

 

““60: more of several poss

\H

. 21Louis W. Stern an
E" Distribution Channels :
WSW Stern (ed.), Distr

”filigéons (Boston: Houqht

 

221mm, pp . 15 7-16

 

l

25

changes: (1) a change that precipitates the conflict rela-

fibnship and (2) a change in response to conflict which
either leads to resolution or to system disintegration.21

Stern and Gorman analyze three conflict areas: (1)

causes of conflict; (2) the types of behavior which result

from conflict; and (3) the possible outcomes of the conflict

process.

According to Stern and Gorman, conflict is caused

by the action of one or more of the following factors:
1. Role Deviance. Role deviance occurs with respect
to rights or obligations which are not dealt with
as one or more of the channel members had expected.
2. Issues. An issue is a scarcity of the same resource
which is desired by two parties. Conflict arises
when two parties wish the same resource.

3. Incongruence of channel member perceptions, ex

I
tations, decisions, goals, or communications.zgec_ [
Stern and Gorman note while discussing the types of behavior
when can result from conflict, that either one party to
He conflict may make an effort to change the behavior of
the other components of the channel system, or the channel

member may alter himself so that the frustrating behavior

can be accommodated.

Conflict resolution. Members of channels may take

one or more of several possible steps to resolve existing

M—

_ 21Louis W. Stern and Ronald H. Gorman, "Conflict
in Distribution Channels: An Exploration," Chapter 10 in
Louis W. Stern (ed.), Distribution Channels: Behavioral
Dimensions (Boston: Houghton Mifflin Company, 1 6 , pp,
156-175.
221bid., pp. 157—161.

 

 

 

conflict. They may engage J
mixing in a third party i
{may attempt to limit tr
hrbehavior or by coming tc
hglhich will serve to curl
aples of such activities a:
culpetition, the employment
55 exclusive territories for
mlude that the outcomes c
fication of the parties to 1
Me relationships which I
This publication by
5ilbstantial contribution to
iftelpt to provide a specifj
icnflict, incorporates the '1
ialists, some of whom have \
their hypotheses .
Another noteworthy
“Channel conflict comes f

analSIZed the roles assumed

andmanufacturer trade asso

 

the conflicts by political

M
23%., pp. 170-17
2["Hemcy Assael, "Th

ciations in Distributive Co
“3!. keting (Chicago: Americ

 

32April, 1968), pp. 21-28

 

F——7

26

conflict. They may engage in a bargaining process, they
may bring in a third party to settle their differences, or
'Umy may attempt to limit the conflict by instituting rules
for behavior or by coming to some sort of prior understand—
hw which will serve to curb the extent of conflict. Ex—
amples of such activities are: collusion to limit price
competition, the employment of franchises, or the awarding
of exclusive territories for distributors. Stern and Gorman
conclude that the outcomes of conflict will be either a uni—
fication of the parties to the conflict, or a disintegration
of the relationships which bind the parties to one another.23
This publication by Stern and Gorman represents a

substantial contribution to the channels literature. Their

 

attempt to provide a specific theoretical understanding of
conflict, incorporates the work of many social science spec—
ialists, some of whom have used empirical data to support
their hypotheses.

Another noteworthy contribution to the literature
of channel conflict comes from Professor Henry Assael, who
analyzed the roles assumed by various retailer, wholesaler,
and manufacturer trade associations in resolving distribu—

tive conflicts by political means.24 The investigation was

“——

23Ibid., pp. 170—172.

_ 24Henry Assael, "The Political Role of Trade Asso—
Ciations in Distributive Conflict Resolution,“ Journal of
Marketing (Chicago: American Marketing Associati$fiT_V3TT
32, April, 1968), pp. 21—28.

 

 

 

Elited to the role of natio;
:s‘cributive issues of natio,
sfederal level. Assael e
:nlict resolution: polici
:cced to power sources ou

self-resolution, trade assoc

.—
:'e:'nannel syscem. . . .43

Assael stresses that
:ibutive conflict is the be
this in the channel syste

-occn conflict and COOPGIa

:.: 7.11 order to survive; ‘now

:5 differ

mg economic goals
Busting literature
«name processes used to

{2m '
lb9h conflict has been 5

1'51 r
poducts, few persons h

won or measurement. 26 T“
We N
asure conflict, but to
:hfli '
Ct ls related to the c

inc
hannel structures

25

I e
@‘v p. 21.
p . In

J, the
Seem of a present stL

so
59% 2(1): containec

examples

 

27

limited to the role of national associations in resolving
distributive issues of national importance, primarily at
Um federal level. Assael examined two distinct modes of
conflict resolution: political resolution, those activities
directed to power sources outside the channel system; and
self—resolution, trade association activities internal to
Um channel system. . . .25

Assael stresses that the underlying source of dis-
Ufibutive conflict is the basic interdependence between com-
ponents in the channel systems. This interdependence results
hlboth conflict and cooperation; units in a system cooper—
ate in order to survive; however, conflict results because
of differing economic goals and ideological motives.

Existing literature has tended to focus on the causes
of and the processes used to resolve conflict in channels.
Although conflict has been studied in the channels for sev—
eral products, few persons have attempted its Operationali-
zation or measurement.26 The present study seeks not only
to measure conflict, but to determine the extent to which
conflict is related to the degree of vertical integration
hlchannel structures.
‘~‘——————

25313., p. 21.

261n the present study, conflict is indexed with

Um aid of a scale contained in each research instrument.
36% Appendix B for examples of the various conflict scales°

l
,
u
‘1

 

 

 

mention in Marketing C‘na

Vertical cooperatior
aristence of channel systems
necessary for a firm to c
ims‘nip with other firms ir
:meel. Baligh and Richart

he set of firms . . . C
economic syscem--a verti
Here membership for a f:
mined by the existence c
lgchac this firm has
ELIE in the sec although
mayor may not exist. I
is that which permits t:
ices among firms within
;0 collusion, and hence
between firms in differe

the sec or vertical marl
Thus, in the analys:
Inducted by Baligh and Rick
tie cement which binds marke
ininiumm amount of coopera?
Alderson agrees wit}

COODeration is as
preva

Egolglpetition. Intern
ing ch:havlor System is‘
0p‘-‘rétti:nej:s CannOt fum
fmm h" n which each
POsite membe;

.l

28

Eggggation in Marketing Channels

Vertical cooperation is crucial to the continued
existence of channel systems. Some writers feel that it
isrmcessary for a firm to demonstrate a cooperative rela-
tionship with other firms in order to be included in the
channel. Baligh and Richartz, for example, suggest that:

The set of firms . . . could also be viewed as an
economic system——a vertical structure, in fact.
Here membership for a firm in the set is deter—
mined by the existence of a cooperative relation—
ship that this firm has with at least one other
firm in the set although competitive relationships
may or may not exist. A cooperative relationship
is that which permits the exchange of goods or serv-
ices among firms within the set. It does not refer
to collusion, and hence is limited to relationships
between firms in different levels or sub—sets in
the set or vertical market structure. . . .

 

 

Thus, in the analysis of vertical market structures
conducted by Baligh and Richartz, vertical cooperation is l
the cement which binds marketing systems together; without
a minimum amount of cooperation there can be no system.

Alderson agrees with this philosophy; he states:

Cooperation is as prevalent in economic activity

as competition. Internal cooperation is required

if a behavior system is to act as a unit. Market—
ing channels cannot function without sustained co—
operation in which each party knows what to expect
from his opposite member. Economic processes in—
volving marketing require cooperation and close co—
ordination of marketing with other business functions
such as production and finance. . . .28

M

27Helmy H. Baligh and Leon E. Richartz, Vertical
Market Structures (Boston: Allyn and Bacon, Inc., 1967),

9 4o
28Wroe Alderson, Dynamic Marketing Behavior (Home-
WOOd, Illinois: Richard D. Irwin Co., 1965), p. 239.

 

 

 

For Alderson, a ch

, segree in order to effectiv

; systems. He states:
. . . The effectiveness
petition will depend no
but on his ability to i
to cooperate with him.
its tire retailers and
team which competes Hit
Firestone in selling ti
In any case, the dynami
be substantial if Fires
up to Goodyear in the p
tion from others in the
after another should fa
firm might achieve domi
itors might be helpless
been unable to maintain
tion. . . .

Alderson provides a
dish parallels the industr
Me refers to the vertica
T'e broiler industry. He
fixed feeds has organized
ideal financing and other
50”he producers of broil

Bruce Mallen, in s
barS cooperate with one an
fional Extension Concept. '

°" c°°Peratimg rather than

leabets has led to the con

W
291mm, pp. 255-2

 

 

29

For Alderson, a channel must cooperate to a high
degree in order to effectively compete with rival channel

systems. He states:

. . . The effectiveness of each manufacturer in com—
petition will depend not only on his own efficiency
but on his ability to induce wholesalers and retailers
to cooperate with him. Thus, Goodyear, together with
its tire retailers and distributors, constitutes a
team which competes with a corresponding team from
Firestone in selling tires to consumers....

In any case, the dynamic effect on the market would
be substantial if Firestone were not able to stand
up to Goodyear in the process of inducing coopera-
tion from others in the channel. If one competitor
after another should fail to measure up, a single
firm might achieve dominance very rapidly. Compet-
itors might be helpless to retaliate if they had
been unable to maintain effective channel coopera—

tion. . . .

Alderson provides an example of channel cooperation

 

much parallels the industry included in the present study
as he refers to the vertical cooperation which is found in

Um broiler industry. He notes that the manufacturer of

mixed feeds has organized the entire channel and has pro—

vided financing and other cooperative support where needed

for the producers of broilers.

Bruce Mallen, in seeking to explain why channel mem-

bers cooperate with one another, has suggested an 'Organiza—

tional Extension Concept.‘ He says: ". . . this emphasis

mlcooperating rather than conflicting objectives of channel

members has led to the concept of the channel as simply an

R
29Ibid., pp. 255—256.

 

 

 

atension of one' s own inte

amples of the application
net suppliers will often
musing or that they wi
;otential or forecast sales
Marketing writers h
.n that cooperation is a
fistea survival. Future re
mining the nature of the
ue‘ channel structure. Thi
”isdirection by exploring

izzegration and cooperation

Channel member innc
fine how competitive a syst
3f the numerous articles
iaebest known is that wri
ing the concepts which rel
fasion of innovations, say

There are four essenti

0f the diffusion of an
and (2) its communicat

\____.
30Mallen, op . cit .

31As explained in

itionalized and measured f
“specially developed 5

 

30

. . . . .30
extenSion of one's own internal organization. . . .“ As

examples of the application of this concept, Mallen mentions
that suppliers will often provide resellers with help in
advertising or that they will help resellers study market

potential or forecast sales volume.

Marketing writers have seldom gone beyond the state—
ment that cooperation is a necessary condition for channel

system survival. Future research should be directed at de—

termining the nature of the relationship between cooperation

and channel structure. This thesis lays some groundwork in

this direction by exploring the relation between vertical
31

integration and cooperation.
Innovativeness in Channel §ystems

Channel member innovativeness may partially deter—
mine how competitive a system is in relation to its rivals.
0f the numerous articles and books on innovativeness, perhaps
Um best known is that written by Rogers, who, while review—

ing the concepts which relate to innovativeness and the dif—

fusion of innovations, says:

There are four essential elements in any analysis
of the diffusion of an idea: (1) the innovation,
and (2) its communication from one individual to

\—
3oMallen, op. cit., pp. 131—132.

. 31As explained in Chapter III, cooperation is oper_
ationalized and measured for each respondent through the use
Of a sPecially developed scale.

 

 

 

 

another, (3) in a soci
W is an idea
vidual. W is
tion spreads. The diff
of a new idea from its
tion to its ultimate us
system is a population
:ionally differentiate
problem-solving behavi
to continue full use 0
process is the mental p
vidual passes from firs
to final adoption. m
which an individual is
ing new ideas than the
system. Adopter categc
of individuals within a
of innovativeness . 32

While most of the f
{IRoqers stemmed from rese
nieducation, much of this
119:0 explain behavior in
i'leculture, norms, and tra
:ion of channel members hav
Vetiveness within the Chan
iiY increase the effective
“pinion leader or Chang
“Option of certain practi

King reviewed the
themarketing field, and n

Asubstantial volume 0
Conducted in marketing
Research has focused o

Practices, television,
tars, birth control pr

\.~_______

32Everett M. Roger
“Pk: The Free Press: 196

 

 

31

another, (3) in a social system, (4) over time. An
innovation is an idea perceived as new by the indi—
vidual. Diffusion is the process by which an innova—
tion spreads. The diffusion process is the spread
of a new idea from its source of invention or crea-
tion to its ultimate users or adopters. A social
system is a population of individuals who are func—
tionally differentiated and engaged in collective
problem-solving behavior. Adoption is a decision

to continue full use of an innovation. The adoption
process is the mental process through which an indi-
vidual passes from first hearing about an innovation
to final adoption. Innovativeness is the degree to
which an individual is relatively earlier in adopt-
ing new ideas than the other members of his social
system. Adopter categories are the classifications
of individuals within a social system on the basis
of innovativeness.32

While most of the findings and conclusions discussed

by Rogers stemmed from research in sociology, anthropology,

and education, much of this information is useful in help-

ing to explain behavior in channel systems. For instance,

the culture, norms, and traditional versus modern orienta-

tion of channel members have a significant impact on inno-

vativeness within the channel. Further, a channel captain

may increase the effectiveness of his channel by acting as

Mlopinion leader or change agent to influence the rate of

adoPtion of certain practices.

King reviewed the diffusion research conducted in

Hm marketing field, and noted that:

A substantial volume of diffusion research has been
conducted in marketing and marketing-related contexts.
Research has focused on the adoption of homemaking
practices, television, durable goods, fall-out shel—
ters, birth control practices, health insurance, new

\-_
32Everett M. Rogers, Diffusion of Innovations (New
YOrk: The Free Press, 1962), pp. 19—20.

 

 

 

 

synthetic fabrics, publ
variety of other market
has concentrated on the
system and the effectiv
strategies in promoting
and technology. Resear
at identifying the inn' 0
innovation within the f
Mast few topics of diff
ire of most crucial relevan
:ionby the members of a di
Several sources of
:elsvere discussed by McCa
iachannel captain or chan
idlinistrator often is an i
:21 product is being market
net channel captains are 11

because they are subject to

Straints and may consequent

   
  
   
   
   

tich would involve a res
51119 with its customers.

Second, and more i

M

. 33Charles W. Kinf,
l“marketing: An Overview,
a?“ TEChnolo and Mark
FQ'ANerican Marketing Ass
“ting Association, 1966) ,

34'Bert C. McCammon
5f Institutional Change an
“feyser (ed.), Toward Scie
thewinter Conference of t
“ember, 1963. (Chicago:
it 477-490. Reprinted in
ktin Channel: A Conce t
iHEY and Sons, 1967), P-

   
 
    
  
 
    
    

  

 

32

synthetic fabrics, public health measures, and a
variety of other marketing-related ideas, Research
has concentrated on the role of the distribution
system and the effectiveness of various marketing
strategies in promoting adoption of new products
and technology. Research has also been directed

at identifying the innovative firm and explaining
innovation within the firm.

Mm last few topics of diffusion research mentioned by King
are of most crucial relevance to the understanding of adop—

tion by the members of a distribution system.
Several sources of innovativeness in marketing chan—

nels were discussed by McCammon, The first of these may be

Um channel captain or channel administrator; "the channel

administrator often is an innovator, particularly when a

new product is being marketed."34 McCammon notes, however,

amt channel captains are not usually highly innovative,
because they are subject to procedural organizational con—
straints and may consequently tend to resist an innovation

much would involve a restructuring of the firm's relation-

ship with its customers.

Second, and more important in explaining innovation

_________________
33Charles W. Kinf, "Adoption and Diffusion Research

An Overview," in Raymond Lo Hass (ed.), Sci—
1966 Fall Proceedings of
American Mar—

in Marketing:
Eflfih_gechnology, and Marketinga
Hm American Marketing Association (Chicago;
keting Association, 1966), p0 6720

34Bert C. McCammon, Jro, "Alternative Explanations
0f Institutional Change and Channel Evolution," in Stephen A.
GreYser (ed.), Toward Scientific Marketing, Proceedings of
the Winter Conference of the American Marketing Association,
December, 1963. (Chicago: American Marketing Association),
Ppo 477-490° Reprinted in Bruce Eo Mallen (edo), The Mar—
£§Eflg_§hannel: A Conceptual Viewpoint (New York: John
wlley and Sons, 1967), p. 80°

 

 

 

 

within the channel, accord'

Lyafirn outside of the e
fires J. C. Penney, Richar
spies within the retailin
:ntented outsiders who be
atechnique of irresistable
Innovativeness as 1
:channel systems is analy
fies of the most recent :33

Luovation36 reveals that n

 

: analyze this relationshi

VERTICAL INTEGR)
The increasing tren
relating systems37 has bee

M

35Ibid., p. so .

   
   
  
   
 
    
  
  
     
   
 
    

_ 36Everett M. Roger
fhmovations. Diffusion
:3: 6 (East Lansing, Michi
image of Communication
my, 1967).

. 37For purposes of
1‘?“ 0f vertical integrati
fllsts when one channel me
3“ Operations in producti
~fltnnel levels) will suffi
David A. Revzan, "
Mel," Wholesalin
9h wiley and Sons, 1961
fUceE. Mallen (ed.),
M (New York:
, H wever, as vertic
“the Present study (see
W’ risk, commitment, an
’5} basis of determining
“Ch Characterizes each c

33

within the channel, according to McCammon, is the role played
by a firm outside of the established channel structure. He
cites J° C. Penney, Richard Sears, and King Cullen as ex-
amples within the retailing field of persons who were "dis—
contented outsiders who believed that they had discovered
a technique of irresistable power,"35

Innovativeness as it relates to vertical integration
hichannel systems is analyzed in the present studya A re—
view of the most recent Bibliography on the Diffusion of

Innovation36 reveals that no studies have been undertaken

to analyze this relationship.

VERTICAL INTEGRATION IN CHANNEL SYSTEMS
The increasing trend toward vertically integrated

marketing systems37 has been frequently discussed in the

________________
35Ibid., p. 80.

36Everett M. Rogers, Bibliography on thgiDiffusion
0f Innovations. Diffusion of Innovations Research Report
N00 6 East Lansing, Michigan: Department of Communication,
College of Communication Arts, Michigan State UniverSity,
July, 1967),

37For purposes of this discussion, Revzan's defini—
ti°n Of vertical integration (ioeo, vertical integration
eXiSts when one channel member controls a number of differ—
ent Operations in production and/or marketing at succeSSive
channel le 11 sufficeo .

DaXEdSA.wRevzan, "Marketing Organization(Thro;ghkthe
Channel," Wholesaling in Marketing Organization Ngw.n°r '
John Wiley and Sons, 1961), ppo 107-142. Reprint: ie tual
Bruce E. Mallen (ed.), The Marketing Channel. A 02: p 7
lkflflfiifli (New York: John Wiley and Sons, Inca, 1? :1?» d.
However, as vertical integration islopggfigiggacgge
inthe present study (see Appgngi§_g) the e em re adopted
trol, risk, commitment, and channel structure ai inte ration
as a basis of determining the degree of vertica 9
Much Characterizes each channel,

 

 

 

 

 

literature. While this d:
meadvantages and disadve
rcoroad base of agreement
gated systems are desirab
:isdiscussion are review

iicCammon and Bates
’:entrally coordinated sys
national marketing charmer
:eianism in the American c
nis‘n among three types of
fine systems are:

Corporate marketing sy:
stages of production at
ownership . . . [admini
formed by individual er
ship and influence or c
(if adjacent firms withi
vertically aligned com;
other to achieve trans;
processing, and adverti
'total' costs within t
systems] emerge when i
levels coordinate the'
basis to obtain system
that could not be achi

M

38Bert C . McCammon

Firgence and Growth of Co
fithe American Economy, " .
\lrowth, Competition and Wm
3ruczténg Association, 196!
View 3 Mallen (ed.), 2:13
W (New York: John

39 ,
39.12.. pp. 287-2

34 y

literature. While this discussion has tended to emphasize
Hm advantages and disadvantages of such systems, there is
no broad base of agreement among writers that highly inte—
grated systems are desirable. The points of view taken in
fins discussion are reviewed below.

McCammon and Bates suggest in a recent paper that:
’bentrally coordinated systems are gradually displacing con—

ventional marketing channels as the dominant distribution

. . . 38 . . .
mechanism in the American economy." These writers distin-

guish among three types of centrally controlled systems;
these systems are:

Corporate marketing systems, which combine successive
stages of production and distribution under a single
ownership . . . [administered systems] . . . are
formed by individual enterprises which exert leader—
ship and influence or otherwise control the behaVior
Of adjacent firms within the channel; as a result,
vertically aligned companies work closely Wlth each
other to achieve transportation, warehoueing, data
processing, and advertising economies which reduce
'total' costs within the system . . . [contractual
systems] emerge when independent firms at different
levels coordinate their activities on a contractual
basis to obtain systemic economies and market impact 39
that could not be achieved through indiVidual action.

x“-

38Bert C° McCammon, Jr., and Albert D. Bates, "The
Emergence and Growth of Contractually Integrated Channels
in the American Economy," in P, Do Bennett (ed.), Economic
Eflflfl£h_22§petition and World Markets (Chicago: American
Marketing Association, 1965), ppo 496-515. Reprigted igual
Bruce E. Mallen (ed.), The Marketing Channel. A oneep
W (New York: John Wiley and Sonsy Inc” 196 . po
287.

39Ibid., pp. 287—288.

 

 

 

 

 

HcCammon and Bate:

robe responsible for the
systems and they note a nu
watinue to confront the m

Centrally controll
important during the last
agree that they are desire
for instance, says that th1

nation is desirable depem
. . 4O .
has or diseconomies. 1
functions occurs:

. . . the costs of the
be spread over more ac1
The integration of func
quence, reduce total cl
it is clear the merger
the elimination of the
it is by no means certe
ate any savings. . . .‘
Sicklin discusses a number
"1th increasing integratio
he to management' 5 inabil
Efficiently. 0n the other

Economies of integrati
useof resources, the
striations upon firm 5.

\
40 ,
St: Louis P. Bucklin

"agile. Institute of Bu
forn' chool Of Business Ad
1a» Berkeley, 1966.

41
M‘s p0 85.

 

35

McCammon and Bates cite a number of factors thought
to be responsible for the growth of these centrally controlled
systems and they note a number of unresolved problems which
continue to confront the members of contractual system.

Centrally controlled systems have become increasingly

important during the last few years; however, not all writers

agree that they are desirable on economic grounds. Bucklin,

for instance, says that the degree to which vertical inte—

gration is desirable depends upon whether it produces econ—

. . . 4O . .
omies or diseconomies. He notes that as integration of

functions occurs:

the costs of the communication function can

be spread over more acts of other marketing functions.
The integration of functional acts may, as a conse-
quence, reduce total channel costs. But, although

it is clear the merger of institutions results in

the elimination of the market-oriented communication,
it is by no means certain that this action will cre—
ate any savings. .

o o

Bucklin discusses a number of diseconomies which may occur

with increasing integration, and which, he suggests, are

due to management‘s inability to control an organization

efficiently. On the other hand, he notes:

Economies of integration may result from the joint
use of resources, the elimination of market re—
strictions upon firm size, and the breakup of

“—

Louis P. Bucklin, A Theory of Distribution Channel
§tructure, Institute of Business and Economic Research, Grad-
Uate School of Business Administration, University of Cali-
fornia, Berkeley, 1966.

4lIbid., p. 85.

 

 

 

 

 

monopolistic pricing .‘

Thus, in Bucklin':
affunctions and firms is
ciies and the diseconomies
‘ntegration; he is not pre
‘ntegration ES se is eith

At least one empir
:echced total costs which
tiarthy and Williams deve
isolation model to calcul
Ensfor producing and marl
vrious market requirement
ion, McCarthy and William.

In the integrated SYS‘U
with buying and sellinl
firms have been elimin.
tion has been centrali:
down to the retail lew
wands. In contrast, i
together in the barny

systems are unable to

available to the integ

The importance of a 111
tion can be seen when
ferent levels of capac
marketing systems. F0
an integrated unit sel

\N—
42 .
5112-: PP- 86'87

43
Won 3- Jerome McCart
led) 0f Production~Marke

I S '
Encef: clence Technolo ‘

roceedin
a . . gs of th
9°» American I"1arlce1:(§.rié11

 

36

monopolistic pricing.42

Thus, in Bucklin's opinion, whether the integration

of functions and firms is desirable depends upon the econ—

omies and the diseconomies which will accrue because of such
integration; he is not prepared to conclude that vertical

integration per se is either good or bad.

At least one empirical study points to substantially
reduced total costs which result from vertical integration.
McCarthy and Williams developed a general purpose computer
Simulation model to calculate the costs of alternative sys—

tems for producing and marketing Grade A large eggs under

various market requirements.43 As a result of this simula—

tion, McCarthy and Williams concluded that:

In the integrated system, many of the costs connected
with buying and selling activities between individual
firms have been eliminated. Also their entire opera—
tion has been centralized under a single management
down to the retail level and is geared to market de—
mands. In contrast, independent units linked loosely
together in the barnyard, part-time and commercial
systems are unable to achieve many of the economies
available to the integrated system.

The importance of a marketing and channel orienta-
tion can be seen when the simulation is run at dif~
ferent levels of capacity for various production—
marketing systems. For example, the total costs of
an integrated unit selling to large retailers is

\—

421bid., pp. 86—87.

438. Jerome McCarthy and Robert J. Williams, “Simu—
lation of Production—Marketing Channels," in Raymond M. Hass
(ed.), Science, Technology, and Marketing, 1966 Fall Confer—
:nce Proceedings of the American Marketi

ago:

_ . n3 Association (Chi-
American Marketing Assoc1ation, l9 6), p. 336.

 

 

 

 

about 36 cents a dozen
at 25 per cent of capac
perdozen. Thus, an 1
operation to market de
over channels where ex
1 at various levels in t

. . . in the end, then
nels. And it appears
advantages over channe
ent firms which all ac

The notion that ch
:een amplified in a more r
tCarthy. They note:

... the members of a
of themselves as membe
less their system will
long as their unique c
fectively than competi

To provide a more
hfition which exists bet
fiStinguish between 'chann
119 systems':

- . . Not all channel
themselves as part of
business firm, for exa
with its own affairs 5
cerned with how its de
have affected the char
have not had a strateg
system. . . .

M

“Ina, p. 343.
45
Donald J. Bower

Planning of Vertical Mark.

inner delivered at the Ch
erican Marketing Associ
Ethodology, Nov. 6-8, 19

H

 

 

 

37

about 36 cents a dozen at 100 per cent capacity while
at 25 per cent of capacity, costs jump to 45 cents
per dozen. Thus, an integrator that has geared his
operation to market demands has a decided advantage
over channels where excess capacity occurs regularly
at various levels in the channel system.

0 o o

. . . in the end, then,

channels compete with chan—
nels.

And it appears that integrated channels have
advantages over channel systems composed of independ—
ent firms which all act autonomously.44

The notion that channels compete with channels has

been amplified in a more recent paper by Bowersox and

McCarthy. They note:

0 9

. the members of a channel system may not think
of themselves as members of a system, but neverthe~
less their system will continue to exist only as

long as their unique combination performs more ef—
fectively than competing channels.

a o 0

To provide a more complete understanding of the com—
Petition which exists between channels, Bowersox and McCarthy

distinguish between ‘channel systems' and ‘vertical market—

ing systems':

0 0

. Not all channel systems have members who see
themselves as part of a system. The traditional
business firm, for example, is primarily concerned
with its own affairs and has not been seriously con—
cerned with how its decisions or others' decisions
have affected the channel system. That is, they
have not had a strategic commitment to a channel
system. .

M

44Ibid., p. 343.

Donald J. Bowersox and E. J. McCarthy, "Strategic
Planning of Vertical Marketing Systems.“ A mimeographed
paper delivered at the Channels Seminar sponsored by the
American Marketing Association Task Force on Marketing
Methodology, Nov. 6—8, 1968, Evanston, Illinois, p. 8.

 

 

 

 

. . . A vertical market
to a traditional or adm
is one in which both
dleman (or middlemen) h
channel system-~a commi
thing to some target In

. . . the potential po
vertical marketing sys
the willingness of all
commit to one over-rid

Vertically integra

anpetitors because of the
.n because of this they a
ent of separation. Under
:stbe carried on by the
trainer as to result in a
ipossible recombining of

ionsvhich may be undert

Enhance, at a lower cost.

The most effective char
transactions may not re
trrangement for exchanq
9r0uping of specializer
achieving a maximum tr.
asecond grouping of ac
mg efficiency in phys;

. . . This division of
of specialized compete}
Concept of separation;
A recent empirical
he lumber manufacturing,

\~_
462331., pp. ll an

47M” p. 20.

 

 

38

. . . A vertical marketing system, in comparison
to a traditional or administered channel system,
is one in which both the manufacturer and the mid—
dleman (or middlemen) have made a commitment to the

channel system-—a commitment to jointly provide some-
thing to some target market.

0 o o

. . . the potential power and effectiveness of a
vertical marketing system, therefore, comes from
the willingness of all the firms in the system to46
commit to one over—riding channel strategy.

0 0 o

Vertically integrated marketing systems are stronger
competitors because of the commitment of system participants,

and because of this they appear to have exploited the con—

cept of separation. Under this concept, the activities which

must be carried on by the channel may be analyzed in such
a manner as to result in a separation of the activities and
a possible recombining of these activities into configura—

tions which may be undertaken with no duplication of effort

and hence, at a lower cost. Bowersox and McCarthy note that:

The most effective channel arrangement for creating
transactions may not represent the most effective
arrangement for exchange fulfillment. Thus, one
grouping of specialized activities might exist for
achieving a maximum transaction creating impact while
a second grouping of activities might focus on achiev—
ing efficiency in physical exchange.

0 0 0

a o o

This division of activities is on the basis

of specialized competence and is referred to as the
concept of separation.47

A recent empirical study of vertical integration in

the lumber manufacturing, hospital supply, and mail order

K‘s—~—

461bid., pp. ll and 13.

471b1a., p. 20.

 

 

 

 

entries was reported by

mm an industry, a mean
abies which were assumed to
negation was calculated.
'.. . an approximate measu
redegree of influence of
ales and industries."48

As a result of his
flitthe following five con
:iaants of vertical integra
éstudied:

1. Threat posed by a c
2. A change in market
3. Member firms with a
4. Channel conflict

5. Entrepreneurial dri

Sturdivant's study
'ical integration . A1 thou
Minted data obtained th
Wary sources, they provi
iIvortance of each factor

M It is important to

\——__.

“956), p- 473.

49
11%, pp. 473-4

 

—7
39

industries was reported by Sturdivant. For each company
within an industry, a mean score for each of fifteen vari—
ables which were assumed to have some influence on vertical
integration was calculated. Sturdivant states that:
‘h . . an approximate measure was available, therefore, for
‘Hm degree of influence of each variable within given com-
pmues and industries."48

As a result of his research, Sturdivant suggests
that the following five conditions were the strongest deter—
minants of vertical integration in the channel systems which
he studied:

1. Threat posed by a competing system

2. A change in market conditions

3. Member firms with a broad horizontal base
4. Channel conflict

5. Entrepreneurial drive49
Sturdivant's study relates several factors to ver—
tical integration. Although his conclusions are based on
GStimated data obtained through a content analysis of sec-
ondary sources, they provide a basis for approximating the
imPOrtance of each factor in determining vertical integra—
tion. It is important to recognize Sturdivant's implicit

M

48Frederick D. Sturdivant, "Determinants of Vertical
IDtegration in Channel Systems," in Science, Technology, and
Marketing, 1966 Fall Conference Proceedings of the American
IVlé-rke’Cing Association (Chicago: American Marketing Associa—
tion, 1966), p. 4730

49

 

Ibid., pp. 473—478.

 

 

 

 

 

when that these fact

The following fact
to: Sturdivant's study of

1. It does not assume
tion. Rather, int
dependent variable
dependent variable
study and the Stair
sis of vertical in

It analyzes behavi
rather than concen
to the firm.

Ix)

w
o

It employs several
variables. These :1
data which meet the
statistical tests.
While most of the n
Ft‘tal integration in gel
hthe field of agriculture
ific characteristics of in
live been concerned with t]
the economic cost of pr
iirfarm product. Others

Mails of contractually i

'119 levels of satisfaction

3hate in contract channe

MAJOR STUDIES 01"
OF AGRIC
Mighell and Jones,

tical coordination in 391-” i
11!

8130f several of the r

 

 

 

4o

assumption that these factors cause integration.
The following factors differentiate the present study
from Sturdivant's study of integration:

1. It does not assume that the variables cause integra—
tion. Rather, integration is manipulated as an in—
dependent variable, and the change in each of the
dependent variables is noted. Thus, the present
study and the Sturdivant study approach the analy-
sis of vertical integration from opposite directions.

2. It analyzes behavioral and attitudinal variables,
rather than concentrating on conditions external
to the firm.

3° It employs several research instruments to index
variables. These indices provide interval level
data which meet the assumptions of parametric
statistical tests.

While most of the marketing literature has treated

vertical integration in general terms, several publications

 

h1the field of agricultural marketing have focused on spe-
cific characteristics of integrated channels. Some writers
have been concerned with the impact of vertical integration
on the economic cost of producing and distributing a particu-
lar farm product. Others have described the operational
details of contractually integrated channels, as well as

the levels of satisfaction experienced by the persons who

Operate in contract channels.

MAJOR STUDIES OF STRUCTURAL ORGANIZATION
OF AGRICULTURAL CHANNELS
Mighell and Jones, in an extensive analysis of ver-
tical coordination in agriculture, present an economic an—

alysis of several of the relationships of vertical integration

 

 

 

and contract production .

ml of vertical integrat
scale which may be realize.
: the particular industry

Particularly relev

fiscussion by Highell and

 

basic beliefs and values.

Farming in this countr
basic beliefs and valu
enterprise and the fam
of operation are often
the advantage of the c.
Even then, force of ha
ficult to overcome. C
ing in which certain m
do not respond immedia
to contract arrangemen
are severe and the inc

As with innovations of
coordination may have

ate difficult problems
reason or another are

a new structural align
adisadvantage, they
production of a partic
tire from farming. Bu
kinds of vertical coor
transitions. In some
may be a transitional
plate vertical integr
may be a way of intro
which eventually will
an open market. In e'
to work out the ultim
nth less frictional

    
 
   
   
  
   
  
   
   

K

  
 

Ronald AMighell

Aricultu

Alivicskilture, Economic Res
on, Agricultural

“63. pp 75-76. EC

0R
Ll d1 t

  
 

  

l———

41

and contract production. These authors present an economic
model of vertical integration which considers economies of
scale which may be realized depending on the cost structure
of the particular industry under study.

Particularly relevant for the present study is the
discussion by Mighell and Jones of barriers to change and
basic beliefs and values. They note that:

Farming in this country early developed a set of
basic beliefs and values around concepts of private
enterprise and the family farm. . . . Customary modes
of operation are often difficult to change unless
the advantage of the change is direct and obvious.
Even then, force of habit and tradition may be dif-
ficult to overcome. Communities and types of farm—
ing in which certain methods have been customary

do not respond immediately to proposals to shift

to contract arrangements unless economic conditions
are severe and the incentives are great. . . .

 

As with innovations of any kind, changes in vertical
coordination may have transitional effects that cre-
ate difficult problems. Those farmers who for one
reason or another are not able to participate in

a new structural alignment will find themselves at

a disadvantage, they may be forced to give up the
production of a particular commodity or even to re-
tire from farming. But the existence of several
kinds of vertical coordination may greatly ease such
transitions. In some instances, contract farming
may be a transitional phase leading toward more com—
plete vertical integration later. Or contracting
may be a way of introducing a new farm commodity
which eventually will be produced independently for
an open market. In either case, it becomes possible
to work out the ultimate adaustment more slowly and
with less frictional cost.5

M

50Ronald Mighell and Lawrence A. Jones, Vertical

 

EQQrdination in Agriculture, United States Department of
Agriculture, Economic Research Service, Farm Economics
Division, Agricultural Economic Report No. 19, February,
1963, pp. 75-76.

 

 

 

 

 

fi

A second major stu
n‘the National Commission
asses and effects of vert
key, and broiler industrie
:5 integration as a means
nggested and conclusions
hstry are listed. As not
:i'ennels which may have th
:t‘ne future will be thos
that commitment. To rei
mission on Food Marketi

. . . many retailers h
quality eggs to be an
they wished to convey
entered into a variety
in order to achieve th
Nearly half of the egg
handling firms were cu
retail organizations h
branded cartons. They
egg packing plants or
specifications packed
tions. As mentioned
have developed vertic
ownership or contract'
duction by the laying
n. a strong determinant
rEtailer's product—market
freSh, high quality eggs
Image.

\~—_—_..__
51

Or anization an

3:3 Industries , Washing to

00d Marketing . Technica

  

 

E::::____________________________________if
42
A second major study, issued as a part of the Report

of the National Commission on Food Marketing, analyzed the
causes and effects of vertical integration in the egg, tur—
key, and broiler industries. Several reasons for the increase
of integration as a means of structural organization are
suggested and conclusions about the future of the egg in—

dustry are listed. As noted earlier in this chapter, the

 

channels which may have the strongest competitive advantage
in the future will be those which display a strong product—
market commitment. To reinforce this view, the National
Commission on Food Marketing Study states that:

. . . many retailers have considered uniform high

quality eggs to be an important part of the image

they wished to convey to consumers, and they have

entered into a variety of procurement arrangements
in order to achieve this objective.

 

 

Nearly half of the eggs sold by respondent egg
handling firms were customer brandedo Most large
retail organizations have sold eggs in their own
branded cartons° They either operated their own
egg packing plants or had eggs meeting prescribed
specifications packed for them under strict condi—
tions. As mentioned earlier, some large retailers
have developed vertical arrangements either through
ownership or contracting which extend back to pro-
duction by the laying flock°51

Thus a strong determinant of egg industry structure is the
retailer's product—market commitment expressed in terms of
fresh, high quality eggs which are congruent with his total
image.

K“—

51Organization and Competition in the Poultry and
E Industries, Washington, DOC.) National Commission on
Food Marketing° Technical Study'N0° 29 June, 1966: P. 44-

 

 

7 7" fi

A disagreement amox
1.10le degree of verti<
hthis chapter. Bucklin,
ion in terms of resultant
hersox and McCarthy, on ‘
degree of vertical integra‘
at only on costs but on 0‘
astbeneed for a strong 14
nedfor a high degree of j
flannel. This latter poin

of the Report of the Nation
hgrowing share of pro
handled by the largest
seem destined to disap
on small firms will co
omies obtainable in la
relatively continuous
available intraplant e
plants handling a very
duction. Further econ
processing, and distri

with centralized cont:
and horizontal functi

  
   
   
 
   
  
 
 
 
 
   
   

Increasingly importan
are in selling and bu
and quality requireme
be obtained from larg
SUppliers can more ea
dependent on a few pr

feasibly maintain hig
information systems t
to areas and outlets

under changing market
chases of equipment,

and supplies of other
may also be able to r
their smaller rivals.

M
521bid., pp. 85-8

    

——_——1
43

A disagreement among writers about what constitutes
aloptimal degree of vertical integration was noted earlier
in this chapter. Bucklin, for instance, analyzed integra—
tion in terms of resultant economies and diseconomies°
Bowersox and McCarthy, on the other hand, state that the
degree of vertical integration which ought to exist depends
not only on costs but on other system—oriented factors such
as the need for a strong leader within a channel and the
rmed for a high degree of product—market commitment in the
channel. This latter point is supported by the conclusions
of the Report of the National Commission on Food Marketing:

A growing share of processing is expected to be
handled by the largest firmso Many small firms

seem destined to disappearo Part of the pressure

on small firms will come from potential scale econ-
omies obtainable in larger plants operated on a
relatively continuous basiso However, potentially
available intraplant economies can be realized in
plants handling a very small share of national pro—
duction. Further economies in physical production,
processing, and distribution appear to be associated
with centralized control in coordination of vertical
and horizontal functions in a large scale enterprise.

Increasingly important advantages of large firms

are in selling and buyingo The volume, variety,

and quality requirements of large buyers can best

be obtained from large suppliersa Moreover, large
SUppliers can more easily avoid becoming strongly
dependent on a few principal customers or buyers
within limited geographic areas° They can also more
feasibly maintain highly developed and elaborate
information systems to help them allocate their sales
to areas and outlets yielding highest net return
under changing market conditionso In quantity pur—
chases of equipment, feed, medication ingredients,
and supplies of other production inputs, large firms
may also be able to realize gains unavailable to
their smaller rivalso

\—
Ibido, ppo 85—860

52

 

 

 

777 fi

REPORTS ON

IN TH

Several agricultur

:‘cal integration in egg c
itegration is a relativel
tdustry. In the past, .

. . . one or another 9
sized firms performing
erted a degree of domi
in other segments of t
For instance, tern
axial in the beginning of
Reuse of their importanc
Ill assembler—grading sta
. innce as they performed n
29 functions. More recer
ancerns have had a major
heindustry.“
Only recently hav
f9ed company, hatchery, p
hgto accept central coo
Several authors h

the details of contractua

M

53Henry E. Larzel
Alricul’cural Economics Re
cultural Economics, Colle
University, East Lansing,

54mm. , pp. 27—2

   

——_—_’
44

REPORTS ON VERTICAL INTEGRATION
IN THE EGG INDUSTRY
Several agricultural economists have discussed ver—
tical integration in egg channels. Larzelere notes that

integration is a relatively recent development within the

industry. In the past, . . .

. . . one or another group of small or moderate
sized firms performing similar functions has ex-
erted a degree of dominance in influencing changes
in other segments of the industry. . . .5

For instance, terminal market handlers were influ—

ential in the beginning of commercial egg marketing partly

because of their importance as egg storers. Later, the cen—

tral assembler—grading stations in the country gained dom—
inance as they performed more of the processing and carton—
ing functions. More recently some local and national feed

concerns have had a major influence in exerting change within

the industry.54

Only recently have the four major industry segments——
feed company, hatchery, producer and buyer——become more will-
ing to accept central coordination as a means of operation.

Several authors have published studies which describe

the details of contractual arrangements for integrating egg

“——
53Henry E. Larzelere, The Egg Industry in Transition.
Agricultural Economics Report, No. 79. Department of Agri—
cultural Economics, College of Agriculture, Michigan State
University, East Lansing, Michigan, March, 1967, pp. 27-29.

54Ibido , ppo 27-290

 

;—¥

 

7 7’" fi

dauels. These publicati
study, and to persons who
Marketing because they
hunch channels operate.
'7yDr. Ralph Baker, who so
are started, who coordina
Int decisions, how many c‘
from the producer to other
hyvhich producers were cc
:iskvas transferred from

the channel.55

REPORTS ON THE I

OF

While several pub]
gration in agriculture, re

analyzed the attitudes he]

Ward integration and to

  
   
   
   
   
   
 
  
 

M the industry. Two exc

Niles, Williams ,

M

55Ralph L. Baker,
laketin in the North Ce
.gftlgrams, North Central R
ho: Ohio Agricultural
ReSearch Bulletin 973. J

For an example of
5tribes the details of co
Md L. B. Darrah, Surve
Res. 165. Ithaca, New Yo
Experiment Station, Febru

      
   

 

45

channels. These publications are relevant to the present
study, and to persons who are unfamiliar with egg production
and marketing because they help to provide a knowledge of

how such channels operate. One such study was published

by Dr. Ralph Baker, who sought to determine why such programs
were started, who coordinated them, who made certain manage-
ment decisions, how many decisions had been transferred away
from the producer to other parties in the channel, the means
by which producers were compensated, and the means by which
risk was transferred from the producer to other parties in

the channel.55

REPORTS ON THE ATTITUDES HELD BY MEMBERS
OF EGG CHANNELS
While several publications describe vertical inte—
gration in agriculture, relatively few publications have
analyzed the attitudes held by persons within the industry
toward integration and toward the structural organization
of the industry. Two exceptions are noted below.

Niles, Williams, and Kohls conducted a study of 45

________________
55Ralph L. Baker, Coordinated Egg Production and

Marketing in the North Central United States. I. Kinds of
Wooster,

PrOgrams, North Central Research Publication 160.
Ohio: Ohio Agricultural Research and Development Center,

Research Bulletin 973. July, l965°

For an example of another empirical study which de—
5cribes the details of contract programs. see James 5° ROCk
and L. B. Darrah, Survey of Integrated Egg Production, A. E.
Res. 165. Ithaca, New York; Cornell University Agricultural
Experiment Station, February, l965°

 

 

 

—

ntract egg producers who
gohcers. These research
~aidby these two groups 0
hes, such as: their sati
nangeneht, their relativ
fives, their economic retu
nangenent, the incentive
Lyvhich various items of
:actices were adopted ove
With respect to tr
2:'Kohls conclude that:
Analysis of the use of
and Ianagement prac tic
tractual programs tenc
newer technology. . .
This study was amc
“ass and vertical integrat

Mills studied innovativene

Channels only, the presen

 
  
   
     

 

it occurs along the entir

A second report 0

\____

56Edwin E. Niles,
. rdinated E Productio
efizlagttEaézes. II. Characte
$th ontract E Prod
MC Publication No. 16
3| Versity Agricultural E

 
 
   
   
  
   
    
   

  
   

 

57
‘ the . Statistical hyp
impreuse nature of the
Wation and vertical 1

  

 

46

contract egg producers who were matched with 45 independent
producers. These researchers were interested in the opinions
kmld by these two groups of producers about a number of is—
sues, such as: their satisfaction with their contractual
arrangement, their relative loss of decision making preroga-
tives, their economic return provided under their contractual
arrangement, the incentive to do a good job, and the speed
by which various items of equipment and various management»
practices were adopted over time.
With respect to the latter point, Niles, Williams,
and Kohls conclude that:
Analysis of the use of certain pieces of equipment
and management practices followed showed that con—
tractual programs tended to bring about the use of
newer technology. . . . 6
This study was among the first to link innovative—
ness and vertical integration. Whereas Niles, Williams, and
Kohls studied innovativeness in independent and contract
channels only, the present study explores innovativeness as
it occurs along the entire vertical integration continuum.57
A second report on the attitudes held by persons

M
56Edwin E. Niles, Robert J. Williams, and R. L. Kohls,

Coordinated Egg Production and Marketing in the North Cen-
tral States. II. Characteristics and Attitudes of Independ—
ent and Contract Egg Producers. North Central Regional Re—
search Publication No. 162. Lafayette, Indiana: Purdue
gniversity Agricultural Experiment Station. May, 1965, p.

57Statistical hypothesis 1.5 in Chapter III states
the precise nature of the predicted relationship between
lnnovation and vertical integration.

 

L___‘c

 

—

m'thin the egg industry 3;:
zanducted by Egg Industry
salesmanager for the Pool:
nic‘n manufactures poultry
hpany conducted a large
gersons within the egg inc
:luded in Pockman's sample
i'echannel, as well as ec
Of interest to the
mey findings which com
...Inall of our 5\
that they were most 3;
return on investment,
were those with integl
operations.
The single profit cen‘
Primary control over .
of activity from prodl
Whatever the arrangeml
Imitation on degree

involved, and the act
of the control point.

 
  
  
 
 
   
  
  
  
     
    

 

In almost every such
eggs to the retailer
1y controlled by the
troduction was direc
in most instances co
agreement . 58

CONCLUSIONS ABOU
CHANN
Several conclusi

1161 literature published

1 58“
‘ 5 Egg Industry
1 IHdustr , August, 19

    

 

47

within the egg industry appeared as a published interview
conducted by Egg Industry magazine, with Richard L. Pockman,
sales manager for the Pockman Manufacturing Company, a firm
which manufactures poultry production equipment. The Pockman
Company conducted a large scale study of the attitudes of
persons within the egg industry, during 1967 and 1968. In-
cluded in Pockman's sample were persons at each level in

the channel, as well as equipment and medicine suppliers.

Of interest to the present study were the Pockman
survey findings which concern the structure of the industry:
. . . In all of our surveys, the producers who stated
that they were most satisfied with their profit and

return on investment, regardless of price levels,

were those with integrated or single profit center
operations.

 

The single profit center is defined as one having
primary control over all or most of the major items
of activity from production to retail distribution...
Whatever the arrangement or control, there is a
limitation on degree and numbers of profit centers
involved, and the activity is conducted on the terms
of the control point.

In almost every such operation the marketing of the
eggs to the retailer and the production were direct—
ly controlled by the same person or organization.
Production was directly related to retail need, and
in most instances covered by a direct marketing
agreement.

CONCLUSIONS ABOUT THE MARKETING AND AGRICULTURAL
CHANNEL LITERATURE SURVEYED
Several conclusions may be drawn about existing Chan-
nel literature published by marketing scholars. First, a

“——

58"Egg Industry Interview with Richard L. Pockman,"

Egg Industry, August, 1968, P0 20°

 

 

—

39h proportion of availab
is is economics-oriented
re directed at the examin
aiables. Second, most p
all, abstract, conceptual
29 variables and thus, a

tetation across writers.

:r-eapirical.

The literature ava
ranting has reflected a
all as an emphasis on des
31y true of the literati.
if agricultural industries
Ail with the egg industry
mlhing attitudes which

115. Further, much of ti

 

:nL ' . .
“duties which are unde

failed to emphasize the 0

 

Sisters .

 

.8 7
\

lfigh proportion of available marketing literature on chan—
nels is economics-oriented. Relatively few publications
are directed at the examination or analysis of behavioral
variables. Second, most publications are written at a gen-

eral, abstract, conceptual level. Few studies operationalize
the variables and thus, a given term varies widely in inter—
pretation across writers. Third, most channel articles are
non—empirical.

The literature available in the field of agricultural—
marketing has reflected a strong economics orientation, as
well as an emphasis on descriptive detail. This is particu—
larly true of the literature which deals with the structure
of agricultural industries. Many articles and monographs
deal with the egg industry; few of these are directed at
analyzing attitudes which are held by members of egg chan—
nels. Further, much of this literature concentrates on the
activities which are undertaken by egg producers and has
failed to emphasize the operation of egg channels as entire

systems.

 

 

C

RESEARCH DESK

I!

The purpose of th:'

spent of the research in:
study, the data collectior
hanalyzing the data. ’1‘]
which presents the genera;
isection on the developmc
idiscussion of the field
hetespondents; and a se<

356d to process and analy:

This section cont

tical, and null hypothese

mthesis 1

When comparing produc
fer in terms of vert'
integrated channels

(a) exhibit a lower

 

I (b) adopt new busine

c)
(d) be composed of y
(e) be composed of m

In order that th'

finalized, several stat

Each statistical hypothe

 

 

CHAPTER III

RESEARCH DESIGN AND ANALYSIS OF DATA

INTRODUCTION

The purpose of this chapter is to explain the devel—
opment of the research instruments which were used in the
study, the data collection procedure, and the methods used
in analyzing the data. This chapter contains: a section
which presents the general statistical, and null hypotheses;
a section on the development of the research instruments;
a discussion of the field procedure used in interviewing
the respondents; and a section which explains the methods

used to process and analyze the data.

HYPOTHESES
This section contains a number of general, statis—

tical, and null hypotheses. These are;

Hypothesis 1

When comparing production—marketing channels which dif—
fer in terms of vertical integration, the more highly
integrated channels will tend to:

(a) exhibit a lower level of conflict and a higher de—
gree of cooperation

ib) adopt new business ideas and practices more quickly

(c) contain larger business units

(d) be composed of younger persons

(e) be composed of more highly educated persons

In order that this general hypothesis may be opera-

tionalized, several statistical hypotheses have been developed.

 

EaCh statistical hypothesis is a prediction of how the

49

 

 

 

‘

statistics used in the an
:ical hypotheses which re
hediately below.

For each of these
fiat there is no relation
II, has been stated. Th.
wthe first statistical

esdof each of the statis

Statistical flypg the se

e515 1

1.1 The value of
negatively re
score,1 and t
cantly from 2

(a) channels
(b) individua

1.2 The value of
negatively re
score and the

, cantly from 2

(a) channels
(b) individua

M

1The conflict rec
scores for each person.
“Sin value of the conflic
vidual, as reported by
eValuated him. For exam
llindividual producer w
eValuations of this prod
sllpplier, bird supplier,
Iein conflict received 5
Iing the conflict receiv
chnnel and dividing by
. Conflict given r
flirt scores which an in
“her persons in his cha

n.

   

 

 

50

statistics used in the analysis will turn out. The statis—

tical hypotheses which relate to Hypothesis 1 are stated

immediately below.

For each of these, a null hypothesis which states

that there is no relation between the variables in the prob-

lem, has been stated. The null hypothesis which relates

to the first statistical hypothesis will be stated at the
end of each of the statistical hypotheses sections.

Statistical Hypotheses Which Relate to General Hypoth—
esis l

1.1 The value of the vertical integration index is
negatively related to the mean conflict received
score, and the relation will differ signifi—
cantly from zero, for:

(a) channels
(b) individuals in the North

The value of the vertical integration index is
negatively related to the mean conflict given
score and the relation will differ signifi—

cantly from zero, for:

(a) channels
(b) individuals in the North

\g—

The conflict received score is one of two conflict
SCOres for each person. Conflict received refers to the
mean value of the conflict scores which relate to an indi_
vidual, as reported by the other persons in his channel who
eValuated him. For example, a conflict received score for
an individual producer would result by averaging the conflict
evaluations of this producer which were given by 'his' feed
Supplier, bird supplier, and processoredistributor. The
mean conflict received score for a channel is found by sum—
mlng the conflict received score for each individual in the
Channel and dividing by N.

Conflict given refers to the mean value of the con-
fliCt scores which an individual reports in evaluating the
other Parsons in his channel.

 

 

1.3 The value of
positively re
ceived2 score
nificantly fr

(a) channels
(b) individua

1.4 The value of
positively re
score and the
from zero, fc

(a) channels
(In) individua

1.5 The value of
positively re
score and the
from zero, fe

(a) channels
(b) individue

1.6 The value of
negatively re
lation will <

(a) channels
(b) individue

1.7 The value of
positively re
and the relai
zero, for:

(a) channels
(b) individu

1.8 The value of
positively r
of operation
significantl

M

2The cooperation
ation scores for each pe
Caived refers to the mea
'hich relate to an indiv
50118 in 'his' channel wh

Cooperation give
C00Dilution scores which
1‘19 the other persons in

   

 

51

1.3 The value of the vertical integration index is
positively related to the mean cooperation re—
ceived2 score and the relation will differ sig-
nificantly from zero, for:

(a) channels
(b) individuals in the North

1.4 The value of the vertical integration index is
positively related to the mean cooperation given
score and the relation will differ significantly
from zero , for:

(a) channels
(b) individuals in the North

1.5 The value of the vertical integration index is
positively related to the mean innovativeness
score and the relation will differ significantly
from zero, for:

(a) channels
(b) individuals in the North

1.6 The value of the vertical integration index is
negatively related to the mean age and the re—
lation will differ significantly from zero, for:

(a) channels
(b) individuals in the North

1.7 The value of the vertical integration index is
positively related to the mean education score
and the relation will differ significantly from
zero, for:

(a) channels
(b) individuals in the North

1.8 The value of the vertical integration index is
positively related to the mean standard score
of operational size and the relation will differ
significantly from zero, for:

M—

2The cooperation received score is one of two cooper—
ation scores for each person and channel. Cooperation re—
ceived refers to the mean value of the cooperation scores
Which relate to an individual, as reported by the other per—
sons in ‘his' channel who evaluated him.

Cooperation given refers to the mean value of the
SOOperation scores which an individual reports in evaluat-
ing the other persons in his channel.

 

 

t

(a) channels
(1)) individua.

Sample Null Hypothesi
theses 1.1 to 1.8

The relationship
integration index
score will not di

(a) channels
(12) individuals 1
Imthesis 2

irnen comparing produc
fer in geographic 1°C
the southern states w
(a) include a produce

by someone else w
(b) operate with a 10

degree of coopera
(c) adopt new busines
(d) contain larger bu
(e) be composed of yo
(f) be composed of mo

Statistical Hypothese

2.1 Southern chan
of control ov
and the diffe
icant.

2.2 Southern cha
received sco
difference w

2.3 Southern cha
given scores
ference will

2.4 Southern ch

significant .

2.5 Southern ch
eration give
the differen

   

 

(a) channels
(b) individuals in the North

Sample Null Hypothesis Which Relates to Statistical
Hypotheses 1.1 to 1.8

The relationship between the value of the vertical
integration index and the mean conflict received
score will not differ significantly from zero, for:
(a) channels

(b) individuals in the North

Hypothesis 2

When comparing production—marketing channels which dif-

fer in geographic location, those channels located in
the southern states will tend to;

(a) include a producer who is more closely controlled
by someone else within the channel.
(b) operate with a lower level of conflict and a greater
( ) degree of cooperation.
c
(d)
(e)
(f)

adopt new business ideas and practices more quickly.
contain larger business units.

be composed of younger persons.
be composed of more highly educated persons.

Statistical Hypotheses Which Relate to Hypothesis 2

2.1 Southern channels will report higher indices
of control over producers than northern channels

and the difference will be statistically signif-
icant.

2.2 Southern channels will report lower mean conflict
received scores than northern channels and the
difference will be statistically significant.

2.3 Southern channels will report lower mean conflict
given scores than northern channels and the dif—
ference will be statistically significant.

2.4 Southern channels will report higher mean coop—
eration received scores than will northern chan—

nels and the difference will be statistically
significant.

2,5 Southern channels will report higher mean coop—
eration given scores than northern channels and
the difference will be statistically significant.

 

 

—

2.6 Southern chan
vativeness sc
the differenc

2.7 The mean age
be less than
and the diffe
icant.

2.8 The mean educ
members will
channel membe
statisticall

 

There will be no
ences between th
southern versus

ilpgthesis 3

There will be a posi
vertical integration
(a) perceives his ch
and low conflict
(b) agrees to accept
and 6).
(c) perceives himself
7 and 8).
(d) displays a markel
10, and 11).
(e) perceives his cha
ments 12 and 13)
(f) rates other memb:
bility, trust, a:
and 16) .
(g) favors the value
(Q statements 17
(h) displays a moder
. 20, 21, and 22).
(1) values knowledge
| in which he oper
J) agrees with the
the good of the
25 and 26).

V

”lo 3111 the analysi
in: are members of two
egration score of th
e}: Participate .

 

A

y
3 53

2.6 Southern channels will report higher mean inno—
vativeness scores than northern channels and
the difference will be statistically significant.

2.7 The mean age of southern channel members will
be less than that of northern channel members
and the difference will be statistically signif—

icant.

2.8 The mean education score of southern channel
members will be greater than that of northern
channel members and the difference will be

statistically significant.

Sample Null Hypothesis Which Relates to Statistical

Hypotheses 2.1 to 2.8

There will be no statistically significant differ-
ences between the indices of control reported by
southern versus northern channels.

Hypothesis 3

There will be a positive relation between a person’s
vertical integration score and the degree to which he:

(a) perceives his channel as having high cooperation

and low conflict (Q statements 1, 2, 3, and 4).
agrees to accept channel control (Q statements 5

 

(b)

and 6).

(c) perceives himself to be an innovator (Q statements

7 and 8).

(d) displays a marketing orientation (Q statements 9,

10, and ll).

(e) perceives his channel as a reference group (G state-

ments 12 and 13).

(f) rates other members favorably in terms of dependa—
bility, trust, and competence (Q statements 14, 15,

and 16).
(g) favors the value of working with a channel captain

(h)

(Q statements 17 and 23).
displays a modern orientation (0 statements 18, 19,

20, 21, and 22).
Values knowledge of the structure of the channel

' in which he operates (Q statement 24).
(J) agrees with the concept of individual sacrifice for
the good of the channel as a whole (Q statements

25 and 26).

.________~._____

3
Who In the analysis for general hypothesis 3, persons
inteare members of two or more channels will receive the
th gration score of the least integrated channel in which
eY Participate.

—¥—\

(1)

 

 

 

—

(k) favors a system 0
price risk among .
27).

     
  
     
    
 

Statistical H othese

3.1 The integrati
positively re
with Q statem
18, 19, 23, 2

3.2 The integrati
positively re
ment with sta
17, 20, 21,

 

   

Sam 1e Null H othesi
theses 3.1 and 3.

  
  

There is no signi
gration score of
agreement to any

Ii'pgthesis 4
The cohesiveness of

predicted from knowl
vativeness, educatio

Statistical Hypo the se

4.1 Measures of i
novativeness,
employed as a
explain a sta
variance in 1
of individual

 

4.2 Measures of
vativeness, «
employed as .
explain a st
variance in
individuals.

4.3 Measures of
novativeness
employed as
explain a st
variance in
of individue

 

——7
54

(k) favors a system of providing a means of sharing of
price risk among all channel members (Q statement

27).
Statistical Hypotheses Which Relate to Hypothesis 3
3.1 The integration score of individuals will be
positively related to the degree of agreement
with Q statements: 1, 3, 5, 8, ll, l3, 14, 15,
18, 19, 23, 24, 25, 26, and 27.
3.2 The integration score of individuals will be
positively related to the degree of disagree—
ment with statements 2, 4, 6, 7, 9, 10, 12, l6,
17, 20, 21, and 22.

Sample Null Hypothesis Which Relates to Statistical

Hypotheses 3.1 and 3.2

There is no significant relation between the inte—
and the degree of

 

gration score of individuals,
agreement to any of the Q statements.

Hypothesis 4

The cohesiveness of a potential channel member may be
predicted from knowledge of his operational size, inno—
educational level, and age.

 

vativeness,
Statistical Hypotheses Which Relate to hypothesis 4

4.1 Measures of individuals' operational size, in-
educational level, and age, when

novativeness,

employed as a team of predictor variables can

explain a statistically significant amount of

variance in the mean conflict received scores

of individuals.

4.2 Measures of individuals' operational size, inno-
vativeness, edUCational level, and age, when
employed as a team of predictor variables, can
explain a statistically significant amount of
variance in the mean conflict given scores of

individuals.
Measures of individuals' operational size, in—
novativeness, educational level, and age, when
employed as a team of predictor variables, can
explain a statistically significant amount of
variance in the mean cooperation received scores

 

of individuals.

 

 

 

 

—

4.4 Measures of in
novativeness ,
employed as a
explain a sta
variance in t
of individual

     
     
   

Sam 1e Null H othesi
' theses 4.1 to 4.4
There is no stati

tween the predict
variable, conflic

El'pthesis 54
Members of similar ch
with one another, and
tistical factor, on t
Q-sort statements.

frptheSis 6

Channels which are ch
integration will tend
other and will load 5
factor on the basis 0
cooperation, innovati

DEVELOPMENT

This study emPlOY
entatives in the egg ind
flannel levels: feed SUP
Processing-distributing :
599 activities performed
'35 necessary to develop

l
M level. Each re sear<

\‘D-l—n—‘y

(a 4As explained in
$021: the data which re
, or analyzed, and COn
Pobeses are appropria

A

 

 

55

4.4 Measures of individuals' operational size, in—
novativeness, educational level, and age, when
employed as a team of predictor variables, can
explain a statistically significant amount of
variance in the mean cooperation given scores

of individuals.

Sample Null fiypothgsis Which Relates to Statistical

Hypotheses 4.1 to 4.4

There is no statistically significant relation be-
tween the predictor variables and the dependent
variable, conflict received.

4

Hypothesis 5

Members of similar channels will tend to correlate highly
and will load heavily on the same sta—

with one another,
tistical factor, on the basis of their responses to the

Q-sort statements.

Hypothesis 6

Channels which are characterized by similar amounts of
integration will tend to correlate highly with one an—
other and will load significantly on the same statistical
factor on the basis of mean channel scores on conflict,
innovativeness, age, and education.

 

cooperation,
DEVELOPMENT OF RESEARCH INSTRUMENTS
This study employed personal interviews with repre—

Sentatives in the egg industry who operate at five different

Channel levels: feed supply, bird supply, egg production,
processing-distributing, and retail. Because of the differ—
ing activities performed by respondents at each level, it
was necessary to develop a separate research instrument for

each level. Each research instrument contained questions

x—
'Limitations' section of Chap—

4

t AS explained in the

f:r I: the data which relate to hypotheses 5 and 6 were
Ctor analyzed, and consequently no statistical or null

hYpotheses are appropriateo

 

 

 

—

tesigned to elicit respor
ables: control, conflict
ion given, cooperation
size, age and education.
lated to the particular
mstrument was directed.
Each of the five
sections. The first of
signed to provide data a
five research instrument
Spect to the exact quest'
it. However, they are
I: the same variables,
ale follow one another

istruments .

The second part c

 

sseries of twenty~seven
Elicit responses which re
eats. The same statemeni

instruments .

LmStruction of Question

To ensure that t
me questions are of
a”(Nieves appropriate f0

N

(at sThese Q—sort 51
”in the chapter.

 

#

7
56

designed to elicit responses for each of the dependent vari—
ables: control, conflict given, conflict received, coopera—
tion given, cooperation received, innovativeness, operational
size, age and education. The content of each question re-
lated to the particular channel level at which that research
instrument was directed.

Each of the five research instruments has two main
sections. The first of these is composed of questions de—
signed to provide data about the dependent variables. The
five research instruments differ from one another with re—
spect to the exact questions asked of each kind of respond-
ent. However, they are similar in that the questions relate
to the same variables, and the questions regarding each vari—
able follow one another in the same order in all research
instruments.

The second part of each research instrument contains
a series of twenty—seven Q—sort statements5 designed to

elicit responses which reflect the attitudes of the respond—

 

ents. The same statements are included in all five research

instruments.

Construction of Questions

 

To ensure that the data provided by the responses
to the questions are of the interval level of measurement

and hence appropriate for parametric statistical tests, the

M

5These Q—sort statements are more fully explained
later in the chapter.

 

 

 

responses allowable for t
he use of pre-specified
questions for each varial
were constructed so that
Later excluded in the its
remaining homogeneous qua
hoax. The topics conta:
the: a review of egg inc
iionsvit‘n personnel in 1

6 Agricultural Economic

instruction of Attitudix
nyloyed in the Q-Method<

Q-methodology is
Stephenson8 to refer to <'
him he developed. It I
55 different individuals
leer. sorted into piles Cé
isked to look at each of
”lid“ had been typed on .
0 determine approximate.
”1th each statement and ‘
\

6This item analy;

Charles sghe aUthor is

State ‘ eppard J
constrU erSit , cm W

——7

57

responses allowable for each question were limited through
the use of pre-specified answer categories. To develop the
questions for each variable, a sufficient number of questions
were constructed so that when non—homogeneous questions were
later excluded in the item analysis,6 the answers to the
remaining homogeneous questions would constitute a valid
index. The topics contained in the questions were selected
after a review of egg industry literature and after discus-
sions with personnel in the Departments of Poultry Science
and Agricultural Economics at Michigan State University.7
Construction of Attitudinal Statements
Epployed in the Q—Methodolqu

Q—methodology is a general name used by William
Stephenson8 to refer to a group of psychometric procedures
which he developed. It involves correlating the responses
of different individuals to a set of statements which have
been sorted into piles called Q—sorts. Each respondent was
asked to look at each of twenty—seven attitudinal statements
Which had been typed on individual cards. He then was asked
to determine approximately how strongly he agreed or disagreed

with each statement and to lay each card in one of the sections

M

6This item analysis is explained in Appendix C.

7The author is grateful to Drs. Henry Larzelere,
Charles Sheppard, John Wolford, and Carl Hoyt of Michigan
State University for their suggestions regarding question
COnstruction.

8William Stephenson, The Study of Behavior (Chicago:
The University of Chicago Press), 1953.

 

 

 

m the nine POint contin
him. It was explained t
:inuum represented a neu
continuum represented 5t
:epresented strong disag
each section of the cont
statements, with the mos
m‘ the least at each en
forced to arrive at a qu
dish is required by the
malysis.

The statements i
31110:" a selection proc
“fiber 0f variables and
“959 Variables. A numb
each Variable and many w
5°‘C1arity and lack of
T .

«sort in that it was be
about eaCh set of variab
undnegatiV91)! phrased s

P

laid Testin the Resear

prior to embarki

Stud
was undertaken to
9Ap
. e ‘
the meChanigsngfx Becont
car

——————y

58

on the nine point continuum which had been placed before
him. It was explained that the center section on the con-
tinuum represented a neutral opinion, the left end of the
continuum represented strong agreement, and the right end
represented strong disagreement. It was also explained that
each section of the continuum required a certain number of
statements, with the most statements in the center section
and the least at each end. Therefore, the respondent was
forced to arrive at a quasi—normal distribution of cards
which is required by the statistical model used in data
analysis.9

The statements included in the Q—sort were the re-
sult of a selection process which started by specifying a
number of variables and then constructing statements about
these variables. A number of statements were generated about
each variable and many were eliminated after being screened
for clarity and lack of ambiguity. This was a 'structured'
Q-sort in that it was balanced to include a few statements
about each set of variables and it included both positively

and negatively phrased statements.

EEELQ Testing the ReSearch Instrument
Prior to embarking upon the full field study, a pilot
StudY was undertaken to test the research instrument. As a

M

9Appendix E contains more detailed instructions on
the mechanics of the card sorting procedure.

   

‘

result of the pilot stud;
:lons included in the re.
able and were relevant t.
respondents; that the 1m
in forty-five minute ti:
he instruments were rele
:sed in the main field 5

reproduced in Appendix B

F

The purpose of t

ed procedures used to s
for this study. The mai
ES the channel which is
Talon either supply inpu
:e'tail eggs which come f1
'35 the determinant of a
llinto the study, the PS

litically included as tht

 

The respondents i
liens: the northern reg.
0"110, and New York and t3
I‘u‘ssissippi, North Carol

L
Te northern region were

liming interviews were

 

7
59

result of the pilot study, it was determined that the ques-
tions included in the research instruments were understand—
able and were relevant to the operations of the particular

respondents; that the interviews could be completed within

the forty-five minute time limit set as a target; and that

the instruments were relatively trouble—free and could be

used in the main field study. The research instruments are

 

reproduced in Appendix B.

FIELD PROCEDURE

The purpose of this section is to indicate the methods
and procedures used to select and interview the respondents
for this study. The main unit of analysis, in this study,
was the channel which is defined as a group of business firms
which either supply inputs to, or process, distribute, and
retail eggs which come from a producer. Thus, the producer
was the determinant of a channel; once a producer was select—
ed into the study, the people with whom he dealt were auto—
matically included as the rest of his channel.

The respondents-were located in two geographic re—
gions: the northern region including Michigan, Indiana,
Ohio, and New York and the southern region including Georgia,
Mississippi, North Carolina, and Arkansaso Respondents in
the northern region were interviewed by the author; the re—

maining interviews were conducted by Agricultural Extension

   

‘

preonnel employed by uni

551216 Size

A total of 176 pe
Leafunction of two fact
were included, and the m;
an each of these produ

The number of pa
region and channel level
relatively small number
:pply, and processing 1
iese respondents repres

i1; these functions in M

TYPE:
~——._..______.____.__.
cflannel Level .. _
Reducer
3ird Supplier
heel Supplier
‘Processor-distribu tor
Retail
I Total

g”

 

M-

. 10A copy of the
filled to each of these
1x51

 

60

personnel employed by universities in the four southern states.10

Sample Size

A total of 176 persons were interviewed; this number
is a function of two factors: the number of producers which
were included, and the number of firms which are connected
with each of these producers.

The number of persons interviewed from each geographic
region and channel level is shown in Table 3.1. While a
relatively small number of persons at the feed supply, bird
supply, and processing levels were included in the study,
these respondents represent most of the major firms perform-

ing these functions in Michigan.

Table 301

 

TYPES OF RESPONDENTS

 

 

 

 

Channel Level North South Total 1
Producer 57 12 69 :
Bird Supplier 12 10 22 g
Feed Supplier 25 7 32 :
Processor—distributor l8 9 27 i
Retail _33; _£1 _ig§

Total 138 38 176

 

 

 

 

 

 

 

M

10A copy of the interview instructions which was
mailed to each of these interviewers is contained in Appen_
dix E.

 

‘

Selection of Producers

 

Producers include
selected on a representat
developed by Agricultural
of producers (and consequ
heincluded was influence
sent each of the major s
Michigan and in the 3
least five complete chan
inthe sample. The numb
type of channel was base
hat channel type in its
and approximate percenta
leagraphic region and ty
1‘: should be noted that t
and independent channels,
cOntract and owner—integz

Farm Bureau Serv:

 

lfflichigan~produced egg:
inlegg-producing Farm BI

services field represent

M

11The market-sha

um"1M by Agricultural
Ulinersity.

12M: . Clyde Spri

~ ‘ V P“: and Mr. Wayne Playi
Seketing Division, prox
erces producers.

 

 

61

Selection of Producers

Producers included from both geographic regions were
selected on a representative quota sample basis from lists
developed by Agricultural Extension personnel. The number
of producers (and consequently, the number of channels) to
be included was influenced by a desire to adequately repre—
sent each of the major structural types of channels found
in Michigan and in the South. As shown in Table 3.2, at
least five complete channels of each major type were included
in the sample. The number of producers selected into each
type of channel was based on the estimated market share of
that channel type in its geographic region.11 The number
and approximate percentage of producers included from each
geographic region and type of channel are shown in Table 3.2.
It should be noted that the North is dominated by contract
and independent channels, and the South depends upon both
contract and owner-integrated channels.

Farm Bureau Services, the largest single supplier
Of Michigan-produced eggs, provided assistance in interview—
ing egg-producing Farm Bureau members.12 Three Farm Bureau
Services field representatives visit each of their producers

M

' llThe market-share for each type of channel was es—
timated by Agricultural Extension personnel at Michigan State
University.

ler. Clyde Springer, Mr. Gene Wyngarden, Mr. Wayland

Fox, and Mr. Wayne playford, of Farm Bureau Services, Egg
Marketing Division, provided introductions to Farm Bureau
Services producers.

 

 

 

 

 

 

 

 

‘

once every ten to fourter
following the egg produc
Farm Bureau Services. '1']
representative for two da
Tue egg producers that we

‘nterviewed .

TYPES OF EGG PRO
CLASSIFIED BY GBOG

.

|
Eecgraphic Type of
. Region Channel

 

North
Independent
Contrac t

r Owner—In tegrat
Total

Independent

Contrac t

Ownerelnte gr at
Total

 

 

g__

interview Procedure
. The non-Farm Bur
‘ sel‘~‘C’ced to be interviev
‘ 9110116. During this com
Se1f as a graduate stud:

doing a study of the 69

states. These persons

 

 

62

once every ten to fourteen days to assist the producer in
following the egg production program as specified by the
Farm Bureau Services. The author accompanied each field
representative for two days on his round of visitations.
The egg producers that were visited during these times were

interviewed.

Table 3.2

TYPES OF EGG PRODUCERS INCLUDED IN THE STUDY,
CLASSIFIED BY GEOGRAPHIC REGION AND CHANNEL TYPE

 

 

 

 

 

Number of Producers Market Share
Geographi Type of Included in Produced by
Region Channel Quota Sample This Type
North
Independent 18 32%
Contract 34 60%
Owner-Integrated _§. 8%
Total 57 100%
South
Independent 0 0%
Contract 7 58%
Owner—Integrated _§ 42%
Total 12 100%

 

 

 

Interview Procedure

The non-Farm Bureau Services egg producers who were
selected to be interviewed were initially contacted by tele—
Phone. During this conversation the author introduced him—
Self as a graduate student at Michigan State University,

dOing a study of the egg industry in Michigan and bordering

States. These persons were told that their name had been

 

 

 

 

 

 

provided by an agent of

lrequest was then made
cases, they agreed to be
eral call-backs were neCr
spondents were occupied r
bet and corn harvest, w.‘
fall of 1968 when the in
Both men and worn
hction operations, espe
:‘nousand hens, the woman
tion while the husband e
:rranaged the corn, whe
Spoudents, in all but a
hair operation in detai

viewer in any way possib

Procedure for In terviewi

During the first
Preducers, each producer
511d birds and who proces
nailed by the producer we
distributor was asked tc
9998 he handled. Freque
eral retailers; when thi
Vere randomly selected 1

“SW dealt with retail:

“Sasked to name the re

 

 

 

63

provided by an agent of the Agricultural Extension Service.
A request was then made for an interview appointment; in all
cases, they agreed to be interviewed. In a few cases, sev-
eral call-backs were necessary because the prospective re-
spondents were occupied with the hunting season and the sugar
beet and corn harvest, which were in progress during the
fall of 1968 when the interviews were conducted°

Both men and women were interviewed. In many pro—
duction operations, especially those having ten to fifteen
thousand hens, the woman of the house managed the egg opera—
tion while the husband either worked part-time elsewhere
or managed the corn, wheat, or sugar beet crops° The re-
spondents, in all but a few cases, were willing to discuss
their operation in detail and to cooperate with the inter—

viewer in any way possible.

Procedure for Interviewing Other Members of the Channel
During the first wave of interviews with the egg

producers, each producer was asked who supplied his feed

and birds and who processed his eggso These individuals

named by the producer were later interviewed. Each processor—

distributor was asked to name the person to whom he sold the

eggs he handled. Frequently, a processor sold eggs to sev-

eral retailers; when this happened, two or three retailers

were randomly selected to be interviewed later. If a proc-

essor dealt with retailers located throughout Michigan, he

was asked to name the retailers located nearest the Lansing

 

 

 

 

 

yea, to reduce subseque

ease.

The four agricul
fixted interviews in the
5:. Howard Zindel, Chair
qultry Science Departme
:0 participate in this s
ifmstructions and inte
13

each interviewer .

Each interviewer

6;; producers: two cont

rated producer. He was

feed supplier, bird SUPP
eachof the three egg Pr
Each containing UP to £0
1M of twelve interview

icu'thern state. To redu

”Y eaCh interviewer ,
not

the

contacted; a total 0

hr -
ted in the four south

AN
The PUrpose of t
¥
3 The
to these inteiivitt
we

 

64

area, to reduce subsequent interviewing travel time and ex—

pense.

Procedure Employed in Interviewing Southern Respondents
The four agricultural extension personnel who con-
ducted interviews in the southern states were contacted by
Dr. Howard Zindel, Chairman of the Michigan State University
Poultry Science Department, and asked if they would agree
to participate in this study. They all agreed. A packet
of instructions and interview materials was then mailed to
each interviewer.l3
Each interviewer was requested to interview three
Egg producers: two contract producers and one owner-inte-
grated producer. He was requested to also interview the
feed supplier, bird supplier, and processor, who dealt with
each of the three egg producers. Therefore, three channels
each containing up to four persons were interviewed. A max—
imum of twelve interviews could have been conducted in each
southern state. To reduce the number of interviews conducted
by each interviewer, the retail level in each channel was
not contacted; a total of thirty—eight interviews were con-

ducted in the four southern stateso

ANALYSIS OF DATA

The purpose of this section is to discuss the methods

M

13The cover letter and interviewing instructions
Sent to these interviewers are included in Appendix E.

 

 

 

 

 

usedin processing the <
analysis involved the f<

l Transform the re

novativene 55 to

l'\)

Compute the med:
question to adj‘

w

Item analyze tm
tiveneSS queSti(

,p.

Compute a score

) cooperation
) COOPeration
) conflict gi\
) confliCt rec

a
b
c
d
‘e) innovativenE
if) operational

m

Compute a score
in step 4-

0»

Compute an int8<_
Each of these da
elplained below and a $1
Cedure appears at the er
table which lists each I

itwill be tested.

hxplanation of Steps in

To Step 1: Conver:

v .
. ativeness to Standarc

in

each of the five char
in

Ch were unique to the
tion '

611 Size of egg prodi

11

who
1' Of hens housed V
’

Sale
sOr tons of feed s<

7
65

used in processing the data collected for the study. The
analysis involved the following steps:

1. Transform the raw data on operational size and in—
novativeness to standard scores.

2. Compute the median for each conflict and cooperation
question to adjust for missing data.

3. Item analyze the conflict, cooperation and innova—
tiveness questions.

4. Compute a score for each individual on:

(a) cooperation given
(b) cooperation received
(c) conflict given

(d) conflict received
(e) innovativeness

(f) operational size

5. Compute a score for each channel on (a—f) mentioned
in step 4.

6. Compute an integration index for each channel.

Each of these data processing steps is more fully
explained below and a summary of the hypothesis—testing pro-
cedure appears at the end of this section in the form of a
table which lists each hypothesis and the procedure whereby

it will be tested.

EXplanation of Steps in Data Processing

Step 1: Conversion of Raw Measures of Size and In—
BQZEEEyggess to Standard Scores. Size and innovativeness,
in each of the five channel levels, were measured in units
Which were unique to that level. (For example, the opera-
tional size of egg producers was indexed in terms of the

number of hens housed, whereas for feed suppliers, dollar

sales or tons of feed sold were employed to measure

 

 

 

operational size.) It

  
  
  
  
   
   
 

data to compare the rel
sons at different chann
innovativeness measure
data were converted to
of z-scores, or stander
size or innovativeness
reflect each individual

other person included i

2.323: Com ut
and Coo eration Questio
research instruments di
questions designed to e
ilict. In some cases a
tothe respondent; in 31.
the question, coded a ze
to the next question.
ing data, remained in t)
data analysis, they woul
“Utas 'no response,‘ b\
the response continuum.
are normally distribute
left end of the distrib
To reduce the disruptiv

on subsequent statistic

With the median re spons

 

—7

66

operational size.) It is inappropriate to use these raw
data to compare the relative size or innovativeness of per—
sons at different channel levels. To provide a size and
innovativeness measure which is channel—level-free, the raw
data were converted to standard scores expressed in terms

of z-scores, or standard deviation units away from the mean
size or innovativeness of each channel level. These z—scores
reflect each individual's size relative to that of every

other person included in the study.

Step 2: Computation of the Median for Each Conflict

 

and Cooperation Question to Adjust for Missing Data. The
research instruments directed at each channel level contained
questions designed to elicit data about cooperation and con—
flict. In some cases a particular question did not apply

to the respondent; in such instances the interviewer skipped
the question, coded a zero in the answer column, and proceeded
to the next question. If these zeros, which represent miss-
ing data, remained in the card columns during subsequent

data analysis, they would be interpreted by the computer

not as 'no response,’ but as a response at the low end of

the response continuum. (If it may be assumed that responses

 

are normally distributed, these responses would lie at the
l9ft end of the distribution and would be highly atypical.)
TO reduce the disruptive influence of these missing values
on subsequent statistical analyses, these zeros were replaced

With the median response. This median lies at the center

 

 

  

of the response distrib

  
   
    
     
  
  
  
  
  
  
  
   
 

orthe value of statist

M: Item
mdlnnovativeness Ques
the statistics used in
uployed to index a res
novativeness were ite
Homogeneity of each
ere which involves a fa
correlation of each que
other question about th
:hde in an index only
one another and load 51

salts of this item anal.

M: Compute
onEach Variable. In 01
individual on each varia
liven, cooperation-race:
ceived, and innovativem
Eich of these mean scor
heach question for a

that number of response

ltional size was the st

Step 5: Com ut
5

ac
M. Once ea

 

 

 

67

of the response distribution and has a minimal influence

on the value of statistics calculated to test hypotheses.

Step 3: Item Analysis of Conflict, Cooperation,

 

and Innovativeness Questions. Prior to the computation of
the statistics used in hypotheses testing, the questions
employed to index a respondent's conflict, cooperation, and
innovativeness were item-analyzed to determine the degree

of homogeneity of each question. Item analysis is a proced—
ure which involves a factor analysis of the coefficients of
correlation of each question about a variable with every
other question about that variable. It is desirable to in—
clude in an index only questions which correlate highly with
one another and load significantly on one factor. The re-

sults of this item analysis appear in Table C.l in Appendix C.

Step 4: Computation of Scores for Each Individual
on Each Variable. In order to provide one score for each
individual on each variable, a mean score on cooperation—
given, cooperation—received, conflict—given, conflict-re—
ceived, and innovativeness was computed for each person.
Each of these mean scores was found by summing the responses
to each question for a particular variable and dividing by

that number of responses. Each individual's measure of oper_

ational size was the standard score computed in step I.

Step 5: Computation of Scores for Each Channel on

Each Variable. Once each individual had been given a mean

 

 

 

 

score on each variable,

conputed for each charm
and dividing by the num
he mean conflict-recei
271 summing the conflict

hthe channel and divi

E3223: Com ut

Each Channel. The inde

 

glies to all the firms

tenined by the manner

acterize each channel:
(a) Vertical contro
('0) Price risk is s

in the channel.
(c) The firms in th

in that particu
(d) The production—n

one firm within
Malxsis of Hypotheses
Hypotheses 1.1 1

late a channel's vertica
leasure on each of the c
SimPle correlations. E

to determine whether it

M

14
is The procedure
‘ exPlanned more fully
”‘5 mOre detailed def
5 four components
of how it was

examPles

 

 

 

68

score on each variable, a mean score on each variable was
computed for each channel by adding the individual means
and dividing by the number of individuals. For instance,
the mean conflict-received score for a channel was found
by summing the conflict—received score for each individual

in the channel and dividing by N.

Step 6: 92mputation of an Integration Inggx for
Each Channel. The index of vertical integration which ap—
plies to all the firms within a particular channel was de-
termined by the manner in which the following factors char—

acterize each channel:

(a) Vertical control is exercised over the producer.

(b) Price risk is shifted from the producer to others
in the channel.

(c) The firms in the channel are committed to remain
in that particular channel.

(d) The production—marketing functions are assumed by
one firm within the channel.

Analysis of Hypotheses

hypotheses 1.1 to 1.8. These hypotheses, which re-
late a channel's vertical integration score to that channel's
measure on each of the dependent variables, were tested with

simple correlations. Each correlation coefficient was tested

to determine whether it differed significantly from zero.

_______________

. 14The procedure for calculating the integration index
15 explained more fully in Appendix A. This appendix con—
tains more detailed definitions of integration and of each

of the four components in the integration index, as well as
examples of how it was calculated in two specific channels°

 

 

 

 

   
  
  
  
  
  
  
  
   
  

{nose correlations which
significance were consid

hwthesis .

H otheses 2.1 t

'ierelationship between
Enamel and that channel
variables, and were test

attest of the differen'

 

ferences for which the 5
statistic was .95 or abo

reject a null hypothesis

H otheses 3.1 a
hoped to test the noti
2dividuals will be rela
hinill be related to d.
3ilple correlations were

hing the F test at the

Hypotheses 4.1 t
aPetite test the suggest.

“€58, operationalized in
“Deflation, could be pred
iittor variables. To te
indivicluals for whom all

Whitted to a multiple

halved the Stepwise de

 

7

69

Those correlations which held at or above the .95 level of
significance were considered sufficient to reject a null

hypothesis.

Hypotheses 2.1 to 2.8. These hypotheses dealt with
the relationship between the geographical location of a
channel and that channel's measure on each of the dependent

variables, and were tested for statistical significance with

 

a t test of the difference of two means. Again, those dif—
ferences for which the significance probability of the t
statistic was .95 or above were considered sufficient to

reject a null hypothesis.

Hypotheses 3.1 and 3.2. These hypotheses were de—
veloped to test the notion that the integration score of
individuals will be related to agreement with some statements
and will be related to disagreement with other Q—statements.
Simple correlations were computed to test these hypotheses

using the F test at the .95 level of significance.

Hypotheses 4.1 to 4.4. These hypotheses were devel—
oped to test the suggestion that an individual's cohesive—
ness, operationalized in terms of low conflict and high Co—
Operation, could be predicted with the aid of a team of pre—

dictor variables. To test these hypotheses, data for 136

 

individuals for whom all questions had been completed were
submitted to a multiple regression computer routine which

involved the stepwise deletion of variables from a least

 

 

 

 

 

squares equation. The

described in Appendix H

fiyathesis 5.1.
sible discovery of type
mess of agreement an
W~ Hyp
:se of factor analysis.
:edural steps involved
M; a more genera
is included here.

After each pers
stems of his degree
responses of each perso
idler person. This ma
Wbmitted to factor anal
was obtained. This was
Which produced orthogona
”We 0f person.

The varimax rout
sUlutions,' each of whic
so'18- Following genera
native factor solutions

leneity with the use of

\.

e terms "ty

th 15%
amJeably in discussin

heses 5.1 and 6.1.

 

——1

70

squares equation. The operation of this LSDEL program is

described in Appendix H.

Hypothesis 5.1. This hypothesis relates to the pos-
sible discovery of types15 of ersons, as reflected by com—
monness of agreement and disagreement with the set of g:
sorted statements. Hypothesis 5.1 was analyzed with the
use of factor analysis. A detailed statement of the pro-
cedural steps involved in this analysis is provided in 527
pendix F; a more general description of these procedures
is included here.

After each person had rank ordered the statements
in terms of his degree of agreement or disagreement, the
responses of each person were correlated with those of every
other person. This matrix of intercorrelations was then
submitted to factor analysis, and a principal axis solution
was obtained. This was then submitted to a varimax rotation
which produced orthogonal factors. Each factor represents
a type of person.

The varimax routine yielded several sets of 'factor

 

solutions,‘ each of which contained a set of types of per-
sons. Following generally accepted procedures, these alter~
native factor solutions were compared subjectively for homo-

geneity with the use of several criteria, as explained in

M
15

 

The terms "type" and "factor" will be used inter—

Changeably in discussing the analyses which relate to Hy-
POtheses 5.1 and 6.1.

 

 

 

 

 

Appendix G. The factor
re maximum between-gr01
groups variance was sele

This seven factc
:mputer routine in 0rd!
:feach type. As a reSL
his solution was descr.‘
:-r.‘: and disagreement w:
able to compare the fac
each farts: most strong:
ims differentiated eaci
L“seven factors was ti

or agree and disagree I

Hypothesis 6 . l .

neisw" ' '
\ nich are Similar 1'

lies based on the commc

.ieiendent variables . H)

p

r
0c6dures employed in t
ind ' -

etall in Appendices

sis
used here was identi
iiidthesi
s 5.1 R
. espons

rela
ted across people (c

 

I_—'

71

Appendix G. The factor solution which appeared to contain
the maximum between—groups variance and the minimum within-
groups variance was selected for further analysis.

This seven factor solution was submitted to a second
computer routine in order to provide attitudinal descriptions
of each type. As a result of this analysis, each factor in
this solution was described in terms of a hierarchy of agree-
ment and disagreement with the statements. It was then pos—
sible to compare the factors to determine which statements
each factor most strongly agreed and disagreed with and which
thus differentiated each factor from the others. Each of

the seven factors was thus described in terms of its strong—

 

est agree and disagree responses.

Hypothesis 6.1. This hypothesis suggests that 2223?
flgig which are similar in integration will form groups or
types based on the commonness of their scores on several
dapendent variables. Hypothesis 6.1 was analyzed with the
aid of a varimax rotation of a principal axis solution. The
Procedures employed in this factor analysis are explained
in detail in Appendices G and H. The method of data analy—
sis used here was identical to that used in the analysis of
hYpOthesis 5.1. Response patterns in both cases were cor-
related across people (or channels composed of people).
Both analyses result in the formulation of types which are
common on some basis. In the first case (5.1) the types

Ware individuals; here (6.1) they are channels.

 

lmayoffflpothesis Te

huorder to brie

‘ ‘ ‘ C
:tefingeacn nypotne.

SUMMARY OF Pfi

h

 

immuis Content of
‘5‘ ' “ hesis
&JX&_
L:ol.8 Integration
related to
dependent

1 variables

3° 2'8 Geographical
location as
related to
dependent

\ variables
x

Integration

related to

\‘ Q‘res OnseS

CoheSivenesg

: Type analij
baSEd On Q‘
\ Statements
.‘ \
9.

/1 Type analysj
\

 

based On

dependent
\_ Variables
\

 

 

 

72

Summary of Hypothesis Testing Procedure

In order to briefly summarize the procedure employed

in testing each hypothesis, Table 3.3 is presented below:

Table 3.3

SUMMARY OF HYPOTHESES TESTING PROCEDURE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Statistical Unit of Analysis
Hypothesis Content of Procedure Each Entire
Numbers Hypothesis Employed Person Channel
1.1 to 1.8 Integration as Simple X X
related to correlation
dependent
variables
2.1 to 2.8 Geographical T test of X
location as difference
related to between two
dependent means
variables
3.1 and Integration as Simple X
3.2 related to correlation
Q—responses
4.1 to 4.4 Cohesiveness Multiple I X \
prediction regression
5.1 Type analysis, Factor X
based on Q— analysis
statements
6~l Type analysis, Factor X
based on analysis
dependent
_ variables

 

 

 

 

 

 

FIN.‘

The purpose of
:fthe various analyses
szated in Chapter III.
sections, each of which

sis of data which relate

was RELATIONS]
SELECTEI
The purpose Of ‘
Cithe analysis of the 1
integration and several
:L'll hypotheses were te:
Ofsimple correlations.
”Nations: (1) 3S cha
t'neNorth and 12 in the
ofWhom were located in
films 0f the two sets <
a“ddiScussed below. Ta
“the two analyses.

H .
0thESls lol: Conflic

Statistical hyp<

 

CHAPTER IV

FINDINGS OF THE STUDY

The purpose of this chapter is to explain the results
of the various analyses employed to test the hypotheses
stated in Chapter III. This chapter is organized into six
sections, each of which discusses the results of the analy-

sis of data which relates to one set of hypotheses.

THE RELATIONSHIP BETWEEN INTEGRATION AND
SELECTED BEHAVIORAL VARIABLES
The purpose of this section is to explain the results

of the analysis of the relationships which exist between
integration and several dependent variables. A number of
null hypotheses were tested for significance with the use
of simple correlations. The data analyzed came from two
populations: (1) 35 channels, of which 23 were located in
the North and 12 in the South, and (2) 137 individuals, all
Of whom were located in the North. The results of the an-
alyses of the two sets of data are presented in Table 4.1
and di5cussed below. Table 4.1 also compares the results

Of the two analyses.

Wis 1.1: Conflict Received

Statistical hypothesis 1.1 says:

73

 

 

 

 

 

 

m0% m0» 0: 0: 0: 02 OS on
huh mwh mwx 0C mmh m0» mwx mm»
1
_ w.H _ b-a ®.H H m.H V.H _ m-H I N.H L H-H \
— ”LAmnE..z Wfimvn—JomyflTa
”UNIV!“

..NH:

OD WHmmcuOQhfl HHJC EMU

NTQ+LQw QL

.N

mmflmmfiuOQhfl ND UMHUHUQHQ
EOHUUQMHU Cfl COHuMHmuHOU mH .H
susom pom. zuhoz zoom
Cd UQUQUOQ WHQQCMSU EOHN MUMQ

HM; HV BVN

144‘“.

MVZHNNT‘QZOU Di wmqm<Hm<> gm0H>§mm
.y N0 “HMKAJ‘ZIV

 

73,-...7. 7..v-v. )1.-

(Halflo 1H<3DJHI>JIQZH QZ< 4J22<<~U 20 an<m M:H|JJWUVH
2:. .....«u ...?

H-Z< 7.11....(vu

311-4.]v‘1.<.4n

 

 

 

 

74

 

 

 

 

 

 

 

 

 

 

 

ma. No. mo. Ha. 0H. «m. mo. Hoo.
mm.H sm.m mm.m mm.m mw.m em. mm.m m¢.NH
Ha. ma. mH.l mH.I ma. mo.l hH.I mm.l
on no» oo oo o: oo no» no»
ooh ooh no» 0: no» 0o now no»
mo. v0. mm. .wo. mm. mm. «mo. mm.
om.m wo.v oo. .mN.m ms. om. .mm.m mm.a
mm. mm. Noo.l om.l ma. Ha. om.l HN.I
no» no» OS on QC oo oo 0c
no» mow mo» 0o no» mo» ooh no»
m.H >.H o.H m.a e.H m.H N.H H.H
Hell

 

 

mumnfisz mHmocquN:

 

Uflumflomum h mo

soaaanmnoua oucmuawecmam .m
Uflumfipmom m mo ooam> .v
pooflUmeoou

coapmamuuoo mo oSHm> .m
«pouuommu

mo mHmocpomma Haas Sou .N
mmflmmcoomhc an oouuwomua
coauumuflp Ga COHpmHouuou mH .a

 

Mano auuoz ca owpmooq
..mamsee>aeqH some when

.Uflymfipmum m mo

suaflenanoud mucmunoacmam .m
oeumflumpm e no mzam> .e
.pcoHUmemou

coauMHouuou mo 05Hm> .m
«Uopuonwu

on mammzpomwn HHSG dmu .N
mmflwocuoahc ho ooyUHUouQ
Goauuouflo CH COHomeuuou mH .H

endow use abuoz euom
ca omumuoq mamccocu Eouw mama

 

 

<H<Q fiDQH>HQZH 92¢. QMZZ<IU ZO Qum<m MBQDWWm MIR. OZHM<mEOU Q24 mmdm<Hm<> ém0H>§mm

QMBUMQMW QZ< ZOHB<MOMHZH ZMHBBWm mHImZOHB’NQMM NIH. DZHBWWB m0 WHQDWWN NEH. .NO WHWNAAWZAN

 

H . .V OHQMH

 

FF-l: *(uf
now“ [on ou

HHSC HHSC
low #0: money

uuonou uuonmu «mwmonuoahs Hand mo

00 30C IGOU 0:0: 0:0: ococ mcoc 30c 30c COHuuman Go coamaumo 08 .N
owmu0> «GOHuMHouuou
Odo: ococ ococ ococ 0C0: Ion woos ococ

M0 coauumufifi OH .H

«mama chdan can» wonumm
Hmsoa>dodH mo who or»
EOHH #Hflwwm m0 CMQU #033

 

 

  

 

 

 

.- MHIZ hmpm H
. . . *§

com N.H mono .mo
noon»: ca . 5Hm> powwow H
mm UH m . QUCM a U W . mu mk/mg Q cm a a o III
.mmuz.wuwwMomam one we woam> ohm pcmammwwmwmam horn mo muHHHQmQOWd oommumwacmMmrwmmznmw m.a
. . . v.H pom m.a mom . map somZDwQ ma mdoa m . . . + Haoz
mgumu mg“. CH “mmuxm a mmnz M#M£UOQ%£ Ugm m mmnz SUHES Sufi N. o A. a ma wwfllflmmmmmqmmmfl gm ml.” mnmfle
o Hoodmcu who no mooaumaouuou Mo ooaumasuwmwoomm 0o ooflnMHou
. . CH.

 

nearing
e + e auaumenaom n no
9 F huHHHQmQOHQ mUGmUHMHomHm OH .m
e e e Nuaumfluoom m mo 03Hm> 0H .w
coaumamuuou

% & wuowfloflwmoou
mo moam> modaonnm 09 .m

aHDQ “ooh

     

whfiwmh [MM Ow... HHSQ HHDC
I on wood

0U 30c Ioowp m boonmu poonou «mammcoomhn HHSC m0
Coo woos woos moon 30c 30c ooauumnmu o0 coamfluoo 05 .m
mmum> mGOHumamuuoo

0 U .
Goo moon moon ocoo moo: Ion oooo moo: Mo Coauumuflo OH .H
amuse Housman carp emanam
ob one

 

 

HMSUH>HUGH Mo m
Eou% uaonom no snow has:

 

 

 

 

 

 

 

 

 

 

1.1 The value of th
negatively rela
score, and the H
from zero, for:

(a) channels
(b) individuals

The null form of th.
The relation be

integration indc
score will not i

(a) channels
(b) individuals

Results of tests:

(a) fail to reje
1.1 (a)

(b) reject the z
The apparent me;
is that when conflict re
Tamar-including higl
nrelation exists. Bu“
tion scores of Northern

that as integration inc1

illpgthesis 1.2: Confli¢
The statistical

1.2 The value of th‘
negatively rela
score and the r«
from zero, for:

\_____

1The null hypot
ineach set of statisti
Md hereafter only the
9595 will be stated.

 

’1

76

l l The value of the vertical integration index is

negatively related to the mean conflict—received
score,

and the relation will differ significantly
from zero, for:

(a) channels

(b) individuals in the North
The null form of this hypothesis1 is:
The relation between the value of the vertical
integration index and the mean conflict—received
score will not differ significantly from zero, for:
(a) channels

(b)

individuals in the North
Results of tests;

(a) fail to reject the null form of hypothesis
1.1 (a)

(b) reject the null form of hypothesis 1.1 (b)

The apparent meaning of these contradictory results

is that when conflict received data is aggregated across

channels-—including highly integrated ones in the South—-

no relation exists° But, when the low and medium integra-

tion scores of Northern channels are analyzed one finds
that as integration increases, conflict decreases.

EXEQEflesis 1°2: Conflict Given

The statistical form of hypothesis 102 is stated

1 2 The value of the vertical integration index is
negatively related to the mean conflict given

score and the relation will differ significantly
from zero, for:

M.

The null hypothesis for each. statistical hypothesis
in each set of statistical hypotheses is similarly stated

and hereafter only the first of each group of null hypoth-
eses will be stated.

 

 

 

 

(a) channels
()3) individuals

Results of tests:
(a) fail to rej

1.2 (b)

(b) reject the

These results p
that the r computed for
though it reached the .
gersons in the North wa
her for channels was
high integration scores
thef'mdings related to
thesis that more highly
lore smoothly, because

one another .

y Ethesis 1.3: Cooper

The statistical

1.3 The value of th
positively rela
score and the r
from zero for:

(a) channels
(b) individu al 5

Results of tests:

fail to reject
(a) and (b)

It was surprisi

Of these hypotheses yié

1Iltegration and cooper a

 

 

 

77

(a) channels
(b) individuals in the North
Results of tests:

(a) fail to reject the null form of hypothesis
1.2 (b)

(b) reject the null form of hypothesis 1.2 (b)

These results parallel those for hypothesis 1.1 in
that the r computed for channels was not significant (al—

though it reached the .07 level), while the r computed for

persons in the North was significant. As in section 1.1,

the r for channels was probably suppressed by the extremely

high integration scores of the southern channels. Generally,

the findings related to hypotheses 1.1 and 1.2 support the

thesis that more highly integrated channels tend to function

more smoothly, because their members get along better with
one another.

fiypgthesis 1.3: Cooperation Received

The statistical form of hypothesis 1.3 is stated:

1.3 The value of the vertical integration index is
positively related to the mean cooperation received

score and the relation will differ significantly
from zero for:

(a) channels
(b) individuals in the North

Results of tests:

fail to reject the null form of hypothesis 1.3
(a) and (b)

It was surprising to find that neither of the tests
of these hypotheses yielded significant relations between

integration and cooperation received. Apparently there is

 

 

 

 

 

no relationship between

[121 and the level of in

"thesis 1.4: Coo r

Hypothesis 1.4
1.4 The value of th
itively related

and the relatio
zero, for:

(a) channels
(b) individuals

Results of tests:
hypothesis 1.4
Neither of the

gration and cooperation
level. The test of hyp
nificant, however, at t
because it indicates a

More highly integrat

hpgthesis l . S: Innova

The statistic a1

1.5 The value of th
positively rela
score and the r
from zero, for:

(a) channels
(b) individuals

Results of tests:
hypothesis 1.5

While neither c

allY significant relatj

 

 

78

no relationship between the amount of integration in a chan-

nel and the level of intra-channel help and assistance.

Hypothesis 1.4: Cooperation Given

Hypothesis 1.4 is stated in its statistical form as:

The value of the vertical integration index is pos—
itively related to the mean cooperation given score
and the relation will differ significantly from
zero, for:

(a) channels

(b) individuals in the North

Results of tests: fail to reject the null form of

hypothesis 1.4 (a) and (b)

Neither of the tests of the relation between inte—

gration and cooperation given were significant at the .05

level. The test of hypothesis 1.4 (b) would have been sig-

nificant, however, at the .10 level. This is noteworthy

because it indicates a trend toward increased cooperation

in more highly integrated channels.

flifipthesis 1.5: Innovativeness

 

The statistical form of hypothesis 1.5 is stated:

1.5 The value of the vertical integration index is
positively related to the mean innovativeness
score and the relation will differ significantly
from zero, for:

(a) channels
(b) individuals in the North

Results of tests: fail to reject the null form of

hypothesis 1.5 (a) and (b)

While neither of these tests produced a statistic~

ally Significant relation between innovativeness and

 

 

 

integration,
anaroached si
hat predicte
The n
integration a
u operationa
that these re
afunction no
:cr, but also
ntion. It s
fiveness is d
iuuson's in
ie length of
Sue adopted
entered the f.
all of the pr
is not consid
Entered the s
sMeinnovati
Erners may ha
ationalize in
nel members In

later than di

Mothesis 1 .

The s

1.6 The V
negat
tion

 

79

integration, it should be noted that in both cases they
approached significance but in the direction opposite to
that predicted!

The negative relation between innovativeness and
integration appears to have resulted from the method used
to operationalize innovativeness and therefore it appears
that these results should be ignored. Innovativeness is
a function not only of present usage of a practice or prod—
uct, but also of how long a person has been using the inno-
vation. It should be explicitly recognized that if innova—
tiveness is determined by the length of usage of a product,
a Person's innovativeness score is necessarily limited by
the length of time during which that person might logically
have adopted the product. Thus, for example, a person who
entered the feed supply business one year ago and who adopted
all of the practices measured on an innovativeness scale
is not considered to be as innovative as another person who
entered the same busineSS two years ago and Who adopted the
same innovations. Thus, the innovativeness scores of south—
erners may have been constrained by the method used to oper—
ationalize innovativeness to the extent that southern chan-
nel members may have entered the egg business relatively

latEr than did their northern counterparts.

H290thesis 1.6: Age
The statistical form of hypothesis 1.6 is stated:
1°6 The value of the vertical integration index is

and the rela—
ne ativel related to the mean age .
tign willydiffer significantly from zero, for.

 

 

80

(a) channels
(b) individuals in the North

Results of tests: fail to reject the null form of

hypothesis 1.6 (a) and (b)°

While the null hypotheses were not rejected, it ap—
pears that young northern individuals tend to be more highly
integrated. For hypothesis 1.6 (b), r = .15 and would have
been significant at the .08 level. The strength of this
inverse relationship between age and integration may have
been suppressed by the rather narrow range of ages repre-
sented in the sample. Ninety—five percent of the respond-

ents were between 36 and 52 years of age.

Hypothesis 1.7: Education
The statistical form of hypothesis 1.7 is stated:
1.7 The value of the vertical integration index is
positively related to the mean education score

and the relation will differ significantly from
zero, for:

(a) channels
(b) individuals in the North

Results of tests: reject the null form of hypothesis
1.7 (a) and (b).
As hypothesized, the positive relation between edu-

cation and vertical integration was statistically significant.

filEEEESélS 1.8: Operational Size
The statistical form of hypothesis 1.8 is stated:

1.8 The value of the vertical integritignréngigrés
' an a
' i 1 related to the mean 5 . '
ggsigezitgonal size and the relation Wlll differ

Significantly from zero, for:

 

 

 

 

 

 

81
(a) channels
(b) individuals in the North
Results of tests: reject the null form of hypothesis

1.8 (a), fail to reject the null form of hypothesis

1.8 (b).

As integration increases, the size of business units
also increases for channels, whereas for individuals located
in the North there is no significant relationship between
the two variables. The large operations of the highly inte—
grated southern channel members seem to make the difference
between the levels of statistical significance in the two
tests. When northern individuals were considered, r was
in the predicted direction but it was too weak to be sig-
nificant.

This lack of significance is probably due to the
fact that many of the relatively highly integrated northern
egg producers bought 'packages' in which a feed mill would
arrange the financing for a 10,000 bird hen house in return
for the producer's agreement to sign a production contract
with that feed mill. This would guarantee the mill a market
for its feed. While the producers who signed these contracts
are relatively innovative, their 10,000 bird houses are rela—

tively small——hence no significant relation was found between

integration and size.

 

 

 

82
A Summary of the Relationships Which Exist Between
Integration and Selected Dependent Variables
As a result of the tests of the preceding hypotheses,
the following relationships were found to be statistically
significant:
As vertical integration increases:

(a) for all channels:

1. education increases
2. operational size increases

(b) for northern individuals;
1. conflict received decreases
2. conflict given decreases
3. education increases
The following relationships are also noteworthy,
in that the significance probability of the F statistic ap—
Proached the .05 level:

As vertical integration increases:

(a) for all channels:
1. conflict received decreases (p=.09)

(b) for northern individuals;

1. cooperation given increases (p=.lO)
2. age decreases (p=.08)

THE RELATIONSHIP BETWEEN NORTHERN VERSUS SOUTHERN
LOCATION AND SELECTED CHANNEL BEHAVIORAL VARIABLES
The purpose of this section is to discuss the results
0f the analysis of the relationships which exist between
geOQraphic location and selected channel behavioral vari—
ables. A number of null hypotheses were tested for signif—

icance with the use of t tests of the difference between

 

 

 

83

two independent means. Hypothesis 2.1 which deals with
control was tested with data concerning 69 individual pro—
ducers, while hypotheses 2.2 to 2.8 were tested with data
from 35 channels. The results of the analysis are presented

in Table 4.2 and discussed below.

Hypothesis 2.1: Control
The statistical form of hypothesis 2.1 is:
2.1 Southern channels will report higher indices of
control over producers than northern channels and
the difference will be statistically significant.

The null form of this hypothesis is:
There will be no statistically significant differ-
ence between the indices of control reported by
southern and northern channels.

Result of test: fail to reject the null form of
hypothesis 2.1.

As was expected, there is no significant difference
between the control over producers in the North as compared
with the South. Egg production in both regions is heavily
based on production contracts which link producers to con—
tractors. The terms of these contracts may vary somewhat
across geographic regions with respect to matters of compen—
sation but these contracts are similar in that they control
many Of the flock management decisions which had formerly

been made by egg producers.

We 2.2: Conflict Received

Hypothesis 2.2 is stated in its statistical form as:

 

 

 

 

84

. occmcu Chmnuzom Cm£# m>H
H may wmmumnz ma %
. opmommdm Coma own mmwmwmcumcuuoc 303m dump map mmsmmwnuwwmwmwmmuam
o whoa on ou.MWMmcp0Q>£ HHDC mcu powwow cmu m3 3 r
3
Oman umoe m

 

 

 

. MPMU HOCCMQU .HOuH

 

 

 

 

 

 

 

 

 

 

 

 

 

 

wmumaoou pom m>d

mm corn umgumu

.pm>OCCH

HMUHuprMpm H50 powwow
warp woman cH..

a
mmu.mp
Q cuwcpuoc muwz mm
. on own cumcusom mums m pom mHmUdoou
.2 mo mNHm may mo meMMMmQMQMHMOCpamHMUSpouQ pw>ao>cfi Houpcou MOM mammE $39.
.mmnz .HHm CH
roan: mo \Jllllllllllllljllllillllll
Hm.m em.a mm,m mm» connmuzom m.m
omflm HOOD H590 owoNV flflsvw OZ wm< boN
cm.c mv. m¢.¢ mm.n ..mww mmmdw>flpm>OCGH mom
.mHm H00. ow m Noam mm.n ..mmw cm>flm coaumummoou mom
.mdm Ho. ms N ow>flmuwu
. .N vmom om.h ..mww GOHumuwQOOU v.m
.mflm moo. mm H mfioH mon 02 C0>Hm #UHHWCOU mom
.m.q so. mm oe.a mm.H mm» ow>nmomu poaflwcoo m,m
. fim.m
.onm mo am.o mm.m am.m oz .Houpcoo H.m
.m.c mm Nauz mwnz «ownownmm omnmoe mowom :Ilmwmmmmtl
Ammumov Uflpmflpmum arson cpuoz mm mflmw manmflum> mammcuoahm
odomaumum mo meo> manonum> lance»: wwwm
. . MU o .
u no no a
mmflpaaflnonoum 05Hm> 2mm:

 

 

 

 

 

mammcpomhm
mo muHSmmm

 

WQHZZ<IU mm MOh mm4m<Hm<> BZWQZHQMD QMBUHQWW QZ<
OHE<UOQ UHIQ<KOOM® ZMMBBmm mHImZOHH<4mm MO WBWMB H
Z

20 Qmm<m mom OH HoN meWIBOmNI MO WBWWB ho

 

N . fl QHQMB

WHADmmm

 

 

 

 

 

5)’ SOL
The d:
“d t}
Meet

l119m}

85

2.2 Southern channels will report lower mean conflict
received scores than northern channels and the
difference will be statistically significant.

Results of test: reject the null form of hypothesis 2.2.
As predicted, southern egg channels are character—

ized by less conflict received than are northern channels.
This finding supports the notion that the success of southern
channels is based on more than differences in production

costs--these channels may function more smoothly because

the members of such channels experience less conflict.

Hypothesis 2.3: Conflict Given

Hypothesis 2.3 is stated in its statistical form as:

2.3 Southern channels will report lower mean conflict
given scores than northern channels and the differ-
ence will be statistically significant.

Result of test: fail to reject the null form of
hypothesis 2.3.

Although the lower amount of conflict given reported
by southern channels was not significant at the 005 level,
the difference would have been significant at the .07 level
and thus partially reinforces results of the test of the
preceding hypothesis. Conflict appears to be reduced in

highly integrated southern channel systemso

H Othesis 204: Cooperation Received
Hypothesis 2,4 is stated in its statistical form as:

2.4 Southern channels will report higher mean coopera—
tion received scores than will northern channels
and the difference will be statistically signif—
icanto

 

 

 

 

 

 

 

1:2:

 

86

Result of test: reject null form of hypothesis 2.4.
The null hypothesis is rejected, but the direction

of the relationship is the reverse of that stated in the
statistical hypothesis. Northern channels report more co—
operation received than do their southern counterparts.
This is especially interesting in view of the earlier find—
ing (hypothesis 2.1) that northern channels also report
higher conflict received. This apparent inconsistency dis-
appears when it is remembered that conflict and cooperation
are not mirror images of one another. High conflict does
n2; automatically equal low cooperation. As operationalized
in the research instruments, cooperation describes 'help'
or 'assistance' given, whereas conflict occurs when 'dis-
satisfaction' arises because the expectations of one person
have not been fulfilled by the actions of a second person.
It thus appears that intra—channel cohesiveness is a complex
phenomenon. When compared with their northern counterparts,
Persons in the South were found to be more satisfied with
the performance of others in their channels, yet they re—

ceived less assistance from other channel members.

EXRQEE§§l§_2.5: Cooperation Given
Hypothesis 2.5 is stated in its statistical form as:
2.5 Southern channels will report higher mean coopera_
tion given scores than northern channels and the

difference will be statistically significant.

Result of test: reject the null form of hypothesis 2.5.

 

 

 

7

87

The test reveals a significant relationship between
geographic location and cooperation given but is the reverse
of that stated in the statistical hypothesis. This result
parallels that of the last hypothesis (2.4) and may be par—

tially explained by the factors discussed there.

Hypothesis 2.6: Innovativeness
Hypothesis 2.6 is stated in its statistical form as:
2.6 Southern channels will report higher mean innovative-

ness scores than northern channels and the differ—
ence will be statistically significant.

Results of test: reject the null form of hypothesis 2.6.

The null hypothesis is rejected but the direction
of the significant relationship between innovativeness and
integration is the reverse of that stated in the statistical
hypothesis. Northern channels are more innovative than
southern channels and the difference is highly significant.
As discussed earlier in the section of this chapter which

dealt with hypothesis 1.5 (a) and (b), to the extent that

southern egg men may have entered the egg industry relatively

 

later than did their northern counterparts, the innovative-
ness scores of these southerners may have been constrained
by the method used to operationalize innovativeness. This
finding therefore is inconclusive, and not until another

study is undertaken to gather more data will it be possible

to precisely state the relationship between innovativeness

and such variables as geographic location and vertical in—

tegration.

 

88

Hypothesis 2.7: Age
Hypothesis 2.7 is stated in its statistical form as:
2.7 The mean age of southern channel members will be
less than that of northern channel members and the

difference will be statistically significant.

Result of test: fail to reject the null form of hypoth—
esis 2.7.

There was no significant difference between the ages
of members of northern versus southern channels apparently
because a rather narrow range of ages was included in the
sample; 95 percent of the respondents were between 36 and
52 years old. Perhaps capital requirements have kept young
people out of the industry. Among northern egg producers
it appeared that the egg business was a part-time activity
managed by the woman of the family. These women spent from
four to six hours a day feeding the hens and collecting the
eggs. Few of these women had pre-school children and it
is possible that most of the producers in the business were
there because they had no small children to care for and

wanted to supplement the family income. These persons would

 

Probably have to be in their late 30's before all of their

Children would be of school age.

 

Hxpothesis 2.8: Education
Hypothesis 2.8 is stated in its statistical form as:
2.8 The mean education score of southern channel members
will be greater than that of northern channel mem-
bers and the difference will be statistically Sig-
nificant.

Result of test: reject the null form of hypothesis 2.8°

 

 

 

1-:13—

 

I1—1

89

The southern channel members appear to be signifi—
cantly more highly educated than northern ones. The egg
industry is relatively new in the South, and it has been
suggested that many persons who have entered the industry
there have specialized educational backgrounds. Indeed,
the data gathered by the Southern interviewers seem to bear
out this notion.

A Summary of the Differences Which Exist Between
Northern and Southern Channels

As a result of the tests of the preceding hypotheses
it was found that when southern and northern channels were
compared, the southern channels were characterized by the
following statistically significant findings:

Southern channels reported:

1. less conflict received

2. less cooperation received
3. less cooperation given

4. less innovativeness

5. more education

These findings seem to indicate that the southern
channels derive some strength because they contain persons
who perform well enough to satisfy other channel members.
Northern channels, on the other hand, may gain some advan—
tage because their members are willing to help one another.
Several northern contract channels were headed by men who
seemed genuinely concerned about the welfare of the people
in their channels. More importantly, the persons in these

channels trusted their channel captains and indicated that

 

 

 

 

 

90

they were willing to work harder because of the help and

special consideration they had received from these leaders.

THE RELATIONSHIP BETWEEN VERTICAL INTEGRATION
AND THE ATTITUDES OF PERSONS

The purpose of this section is to discuss the results
of the analysis of the relationships between vertical inte—
gration and the attitudes of 148 persons who were members
of the channel systems in this study. A number of null hy—
potheses were tested for significance with simple correla-
tions. The results of the analysis are presented in Table
4.3 and are discussed below.

Each of the 27 O-statements is listed in Table 4.3
under one of several headings. Beside each statement are
3 columns which tell: the value of the simple correlation
between integration and the responses to that statement;
the significance level of the correlation; and whether the
correlation was in the direction predicted in hypotheses

3.1 and 3.2.

 

Discussion of the Results of Tests of Hypotheses 3.1
and 3.2

‘—

The statistical form of hypotheses 3.1 and 3.2 is
stated as follows:

3.1 The integration scores (of individuals) will be
positively related to the degree of agreement with
Q statements 1, 3, 59 89 119 139 l4, 157 18, 19,
23, 24, 25, 26, and 27.

 

 

 

 

 

91

w

THE RELATIONSHIP BETWEEN VERTICAL INTEGRATION
AND THE ATTITUDES OF 148 PERSONS

 

Is the Correla-
Simple Level of tion ‘n the
Correlation Statistical Hypothesized
Q-Sorted Attitudinal Statements Coefficient Significance‘ Direction?

 

Statements about Conflict:
Very rarely do my suppliers or cus-
tomers disagree with me on business
matters. —.05 n.s. Yes
2. A big problem with the egg industry
is people—-they're too hard to get
along with. .00 n.s. No
Statements about Cooperation:
3. When

I have a problem, I can most

 

the people I usually deal with. -.O4 n.s. Hes

from anyone in this business these
ays. —.02 n.s. ' No
Statements about Control:

. More profitable egg businesses re—
sult when persons who help with
financing or marketing share in the
decision making. —.11 n.s.

6. Today, egg men are too closely con—
trolled by the people they do
business with. .03 n.s. Yes
Statements about Innovativeness:
7. It is unw1se to begin using a new
product or practice until it has
been proven effective through wide-

spread use. .04 n.s. Yes
8- A man's success often depends on

his acceptance of new ways of doing

things. .03 n.s. No

Statements Which Reflect a Production or
Market ng Orientation:
9. It would be a good thing if the egg
industry would concentrate on pro-
10 duelng more eggs per pound of feed. —.08 n.s. No
- Money spent on egg promotion could

11 stamp out diseases among laying hens. .04 n.s. Yes
. People in the egg industry ought to

ailers and to the public. —.02 n.s. Yes

ret
Statements Concerning the Channel as a
BSEEEEIEELJEEQEE:

 

 

 

 

 

92

12. It is best to forget what 'ot

13. When making most decisions, ry
to consider how they will affect the
I do business with. —.06 n.s.

Yes
Statements Concerning Intra—Channel
Ratings
14. I know that most of the people I
do business with are dependable. —.23 .005 Yes
15. I can trust most of the men that
I deal with. —.17 .037 Yes
16. The majority of the people I deal
with don't know as much about the
egg business as they ought to. .27 .001 Yes
Statements About the Value of Independence:
17. It is better to be your own boss than
to operate under a contract or agree—
ment which ties you to someone. .19 .024 Yes

23. People in this business should work
together with someone who
coordinate all of their production
d marketing efforts. .07 n.s. No
Statements Which Reflect a Traditional
or a Modern Orientatio
8. I usually try ton see the other per—

 

19. I try to be aware of what other

people in this business are doing

in different communities. .06 n.s. No
20. I concern myself with the people

and activities of m‘ own community

rather than those outside it. .04 n.s. Yes

21. Ptopln in this country have become

too specialized. 'J .01 n.s. Yes
22. To be successful in this business

one should stress friendliness and

tradition first, and planning and

cision making sec d .05 n.s. Yes
Statement About knowledge of the Channel
Structure'

 

It is important for me to n

NO

rovement of the Lar er
5. Everyone would be better off if they
would forget their own wishes and
work to accomplish goals which would
help the egg industry as a whole. .03
Often the best solution to a problem
will require that some one individual
must sacrifice for the good of t e
group as a whole. --05
27. We ugh t to work toward a system
which would allow all the people
who produce and market eggs to
make a Ml w<rnt living, regardless
of whrth»: prices are high or low. —.02

n.s. No

0
For Purposes of testing hypotheses 4.1 and 4. 2, those correlations for which

the Significance probabilities were .05 or less were considered sufficient to
reject a null hypothesis.

 

 

 

 

 

 

93

3.2 The integration score (of individuals) will be
positively related to the degree of disagreement

with Q statements 2, 4, 6, 7, 9, 10, 12, 16, 17,
20, 21, and 22.

The null form of these hypotheses is:

There is no significant relation between the inte-
gration scores (of individuals) and the degree of
agreement with any of the Q statements.

Result of Test: Fail to reject all null hypotheses,
except those dealing with the relation between
integration and statement numbers 14, 15, 16,
and 17.

As shown in Table 4.3, the responses of 18 of the 27 state-
ments correlated with integration in the direction predicted.
Of these 18 statements, the correlations of the responses

of statements 14, 15, 16, and 17 were statistically signif—

icant. These results indicate that as integration increases:

(a) channel participants will rate the other members
of their channel as being:

(1) dependable
(2) worthy of trust
(3) competent

(b) persons agree that it is better to operate under
a contract or agreement than to be independent.

At least two factors help to explain the lack of
Significance in the correlations of integration and the
other attitudinal responses. These factors are:

(l) The attitudinal response data were collected
primarily for the factor analysis of hypothesis 5, which
dEals with types or persons. Factor analysis requires nor—
malized data and hence the respondents were forced to sort
the Q-statements into a nine category quasi—normal distri-

bution. The number of statements allowed in each response

 

 

 

94

category was:
response categories: 1 2 3 4 5 6 7 8 9
number of statements: 2 2 3 4 5 4 3 2 2

This procedure required almost half of the responses
to fall within three response categories, and thus limited
the degree to which correlations computed from these data
could differ from zero. This is true because the magnitude
of a correlation coefficient varies inversely with the de-
gree to which the range of one variable has been curtailed.2

(2) The integration scores of all persons in each
channel are based on characteristics of the producer. If
the producer receives a high integration score, so do his
bird and feed suppliers as well as his processor-distributor
and retailer. In a few cases, all of the members of a chan—
nel may hold identical attitudes. In the majority of chan—
nels, however, the differing background and experience of
channel members will be reflected in non—similar attitudes.
When all the members of one channel are given the same inte—
gration score and yet have different and possibly conflict—
ing attitudes, it is likely that the correlation between
integration and these attitudes will be lowo

In summary, two—thirds of the correlations between
integration and attitudinal responses were in the direction
M

2Quinn McNemar, Psychological Statistics (New York:
i2? Wiley and Sons, Inc., Third Edition, 1962), pp. 144,—

 

 

 

 

 

 

——7

95

predicted by hypotheses 3.1 and 3.2. Of these, the associa—
tions between integration and intra—channel ratings of com-
petence, trust, and dependability, and independence as a
value, were statistically significant. The remaining cor—
relations were not significant; this was partially explained
by two factors: the high probability that the members of a
single channel who have the same integration score will have
dissimilar attitudes; and the limited range of allowable

categories for the attitudinal statements.

PREDICTION OF CHANNEL MEMBER COHESIVENESS
The purpose of this section is to discuss the results

of the analysis concerned with the prediction of the future
cohesiveness of a prospective channel member, based on some
known measures of other variables which characterize him.

Cohesiveness, in this analysis, is operationalized in terms
of low conflict and high cooperation. Four null hypotheses
were tested for significance with the use of stepwise dele-
tion multiple regression analysis. The results of this an—

alysis are presented in Table 4.4 and discussed below.

Hypothesis 4.1: Prediction of Conflict Received

 

The statistical form of hypothesis 4.1 is:

4.1 Measures of individual’s operational size, inno-
vativeness, educational level, and age, when em—
ployed as a team of predictor variables explain
a statistically significant amount of variance in
the mean conflict received scores of individuals.

The null form of this hypothesis is:

 

 

 

 

 

 

 

96

 

 

.mflmmbuoawfl Had: mcw poommu ow bcmflUfimew owumbflwcou mm3
mmwa no mo. mH xbflaflflmboud wUGMUflchmHm HmUHbmemwm orb £0H£3 Mom UflbmflHMpm m :<.

 

 

 

Hoo. mm<
mmmc
oao. Im>Hbm>occH mflmmcpoawm cw>ao
mac. mNflm A.mflmv Hasz COmequOdU
woo. mmo. COHbMUdom mo. wuwmwm v.v mHmwrvOme
mmmc
ooo. Iw>me>occH mflmmnpommm
H00. COHMMUSUM Hafiz Uw>kuwm
moo. mNHm A.oflm poco puwmmm COflbmuonou
boo. woo. oa< mm. on Hams m.a mammau0dsm
Noo. mo<
HHo. connmuscm
oao. wNflm mHmme0Q>m cm>H0
mmmc A.mflmv Hasz uboflflmcoo
moo. owo. Im>flbm>OCCH moo. buwnwm N.v mammbboa>m
moo. mo<
woo. wNHm
mHo. coaumusom mammzboaxz om>fioumm
mmmc A.mflmv Hasz IpUHHmcoo
Ono. moo. l®>flbm>occH mmo. pomnmm H.v memnbomhm
meQMHum> EmwB mucmflum> . Uflbmfibmbm umwe mflmmnuOme comm
uowoflomum COHbUHUwum OCHCHMHQXm m mo mammnp0Q>m CH om>OHQEm
w Ham >9 waflm Op mHQmHum> ca Qbocmubm mmeHHflQMQoum mo uazmwm mmwcw>flmmnou
Iflum> Goebwbfluu noboflomum comm mo umouo mUCMUHchmHm mo COHuwbHuU
CH omcflmfidxm mo COHbfioo< oCHUcwummo
mUCMfium> Mo bcmu >2 UMCHMHme CH Umpmflq
ammo .COflbmcHE wocmflum> mmHQMHum>
luwvmo mHQfipHDE wo bcmuumm MObUHUmum
Mo bcmflUHmwmoo
“mil
mmmZmZ QMZZ<IU A<HHZMBOQ

< m0 mmmZm>HmmIOU mmDBDh mIB m0 ZOHBUHDmmm

 

¢.¢ wHQMB

 

 

 

 

 

 

 

97

There is no statistically significant relation
between the predictor variables and the dependent
variable, conflict neceived.
Result of test: Reject the null hypothesis.
This team of predictor variables explains a small
but statistically significant amount of variance in conflict

received. All four predictor variables, acting together,

explain seven percent of the variance in conflict received.

Hypothesis 4.2: Prediction of Conflict Given
The statistical form of hypothesis 4.2 is:

4.2 Measures of individual's operational size, inno—
vativeness, education level, and age, when em—
ployed as a team of predictor variables, explain
a statistically significant amount of variance in
the mean conflict given scores of individuals.

Result of test: Reject the null hypothesis.

Again, all four variables acting together explain

about seven percent of the variance in conflict given.

Hypothesis 4.3: Prediction of Cooperation Received
The statistical form of hypothesis 4.3 is:

4.3 Measures of individuals" operational size, inno—
vativeness, educational level, and age, when em—
ployed as a team of predictor variables, explain
a statistically significant amount of variance in
the mean cooperation received score of individuals.

Result of test: Fail to reject the null hypothesis.
Perhaps the most logical explanation for the fact

that hypothesis 4.3 is not significant stems from the fact

that cooperation received is the mean value of the coopera—

tiveness of an individual as reported by the other persons

 

 

 

r________________________________________i
98

in his channel. These intra—channel evaluations may vary
rather widely from one another, and the variation itself
may have contributed to the lack of significance in hypoth—

esis 4.3.

Hypothesis 4.4: Prediction of Coopegation Given

 

The statistical form of hypothesis 4.4 is:

4.4 Measures of individuals' operational size, inno—
vativeness, educational level, and age, when em—
ployed as a team of predictor variables, explain
a statistically significant amount of variance in
the mean cooperation given scores of individuals.

Result of test: Reject the null hypothesis.

All four variables, acting together, explain about

four percent of the variance in cooperation given.

 

A Summary of the Relationships Which Exist Between the
Criteria of Success and the Predictors of Success

Innovativeness, education, size, and age, when em-
Ployed as a team of predictors, explained a low, but signif—
icant degree of variance in conflict received, conflict given
and cooperation given. A channel captain who wished to fore—
cast the probable future success of a potential channel mem-
ber could use measures of these variables as a first step
in the evaluation process. The predictions of the channel
Captain could possibly be improved if he added other vari-
ables to the prediction team. The candidates' responses
t0 the 27 attitudinal statements discussed earlier, if added
to the set of predictor variables, would perhaps increase

the probability of the prediction of success because responses

 

 

[+7
99

to these statements would yield a measure of a person's

willingness to work with the others in his channel system.

THE DIFFERING ATTITUDES OF TYPES OF PERSONS

The purpose of this section is to discuss the results
of the analysis of the Q-sorted attitudinal statements which
relate to hypothesis 5.1. As a result of the analysis, seven
groups of individuals were isolated in terms of their dif-
fering responses to a set of attitudinal statements. The
analytical procedures employed to generate and describe the
seven factor types were discussed in general terms in Chap—
ter 3 and in detail in Appendices F and G. Basically, the
responses of all individuals were factor analyzed. Of the
several sets of groupings produced by the factor analysis,
one was chosen which contained seven types, each of which
was distinct in terms of degree of integration, channel type,
and channel level. Thus, each group contained individuals
who had similar attitudes, and who participated in the same
type of channel; one group might be made up of people in
Contract channels, another group might be composed of inde—
pendent operators, and so on. Each of the seven types of
Persons was then described in terms of attitudinal responses.
The description of each factor type, in terms of the persons
who make up each factor, and in terms of attitudinal response

patterns, is summarized in Table 4.5 which is explained

be10w.

   

 

 

lOO

Explanation of Table 4.5
Table 4.5 describes each factor in terms of: (a)

physical characteristics, such as degree and means of inte—

 

gration, most representative channel type, and most repre-
sentative channel level, and (b) attitudinal responses.
Each factor is described in terms of the attitudinal state—
ments to which the persons in that factor agreed or disagreed
nmre strongly than did the persons in any other factor. A
column designated by the number of each factor contains the
words 'agree' or 'disagree‘ beside those statements which
most strongly describe each factor. The term ‘agree' is
used to describe a factor when that factor agreed with a
particular statement to a greater extent than did the other
factors. Similarly, the term 'disagree' is used to describe
a factor when that factor disagreed with a particular state—
ment to a greater extent than did the other factors.

Each of the 27 statements was most strongly agreed
with by one factor and most strongly disagreed with by an—
other factor. Thus, beside each statement the words 'agree'

and 'disagree' each appear once.

Description of Each Factor Type

 

Factor 1, 2, and 3, at the low end of the integra-
tion continuum, will be discussed first. Then, factor 4,
WhiCh lies at thelmiddle of the continuum, will be described.
Last, factors 5, 6, and 7 at the high end of the integration

will be discussed. Thus, the structural characteristics of

 

 

 

 

 

 

101

Table 4.5

 

REPRESENTATIVE TYPES OF PERSONS IN EACH FACTOR TYPE, AND ATTITUDINAL

RESPONSE PATTERNS WHICH MOST STRONGLY CHARACTERIZE EACH

Factor Types, In Ascending Order of Integration

 

Description' of Each
Factor In Terms of:

 

 

1. Integration

2. Means of integration.
'f

not independent

3. Channel type
4. Channel level

Description of Each
Factor In Terms of
Responses to Attitu—
dinal Statements:"

1. Very rarely do my
suppliers or cus—
tomers disagree with
me on business
matters.

2. A big problem with
the egg business is
people-~they're too
hard to get along
with It.

 

3. when I have a prob—
I can most often

get good advice from
one of the people I
usually deal with.“

4. It's tough to get
much cooperation from
anyone in this busi-
ness these days.“’

5. More profitable egg
businesses result

decision making.“‘

6. Today, egg men are

business with."'

7. It is unwise to begin
using a new product or
practice until it has
been proven effectiv
through widespread u F."’

 

 

 

 

 

 

 

 

 

agree

2 3 4 5 6 7
non low low to medium medium medium high
inte— medium to high
grated
60—day con— con— con— contract contract
sponsored agree— tract tract tract & owner— & owner—
quality ment inte— inte—
control with grated grated
program co—op
B C E no F G & H S
,dominant
channel
pro- pro— proc.— pro— pro- bird pro—
ducers ducers dist. ducers ducers, ducer
retailers retailer
agree dis—
agree
agree dis—
agree
dis— agree
agree
agree dis—
agree
dis— agree
agree
agree dis—
agree
agree dis—

 

 

 

 

 

 

 

102

B. A man's success depends

on his acceptance of new dis— agree
ways of doing things."' agree

9. It would be a good .
thing if the egg indus— agree dis—
try would concentrate agree

per pound of feed."‘

10. Money spent on egg
promotion could be

better spent on re— dis— agree
search to stamp out agree

diseases among laying

hens.

 

 

 

11. People in the e g

harder at promoting agree dis—
ggs to retailers and agree
to the pubi

 

this buSiness might agree dis—
think of me when I make agree
my own decisions."‘

 

13. When making most de—
cisions, I try to con-

 

 

sider how they will I agree ag::e
affect the people I do
business with
14. I know that most of
the people I do busi— is— agree
ness with are depend— agree
able
15. I can trust most of dis— agree
the men that I deal agree
with. "'

16. The majority of athe
d

with don't know as agree dis—
much about the egg agree
business as they
ought to.

17. It is better to be
your own boss than to

operate under a con- agree d15-
tract or agreement agree
which ties you to

some eone.

18. I usually try to see
t

dis-

a ree
agree 9

 

 

 

 

 

always try to see
i {3.00.

 

 

 

 

 

19. I try to be aware of

. dis— a ree
this business are agree g

Communities."

   

 

 

 

 

[+7
103

20. I concern myself
with the people and _
activities of my own agree dis
comm ty agree
those outsi e t.

C
:3
p
'1
DJ
rt
3‘
m
'1
n.
3"
[ll
5

21. People in this coun—
have become too agree dis—
specialized."' agree
22. To be successful in
this business one
should stress friend— dis— agree
liness and tradition agree
first, and planning
and decision making
second.
23. People in this busi—*

ness should work
together with someone dis— agree
who can coordinate all agree
of their production &
marketing efforts."‘
24. It is important for
me to know who handles
e ggs, from pro— agree dis-
agree

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

the consumer's table.

25. Everyone would be

 

would forget their
' agree dis—
agree

egg industry as a

w o e.

26. Often the best i
solution to a problem I
will require that some dis—
one individual must
sacrifice for the agree
good of the group as
a whole.“‘

27. We ought to work
t

 

 

 

agree

 

dis— agree
market eggs to make a agree

 

 

 

 

 

are high or low."'

 

less of whether prices J

 

‘Descriptions of channel type and channel level are in terms of the channel type and
Channel level category which contained the greatest number of people in given factor.
n ‘ ‘ f i L 3 L‘ re based on the median integration value of the persons

r o
in each factor.
“Responses to each statement were in terms of a point along an agreement—disagreement
Continuum. The arm 'agree' is used to describe a factor when that factor agreed with
a particular statement to a greater extent than did the other factors. Similarly, the
term 'disagree' is used to describe a factor when that factor disagreed with a particular
Statement to a greater extent than did the other factors.
"‘The relation between vertical integration and the responses to this statement is
Predicted in statistical hypotheses 4.1 nd 4.2. The responses to twenty of the twenty—
seven statements were in the predicted direction.

 

 

 

 

E::T_______________________________________i

104

each group of channels, as well as those statements which

most strongly characterize that set of channels, will be

discussed, starting with the least integrated groups, and

progress toward more highly integrated ones.

Factor 1. This is the least integrated of all the groups.

It is represented by channel type B, which is composed of

independent producers who traditionally have purchased feed
from and sold eggs to a combination feed mill and process—
ing planto The processor sells eggs to independent jobbers
who sell to independent retailers. The attitudinal responses

of the persons in the channels which make up factor 1 are:

Do not View the channel as reference group
Value independence as opposed to ties with a system
3. Localite——concerned with own community

Are opposed to a procedure by which price risk would
be borne by all members of a system

5. Low level of disagreement among channel members
Knowledge of channel structure is important for

 

60
these persons
Factor 2. This factor is slightly more integrated than the

Preceding one. It is composed of several previously inde—

pendent producers who still favor the value of independence,
bUt have linked themselves by means of a short term written
agreement to a cooperative which sells feed and processes

8995- The following rather independent—minded attitudes

describe factor 2:
Dislike centralized coordination or production and

la

marketing efforts
2. Knowledge of the channel structure is unimportant
3o Disagree with the concept of individual sacrifice

for the good of the group as a whole

 

 

 

 

EZZT_—————————————————————————————————————"
105

Favor a procedure by which price risk would be borne

4.
by all members of a system

Factor 3. This factor is composed of members of the least
integrated of the contract channels, and the members are

more integrated than those in factors 1 and 2. Usually,

a contract links one or a few producers to one of several
which have often resorted to contracts

small feed companies,
The

as a means of collecting past due accounts-receivable.

eggs are sold to small processors, who in turn sell to any

retail outlet who will buy. The attitudes of the persons

in factor 3 are:
l. Disagree with centralized coordination of the chan-
nel

2. Not innovative
3. Low ratings of intra-channel competence

4. Are concerned with activities outside the local

community
Consider the effect of decisions on the rest of

the channel

5.

 

This factor type is the least homogeneous of any

Factor 4.
‘

Of the seven° It contains one or a few persons from several

channel types and levels and although some of the attitudinal

reSponses which describe it form a pattern, others are con—

fusing and contradictory. This, however, was expected be—

Cause it is generally true in type analysis that the types
at either end of a continuum will be relatively well defined
and unambiguous, and those types, such as factor 4, at the

Center of a continuum tend to be heterogeneous and difficult

to describe.

 

 

———i '

106

Factor 5. Factor five is principally composed of persons

in channel type F, which employs production contracts to
link one privately owned feed mill with several egg pro—

ducers° These eggs flow through one processor to one major

retail chain. The high degree of efficiency and smooth op-

eration of this moderately integrated channel observed as

the data were collected is reflected in the following atti—

tudinal description of factor five:

1. Low intra—channel conflict
2. High intra—channel cooperation
3. High innovativeness

4. Market-orientation

5. Perceive the channel as a

decision making
6. Agree with the concept of individual sacrifice

for the good of the group as a whole
7. Tradition-orientation, as reflected by emphasis
on friendship and tradition

reference group in

 

Factor 6. This factor is composed of the most highly inte-

grated persons who operate in the North.

 

They participate

in two types of channels; one in which several producers

are contractually linked to a cooperative which operates

its own processing plants, and markets its eggs to two major
retail chains; and (2) one composed of owner—integrated op-
erators who have larger than average flocks, operate their
Own feed mills, and may own additional channel level facil—

ities. The following progressive attitudes describe this

faCtor:

1. High innovativeness

2° Market orientation
3° High ratings of intra—channel trust

 

 

 

 

107

4. Modern orientation versus traditional orientation
as expressed by

(a) high empathy
(b) cosmopoliteness-—interested in egg industry
activities in other locations
(c) support the value of occupational specialization
Factor 7° This most highly integrated group of persons is
located in the South and participates in contract or owner—
integrated channels. Factor seven is described in terms of
the following responses;
1. A high degree of intra-channel cooperation
2. Agreement that existing intra—channel control is
not excessive
3. High ratings of intra—channel dependability and
competence
On the other hand, a few of the responses of persons in fac—
tor seven were surprising in that they indicate:
19 A production orientation, rather than a marketing
orientation
2. An unwillingness to subvert their own wishes and
work to improve the whole egg industry
A Summary of the Differing Attitudes of Persons
Hypothesis 5 appears to have been substantially borne
out by the data collected in this study. The responses of
members of similar channels tended to correlate highly with
one another, especially when the attitudes of the persons

at either end of the vertical integration continuum were

considered. For instance, the most highly integrated per-

 

SODS (in Factors 6 and 7) were described in terms of: high
intra—channel cooperation; acceptance of intra—channel con-

tr01; high ratings of intra—channel dependability, competence

n___

 

 

 

 

7
108

and trust; a market orientation; a modern orientation ex—
pressed in terms of high empathy, cosmopoliteness, and agree—
ment with the value of occupational specialization; and high
innovativeness. On the other hand, the least integrated
persons (in Factors 1 and 2) were described in opposite

terms: fail to view the channel as a reference group; value
independence; localite; dislike centralized channel coordina-
tion; disagree with concept of individual sacrifice for the

good of the system.

THE DIFFERENT BEHAVIOR PATTERNS OF CHANNEL TYPES

The purpose of this section is to discuss the results
of the analysis of hypothesis 6.1. The analytical proced-
ures employed to generate and describe channel types were
similar to those employed to analyze the data for hypothesis
5.1. Mean measures on conflict, cooperation, innovativeness,
age, and education for each of 35 channels were factor an—
alyzedo

Of the several sets of channel groupings produced
by the factor analysis, one was chosen which contained four
tYpes. Ideally, each of the four factor types should have
been distinct from the other factors, in terms of vertical
integration. The factors generated in the present analysis
were relatively heterogeneous on integration-—each factor
Was somewhat like the others. The description of each fac~
tor type in terms of the channels which make up each factor

and in terms of behavioral variables is summarized in Table

4.6 and explained below.

 

 

 

 

 

109

Table 4.6

BEHAVIORAL DESCRIPTION OF CHANNEL TYPES

 

 

 

 

 

 

 

Channel Types in 1
Order of Integration
1 2 3 4 ‘
Integration Range‘ Low Low to Medium High
medium to
high
Behavioral Description:’*
1. Innovativeness High Low
2. Education High Low
3. Age Low High
4. Conflict Given Low High
5. Conflict Received LOW High
6. Cooperation Given High LOW
7. Cooperation Received High Low

 

 

 

 

 

 

‘Descriptions of integration are based on the
median integration of channels with positive factor load—
ings.

‘*Measures on each variable were standardized to
insure comparability. These measures range from low to
high. The term 'low' is used to describe a factor when
the channels with positive loadings on that factor had
lower mean measures on a particular variable than did all
other factors. The term 'high‘ is used to describe a fac—
tor when the channels with positive loadings on that factor
had higher mean measures on a particular variable than did
all other factors.

 

 

Ji___

 

 

 

 

110

The four factors in Table 4.6 are listed in order
of integration from the lowest to the highest, based on the
median integration score of the channels in each factor.

As shown in Table 4.6, each variable strongly de—
scribes two factor types. For instance, innovativeness most
strongly describes Factors 2 and 4. Factor 2 is described
as highly innovative and Factor 4 is non—innovative. The
description of each factor becomes slightly more complicated
by the fact that within each factor in this analysis some
channels have positive factor loadings and others have nega—
tive loadings. The difference means that the two sets of
channels within a factor are opposite with respect to each
0f the variables which describes that factor. Thus, within
Factor 2, one group of channels is highly innovative and
another group is low on innovativeness. Also, within Fac-
tor 4, one group is highly innovative and one group is not.
Thus innovativeness (and each other variable) describes four
groups (A, B, C, and D) of channels in two factors. This

may be depicted as follows:

Variable: Innovativeness

Sign of Factor Loading

 

 

 

 

Positive Negative
Group A Group B
2 innovative non—innovative
4 Group C Group D
innovative non—innovative

 

 

 

 

 

 

L___ i

 

 

111

The results of the channel type analysis will be
presented in two forms. First, each of the four factor types
will be described in terms of behavioral variables. Second,
the types of channels which describe each variable will be
isolated, in order to show the extent to which integration

appears to be related to each variable.

Behavioral Description of Each Factor Type

Each of the four factors will be described in terms
of its strongest behavioral characteristics, starting with
Factor l——the least integrated, and ending with Factor 4——
the most highly integrated.

Factor 1. The behavior of the two differing groups

Of channels in Factor 1 can be illustrated as follows:

 

; Low Education High
LgHigh Conflict Received Low ‘

 

The channel which is de—
scribed in terms of this
end of the behavioral
continuua is:

(a) a contract channel of
medium integration in
which several producers
are linked to a pri-
vately owned feed
company

Those channels which are
described in terms of this
end of the behavioral
continuua are:

(a) non-integrated,
independent channels

(b) an owner—integrated
channel of medium
integration

Factor 2. The behavioral description of the two

differing groups of channels in Factor 2 may be illustrated

as follows:

 

 

 

 

 

112

 

 

 

 

 

AyLow Innovativeness High
< Old Age YoungA

High, Conflict Given Low
tee-e-r 47

1 Low Cooperation Given High
< Low Cooperation Received High 7

The channel which is described

in terms of this end of the
behavioral continuua is:

(a) southern, owner—
integrated channel

Factor 3.

behavioral continuua:

Those channels which are
described in terms of this
end of the behavioral
continuua are:

(a) a few independent chan-
nels of low integration

(b) a channel containing
a large number of
producers who are
linked to a coopera—
tive by a 60 day
written agreement

(c) a channel containing
a large number of con—
tract producers who
are linked to the
cooperative mentioned
in (b) above.

Factor 3 is described in terms of these

 

 

 

 

 

 

 

 

4 Low Education High :
/ Old Age Young

4 Low Cooperation Given High_>
1 Low Cooperation Received High :
\

Those channels which are
described in terms of this
end of the behavioral con-
tinuum are:

(a) non—integrated, inde—
pendent channels

(b) contract channels which
are slightly integrated

Those channels which are
described in terms of this
end of the behavioral con-
tinuum are:
(a) highly integrated
southern contract
and owner integrated
channels

 

 

 

 

 

EI:I:_____________________________________i

 

 

113
Factor 4. The behavioral variables which most strongly
describe the channels in factor 4 are:
e Low Innovativeness Highyg
High Conflict Giv§n__fi Low
é High Conflict Regeived Low .s

 

Those channels which are
described in terms of this
end of the behavioral
continuum are:

(a) non-integrated inde—
pendent channels

(b) contract channels which
are slightly integrated

Types of Channels Which Are
Described by Bach Variable

Those channels which are
described in terms of this
end of the behavioral
continuum are:

(a) southern channels based
on either contracts or
owner—integration

As stated earlier, each variable describes four

groups of channels——two which are "high" on the variable,

and two which are

"low" on it.

More about the relation be—

tween vertical integration and each variable may be learned

by examining the integration scores of each of the four groups

Of channels which were described by each variable in the

type analysis discussed above.

This information is shown

in Table 4.7, and explained below.

Explanation of Table 4.7.

The relation between ver—

tical integration and each variable is shown in Table 4.7.

For each variable, the median integration scores of the 2

groups of channels that were:

and (b) highest on that variable are given.

(a) lowest on that variable;

For instance,

the two channel groups which were least innovative had median

 

 

 

 

 

 

114

Table 4.7

MEDIAN INTEGRATION SCORES OF CHANNELS

THE EXTREME ENDS OF EACH BEHAVIORAL CONTINUUM

 

l—"

Median Integration
Score of

Channels Described
by Low End of

 

Median Integration 1
Score of

Channels Described

by High End of

 

 

 

 

 

 

 

 

 

 

 

 

Factor Continuum for Continuum for
Number Each Variable Variable Each Variable
2 100 Innovative— 39
ness

4 50 97.5

1 10 Education 56

3 41 74

2 100 Age 39

3 41 74

2 39 Conflict lOO
Given

4 97.5 50

l 56 Conflict 10
Received

4 97.5 50

2 100 Cooperation 39
Given

3 41 74

2 100 Cooperation 39
Received

74

 

 

 

 

 

 

 

 

EI::____________________________________i
115

integration scores of 100 and 50. The two most innovative
channel groups had integration scores of 39 and 97.5. Thus,
the relation between integration and innovativeness, as shown
by the type analysis is mixed.

As shown by the type of analysis of channels, the
relation between integration and age, conflict given, coop—
eration given, and cooperation received is also mixed. In
other words, there were both high and low integrated chan—
nels at each end of the continuum for each of these variables.

Clear relations did emerge for education and conflict
received, as shown by the fact that highly integrated were
highly educated and displayed low conflict received, whereas

low integrated channels were described in opposite terms.

A Summary of the Differing Behavioral
Descriptions of Channel Types

Hypothesis 6.1 must be rejected. It was found that

 

Channels which correlated highly with one another on the
basis of their scores on each variable were characterized
by Widely differing integration scores.

The integration scores of each of the four factor'
types were quite similar and when behavioral variables were
used to describe the groups within each factor the results
were contradictory.

This means that often channels in a given integra_
tion range are dissimilar. Highly integrated southern chan—
nels may behave more like low integrated northern ones than

like highly integrated northern channels. Similarly, some

 

 

 

 

 

116

contract channels in the North appear to behave much like
some northern independent channels. More will be said about

this in Chapter V during the 'Evaluation of Competing Chan-

nel Systems.‘

 

 

 

 

 

 

CHAPTER V
SUMMARY OF RESEARCH APPROACH AND FINDINGS, EVALUATION OF

COMPETING CHANNEL SYSTEMS, AND CONCLUSIONS

INTRODUCTION
The purposes of this chapter are: to summarize the
approach and findings of this research project; to evaluate
the characteristics of three types of competing channel sys-

tems; and to draw conclusions on the basis of this analysis.

SUMMARY OF RESEARCH APPROACH
Purposes of the Study

While the present trend toward increasingly inte-
grated production-marketing systems has been due in part
to economies of scale which produce lower costs, the increase
in the efficiency of such systems may also be the result
of certain attitudinal and behavioral patterns displayed
by the members of these systems.

The purposes of this study were: (1) to study the
relationships between vertical integration and selected at—
titudinal and behavioral variables in competing channel sys—
tems; and (2) to develop operational measures for each of

the channel system variables under study.

Literature Reviewed
Chapter II reviews selected literature which deals

With the concepts and variables studied in this research:

117

 

 

 

 

7
118

vertical control, vertical conflict, vertical cooperation,
innovativeness, and vertical integration. These variables
are defined in Appendix A. Although many persons have con—
tributed economics—oriented, or general and non—operation—
alized publications, few people have reported empirical chan-
nel analyses.

Such authors as Alderson, McCammon, Little, Fisk,
Breyer, Ridgeway, and Stern have argued that a marketing
channel is an operating system with an identifiable pattern
of behavior. In discussing the systemic nature of channels,
McCammon and Little note that a channel is a set of inter-
related components which undertake sequential activities
in order to attain mutually acceptable objectives. Partic—
ipation by the components is voluntary; behavior of the chan-
nel members is regulated by a code; and the activities of
the components of the system are frequently regulated by
some single entity in a position of centralized control.

The contribution of vertical control to channel ef—
ficiency has been discussed in the literature by several

persons. While Little and Mallen suggest that some central

 

direction must exist in channels if their members are to
be efficient, others have suggested that authoritarian re-
lationships within channels will not invariably improve chan-
nel performance.

Vertical conflict has also received considerable
attention in the marketing literature. Palamountain, for

instance, studied conflict in the distribution of several

 

 

 

7
119

commodities, focusing on the legislative efforts of the mem—
bers of these channels to remove sources of conflict. The
role assumed by various trade associations in the political
resolution of channel conflicts was recently studied by
Assael. The causes and results of conflict were extensively
reviewed by Stern and Gorman, who undertook an analysis of
vertical conflict which incorporated the work of many social
scientists, several of whom had used empirical data to sup-
port their hypotheses.

The effect of vertical cooperation on the existence
and effectiveness of channel systems was discussed by Baligh
and Richartz. These authors suggest that marketing channels

cannot function without the sustained cooperation of all

 

the parties in a channel.

 

 

A high degree of innovativeness among the members

Of a channel system may lead to reduced costs and increased

 

efficiency and may therefore partially determine how compet—
itive that channel is. McCammon feels, however, that most
innovations do not come from channel captains or from com—
ponents within a given system. Rather, the innovator is
usually a firm outside of the established channel structure.
A review of the most recent Bibliography on the Diffusion

0f Innovations, however, reveals that no studies have been
Undertaken to analyze the relationship between integration
and innovativeness.

While the increase in the efficiency of production—

marketing systems may be the result of certain attitudinal

 

 

 

120

and behavioral patterns displayed by members of these sys—
tems, these behavioral dimensions have not been studied em-

pirically until this research project was begun.

Research Desigp and Methodology

Chapter III discusses the general, statistical, and
null hypotheses developed to guide the analysis of the data
collected for the study; the development of the research
instruments, the field procedures employed in the study; and
the methods used to analyze the data. Each of these topics

will be reviewed below.

General Hypotheses. Six general hypotheses concern—
ing vertical integration were developed. Briefly, these are:

1. As vertical integration increases in marketing
channels these channels will tend to be characterized
by: a higher amount of cooperation, a lower amount
of conflict, a greater degree of innovativeness,
larger operational units, younger persons, and more
highly educated persons.

2. When comparing those channels which are located

in the South rather than in the North, we will find:

a greater amount of control, a lower amount of conflict,
a greater degree of cooperation, larger operational
units, a higher rate of innovativeness, younger persons,
and more highly educated persons.

3. Certain attitudes of persons will be related to
the amount of integration which characterizes the
Channels in which they operate.

4. Knowledge of an individual's operational size, in—
novativeness, educational level, and age will allow the
prediction of the amount of conflict and cooperation
which would tend to characterize that person's channel
relationships.

5. Members of similar channels will tend to group to-
gether into 'types' based on their attitudinal responses°

 

 

 

 

 

121

6. Channels which are characterized by similar amounts
of integration will tend to group together into 'types'
on the basis of a number of measures on selected be—
havioral variables.

Several statistical hypotheses were developed for
general hypotheses l, 2, 3, and 4. Each statistical hypoth—
esis specified the means of operationalizing each variable
and the direction expected in the relationship between vari—
ables. A null hypothesis, corresponding to each statistical
hypothesis, was stated and tested for statistical signifi—
cance at the .05 level.

As explained in Chapter III, no statistical or null
hypotheses were formulated for general hypotheses 5 and 6;

these general hypotheses were tested with non-inferential

'type' analysis of attitudinal and behavioral data.

Research Instruments. This study employed personal
interviews with 176 egg industry members who operate at five
different channel levels: feed supply, bird supply, egg
production, processing—distributing, and retail. Because
of the differing activities performed by respondents at each
level, it was necessary to develop a separate research in—
strument for each level.

Each research instrument has two sections. The first
Of these is composed of questions designed to provide data
about the dependent variables. These five research instru-

ments differed from one another with respect to the exact

questions asked of the reSpondents at each channel level.

However, each research instrument is similar in that the

 

 

 

 

 

122

questions relate to the same variables.

The second part
of each research instrument contains a set of 27 Q—sort
statements which each respondent sorted according to his

degree of acceptance of each statement.

This procedure was
designed to elicit responses which reflect the attitudes

of the respondents.

The same statements were included in
all five research instruments.

These Q-sort statements are
reproduced in Appendix B.

Operationalization of Variables. To provide data of

the interval level of measurement required by the assumptions
of the parametric statistical tests employed, each variable
was rigorously operationalized with the use of several in—
ternally consistent, closed—ended questions. The specific
means of operationalizing each variable is discussed below.
Control, Conflict, and Coopgpation. A quantitative
measure on each of these variables for each person was deter—

mined by computing the mean value of his responses to the

Innovativeness.

 

questions selected to operationalize each of these variables.

The standardized innovativeness
score for each person was determined by:

first, reflecting
{subtracting from the year 1969) the year of adoption of
each practice included in the item~analyzed innovativeness
scale; second, computing the mean of these reflected scores;
and third, converting this raw score to a standardized score
50 that the innovativeness of a person at one channel level

Could be compared with that of a person operating at a dif-
ferent channel level.

 

 

 

 

 

 

E:::_______________________________7

123
Operational Size.

The size of each business unit1
operating at each of the five channel levels was measured

in terms of its volume of physical units handled, i.e.,
of feed, pullets sold,

tons
hens housed,

cases of eggs processed,
and cases of eggs sold.

These raw measures were standard—
ized with a normalization approach to allow the comparison
of size across channel levels.

Vertical Integration.

An index of vertical integra-
tion which applies to all the firms within a particular chan—

nel is determined by the extent

to which the following fac—
tors characterize each channel:

(a)
(b)

Vertical control is exercised over the producer

Price risk is shifted from the producer to others

in the channel

(c) The firms in the channel are committed to remain
in that particular channel, and

(d)

The production—marketing functions are assumed
by one firm within the channel

Internal Consistency of Questions.

The unidimensionality
of the questions was determined by factor analysis; the fac—

tor loading of each question on a common factor indicates
how strongly each question is like the dimension which under—
lies the questions in its group. Questions with high factor

loadings of the same sign on a common factor were selected
to index each variable.

Operationalization of Attitudes. The selected attitudinal
______________

Size is measured in terms of volume of business done
by each entire business firm, rather than volume of a partic—
U1ar unit at one location.

 

 

 

 

 

 

124

dimensions involved in this analysis were operationalized
and measured with Q methodology,

a group of psychometric
procedures which involve correlating the responses of dif—

ferent individuals to a set of statements which have been

sorted into several response categories.

The statements
included in the Q—sort are the result of a selection process

which started by specifying a number of attitudinal dimen—
sions such as production orientation versus marketing ori—
entation;

group;

ability to perceive the channel as a reference
and so on,

and then constructing statements about
these dimensions.

A large number of statements were gener—
ated about each dimension and many were eliminated after

being screened for clarity and lack of ambiguity. This Q
sort was structured, in that it was balanced to include a
few statements about each variable and it included both pos-
itively and negatively phrased statements.

Each respondent sorted these statements, according

to the extent to which he agreed or disagreed with each one,

into a specific number of rank ordered pileso

This forced
the responses into a quasi—normal distribution in order to

 

standardize the data and to insure transitivity as required

by the assumptions of the factor analysis procedure used
later.

Pilot Study.

Prior to embarking upon the full field

StudY, a pilot study was undertaken to test the research
instrument and the interviewing procedures“

As a result,

 

 

 

 

|_’—'

125

it was determined that the questions included in the instru—
ments were understandable and relevant to the operations of

each of the types of respondents. Each of these pilot study
interviews was completed within the 45 minute time limit set

as the target.

Selection of Respondents. 176 persons were interviewed-—
138 interviews were conducted in the North and 38 in the
South. A quota sample of persons was employed to represent
persons in independent, contractually integrated and owner—
integrated channels, in proportion to the volume of business
handled by each of these types of channels in the two geo—
graphic regions under study. Because of the limited number
of feed suppliers, bird suppliers, and egg processors and
distributors in Michigan, the respondents at these channel
levels constitute almost a complete census of the population.
Entire production-marketing channels were traced by
asking each egg producer, selected from a list developed by
Agricultural Extension personnel, to name his feed supplier,
bird supplier and processor. Each processor was asked to
name the principal retailers to whom he sold his eggs and
the largest of these retailers was selected to be interviewed.
Field Procedure. The channel is the main unit of

analysis in this study and was defined as a group of business

 

firms which supply inputs to, or process, distribute, and
retail eggs which come from a producer. The producer was

the determinant of a channel; once a producer was selected

 

 

 

 

126

into the study, the people with whom he dealt were automat—
ically included.

The respondents were located in two geographic re—
gions: the North, including Michigan, Indiana, Ohio and
New York; and the South, including Georgia, Mississippi,
North Carolina, and Arkansas. Northern respondents were
interviewed by the author while the remaining interviews
were conducted by Agricultural Extension personnel employed
by state universities in the four southern states.

Analysis of Data. The procedures employed to an-
alyze the data relevant to each hypothesis are presented
in Table 5.1. The data analysis procedures included: simple
correlations, t tests of the differences between two means,
stepwise multiple regression, and factor analysis of 'types'
(P analysis). The .05 level of statistical significance
was used to reject null hypotheses which were tested with
the first three analytical procedures.

Both persons and channels were used for statistical
analysis. The reasons for the use of each population are
discussed below, by hypothesis number.

Hypothesis 1. Data from individuals as well as from
channels were used in order to help to isolate the effects
Of the interaction between the highly integrated southern
scores and the low and medium integrated northern scores

on the values of the correlation coefficients.

 

 

 

 

 

 

127

Table 5.i

STATISTICAL PROCEDURES USED lN THE

ANALYSIS OF HYPOTHESES

 

 

General Content of Statistical Unit of Analysis
Hypothesis General Procedure Each Entire
Numbers Hypothesis Employed Person Channel
1 Integration Simple X X
as related correlation

to dependent
variables

2 Geographic t test of X
location as difference
related to between
dependent two means

variables

3 Integration Simple X
as related correlation
to Q
responses

4 Cohesiveness Multiple x
prediction regression

5 Type analysis Factor X
based on analysis
Q—sort

6 Type analysis Factor x
based on analysis
dependent

variables

 

 

 

 

 

 

 

 

 

 

 

 

 

_ 'l
1 28 '\
Hypothesis 2. Mean channel data were used, in order
to allow a comparison of entire competing channel systems

across differing locations.

Hypotheses 3 and 5. These hypotheses dealt with

 

the attitudes of individual persons. To combine these data
across all persons in each channel would have caused a re-

gression of existing extreme individual differences toward
some, less meaningful, average value. Thus individual data
were analyzed.

Hypothesis 4. The goal of the analysis concerned
with hypothesis 4 was to determine the extent to which the
intra-channel cohesiveness of a prospective channel partici—
pant could be predicted from knowledge of certain aspects
of his behavior. Since the cohesiveness of individuals,
rather than that of whole channels is being predicted, data
from individuals were analyzed.

Hypothesis 6. Mean channel data were analyzed be-
cause the goal of this analysis was to determine the extent
to which persons in similarly integrated channel systems

behaved alike.

 

SUMMARY OF FINDINGS
Chapter I posed six questions to be answered by this
StUdy. This section will indicate how the findings answer
each of these questions.

1. Is there a relationship between the amount of
integration in a channel system, and the:

 

 

 

 

 

7
129
(a) amount of cooperation and conflict displayed
by its members?
(b) speed with which its members adopt new ideas
and practices?

(c) size of the business units in the channel?

(d) age of its members?

(e) education of its members?

la. Both conflict given and conflict received were
significantly reduced as vertical integration increased.
Cooperation given increased significantly as vertical inte—
gration increased. And, although not statistically signif—
icant, there was a tendency toward higher cooperation received
in the more highly integrated channels. Thus, it appears
that more highly integrated channels operate more smoothly
than their less integrated rivals, that is, they contain
people who get along better with one another.

lb. A significant relation existed between integra-
tion and innovativeness. However, contrary to the expecta-
tions of the researcher, highly integrated channels tended
to be significantly less innovative than less integrated
channels. As explained in Chapter IV, this finding appears
to be an artifact, caused by the fact that normal methods
Of measuring innovativeness are time related, and vertical
integration in the South is relatively new. Thus, before
a more precise statement of the relation between these vari—
ables can be made, new methods of measuring innovativeness
are required.

lc. As integration increased, channels were composed

of only slightly larger business units. As explained later

 

 

 

 

 

 

 

EI::T___________________________________i
130

in this chapter in the 'Evaluation of Competing Channel Sys—
tems,’ the relationship between integration and operational
size varies widely among channel levels.

ld. Age was not related to integration in a clear—
cut manner, perhaps because of the very narrow band of ages
of the respondents in the sample. When data on age from
channels were correlated with integration, no relation was
found. However, a negative correlation between age and in—
tegration emerged, as predicted, when data from northern
individuals were tested separately.

le. A positive significant correlation between edu—
cation and vertical integration resulted from the tests of
data from channels as well as from tests of data from northern
individuals.

2. Do the attitudes and behavior of channel members
seem to vary depending on geographic region?

As shown by t tests of the differences between

 

northern and southern channels, it was found that northern
channels were characterized by significantly higher amounts
of conflict received, cooperation given and received, sig—
nificantly higher innovativeness scores, and significantly

. 2
lower education scores.

‘~———_——_—

2Conflict and cooperation need not be opposites.
AS Operationalized in this study, cooperation describes
'help‘ or 'assistance given,‘ whereas conflict occurs when
'dissatisfaction’ arises because the expectations of one
person have not been fulfilled by the actions of a second
98rson.

 

 

 

 

 

 

131

3. Is there a relationship between the amount of
vertical integration in a channel system and the
attitudes held by the members of that system?

A positive and significant relationship was found
between integration and the responses to four statements.
These responses indicated that highly integrated respondents
tended to believe that the persons in their channels were
dependable, trustworthy, and competent. These persons also
agreed with the concept of contractual integration versus
independent operation.

4. Is it possible to predict the cohesiveness of an
individual with the others in his system before he
joins that system?

The analysis revealed that the conflict given and

received and the cooperation given score of an individual

 

could be partially predicted on the basis of that person's
scores on innovativeness, education, age, and operational
size. These four predictors explained 7 percent of the vari—
ance in the criterion variables.

5. Do the members of similar channels have common
attitudes?

Yes. The analysis revealed that the most highly
integrated types of people were characterized in terms of
the following attitudinal responses: High intra-channel
cooperation; acceptance of intra-channel control; high rat-
ing of intra—channel dependability and trust; a market ori—
entation; a modern orientation expressed in terms of high

empathy, cosmopoliteness and agreement with the value of

OCCUpational specialization; and high innovativeness.

 

 

 

 

132

The least integrated types, on the other hand, were
described in terms of the following opposite set of attitud—
inal responses; failed to view the channel as a reference
group; value independence; localite; dislike centralized
channel coordination, and disagree with the concept of indi—
vidual sacrifice for the good of the system.

6. Do channels which are alike in terms of integration
tend to be alike in terms of other variables?

No. While a few channels of medium and high inte-
gration were described in terms of:

1. high innovativeness

2. low conflict given and received

3. high cooperation given and received

4. young ages
and while some non—integrated and low-integrated channels
were described in opposite terms, there was no consistent
pattern of behavior that applied to the majority of highly
integrated channels or to most non-integrated ones. Thus,
channels which were similar in integration did not appear
to share similar patterns of behavior, even though some sig-
nificant correlations between integration and these variables
were found in the analysis of Hypothesis 1.

The first objective of this study—-to study the re—
lationships between vertical integration and selected atti—
tudinal and behavioral variables in competing channel sys—
tems——was achieved relatively successfully. Several hypoth—
eses drawn from suggestions in the literature were supported

by the statistical findings. Some hypotheses were not borne

 

 

 

 

 

 

133

out—-perhaps because of the complexity of the behavior be-
ing analyzed.

The second objective of the study—-to develop opera—
tional measures for each of the channel system variables
under study——was achieved with an even greater degree of
success. These variables were defined, operationalized,
and measured to provide the level of data necessary for the
analytical methods employed.

The fact that this analysis did not bear out all
hypotheses points up the need for more sophisticated theory
and analysis. Based on the quantified data gathered to test
the six general hypotheses, and on supportive information
gathered during the three months of interviewing, the fol—
lowing alternate formulation seems appropriate to summarize

what has been learned.

EVALUATION OF COMPETING CHANNEL SYSTEMS

The purpose of this section is to evaluate several
rival channel systems employing both the quantitative find—
ings of the study and the researcher's subjective appraisal
of the persons and channels being analyzed. This evaluation
will show that large differences exist among competing chan—
nel systems—-differences which appear to result more from
behavioral and attitudinal factors than from physical char—

acteristics which had been previously thought to affect

measures of system effectiveness.

 

 

 

 

III-I::::

 

 

 

134

Physical Characteristics of Egg Channel Systgmg

The egg channels included in this study differed
in terms of both the operational size of system members and
the system structure. Each of these characteristics is dis—
cussed below.

Operational Size of System Members. As explained earlier

in this chapter in the section titled "Operationalization
of Variables," the operational size of each system member
was measured in terms of units of some commodity handled
by that firm during the last fiscal or calendar year of the
firm's operation. The measure of size for each channel level
was:

Channel Level Measure of Operational Size

Feed Suppliers Tons of feed milled for laying hens

per year
Bird Suppliers Started pullets sold per year
Producers Hens housed per cycle
Processors 3O dozen cases of eggs handled per
week
Retailers 3O dozen cases of eggs sold per
month

System Structure. The structure of the egg channels
included in this study may be described in terms of (a) the
degree and means of vertical integration; and (b) the number

Of firms at each channel level.

Other Characteristics of Systems
As noted in Chapter II, several writers have asserted
that marketing channels may be characterized as systems which

differ in terms of relative openness. The openness of a

 

 

 

135

system is a function of that system's accessibility and re—

sponsiveness to the external environment.

The widely differing levels of competitive effective-
ness of the channels included in this study seem to be di—

rectly related to systemic openness. It appeared that a

small number of relatively open systems consistently out—
performed the remaining large group of less open channels.
The apparent relationship between a system's openness and

its effectiveness will be discussed below in terms of the
following intra-system attributes:

ggntral Coordination. Frequently referred to as
'integration' in the systems literature, central co—
ordination exists when a single decider such as a

channel captain exerts primary control over a system.
Feedback. Feedback involves the transmission of
information about the system‘s effectiveness from the
outside environment to the source of output within a
system, in order to reduce existing error or deviation

in output from some criterion or reference which the
system seeks to maintain.

Cohesiveness.

 

As used in the systems literature,
cohesiveness is a function of social or physical prox-

imity, and describes the tendency of the components of
a system to maintain sufficient physical contact to en—
able them to combat disruptive forces. In the present
evaluation, the meaning of the term 'cohesiveness' will
be broadened to describe the level of conflict and co—
operation which occurs within a system, as operation—
alized by the cooperation and conflict scales in each
research instrument.

Attitudes. Information on the attitudes of the
Persons included in the channel systems under study
was gathered through: (a) the responses to the Q-
sorted statements in the research instrument (the re—
SpOnses which most strongly characterize each particu—
lar type of system were yielded by the factor analysis
Of individual data) and (b) the explanatory comments
of the respondents as they answered the size, control,
conflict, cooperation, and innovativeness questions in

 

 

 

 

 

EI:T_________________________________7

136

the research instruments.

These candidly volunteered
supplementary responses were extremely valuable because

of the background information they yielded about the
respondent's perception of a given situation.

Conceptually, it is possible to locate any channel

system on a continuum which runs from relatively open to
relatively closed. The present discussion will, however,
be confined to 14 channels whose summed ratings in Table

5.2 resulted in three groups of systems which represent the
'relatively open,‘ 'moderately closed,’ and 'relatively

closed' positions on the following continuum.

Relatively

Moderately Relatively
Open Closed Closed
Systems Systems Systems

 

To illustrate the seeming relationship between sys—
tem openness and effectiveness, the characteristics of sev-

eral selected channels will be discussed below. These par—

ticular channels are included because they differ widely

in terms of central coordination, feedback, cohesiveness,
and attitudes, and thus occupy divergent points on the open-
ness continuum.

As indicated in Table 5.2, each channel was subjec—
tively rated on each openness characteristic. Since the
ion: separate ratings for each channel are monotonic—~all
ratings run the same way, and because it is not unreasonable

to weight each rated openness characteristic equally, the

ratings were summed to provide an ordinal level total rating

for each channel. This rating has a range of 4 to 12, and

a midpoint of 8.

After summing the ratings for each channel,

 

 

 

 

137

 

 

 

.pxwp map CH comm:
.munmocommmu Hm3pH>HUCH Ho muapcwoa wmwcmMMMOMWm
. . . m

pmmm wmmzb Mo woe
Hp UHo>m Op vows.

#MHkuUmum 0
Ohm meQEDM»uwflu #DQ
mumpHHHumm on cum .mEmumNm mcflpmmfiou may mo mcmwoccmcu umHHEHm
#zmflu mzb CH omEESm mum HMQCMQU nomm How mmambmwuo HHwnw>o G
usom may mo comm so .chuH no .HmupsmcuN .doomumsmw one
I Umbmu

o
mmeMMMWMm Mo COHmmDUMHU
.mUH w ou QESHOU pad
.# Humuumumgu mmwc r
wH chcmcu comm ammo
i

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

w H H
w H H M H llllllllllll
“V H H H H Wm
m N H H 5H
H H mo
OH
s N H N N "meumhm cmmOHU mHm>flhmHmm
b H N N N mH
h N N H N MH
m N N N N mN
. N N N N m
m N N
m N .8
"mEmumMm pmmOHU mepmumpoz
HH m m N m
cm
NH m m m m NN
NH m m m m eH
NH m m m m mo
"mambmmm ammo >H0>HbmHmm
lilllllllllllllllNl
mmcflpmu mmpspflbu< mmmcw>flmmaou xomnoomh GOHMMGHUHOOU ..HwnEDZ Hmscmcu 9
may Mo Edm Hmupcmu pmuMUHch mfimummm
Hmccmno mo mmsouw
will. mmmccmno mo mUHbmflumpumuan

 

  

 

 

 

mmmzzmmo EMBmNm ho <HMMBHmU mDOh m0 EU<W
ZO .Qma<m mqmzz<mu ho mmbomw mmmmfi

llillllllu
N N. m GHQMB

 

 

 

 

7

138

the individual channels were grouped on the basis of simi-
larity of total ratings. Channels with ratings of 12 and
ll are called relatively open systems, those with 8 and 7
are moderately closed systems, and those channels which re—
ceived ratings of 4 or 5 are termed relatively closed systems.
Eggpeting Channels Described in Terms of
Physical and Other System Characteristics

This section will analyze and compare selected char-
acteristics of channels which represent each of the three
types of channel systems.

Relatively Open Systems. Of the three groups of systems
under discussion, this group of channels appears to be the
most open and responsive to the external environment. Al—
though these channels do not differ appreciably from the
others in terms of the size of the business operations they
contain, the members of these channels are unique in their
ability to work well with one another, their acceptance of
Change, their market orientedness, and their willingness to

consider the other system members and the goals of the sys—

 

tem as important and relevant to their individual operations.
Several characteristics of these relatively open systems are
discussed below.

1. Central Coordination. Without exception, the
Channels in relatively open systems were administered by
Single firms, or in a few cases by pairs of firms. The

channel captain in mest cases was a feed milling enterprise,

 

 

 

 

 

 

139

although in a few instances the feed mills and processor-
distributors operated together to formulate policy and make
decisions.

More important than who took command of these chan-
nels is the fact that someone had taken the initiative to
aggressively seek out new markets, schedule production, in
anticipation of future prices, and to ensure that the sys—
tem is technologically up to date. Channel captains in these
relatively open systems had frequently developed formalized
arrangements with retail buyers to sell specified quantities
of eggs at prices given by a predetermined formula. In re—
turn for this orderly marketing agreement, a channel captain
usually provided the retailer with merchandising assistance
in the form of point of purchase display materials, coopera—
tive advertising funds, egg carton stuffers, or reduced prices
during special promotional events.

2. Feedback. The communication which flowed from
the market and from system components in relatively open
Systems was both regular and plentiful. Information moved
from retail egg buyer back to the processor—distributor,
and then to the feed mill who coordinated production activ—
ities, because all parties knew one another on a face-to-face
basis, and because they talked to one another frequently by
telephone. One example of the type of feedback which strength—
ened these relatively open systems occurred when the auto—

mated conveyor—belt egg gathering equipment of an egg producer

 

 

 

 

 

 

 

 

 

 

140

began to malfunction and produce rust stains on the eggs.
When this happened, the feed mill and the processor sent
technical men out to the farm in an attempt to correct the
situation. They returned several times before the machinery
suggested

was finally repaired.

The processor, meanwhile,
several remedial measures to the producer, and tried several

different cleaning solutions in his automated egg cleaning
The retailer was alerted before the stained eggs

equipment.
and was able to plan a special price event to

reached him,

move the eggs effectively.
The excellent feedback which characterized these

 

systems appeared to have been fostered by a mutual trust,
and by a feeling that the channel captain was attempting

to strengthen the entire channel and thereby improve the

profits of all system members.
the mem—

3. Cohesiveness. As shown in Table 5.3,

bers of channels in relatively open systems displayed a
higher degree of cooperation and a lower degree of conflict

than that found in moderately closed or relatively closed

Systems.
As revealed by the factor analysis

4. Attitudes.
0f the Q—sort data, the attitudes of the members of rela—

tiVely Open systems are described in the following terms:

(a) Cooperative

Disagree: "A big problem with the egg business is

people—~they're too hard to get along

With 0 "
"It's tough to get much cooperation from

Disagree:
anyone in this business these days.“

 

 

 

 

 

 

 

141

Table 5.3

COHESIVENESS IN COMPETING
TYPES OF CHANNEL SYSTEMS‘

 

 

 

 

 

L_> Type of Channglr§ystem
Relatively Moderately Relatively
ariable Openr Closed Closed
ooperation“ High Medium Medium
(7.50) (6.59) (6.32)
onflict“* Low Medium Medium
(1.40) (1.98) (2.09)

 

 

 

*Data in cells are means computed from cooperation
and conflict responses of persons in the 14 channels.

as operationalized by the cooperation

‘*Cooperation,
scale, runs from 2.0 = lowest cooperation, to 8.0 = highest

cooperation.

“‘Conflict, as operationalized on the conflict scale,
runs from 1.0 = lowest conflict, to 5.0 = highest conflict.

(b) Innovative

Disagree: "It is unwise to begin using a new product

or practice until it has been proven ef~
fective through widespread use."
”A man’s success depends on his accept—

Agree:
ance of new ways of doing things."

(c) Market Oriented
Disagree: "It would be a good thing if the egg in—

dustry would concentrate on producing
more eggs per pound of feed."

"Money spent on egg promotion could be
better spent on research to stamp out

diseases among laying hens."

Disagree:

(d) Views the Channel as a Reference Group
Disagree: "It is best to forget what other people
in this business might think of me when

I make my own decisions."

(e) Trusting
Agree: "I can trust most of the men that I deal
with."

 

 

 

 

 

 

142

(f) Empathic
Agree: "I usually try to see the other person's
point of View even though he doesn't
always try to see mine."
(9) Cosmopolite
Agree: "I try to be aware of what other people
in this business are doing in different
communities."
(h) Modern Oriented
Disagree: "People in this country have become too
specialized."

V

(i Willing to Work for the Group

Agree: "Often the best solution to a problem
will require that some one individual
must sacrifice for the good of the group
as a whole."

This attitudinal description was reinforced by com—
ments made by respondents as they were interviewed. Many
examples of these comments could be cited; one which seems
especially relevant because of its explicit recognition of

the dependence of the system on its external environment is

as follows:

 

The egg business has been in the hands of the farmer
and the egg peddler for too long. Their primary
concern was to sell a load of eggs to a store. Our
primary concern is to satisfy the customer. The
farmer and the egg peddler don‘t understand today's
customer. Our job is to change eggs from a 'com—
modity' to a brand that has 'take-me—home' appeal.

5. System Structure. The systems are composed of
both contractually integrated and owner integrated channels
Which vary widely in structure. For instance, one contract
Channel contains a single feed mill, several producers, one
bird supplier, one processor, and one retail supermarket

chain. Another contract channel is composed of many feed

 

 

 

EI:f___________________________________i
143

mills, many bird suppliers, many producers, a few process—
ing plants, and a few retail chains. The owner—integrated
channels are most often built around one enterprise which
mills its own feed, and produces, processes, and distributes
its eggs. Its birds are supplied by several hatcheries and
its eggs flow through a variety of medium—sized supermarkets.
The structure of these systems therefore does not appear to
be a key factor in contributing to their relative openness.
6. Operational Size. As shown in Table 5.4, the
mean operational size of the feed suppliers, bird suppliers,
processing plants, and retail outlets in relatively open
systems is smaller than the corresponding facilities in mod—
erately closed or relatively closed systems. The egg pro—
duction facilities in these relatively open systems are,
on the average, equal in size to those in relatively closed

systems, and only slightly larger than those in moderately

 

closed systems.

It thus appears that the largest non—egg producing
facilities tend to be less responsive to the market and to
industry change than their smaller competitors. Some of
the probable efficiencies which derive from economies of
Scale in these large firms may be partially offset by the

lack of openness which characterizes the systems in which

they participate.

 

 

 

 

Table 5.4

MEAN SIZE OF BUSINESS UNITS IN
COMPETING TYPES OF CHANNEL SYSTEMS

Exhibit A

 

Channel Level

 

Processing Retail

Bird Production

Feed

Size of

o
0

Business Unit

Large

Medium

144

  
   
  
   
   

Small

Relatively
Moderately
Relatively

Relatively
Moderately
Relatively

Relatively
Moderately
Relatively

Relatively
Moderately
Relatively

Relatively
Moderately
Relatively

 

closed
closed

open

closed
closed

open

closed
closed

open

closed
closed

open

closed
closed

Open

Processing Retail

Feed Bird Production

.
a

Channel Level

 

 

 

 

__._J

 

 

145

 

 

 

 

 

 

 

 

Acucoe\mmmmuv
. HHmumm
+ ooo.on mmm.muooo.m mam eno
H3m03\mmmmuv
+ ooo.o mmm.mlooo.m mmm.~uo scammmuoum
AmH0>U\mcwmv
+ ooo.oN mmm.mauooo.oa mmm.muo coauuseoum
Humm>\mumHHomv
+ ooo.ooo.a mmm.mmmlooo.00m mam.mmeuo maaasm seam
Aumm%\coav
+ mam.wa mmm.sanooo.m mmm.mno seaasm emmm
magma ,Ilmmmmmz immmemlll. Hmwmmll
mNHm Hmccmcu

 

 

 

 

 

 

mwmdq m0 .ZDHQMZ .QH4Zm m4 Hm>ma Hmzz<mu mU<m ZH mBHzD HEB
MMHmm<HU OB QMMOHQZM WMHmmee<U MZDHO> mWMZHmDm

m panaaxm

AcchHucouv

w.m wHQMB

 

 

 

 

 

146

 

 

 

 

 

 

 

  
 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

000 H 00m 004 000.mm 000.HH om 000.0H 0N0.m eaom 000m mo
mmmmU GWNOU OM
Hamumm

00m 000. . sum: um
m 000.m mme 000.0H 00H.a 000.H 00m.m 000 a emanamr momm mo

mmmmu cmnoc om

mcewmmuoum
If ll‘l'l‘I-IIIIIIIII‘IIIII‘II‘IIII-
oomNN IIIJIIIIIJ s mmaor mama
000 we 000.0N 000.0H 000.Nm 000.ma 000.0 000.00 000.0N mau U\eaoauuseoum
ill!!!" III\
000 0mm 0 . N ham» umd 30m
00 00m H 000.msm 000.000 000.00s 000N0e0 000.0Nm 000.0ms 000.msm mundane embumum
ouam

m
000 N 000.m~ . . nmm>\meme
000 as . . . . 0
00a m 000 am 00m s own 000 on 000 acasma mom emaaas
cmmm mo mCOb
pmmm
.0!" lllll‘l‘l'lll
CHE filldez com: .CHS .xmz new: .caz .xmz cmmz pHcD bamEmMMmMMM
. ON.

cwmoHu NHm>Hbmem cwmoHU mebmemooz cm 0 >H0>HbmHmm Hm>mq Hmccmcu
Embwmm Hmccmcu M0 maze LIIL

 

 

 

 

zmsmwm qmzzeru no mass to<m 2H qm>mq qmzzeru foes
mm omqnzer mzsqo> ZDSHZHZ oz< .zsszez .zemz

U

HUMSCHuCOUV

«V.

panarxm

m OHQMB

 

 

 

 

 

E::I______________________________________i
147

EXPLANATION OF TABLE 5.4

Exhibit A of Table 5.4 relates the mean size of feed
and bird suppliers, producers, processors, and retailers,
to the types of channel system in which they participate.
Thus, for example, Exhibit A shows that feed suppliers in
relatively open systems are slightly smaller, on the average,
than the feed suppliers in moderately closed systems, and
approximately one—fourth the size of the feed supplier in
the relatively closed system.

Exhibit B indicates the business volume categories
employed to classify the units in each channel level as small,
medium, or large. For instance, those feed suppliers which
milled between 0 and 8,999 tons of feed for laying hens dur—
ing their last fiscal or calendar year are classed as 'small';
those which milled between 9,000 and 17,000 tons are 'medium';
and those which milled 18,000 tons or more are 'large.‘

Exhibit C indicates the mean, maximum, and minimum
volume handled by each channel level in each type of channel
system. For instance, in relatively open systems, the larg—
est feed supplier milled 30,000 tons per year, the smallest
milled 160 tons, and the average supplier milled 6,600 tons.

 

Moderately Closed Systems. This group of channels ap—
pears to be relatively less responsive to the market, and
to the actions of competitors than the systems previously
discussed. Although these channels are structurally similar

to several of those in the previous group, the business units

 

in these moderately closed systems are larger, on the aver—
age, than those found in relatively open systems. The atti-
tudes of the members of the moderately closed systems are
radically different from those discussed above. Each of
the attributes of the moderately closed systems is discussed
below.

l. Central Coordination: Little or no centralized

Channel direction or guidance was evidenced among channels

 

 

 

 

 

7

148

in the moderately closed systems. Feed companies, in a few
cases, influenced some of the flock management decisions of
producers—~this control appeared to be exercised out of a
concern for the egg production from the feed mill—owned lay-
ing hens, rather than from a desire to improve the long run
competitive effectiveness of the channel as a unit.

Sub—optimal channel effectiveness appeared to be
partially caused by the contractual arrangements in these
systems. For instance, some contracts reportedly were ten-
dered by one feed mill on the condition that the producer
purchase his hen house and equipment from that feed mill. 1 “
Several producers reported that once they had entered the
business, the feed mill had attempted to discontinue the
contract, or lower the compensation to the producer. Sev-
eral other feed mills had been 'forced' into contracts in
order to help them collect a portion of their accounts re-
ceivable which had been run up by former non—contract feed
accounts during times of low egg prices. The goal of the
feed mills in these situations was to increase sales of hen
houses, equipment, or feed——not to strengthen the system as
a whole.

Whereas in the relatively open systems a central
coordinator had aggressively and selectively sought new re—
tail accounts, had worked to produce a compensation scheme
Which would share profits and losses on a system—wide basis,

and had provided a planned program for efficient egg

 

 

 

 

 

 

7’

149

production and marketing--these matters were virtually ig-
nored in moderately closed systems.

2. Feedback. The relatively infrequent communica—
tion within moderately closed systems flowed largely between
producer and feed mill. Commonly, little information moved
directly back to the producer or feed mill from the retail—
ers--in most cases the retailer did not know who produced
his eggs, and the producer did not know who retailed the
eggs from his hens. Such feedback as did occur typically
was confined to occasional complaints from the retailer which
were transmitted indirectly via the delivery man of the proc-
essor to the producer.

3. Cohesiveness. The members of channels in mod-
erately closed systems displayed a medium degree of cohesive-
ness. As shown in Table 5.3, these systems were character—
ized by a lower degree of cooperation and a higher amount
of conflict than was true in relatively open systems.

4. Attitudes. The attitudes of the members of mod—
erately closed systems are quite different from those encoun-
tered in the relatively open systems discussed above. As
yielded by the Q—sorted responses, these attitudes are:

(a) High Conflict
Agree: "A big problem with the egg business is
people——they're too hard to get along with."
(b) Low Cooperation
Disagree: "When I have a problem, I can most often
get good advice from one of the people
I usually deal with."

Agree: "It's tough to get much cooperation from
anyone in this business these days."

   

 

 

 

 

 

150

(c) Disagree with Central Coordination
Agree: "Today, egg men are too closely controlled
by the people they do business with."

(d) Not Innovative
Agree: "It is unwise to begin using a new prod—
uct or practice until it has been proven
effective through widespread use."
Disagree: "A man‘s success often depends on his
acceptance of new ways of doing things."

(e) Low Ratings of Intra—Channel Dependability, Trust
and Competence

Disagree: "I know that most of the people I do
business with are dependable."

Disagree: "I can trust most of the men that I deal
with."

Agree: "The majority of the people that I deal

with don't know as much about the egg
business as they ought to."

V

(f Tradition Oriented

Agree: "People in this country have become too
specialized."
(9) Low Empathy
Disagree: "I usually try to see the other person's
point of View even though he doesn't
always try to see mine."
5. System Structure. These systems are composed
of contractually integrated channels which are somewhat alike
in basic structure. Typically, several producers are linked
Contractually to a single feed mill. The eggs flow to one
or more processors, who send them to a wide variety of re—
tail accounts. The only link between channel levels occurred
between producer and feed level, and exists for approximately
one year. Thus, whereas in relatively open systems a rela-
tively stable structure exists because of the permanent and
semi—permanent links between channel levels, in the moder—

ately closed channels the temporary relationships produce

relatively unstable channel structures.

 

 

 

 

 

 

151

6. Operational Size. Exhibit A of Table 5.4 shows
that in moderately closed systems, egg production units are
larger, on the average, than they are in the previously dis—
cussed open systems. The eggs from several moderately closed
systems move through a supermarket chain which purchases
more eggs than any other single retail chain included in
the study. Except for this difference in the mean retail
volume, the size of the business units in these systems is
not an important factor in distinguishing moderately closed

systems from relatively open or closed ones.

Relatively Closed Systems. The competitive effective—

 

ness of this group of channels appears to have been severely

 

limited by the reduced coordination, feedback, and cohesive-
ness found in the system, and by the independent attitudes
Of the system members. Selected attributes of several rela—
tively closed systems are discussed below.

1. Central Coordination. Virtually no centralized
control was evident in these systems. Producers managed
their flocks independently, and as a result, product qual—
ity and system output fluctuated widely. A major weakness
of such systems stemmed from the fact that independent proc—
essing and feed milling operations considered themselves
to be entirely separate operations, and not part of any

larger channel group. Such processing operations seemed

 

to be concerned with the eggs only until they were cleaned

and cartoned. Instead of developing a long range marketing

 

 

 

 

 

Ii—
152

program involving retail and consumer promotion, advertis-
ing, development of a brand image, and the solicitation of
new accounts, these processors forfeited marketing responsi—
bility for the eggs by dumping them at reduced prices on
several independent jobbers.

2. Feedback. Feedback in these jobber—dominated,
independent systems was poor, as was the case before World
War II, when this type of system was the prevalent means
of egg distribution. Few people in these systems knew the
others in the system, and the only means of contact between
the retail and production levels was the jobber and processor.
Only infrequently-~when prices were too high at retail or
too low at producer or processor level, or when egg quality

deteriorated markedly——did a felt need for intra—system com-

 

munication appear. 1
3. Cohesiveness. A medium amount of both coopera-
tion and conflict was apparent within these systems. Per-
haps because participants preferred to be independents, they
did not expect much cooperation from suppliers or customers.
The lack of contact among system members probably helped
to reduce the degree of reported conflict. Perhaps more
conflict would have occurred had the persons within the SYS—
tems attempted to work together as a group to achieve a com—
mon purpose.
4. Attitudes. The attitudes of the members of
these systems reflect a high degree of independent—minded—

ness, a concern for local versus national matters, and a

 

 

 

 

7
153

lack of interest in the welfare of others in the system.
As described by the Q—sort responses, these attitudes are:

(a) Failure to View the System as a Reference Group

Agree: "It is best to forget what other people
in this business might think of me when
I make my own decisions."

(b) Independent—Mindedness

Agree: "It is better to be your own boss than
to operate under a contract or agreement
which ties you to someone."

Disagree: "We ought to work toward a system which
would allow all the people who produce
and market eggs to make a decent living,
regardless of whether prices are high
or low."

(c) Localite Orientation
Agree: "I concern myself with the people and
activities of my own community rather
than those outside it."
These specific attitudes were supplemented by sev—
eral more general comments of persons as they were inter-

viewed. Such comments tended to stress the feeling that

a person at one channel level was competing with the others

 

in his channel and that the other system participants were
unresponsive to his needs and were, in fact, working to pre—
vent him from attaining his own goals. For instance, feed

milling operations often appeared to be chiefly concerned

with increasing the volume of feed sales and with keeping

feed accounts paid up. Many producers felt that the feed

mills really did not care about them. One group of producers

rePorted that on several occasions they had requested and

net received feed supplements and medicants mixed with their

 

feed to combat outbreaks of disease among laying hens.

 

 

 

 

 

154

Several producers were also critical of the markups taken
by jobbers and retailers in their systems. These markups
were alleged to be excessive and were perceived by the pro—
ducers as an important reason for the reduced demand for
their eggs. Retailers, on the other hand, tended to crit-
icize the processors, and jobbers for failing to provide
cooperative advertising money or other promotional assistance.

This bifurcation of interests was not apparent in
relatively open systems, and was rarely reported in moder—
ately closed systems. The independent attitudes of these
system members, together with the limited cohesiveness, feed-
back, and coordination which characterize relatively closed
systems combine to severely limit the effectiveness of these
systems in competing against its more open rivals.

5. System Structure. These relatively non—inte-
grated systems are composed of a large number of independent
egg producers who have, by tradition, purchased feed from
one of several local feed mills. Eggs are sold to one of
several independent processing plants which sell to several
distributors and/or to a wide variety of retail outlets.

6. Operational Size. While the production and proc-

essing facilities in the three types of systems are similar

 

in average size, the feed and bird supply facilities are

much larger, and the retail facilities are much smaller in
these relatively closed systems than the corresponding facil—
ities in the other types of systems. However, it again ap—

pears that the size of the business units in a system is

 

 

 

 

155
not related to the relative openness of the system.

CONCLUSIONS AND REFLECTIONS
Whereas relatively few differences in size and struc-
ture appear to exist among competing channel systems, these
channels differed widely in attitudes and behavior. The
apparent meaning of this research for egg channels as well

as for other types of channel systems is discussed below.

Mganing of This Research for Egg Channels

As a result of this research it appears difficult
to continue to cite lower costs and economies of scale as
the principal reasons for the movement toward increasingly
integrated egg systems. Surprisingly, some of the largest
processing plants and feed mills were found in the less in—
tegrated channels, while few differences existed across sys—
tems in the size of the average bird supplier or producer.
Only at the retail level were the smallest units found in
the less integrated systems.

Instead, it seems more likely that non-integrated
and semi—integrated channels are disappearing because of
the inability or lack of desire on the part of system par-
ticipants to: adapt to a rapidly changing market situation,
maintain and foster an attitude of intra—system feedback
and communication, and to operate as a team with a minimum
of divisiveness and a maximum of intra—channel assistance.

The strongest differences in attitudes and behavior

 

 

 

 

 

7

156

occurred when the most highly integrated channels were com—
pared with the least integrated ones. The most highly inte—
grated systems were open, cohesive, well coordinated, and
highly communicative. Members of these systems displayed
a high degree of trust, dependability, competence, empathy,
cosmopoliteness, and innovativeness, as well as a strong
market orientation. The fact that the least integrated sys—
tems were described in the opposite terms seems to place
severe limits on the ability of such systems to continue
to operate in a dynamic environment.

Less clear—cut differences in attitudes and behavior
emerged among the systems in the middle ranges of integra-

tion. Some non—contractual systems of moderate integration

 

appeared to operate more openly and smoothly than some of

the more highly integrated contractual channels. The most
effective systems in the middle range of integration appeared
to be those which contained someone willing to exercise a
directive force over the channel. Even when such a channel
contained persons who did not cooperate well, or whose atti—
tudes were tradition oriented, it appeared that a strong
channel captain, if possessed with a planning orientation

and a market orientation, could achieve at least a moderate
degree of success in the market.

Meaning of this Research for Channel Systems
92E£§ting in Othgr Industrial Settings

There appears to be little difference between egg

Channels and the channels used to produce and market many

 

 

 

 

 

 

 

L’—'

157

other consumer and industrial goods. Egg channels were
selected for research in this study because they contain
merchant wholesalers, manufacturing operations, consumer
goods retailers, and processors which are standard U.S. Cen—
sus of Business categories. Therefore, it appears that much
of what was learned about the operation of egg channels may
apply to other types of channel systems.

As a result of this research, it appears that sev—
eral factors contribute to the competitive effectiveness of
channel systems. Some of these factors are: (a) the lower
costs and economies of scale inherent in some large facil—
ities; (b) the relative openness of some systems; (c) the
degree of product—market commitment, i.e., the extent to
which all channel members are strategically committed to
their system and to their target market(s); and (d) the
degree to which the system is vertically integrated.

Moreover, there appears to be a trade off between
system size and system Openness. The specific nature and
exact circumstances under which this trade off operates are
not yet clearly defined. Potential economies of scale seem
to exist in many different structural types of channel sys—
tems; the degree to which these potential economies are actu-

ally achieved appears to be influenced by the relative open-

ness of these systems.

 

 

 

 

 

 

 

CHAPTER VI
MANAGERIAL IMPLICATIONS AND OPPORTUNITIES

FOR FUTURE RESEARCH

The purposes of this chapter are to suggest how the
research approach of this study may be used by channel man—

agers, and to suggest several directions for future research.

Managerial Implications of this Research Approach

The present research should be useful to channel
captains who are faced with the following managerial tasks:

1. Recruitment and Selection of Future Channel Mem—
bers. Captains of future or on—going channels probably are
aware that substantial differences exist among the attitudes
and patterns of interaction of system members. Since these
differences appear to affect the long run competitive effec—
tiveness of a system, it appears that a manager who wishes
to improve the performance of his system beyond that given
by low costs and economies of scale must selectively recruit
system members. Once this channel captain has specified
Or described the types of people he desires, it will be pos-
sible to administer research instruments like those used in
the present study to prospective system members. The atti-
tudes and behavior of each candidate may then be compared
With the desired types in order to decide which persons to

bring into the system.

158

 

 

 

 

 

IlIII:____________________________———I

 

159

Many standardized personality inventories are now
used for employee selection by personnel departments of cor—
porations. Relatively few corporations, however, build en-
tire distribution networks on this basis.

2. Monitoring the Attitudes of Persons in Existing
Systems. Poor upward and downward communication is commonly
cited as a cause of inefficiency in organizational hier—
archies. This communication breakdown also occurs between
the channel levels in production—marketing systems. Intra-

system effectiveness may be drastically increased when regu—

 

lar audits of system member attitudes are conducted by the
Channel captain to determine:
(a) The content of attitudes——what sorts of problems

are perceived by members of different channel levels, and

 

what sorts of solutions do they suggest? Do most persons

mention the same problems or does each channel level seem

to have its own problems? To what extent are the problems
of one channel level caused by the actions (or inactions)

0f another channel level?

(b) The degree of divergence or convergence of atti—
tudes held by people within a given channel level, and be—
tween channel levels. Perhaps more important than what
People are thinking is the extent to which their attitudes
differ and/or are incompatible. Divergent attitudes in the
areas of dealer training, product pricing, advertising, prod-
uCt development, and so on may be discovered with the aid

of interviews using psychometric devices such as Q-methodology,

 

 

 

 

 

Li—
160

Likert scales, and rating, ranking, or sorting procedures.
During such interviews, respondents might be first asked

to describe the ideal pricing, product, or promotion situa—
tion, and then to describe present conditions in these areas.
The difference in scale values between 'ideal' and 'actual'
would indicate both the direction and magnitude of the re-
quired changes.

Similarly, information on divergent attitudes about
present versus ideal system goals may be gained by comparing
lists of goals which have been rank ordered by persons at
different channel levels.

(c) The flow of vertical communication. Questions

about the frequency, content, and direction of communication

 

could be asked of people at each level, and these responses
compared to note those nodes within the system where great
differences existed between communication sent and communi—
cation received. These responses could also be compared
with attitudinal responses about what the ideal communica—
tion situation would be like, at each channel level. This
type of analysis is relatively inexpensive and might be used
as a first step in channel-wide troubleshooting efforts.
Whenever actual versus ideal evaluations are widely differ—
ent, a more thorough (and more costly) investigation could
be launched.

3. Recognition of the Complexity of the System.
The trend toward multi—product conglomerate corporations

is accompanied by a trend toward vastly more complex

 

 

 

 

 

EII:__________________________________i
161

distribution systems. Management may be unaware of which
markets are presently being served, and how the products
flow to these markets. An audit of present distribution
systems may result in the elimination of small and unprofit-
able accounts, as well as the selection of new retail out-
lets which are better able to serve each market segment.
When management is forced to trace the channels through
which its products presently move, it will gain insight

in the same way that a problem solver is better able to
solve his problem when he must systematically spell out

each step in the formulation of a computer program. He is

 

forced to state his assumptions about existing relationships .
and to consider a wide variety of possible solutions to his

problem.

Qppgrtunities for Future Research

This study is a beginning step in the area of chan—
nel system research. Although much was learned about the
actions and perceptions of people in different types of chan—
nel systems, a great many areas are as yet unexplored. Some
Of these are listed below.

1. The relative importance of system structure ver-
sus the attributes of system members ought to be studied
more intensively. Such research would investigate questions
such as: Which types of channel structures appear to oper-

ate most successfully under which conditions? Which person—

 

ality or attitudinal characteristics seem to improve or hinder

 

 

 

 

 

162

the effectiveness of competing systems?

2. Does cohesiveness cause integration, or does
integration cause improved cooperation? Sturdivant assumes
that high conflict causes integration. But does integration
cause low conflict? This admittedly is a complex question,
but it might be answered with before and after studies or
with the use of experimental research designs.

3. How is cohesiveness related to intra—channel costs
and efficiencies? When can a great degree of cohesiveness
overcome the competitive disadvantage of high cost?

4. Must cohesiveness be uniformly distributed up
and down a channel system? Or are there crucial channel
levels for which high cohesiveness is a must, and other chan—
nel levels in which it is unimportant?

5. How are 'manifest' measures on a behavioral vari-
able related to attitudes toward that variable? That is,
do those persons who agree that regular feedback is desir—
able also provide such feedback in actual practice?

6. Which behavioral variables, in addition to those

 

included in this study, seem to bear directly on systemic ef-

fectiveness? Potential candidates for research include: use 5

and bases of power, causes of and methods of resolving con-

flict, presence of intra—channel norms, and role prescriptions. ;
7. How can innovativeness and receptivity to change

be Operationalized in a manner that is not constrained by

 

the number of years in business?

 

 

 

 

 

 

APPENDIX A

GLOSSARY

Channel of Distribution

 

The term 'channel of distribution' has come to de—
scribe a group of business firms which operate together to
move a product from its point of production through distri-

bution to the customer or final user. In this stud , a

 

channel includes the input—supply firms (feed mills and
hatcheries) and the output—receiving firms (processors,
distributors, and retailers) along with the egg producer
with which they are associated. Thus, the channel begins

with input supply and ends with the retail operator. It

 

does not include the consumer, nor such facilitating agen-
cies as financial institutions, common carriers, or public

warehouses.

Channel Level

The vertical positions of the members of a channel
are referred to as 'levels'—-the feed supply level, the
bird supply level, the producer level, the processor—dis-

tributor level, and the retailer level are included in this

study.

Innovativeness

 

Innovativeness is the degree to which an individual

is relatively earlier in adopting new ideas than other members

163 I

 

 

 

 

 

 

EI:f__________________________________i
164

of his social system. The social systems most relevant to
this study are the five channel levels which are concerned

with the production and marketing of eggs.

System

A system is a set of objects together with the rela-
tionships between the objects and their attributes. Systems
differ from one another in terms of the degree to which they
are relatively open or relatively closed. The relative open—
ness or closeness of a system is a function of that system's

accessibility and responsiveness to the external environment.

 

This relative openness may be indicated by many intra—system
factors, a few of which are: central coordination, feedback,

cohesiveness, and attitudes toward the external environment. a

Product—market Commitment
Product—market commitment is the willingness of all
the firms in the system to commit to one over-riding channel

strategy aimed at providing something to some target market.

 

Q—sort

The act of sorting attitudinal statements into groups,

 

on the basis of a criterion such as agreement or disagreement.
The Q—methodology, of which the Q-sort is a part, was devel-

°Ped by psychologist William Stephenson.

Vertical Conflict

Vertical conflict is the disagreement or struggle which

may exist between members of different levels of channel systems,

 

 

 

 

 

 

 

EI:T__________________________________i
165

(a) Conflict received--The 'conflict-received' score
is one of two conflict scores for each person.
Conflict-received refers to the mean value of the
conflict scores reported about an individual by the
other persons in his channel who evaluated him.

For example, a conflict—received score for an in-
dividual producer would result by averaging the
conflict evaluations of this producer which were
given by his feed supplier, bird supplier, and
prooessor-distributor.

The mean conflict—received score for a channel is
found by summing the conflict-received score for
each individual in the channel and dividing by N.

(b) Conflict given-—The 'conflict—given’ score refers
to the mean value of the conflict scores which an
individual reports in evaluating the other persons
in his channel.

Vertical Cooperation

 

Vertical cooperation is collective action which may I
be undertaken for common benefit, by members of different
levels of channel systems.

(a) Cooperation received-~The 'cooperation—received'
score is one of two cooperation scores for each
person and channel° Cooperation—received refers
to the mean value of the cooperation scores re—
ported about an individual by the other persons
in his channel who evaluated him.

(b

v

Cooperation given--The 'cooperation—given' score
refers to the mean value of the cooperation scores
which an individual reports in evaluating the
other persons in his channel.

 

Vertical Control

Vertical control is the exercise of directing, guid—
ing, or restraining power by a firm at one level in a channel
over a firm at another level in the channel. Vertical con—
trol is exerted when a firm at one level influences or makes

Some decisions traditionally made by a firm at another channel

lGVel.

 

 

 

 

 

 

 

166

Vertical Integration

Vertical integration exists when a firm at one level
in a channel controls, through ownership or contractual means,
a number of production and/or marketing operations of a firm
at another level of the channel. Vertical integration is a
continuous variable, the value of which depends not only
upon how much control is exerted by one firm over another,
but also upon: the degree to which price risk is shared
by several channel members; the long-run commitment of each
channel member to his channel system; and the degree to which
one or more production or marketing functions are assumed

by an individual channel member.

 

 

 

 

 

 

 

 

 

APPENDIX B

RESEARCH INSTRUMENTS

This appendix contains the research instruments used
for each of the five channel levels in the study. Each in—
strument is similar to the others in organization. Each
instrument contains questions used to index each variable,
with the exception of vertical control, which is measured
only at the producer level. The Q—statements, which were
part of each research instrument, appear only in the feed

SUpplier instrument for the sake of brevity.

 

 

 

 

 

 

 

O
O
L'"

l

 

bra

HHHI

8-l3
l4—l9
20—22

I!!!

168

EGG CHANNEL MEMBER SURVEY

FEED SUPPLIER RESEARCH INSTRUMENT

number:
questionnaire number name:
01 feed supplier
02 pullet supplier address:
03 producer
O4 processor position:

05 distributor
06 retailer

location:

01 Michigan

02 Indiana

03 Ohio

04 New York

05 Mississippi

06 North Carolina
07 Georgia

08 Arkansas

data card number

(thousands $) Gross sales of operation in 1967?
(thousands) Tons of feed sold for chickens in 1967?
Number of egg producers that are your customer now?

 

 

 

 

 

 

 

 

169

FEED SUPPLIER

with: (Producer no.: )

CONFLICT: card b

"How satisfied or unsatisfied have you
been with..."

 

the price that he paid for the feed

 

 

1 that you have recently delivered to
him?

2 the terms of payment which relate to
these recent feed deliveries?

3 his promptness in paying for this
feed?

 

his demands for service; in other

4 words, are you satisfied that he
doesn't demand more service than he
is entitled to?

the consistency with which he buys ,
feed from you; that is, is he a l
regular customer that you can count ‘
on for repeat business, or does he

change suppliers frequently?

 

 

 

5 this man as a person to do business
with?

 

 

 

 

 

 

 

COOPERATION: card c "How would you rate this person in
terms of..."

his willingness to cooperate with you?

his willingness to follow any sug-
gestions or advice which you might
give him?

his willingness to make suggestions
to you or to provide information to
you? Feedback

his willingness to accept responsi-
bility for mistakes or poor results?

 

 

 

 

 

 

 

 

 

 

170

FEED SUPPLIER

 

INNOVATIVENESS
Year started in feed supply business:
COL no when
23—25 bulk feed trucks for delivery
26-28 offer contracts
29-31 affiliate with a brand of feed
32—34 attend a feed mill management training
course, such as the MSU short course
35-37 attend other feed mill courses of
shorter duration
38—40 micro—mixing equipment for pre—mix
(1:2000 ratio) (horizontal mixer)
59 Ol 08 years What was the last grade you
02 l—2 years high attended in schools?
school
03 3-4 years high
school

04 1-2 years college
05 3-4 years college
06 more than 4 years college

I

 

60-61 What is your age?

62 01 Male
02 Female

 

Q-SORT STATEMENTS

1» Very rarely do my suppliers or customers disagree with
me on business matters.

2° A big problem with the egg business is people—-they're
too hard to get along with.

 

3. When I have a problem, I can most often get good advice
from one of the people I usually deal with.

4° It's tough to get much cooperation from anyone in this
business these days.

5° More profitable egg businesses result when persons who
help with financing or marketing share in the decision
making.

6°.__~;T0day, egg men are too closely controlled by the people
they do business with.

7° It is unwise to begin using a new product or practice un-
til it has been proven effective through widespread use.

 

8° A man's success often depends on his acceptance of new

 

ways of doing things.

 

 

 

 

 

 

 

 

 

 

171

9. It would be a good thing if the egg industry would
concentrate on producing more eggs per pound of feed.

 

10. Money spent on egg promotion could be better spent on
research to stamp out diseases among laying hens.

 

ll. People in the egg industry ought to work harder at
promoting eggs to retailers and to the public.

 

12. It is best to forget what other people in this bus—
iness might think of me when I make my own decisions.

13. When making most decisions, I try to consider how
they will affect the people I do business with.

14. I know that most of the people I do business with are
dependable.

15. I can trust most of the men that I deal with.

16. The majority of the people that I deal with don't
know as much about the egg business as they ought to.

17. It is better to be your own boss than to operate under
a contract or agreement which ties you to someone.

18. I usually try to see the other person's point of view
even though he doesn't always try to see mine.

19. I try to be aware of what other people in this bus—
iness are doing in different communities.

 

20. I concern myself with the people and activities of my
own community rather than those outside it.

21. People in this country have become too specialized.

22. To be successful in this business one should stress
friendliness and tradition first, and planning and
decision making second.

23. People in this business should work together with
someone who can coordinate all of their production
and marketing efforts.

 

24° It is important for me to know who handles the eggs,
from production all the way to the consumer's table.

25° Everyone would be better off if they would forget
their own wishes and work to accomplish goals which
would help the egg industry as a whole.

26. Often the best solution to a problem will require

that some one individual must sacrifice for the good
Of the group as a whole.
27° ____We ought to work toward a system which would allow all

the people who produce and market eggs to make a decent
living, regardless of whether prices are high or low.

 

 

 

 

 

 

 

 

 

 

172

EGG CHANNEL MEMBER SURVEY
BIRD SUPPLIER RESEARCH INSTRUMENT

1—3 questionnaire number name:
4 ~01 feed supplier
:02 pullet supplier address:
_03 producer
O4 processor position:
05 distributor

06 retailer

5-6 location:

01 Michigan

02 Indiana

03 Ohio

04 New York

05 Mississippi

06 North Carolina
07 Georgia

08 Arkansas

7 data card number

8—13 _gross sales of birds in 1967? (thousands of S)

l4-l9 (thousands) number of pullets hatched and sold
in 1967?

20-22 number of egg producers that are your customers
now?

CONFLICT: card b with: (producer no: )

 

1 2 3 4 5 "How satisfied or unsatisfied have you
been with..."

the price that he paid for the last birds
that you supplied to him?

the terms of payment which relate to these
birds which were recently delivered to him?

his promptness in paying for these birds?

his demands for services; in other words,
are you satisfied that he doesn't demand
more service than he is entitled to?

the consistency with which he buys birds
from you; that is, is he a regular customer
that you can count on for repeat business,
or does he switch suppliers frequently?

this man as a person to do business with?

the care that this man has given to the
birds, once they were delivered to his
hen housing? (as reflected by egg size,
production, and mortality rates)

this man's feeling about when your respon-
sibility for the birds should end and
when his responsibility should begin?

 

 

 

 

 

 

 

 

 

 

 

173

BIRD SUPPLIER

COOPERATION: card c "How would you rate this person in

 

INNOVATIVENESS Year started in bird supply business:

23-25

26—28

29—31

32-34

35-37

38-40
41—43
44—46

47-49

59

60-61

62

 

 

terms of..."
his willingness to cooperate with you?

his willingness to follow any suggestions
or advice which you might give him?

his willingness to make suggestions to
you or to provide information to you?
Feedback.

his willingness to accept responsibility
for mistakes or poor results?

started in the pullet growing business

breeder flocks physically separated from
hatchery

use new chick boxes every time——don't use
same boxes to deliver chicks in time after
time

all the birds in the same house come in and
go out at the same time

 

wash and disinfect pullet delivery crates
between usages

use franchised birds
member of National Poultry Improvement Plan
use several feeds rather than just one

rely on or ask assistance from poultry sci-
ence extension people for feed information,
lighting program, or force molting infor—
mation

What was the last grade you completed in

01 0-8 years SChOOl? ;
02 1-2 years high school |
_ 03 3-4 years " " f
04 l-2 years college
N

05 3-4 years
06 more than 4 years college

I

 

What is your age?
01 male

h—_

02 female

 

 

 

 

174

EGG CHANNEL MEMBER SURVEY

PRODUCER RESEARCH INSTRUMENT

O
O
L'“

-3 questionnaire number
01 feed supplier

pullet supplier

O3 producer

O4 processor

05 distributor

O6 retailer

HHSI

I

5-6 location:

Ol Michigan

02 Indiana

03 Ohio

04 New York

05 Mississippi

06 North Carolina
07 Georgia

08 Arkansas

7 data card number

 

14-19 Could you describe your operation to me? For
instance...(in thousands) How many hens do you
have in production now?

INTEGRATION INDEX INFORMATION

l 01 yes Do you have an agreement or a contract
02 no with some person or organization which
03 dk covers the financing, marketing, or some
other aspects of your operation?

 

/ IF YES, PROCEED TO #2, OTHERWISE, SKIP TO #21 /

 

2 With whom do you have this agreement?
3 What type of agreement is this; that is, what is
this agreement called?
4 01 Written Is this agreement written or oral?
~02 Oral

 

 

 

 

 

 

175

Does this agreement require that...

someone other than you owns the hens?

you supply and keep the buildings and equip-
ment in working order?

you personally care for the hens?

you gather and oil the eggs in a specified
manner?

you carry out a fly control program?

you restrict the entrance of persons to the

10
areas where the birds are housed?

you keep flocks separated that vary in age
more than 60 days and keep the eggs separated
on the same basis?

11

you maintain daily inventory records of all
chickens?

all eggs, except for those used in your own
household for cooking, be marketed by the
contractor, or...

in the alternative, to market through the
contractor a stated minimum number of cases
per week throughout the year?

the eggs produced by chickens covered by the
agreement must be kept separate from eggs

by other chickens?

crate and load out old hens at the end
the flock year?

install a refrigeration unit in the egg
lding room capable of cooling the eggs to
specified minimum temperature?

contractor will supply vaccine and water
tation supplies?

contractor will supply directions, instruc—
and information to be of aSSistance

12

13

    

14

15

 

l6

 

  

l7

 

18

 

19

 

7

you?
contractor be permitted to deduct amounts
your check for contributions to egg pro—

on organizations?
21 What is the basis of the method of payment to you as a
producer?

01 receive entire price for the eggs
02 marketing agency pays premium price for the eggs

which are high quality

 

2O

 

 

 

 

 

 

176

O3 guaranteed price

fixed rate per dozen or per bird

OS incentive clauses for "good" feed use per dozen
eggs

6 profit—sharing; receipts less the expenses divided
on some basis between producer and coordinator

O7 rates tied to various well known price quotations

08 percentage of the egg check

I I;

L

l

l

 

 

 

 

 

 

09 other:
22 Who supplies your feed?
23 01 Buy Do you buy this feed, or is it sup—
02 Supplied/ plied by the contractor?
free
03 Other:
24 Who supplies your birds?
25 01 Buy Do you buy these birds, or are they
02 Supplied supplied by the contractor?
free
03 Other:
26 Who do you sell your eggs to?
27 Who does he sell these eggs to?

 

I

Do you...

Yes No
l 2

raise pullets?
manufacture own feed?
grade eggs?

candle eggs?

carton eggs?

deliver eggs?

grow grain or corn?

   

 

 

 

 

177

/CONTROL/

 

Now I would like to ask you about a number of decis—
ions commonly made by persons in the egg business. It would
be helpful if you could tell me the extent to which you or
someone else decides about these matters. (HAND CARD A) On
this card are listed a number of statements which describe
who might make a given decision. For example; (READ ALTER—
NATIVES) Now when I ask about each decision, could you please
answer in terms of the statements on the card?

Col 1 2 3 4 5
1

Who decides when the birds come in and
go out?

2 3
2 ---- Who decides which brand of feed, or feed

ingredients will be used?

3 E:E:]::I::I:j Who decides which breeds or strains of
hens are used?

4‘ E:E:1::I::I:] Who decides to whom the eggs will be sold?
5 am Who decides on the details of the vac-

cination program?

5 [:1::I::E::[:j Who decides how many laying hens are in

production at one time?
7 E:I::I::I::[:j Who decides what type of equipment will
be used in the laying house?

8 [:I::I::I::I:j Who decides whether you must accept the

Ibirds that are sent?

9 E:I::[::[::E:] Who decides what flock records are to be

kept?

10 E:I::E:]E::E:] Who decides on the method and time of
payment for the eggs?
M
/CONFLICT/ . .
Now let's change the subject just slightly. It is
Often true that when people in the egg buSiness deal filth
one another, they will be dissatisfied With the way t 129:
have been handled. They may feel more satisfied if cerl: n
Changes were made. It would be helpful if you could teh me
how satisfied or dissatisfied you are with respectdtg tre
following questions. (HAND CARD B) Listed on carin grim
a number of possible answers to each question rang 9

Very satisfied to very unsatisfied.

 

 

 

 

 

178

Now, for example, how satisfied or unsatisfied...

Co 1 2 3 4 5

Tim:

S

are you with your share of the return
for the eggs under the present day egg
price levels? (Oct. 1968)

would you be with your share of the re—
turn for the eggs, if egg prices rose
substantially from present levels?

would you be with your share of the re-
turn for the eggs if egg prices fell
substantially from present levels?

have you been with the prices charged
for the pullets you have recently re-
ceived?

have you been with the mortality rates
of the pullets you have recently re—
ceived?

have you been with the egg production
rates of the pullets you have recently
received?

have you been with the prices charged
for the feed which you have recently
received?

have you been with the quality of the
feed which you have recently received?

have you been with the actual perform-
ance of the hen house equipment compared
with what you had expected from it, per-

formancewise?

If independent, skip to next section; if non—independent,

Proceed to #10.

1 2 3 4 5
10

11:13:33

have you been with the amount and quality
of advice given to you by the contractor?

have you been with the terms of the agree-
ment, in general?
have you been with the process by which

' ' her renewed or not
our agreement is eit .
Zenewed at the end of the laying cycle?

(Collect Card B and hand out Card C)

 

 

 

 

179

/COOPERATION7

Now I would like to ask about another topic. People
differ in terms of the ways in which they cooperate with
one another. That is, they may help each other a great deal,
or they may not be helpful at all. I would like to ask you
about the kinds of cooperation you may or may not get from
the people in the egg business with which you deal. With
respect to the following questions, could you rate the co-
operation given by the other party as being "very good,"
"good," "fair," "bad," or "very bad." For example, hgw
would you rate the...

f b vb

feed supplier in terms of his willingness
to cooperate with you?

feed supplier in terms of his help in
financing your feed purchases?

feed supplier in terms of his help with
your business management problems?

feed supplier in terms of his aid in find-
ing a satisfactory market for the eggs?
pullet supplier in terms of his willing-
ness to cooperate with you?

pullet supplier in terms of his help in
financing your pullet purchases?

pullet supplier in terms of his help with
your business management problems?

pullet supplier in terms of his aid in
finding a satisfactory market for the eggs?
marketing agency in terms of his willing—
ness to cooperate with you?

marketing agency in terms of his help with
your business management problems?

marketing agency in terms of his aid in
finding a satisfactory market for the eggs?

 

(Collect Card C)

W

Now I would like to ask about your use of certain
practices. For each item which I mention, I would like :0 t d
know whether you presently use it, and if so, when you 3 ar e
“Sing it. Could you please tell me if you use...

 

 

 

 

 

180

year started in egg business:

23—25
26—28
29-31
32-34
35—37
38-40
41—43
44—46
47-49
50-52
53-55

a contract or agreement

a cage system for confining hens

bulk feed

started pullets, rather than chicks
chemical fly control program

farm—account record keeping

air circulation equipment in hen housing
controlled lighting program

quarterly (or more frequent) manure disposal
medication in water

monthly (or more frequent) records of flock
mortality

monthly (or more frequent) records of flock
productivity

56-58

 

What was the last grade you completed in
school?

59 01 0-8 years

02 l-2 years high school

03 3—4 years high school

04 1—2 years college

05 3-4 years college

06 more than 4 years college

 

 

 

I

 

 

 

 

60-61 What is your age?
62 01 male
02 female

 

(Plus Q—sort statements-—see Feed Supplier Research instrument)

 

 

 

 

 

 

n
O
L"‘

brdFJ
l
WU)

 

181

EGG CHANNEL MEMBER SURVEY

PROCESSOR-DISTRIBUTOR RESEARCH INSTRUMENT

questionnaire number

questionnaire number

01 feed supplier

 

 

 

 

02 pullet supplier
O3 producer

04 processor

05 distributor

 

O6 retailer

 

location:

 

7

8—13

14—19

20—22

Michigan
Indiana

Ohio

New York
Mississippi
North Carolina
Georgia
Arkansas

data card number
(thousands $) Gross sales of eggs in 1967?

(thousands) number of cases of eggs handled per
week

number of producers that are your customers now

What do you do to the eggs?

 

pick up eggs from producer
grade

carton
deliver to customer

which (retail or other) outlets do you sell your eggs to,
where, and who?

 

 

 

 

 

 

182

PROCESSOR-DISTRIBUTOR
CONFLICT: card b with: (producer no:
_______________
“How satisfied or unsatisfied have you been with..."

1 2 3 4 5

 

the prices that you have had to pay this
man for eggs recently?

 

the quality of the eggs that you have
recently bought from this man?

 

the ability of this man to supply you
3 with the number and grades of eggs that
you want to sell?

the consistency with which he supplies
eggs to you; that is, is he a regular

 

 

 

 

4 supplier of eggs, or does he send them
to you on an irregular basis?
5 ~ this man as a person to do business with?
his complaints about his grade—out slips?
6 (number of eggs per grade, and number of

 

 

 

 

 

 

grades being used)

 

COOPERATION: card c "How would you rate this person in
terms of..."

his willingness to cooperate with you?
his willingness to follow any suggestions
or advice which you might give him?

his willingness to make suggestions to
you or to provide information to you?
Feedback.

his willingness to accept responsibility
for mistakes or poor results?

   

 

 

 

183
PROCESSOR-DISTRIBUTOR

CONFLICT with: (retailer no: )

"How satisfied or unsatisfied have you been with..."

the prices this man pays you for your

eggs?

the physical facilities he has for display—
ing and selling these eggs?

the usual daily price he charges his cus-
tomers for these eggs? (including the
retailer's margin)

his promptness in paying you for the eggs
he buys?

his demands for service; in other words,

are you satisfied that he doesn't demand
more service than he is entitled to?

the consistency with which he buys eggs
from you; that is, is he a regular customer,
or does he buy on an irregular basis?

this man as a person to do business with?

 

COOPERATION card c "How would you rate this person in
terms of..."

his willingness to cooperate with you?

his willingness to follow any suggestions
or advice which you might give him?

his willingness to make suggestions to
you or provide information to you?
Feedback.

his willingness to accept responsibility
for mistakes or poor results?

   

 

 

 

 

 

23-

26—

29—
32-
35—
38-

41—

44—

47-

50-

59

60—

62

184

PROCESSOR-DISTRIBUTOR

Year started in processing business:

25 promote tie—in promotional programs with

other products (eggs n' hash)

28 offer use to retailers of a vertical egg

merchandiser
31
34
37
40

give advertising allowances to retailers
use 2 l/2 or 3 dozen cartons

mass candler

pre—cool egg cartons and cases before
cartoning eggs

mechanical egg packing equipment (vs. hand
packing

43

46
49

private brand eggs

either a long—term breaking agreement, or
their own egg breaking and broken egg
marketing operations

52 tied in with an integrator

 

"What was the last grade you completed in
school?"

01

O 8 years
02 l—2 years high school

03 3—4 years high school

04 l 2 years college

05 3~4 years college

06 more than 4 years college

 

 

 

 

 

 

61 What is your age?

 

01 male
02 female

 

I

 

 

 

 

 

7

8-13
14—19
20—22

CONFLICT

 

 

 

 

185

EGG CHANNEL MEMBER SURVEY
RETAIL RESEARCH INSTRUMENT

number:
questionnaire number ,
Ol feed supplier name.
02 pullet supplier
O3 producer address:
04 processor . .
pOSition:

05 distributor
O6 retailer

 

 

location:

 

Michigan

Indiana

Ohio

New York

Mississippi

Alabama

Georgia

Arkansas

data card number

(thousands S) gross sales of eggs in 1967
number of cases of eggs sold per month

number of stores represented

 

 

 

 

card b with: (producer no:
/or/ with: (processor-distributor no:

"How satisfied or unsatisfied have you been with..."

the prices that you have had to pay this
man for his eggs recently?

the quality of the eggs that you have
recently bought from this man?

the ability of this man to supply you
with the number and grades of eggs that
you want to sell?

the consistency with which he supplies
eggs to you; that is, is he a regular
source of eggs, or does he send them to
you on an irregular basis?

this man as a person to do business with?

)

 

 

 

 

 

COOPERATION

 

INNOVATIVENESS

23-25
26-28
29-31
32—34
35-37
38-40
41—43
44—46

 

 

 

 

 

 

card c

 

186

RETAIL

"How would you rate this person in
terms of..."

his willingness to cooperate with you?

his willingness to follow any suggestions
or advice which you might give him?

his willingness to make suggestions to
you or to provide information to you?
Feedback.

his willingness to accept responsibility
for mistakes or poor results?

Year started in retail business:

of tie—in promotions
of a vertical egg merchandiser
of 2 1/2 or 3 dozen cartons

of eggs in week-end specials

point of purchase promotional material

wire baskets in egg cases

records of sales of eggs in dozens per
$1,000 (or other such unit) of grocery

sales

47-49 use of bids to purchase eggs
50-52 contract with processor or distributor
53‘55 ownership of egg production facilities
56‘58 private (retail) branded eggs
"What was the last grade you completed in school?"

59 01 0—8 years

02 l—2 years high school

03 3-4 years high school

04 l-2 years college

05 3—4 years college

06 more than 4 years college
60‘61 What is your age?

 

62 Ol male
02 female

 

 

in—store demonstrations or sampling of eggs

 

 

 

 

APPENDIX C
ITEM ANALYSIS RESULTS
AND OPERATIONALIZATION OF VARIABLES
The purpose of this section is to discuss the reasons
for and the methods used in the item analysis employed in
the study, and to discuss the means by which each variable

was operationalized.

Item Analysis

The purpose of the item analysis undertaken in this
study was to provide an index of validity of each question
in order to select a set of unidimensional questions to meas-
ure each variable. The term validity is used here to mean
a measure of how well the item measures or discriminates in
agreement with the rest of the questions used to measure each
variable. The index of validity used in this item analysis
was the factor loading of each item on a common factor.1
The factor loadings indicate how strongly each question is
like the dimension which underlies the questions in its group.
Questions with high factor loadings, of like sign on a common
factor, were selected for use in measuring each variable.

The factor loadings of each question used to index
each variable after eliminating some in each research instrument

M—

lJ. P. Guilford, Psychometric Methods (New York;

 

Eggraw—Hill Book Co., Inc.; Second Edition, 1954), pp. 417-

o

 

 

 

 

 

7

188

are listed in Table C.1. As noted in this table, more than
seventy percent of the questions used to index all the vari-
ables had factor loadings of .60 or higher. In some cases,
questions with factor loadings lower than .60 were used to
measure a variable because almost all persons had answered
these questions, whereas practically no one had answered

the rest of the questions which dealt with the variable.
When selecting questions to measure each variable, it was
relatively more important to choose a single question which
had a 100% response and which might have been relatively
unlike the other questions in the set, than to choose two

or more questions which were seen as relatively alike but
which had very low rates of response. Each question used

to index each variable for each research instrument is listed

in Appendix D.

Operationalization of Variables

The purpose of this section is to discuss the pro—
cedures used to provide quantitative measures for each per—
son on each variable. These procedures will be discussed

as theY relate to each variable.

Cooperation. Each person's measure on cooperation
was the mean of his responses to the operational questions
Which he answered. For instance, questions 1 and 2 of the
feed suppliers' instrument were used to index this variable.
The values of the response categories for cooperation ques-

tions ran from 2, which indicated very bad cooperation, to 8,

 

 

 

 

 

7

189

which indicated very good cooperation.2

Assume that a feed supplier's responses were:

question
number response
1 8
‘2 6

This individual's cooperation score was the mean
of his responses, (8 + 6)/2 or 7.

Conflict. Each person's measure on conflict was

the mean of his responses to the conflict questions which

he answered. Table C.1 also indicates the numbers of the
questions used to index conflict in each of the five instru—
ments. For instance, the feed supplier's conflict measure
was the mean value of his responses to conflict questions

3, 4, 5, and 6. The values of the response categories for
conflict questions ran from 1, which indicated that the re—
spondent was very satisfied, to 5, which indicated that he

was very unsatisfied.3

Innovativeness. Each person's innovativeness meas-

ure was the mean of his responses to the relevant innovativeness

——-———_.____

2The response categories received the values: 8=very
900d, 7=good, 5=fair, 3=bad, and 2=very bad. These values
were selected on the basis of a study of the choice of words
and phrases for use in questionnaires and public opinion
POlls, conducted by: Stuart C. Dodd and Thomas R. Gerbrick,
"Word Scales for Degrees of Opinion," Language and Speech,
Vol. 3, Part 1 (January—March, 1960), pp. 18-31.

3This is a modified 5 point Likert scale. The values

in the scale were: l=very satisfied, 2=satisfied, 3=unde-
cided, 4=unsatisfied, and 5=very unsatisfied. For more in—
fOrmation on the Likert method, see Guilford, op. cit., pp.
459—460.

 

 

 

7

190

questions which he answered. The numbers of the items in
each of the five innovativeness scales are indicated in
Table C.1. Innovativeness of feed suppliers was the mean
value of the reflected year of adoption of the items listed
in questions 1, 4, 5, and 6. The reflected year of adoption
was found by subtracting the actual year of adoption from
1969. Thus adoption in the year 1968 would equal 1, and
adoption in the year 1965 would equal 4. A higher mean
value indicates earlier adoption and hence greater inno-

vativeness.4

Control. Control was operationalized in the manner
suggested by Baligh.5 The extent to which others in the
channel control a producer was determined by the proportion
of the latter firm's decisions which the former made. Ten
questions, which appear in the control section of the pro—

ducer instrument, were asked to each producer about who

“——

4For a discussion of the validity, reliability, in-
ternal consistency, and unidimensionality of innovativeness
scales, see: Everett M. Rogers and Edna Rogers, "A Methodo-
lOgical Analysis of Adoption Scales," Rural Sociolo , Vol.
26, No. 4 (Dec., 1961), pp. 326-336, and Everett M. Rogers,
"Categorizing the Adopters of Agricultural Practices," Rural
Sociology, Vol. 23, No. 4 (Dec., 1958), pp. 345—354.

5Helmy H. Baligh, "A Theoretical Framework for Chan-
nel Choice," in P. D. Bennet (ed.), Economic Growth, Compe-
tition, and World Markets (Chicago: American Marketing As—
sociation, 1965), pp. 63l~654. Reprinted in Bruce E. Mallen,
The Marketing Channel, A Conceptual Viewpoint (New York:
JOhn Wiley and Sons, Inc., 1967), pp. 204-217.

 

   

 

 

 

lZII:T___________________________——__I
191

controlled certain decisions. The values for the response
categories ran from 1, in which the decision was completely
controlled by the producer, to 5, in which the decision was
entirely made by someone else in the channel. The producer's
control measure was the mean value of his responses to con-

trol questions 1—10.

Operational Size. A different measure of operational
size was employed for each of the five channel levels. Raw
data from each respondent was standardized across his channel
level; each respondent's size was thus indicated in terms
of a z—score—-which indicates the distance away from the
mean of his channel level in standard deviation units. This
procedure provided a channel—level-free size measure which
would allow the comparison of each person in the study with
every other person.

The size of feed suppliers was measured in terms

Of feed sold for chickens in 1967. Bird Suppliers' size

 

was measured in terms of the number of pullets hatched and
Sold during 1967. Producers' size was measured in terms

Of the number of hens in production at the time of the inter—
view. Processor—distributors' sizes were operationalized

in terms of the number of 30 dozen cases of eggs handled

per week during the past year. Finally, retailers' sizes
were operationalized in terms of the number of 30 dozen cases

Of eggs that were sold per month during the last year.

Age. Each respondent was asked his age.

 

 

 

 

77

192

Integration. The procedure by which the integration
index was operationalized for each person and channel in

the study is discussed in Appendix D.

 

 

 

 

 

EIII:_______________________________——_I

193

Table C.1
VALIDITY OF QUESTIONS USED IN EACH INDEX FOR
EACH RESEARCH INSTRUMENT, AS DETERMINED BY FACTOR ANALYSIS

Question Factor
Ngmber‘ Loading“

Variable Channel Level

 

Coopera— Feed Supplier's Rating of

 

l .70

tion Producer 2 93

Bird Supplier's Rating of 1 .85

Producer 2 .85

Producer's Rating of Feed Supplier 1 .75

2 .93

3 .84

4 .92

Producer's Rating of Bird 5 .40

supplier 6 .73

7 .87

Producer's Rating of Processor 9 .64

or Retailer 10 .82

11 .43

Processor's Rating of Producer 1 .70

2 .25

3 .88

4 .83

Processor's Rating of Retailer 1 .23

2 .75

3 .87

4 .87

Retailer's Rating of Processor 1 .53

or Producer 2 .65

3 .83

Conflict Feed Supplier's Rating of 3 .21

Producer 4 .87

5 .42

6 .86

Bird Supplier's Rating of 4 .62

Producer - 5 .63

6 .78

7 .65

8 .75

Producer's Rating of Processor, 1 .89
Retailer, or Contractor 2

.89

 

 

 

 

194

Table C.1 (continued)

 

Question Factor
Variable Channel Level Number‘ Loading"

Conflict Producer's Rating of Bird 4 .86
(cont'd.) Supplier S .48

Producer's Rating of Feed 7 .11
Supplier 8 .12

Producer's Rating of Contractor 10 .57
11 .82
12 .83

.63
.87
.81
.34
.58

.36
.56
.68
.16
.74

.78
.86
.70
.79

.62
.51
.58
.65

.43
.80
.74
.77
.64
.34

.35
.68
.73
.74
.87
.85

Processor's Rating of Producer

Processor's Rating of Retailer

 

Retailer's Rating of Processor
or Producer

Innova- Feed Supply
tiveness

Bird Supply

Producer

NHCDQUOH \Dm’x'lO‘IU‘H—l 03th U11>NH \JONU'IIbl-J OWU'IDPWN

F‘H

 

 

 

 

 

 

 

195

Table C.1 (continued)

 

Question Factor
Variable Channel Level Number‘ Loading“

Innova- Processors
tiveness
(cont'd.)

.42
.82
.76
.75
.75

.51
.37
.65
.67
.74
.79
.79
.45

bWNl-J

Retailers

NI‘\OGJQLHUJR) w

H14

 

‘The questions are listed in Appendix B, which in-
cludes copies of each research instrument.

"The factor loadings indicate how strongly each ques-
tion is like the dimension which underlies the questions in
its group. Questions with high factor loadings, of like sign,
were selected for use in each index, to move each index to-
ward unidimensionality.

 

 

 

 

 

 

 

APPENDIX D
FORMULATION OF THE VERTICAL

INTEGRATION INDEX

The index of vertical integration which applies to
all of the firms within a particular channel is determined
by the degree to which: (a) control is exercised over pro—
ducers; (b) the price risk is shared among system members;
(c) the members are committed to their channel and its mar—
kets; and (d) several production and marketing functions

have been assumed by a single channel member.

Components of the Vertical Integration Index

1. Vertical Control. Vertical control is exerted

 

over the producer by persons at other channel levels. This
control is most often exercised by the contractor, and re—
lates to decisions made about the management of the egg
Production facility. Vertical control is measured by an
index (explained below) which is based on questions included
in the producer research instrument. A high degree of ver—
tical control is consistent with a high degree of vertical
integration.

2. Price Risk. The risk which results from drastic
fluctuations in the price of eggs is known as price risk.
Traditionally, the risk of low profits or even losses has

been borne solely by the producer. We recognize two

196

 

 

 

 

 

 

 

 

III[IIIIT_____________________________I

197

dimensions regarding price risk. The first of these involves
the number of parties who share the price risk. The second
involves the degree to which price risk is shifted from the
producer to other channel members. A high degree of risk
shifting would usually be associated with a high degree of
vertical integration.

3. Commitment to Channel. This refers to the na-
ture of the commitment of an individual or business firm
to a particular channel. This commitment is operationalized
in this study by the type of agreement or contract that ex-
ists. At the low end of the commitment scale is the situa—
tion in which no agreement exists to link the members of a
channel. A longer—run commitment exists when the producer
agrees to participate in a production contract with an inte-
grator for a year or more. The greatest commitment is dis-
played by owner—integrated operators who invest in special—
ized physical plants, and who incur heavy fixed costs in
order to gain a high degree of specialization. A high de-
gree of commitment is consistent with a high degree of ver-
tical integration.

4. Structural—Functional Configuration. This term
refers to the number of persons who constitute a channel
and the degree to which the various production and market-
ing functions are assumed by a single channel member. In—
C1uded in this definition are the following channel levels,

each of which has been traditionally characterized by the

 

 

 

 

 

 

 

 

Ii—
198

performance of a unique function: bird supply, feed supply,
egg production, and processing and distribution. When com-
puting the structural—functional measure for each channel,
the retail function was deleted because similar types of
retailers sell eggs which come from widely differing types
of channels.

The structural functional measure for each channel
is operationalized by computing the value added by each of
the several functions which have been assumed by a single
firm in the channel. The term 'value added' refers to the
cost of producing and marketing a commodity at each level
in the channel, excluding the retail level.1 Thus, when a
dozen eggs are purchased by a supermarket from a distributor,
the price paid represents the cumulative value added by all
the individuals involved in egg production and marketing
up to the retail level. The value added to a dozen eggs
along a channel has been estimated in percentage terms as
follows:2

_—~—_—___—

1Value is always added at each level in a channel.
However, the present study does not consider the value added
by the retail level because any one retailer may sell eggs
which come from many different channels which vary widely
in vertical integration.

 

2These estimates of value added are provided by:
Dr. Henry E. Larzelere, in "Current Styles and Shapes in
Egg Contracting—II," Agricultural Economic Report No. 39,
Feb., 1966, Dept. of Agricultural Economics, Michigan State
University, East Lansing, Michigan, p. 17. These estimates
Of value added apply at an average price level, and will vary
Slightly with changing price levels.

 

 

 

 

 

 

 

 

Ii—
199

Function Performed by Channel Level Value Added

feed supply function 44%
bird supply function 23%
egg production function 14%
processing and distribution 19%

total cost to retail level 100%

Weighting of the Components of the
Vertical Integration Index

The integration index is based on these four com—
ponents: control, risk, commitment to the channel and
structural—functional configuration. Lacking evidence that
any single component is more important than the others in
contributing to our understanding of vertical integration,
each of these components receives equal weight in the index.
The value of this index can run from O to 100 and each of
the four components has a maximum of 25 points.

The values used in computing each component of the

integration index are listed in Table D.1.

Explanation of the Assignment of Values to Components
1. Price risk assumption by others in the channel.
This component is broken into two dimensions: (a) the num-

ber Of parties who share the risk, and (b) the degree to

 

which the risk is shifted from producers to other channel
members. The weights assigned for various possibilities
along each dimension are given in Table D.l. Both dimen—
sions, when summed together, have a maximum possible value
0f 25 points. Of these 25 points, dimension (a) receives

40 percent of the possible points while dimension (b)

 

 

 

 

 

 

 

 

 

200

Table D.l

 

VALUES EMPLOYED IN COMPUTING INDIVIDUAL INTEGRATION
INDEX MEASURES FOR EACH CHANNEL

 

Name of Integration

Characteristic of Channel Member

 

Index Component Value
1. Price Risk (a) Number of parties who share price risk:
Assumption by 0.0 oducer only
Others in 5.0 producer and one party
10.0 producer and two parties

Channel

(b) Degree to which price risk is shifted:
producer receives entire egg check
premium price paid for AA quali
producer receives percentage of egg check
fixed rate per dozen eggs produce
fixed rate per bird h ous

guaranteed price
risk spread over entire owner— integrated operation

 

2. Channel Control

(a) For L t J ‘ ntrol
index values based on producer questionnaire:

 

 

Over Producer
0.0 1 O
5.0 2.0
10.0 3.0
15.0 4.0
20.0 5.0
(b) For owner—integrated operations, control is
25.0 exerted by the total system
3. Structural 3.5 no combination of functions, value added by
Functional production is assigned 4/4=3,5
Configuration 8.25 production, proc—dist 14+19=33/4=s_25
.25 production, bird 14+23=37/4=9.25
10.25 proc-dist, bir 19+23=42/4=10-5
14.0 production, proc-dist, bird 14+19+23=56/4=14_0
14.5 feed, production 44+14=58/4=14_5
15.75 proc-dist, feed l9+44=63/4=15_75
16.75 bird, feed 23+44=67/4=16_75
19.25 feed, production, proc—dist 4 +14+19=77/4=19_25
25.0 feed, bird prod' n, proc— —dist 44+23+14+19=100/4=25,o
4. Commitment 0.0 no agreement or contr
to Channel 1.0 oral agreement, no specific length
7.0 written agreement, 60 day notice to quit
16.0 written agreement, 1 year or 1 cycle
17.0 written agreement, 2 year or 2 cycles
19. written agreement, 3 year or 3 cycles
21.0 written agreement, 5 year or 5 cycles
23.0 Written agreement, 7 year or 7 cycl
25.0 OWner— —integrated operators: Commitment in development
long- run investment

of specialized expertise,
physical facilities

 

 

 

 

 

 

 

 

 

7
201

receives 60 percent of the points possible in the price risk
component of the integration index.

Dimension (a) receives less weight than dimension
(b) to emphasize that the number of people who share the
risk contributes relatively less to channel stability than

is contributed by the means by which that risk is shared.

 

2. Channel control over producer. Vertical control

 

is indexed as a variable in the producer research instrument.
The possible vertical control scores in the research instru—
ment run from 25 which occurs when the producer has total
control over the decisions, to a possible 0, which occurs
when the producer has no control over selected decision areas.
3. Structural-functional configuration. The values
listed in Table D.1 for various alternative structural-func—
tional configurations were derived by first summing the
values added by each function when two or more functions
are performed by one operator, and then dividing by 4.
The following rules are used in computing the struc—
tural-functional value for each channel. If each level in

the channel is independent, the structural—functional (S—F)

 

index which applies to the egg production function (3.5) is
assigned to the channel. If two functions are performed by
One firm the values added by each of those functions are

summed and divided by four° For instance, if one business
firm supplies both the feed and the birds to the producer,

the value added is 23 + 44, which equals 67. This, when

 

 

 

 

 

I———
202

divided by 4, gives a S-F score of 16.75 (see Table D.1).

4. Commitment to channel. This factor is operation—
alized by the existence or non-existence of an oral or written
contract between the producer and some other party in the
channel. If the producer has entered into a contract or
agreement, then the length of time and the oral-versus—written
nature of the agreement determine the value of the commitment
score (see Table D.1).

The commitment scale runs from O to 25; the low end
of the scale represents a situation in which all channel
members are independent of one another. The middle values
on the scale represent situations in which a written contract
links channel members--the longer the contract runs, the
higher the value it receives on the commitment scale. The
high end of the scale represents owner-integrated channels,
which because of their long run high fixed costs are highly
committed to the egg business.

The values included in the commitment scale are the
author's attempt to estimate the approximate degree of inte—

gration inherent in each type of agreement studied. A pro-

 

ducer who is independent, or who sells eggs to a processor
by means of an oral agreement, is free to switch processors
at will. His integration is very low and hence he is given
a commitment value of O or 1. Alternatively, the producer
who has signed an agreement to deal with a channel for one

or more years displays a relatively high degree of integration__

 

 

 

 

 

203

his score ranges from 16 to 23, depending on the length of

his agreement.

Examples of the Calculation of tag Integration Index

The procedure for calculating the integration index
is illustrated in the following examples:
Example 1

Producer 015 buys feed from a local feed mill and
birds from a local hatchery on an independent basis. He
usually sells his eggs to a local processor who, in turn,
usually sells the eggs to the purchasing agent for a major
food chain. None of these relationships is based on any
contract or oral agreement.

Producer 015's integration score (which also applies
to the other members of his channel) is calculated as follows:

Component Egggg

1. Price Risk

a. Producer bears entire risk
b. Producer receives entire egg check

2. Channel Control

a. Vertical control index on producer instru-
ment is 1.0; reduce this by 1.0 and mul-
tiply by 5.0, result is 0.0 0.0

OO
00

3. S—F Configuration

No combination of functions, value added by
production is assigned 3,5

4. Commitment

No agreement or contract

0

.0

-

Total Integration Index for Producer 015

(J!

Example 2

Producer 018 signed a written contract for one year

 

 

 

 

 

 

204

with a local feed company. The feed mill owns the hens and
the eggs and pays the producer a fixed rate per bird housed.
The eggs are picked up by a processor who sells to other
processors and to retailers.

The integration index for producer 018 is computed
as follows:

1. Price Risk

a. Shared by producer and one party 5.0
b. Producer paid fixed rate per bird housed 11.0

2. Control

a. Vertical control index on producer in-
strument is 4.6; reduce this by 1.0 and
multiply by 5.0, result is 18.0 18.0

3. S—F Configuration
a. No combination of functions, value added

by production is assigned 3.5
4. Commitment
Written agreement, one year 16.0

Total Integration Index for Producer 018 53.5

 

 

 

 

 

APPENDIX E

INSTRUCTIONS FOR INTERVIEWERS

(Cover Letter)

December 20, 1968

Dear Sir:

Thank you for agreeing to help me with this study of the
poultry industry. I appreciate your aid in gathering infor—
mation on egg production and marketing in your state. Your
area has become such an important factor in the egg industry
nationally that it is most advisable to include the data-
which you can supply to lend perspective to the information
I have collected about egg producers and handlers in this
area.

This study is an attempt to learn more about the attituges
and behavior of different types of people Within th: inrgs—
try. It thus includes interviews With not only pro uce_ ,
but with feed suppliers, chick and started pullet suppliers,
and processors and distributors.

Within the State of Michigan I drew.a mOdlfled qu°§2 Eaifiif
0f 60 egg producers who represent diffeFentltyPiie ratgd
tionS. which vary accorggna.t:iggg ;::§::::syw:s'agked the
the re. Ea h of the 1C - -

ham: :nd addrgss of the parties who suppliedfhtie::egoagged
birdS, and to whom he sent his eggs- each 0 was then sought
mills, hatcheries, and prOCeSE’m-“C‘is’trlbutors

Out and interviewed.

' I und
The interview generally took 20 to 35 minutes, :ngonv::ient
that by telephoning each respondent to arrange

time and place to meet they were all willing tpngieigggrgy
Viewed. I believe that their cooperation was

205

 

 

 

 

 

 

206
page 2

the fact that I told them that: (a) no one is financing
this study and therefore there is no vested interest or axe
to grind; (b) anything they told me was absolutely confiden-
tial and would not be associated with their name or with

the name of their firm; and (c) this study is being done

for a thesis in a graduate school and would thus hopefully
lead to a better understanding of the egg industry because
the results will be published.

While the following instructions may at first look compli—
cated, they should be relatively easy to follow. I have
tried to make them detailed and comprehensive so that you
will be able to complete these few interviews in the short—
est possible time.

Again, thank you very much. I would appreciate it if you
could complete these interviews and return the material to
me in the stamped, self-addressed envelope by January 15th.

Sincerely,

James L. Wiek

Ph.D. Candidate in Marketing
Graduate School of Business
Michigan State University

JLW/ls

Enclosures
Under separate cover

 

 

 

207

(INSTRUCTIONS FOR INTERVIEWING)

This packet includes:
12 interview schedules: 3 for producers, 3 for feed
suppliers, 3 for bird suppliers and 3 for processor—
distributors;
1 set of 27 3x5 index cards (called Q-sort cards), each
of which has a short statement written on it;
1 manilaecolored piece of cardboard divided into 9
spaces for the respondent to place the Q-sorted cards
upon;
2 sets of response cards, a, b, and c;
1 stamped, self-addressed envelope to use in returning
these materials to me.

Sampling Procedure

Egg Producers——we would like to have you interview
3 egg producers in your area. If possible it would be de—
sirable to select one of each of the following types of
operations:

A. Contract producers who bear no price risk. These
producers are linked to a contractor by means of an oral
or written agreement which covers at least one flock cycle.
They are paid a flat fee per bird, per dozen eggs, or per
house capacity. The birds are owned by the contractor who
also supplies the feed. Housing, labor, and certain other
inputs are supplied by the producer. . .

B. Contract producers who share the price risk.
This form of production differs from (A) above in that the
producer is paid a percentage of the egg check or is paid
On some basis whereby the amount of his compensation(depends
on the market price of the eggs which he produces. Ygur
region may not have both types (A & B) of contracttpr: uc:rs.
If so; then please interview 2)contract producers a ar
0f ' available.

the tg?eO§:::—i:tegrated producers. These producers fie

highest in terms of the extent to which they are Kertigao y
integrated. Not only do they produce eggs, but t ey a s1
assume one or more of the following functions. 1 iupghgir
their own feed through their own feed mill, (2) :ugphyr
Own chicks or started pullets through their own_ a c :shing
(3) processes their eggs through their o?2)ggigtggbu:e the ,
candling, and cartoning facilities, and 'obbers on their
eggs to retail stores or institutions or to J

own trucks.

Other Channel Members"-While igteereglngoegggcggges
ducer, ask who Supplies his feed and blrds an gind that one
and distributes the eggs he produceso You may In the case
firm will perform 2 or more Of these f9nCti°n§$rm one or
Of the owner-integrated producer, pg will Perh of these
more of the functions himself. Interview eac

named persons.

 

 

 

 

 

208

Interview Schedule Contents

Each interview has the following four sections:
1. Questions about the respondent. The type of contract
he may have, the terms of the contract and the size of his
operation, who deals with him, etc.
2. Questions about his attitudes and behavior. For the
producers, these ask about who makes certain decisions, how
satisfied he is with respect to certain matters and how much
cooperation he gets from the people he deals with.
3. Questions about the respondent's use or lack of use of
certain practices or innovations.
4. A series of 27 statements about the egg industry to be
agreed or disagreed with, according to Q-sort procedure.

Instructions for Q—Sort

Read the following to the respondent:

"This is a procedure for determining the nature of
each respondent's attitudes about the egg industry. It con—
sists of a set of statements written on a 3x5 card. There.
are no right or wrong answers, we just want to find out which
statements each person agrees with and which ones he disagrees
with. Please look at each statement and decide whether you
agree or disagree with it. To make this easy for you, I
will lay this piece of cardboard down for you to place each
statement on.

(Note-~1ay the cardboard down so that the respondent can.
see it and don't hand the cards to him yet or he won't listen
to the rest of your instructions.)

"You will see that this piece of cardboard is diVided
into 9 separate sections, each of which represents a differ—
ent degree of agreement or disagreement. "For instance, Eze—
this left end, space number 1 represents very strong air n
ment." The right end space number 9 represents verytst 0 its
disagreement." Space number 5 in the middle is for sta em:
that YOU are neutral about. Now do you have any ques logs _
O.K. Read each statement, decide how much you agree or tis
agree with it and then place the statement, face—up, on e
number.
One other thing, after you have looked atte:ChC§::fi
and Placed it on a numbered section, you will wan aph sec—
and see that the number of statements you havihop :ection.
tion is equal to the number at the bottom of a section 7
So YOU will have 4 cards in section 6, 3 card: inn the mid-
and $0 on. You will finish with the most car stions? O.K."
dle, the least on the two ends. Any other(gu:sof instruc—

(Hand the respondent the cards.) n ndent cannot
tions to be read to respondeifi.) téliimh? respo
read the ar 5 lease read em .

Afteg th: respondent has seen all the giggimigtzee
and has placed them on the numbered sections,

 

 

 

 

 

 

209

that the proper number of cards is in each section. If not,
then please explain that he must have the number of cards

in each section that is designated at the bottom of that
section. Suggest to him that perhaps a card out of an ad-
joining section might fit into this section as well and help
him to select the right number of cards so that each section
contains the desired number. This is most important, so I
urge you to encourage the respondent to consider the cards
until they are in the order that he wants them.

When he has all the cards appropriately sorted, you
are finished with him. Then you pick up the cards in each
section, one section at a time, turn them over and find the
code number at the bottom. This number is merely the number
that corresponds with the same statement on the questionnaire.
You might find, say, a number 124 in section 9. You then
transfer the number (9) of the section from which you picked
up the card to the blank beside the statement in the ques—
tionnaire that corresponds with the number you found on the
back of the card (124). (Question number 124 would now have
a 9 in front of it to show that you found card number 124
in section 9 and so on for all the other cards.) That is
all for the producer questionnaires. The other question—
naires follow the pattern of the producer questionnaires,
so they should be even easier to give.

 

 

 

 

 

 

APPENDIX F

PROCEDURES EMPLOYED IN THE TYPE ANALYSES

The purpose of this appendix is to explain the de—
tails of the procedures used to analyze the data which re-
late to hypotheses 5.1 and 6.1.1 This appendix is organized
into three sections: (1) factor analysis procedures, (2)
criteria for selecting a factor solution for further analy-

sis, and (3) procedures for describing each type.

Factor Analysis Procedures

The principal mode of analysis of the data which
relates to hypotheses 5.1 and 6.1 was factor analysis. Two
such analyses were run: (a) analysis 5.l--the analysis of
the Q-sorted responses of 94 individuals, (b) analysis 6.1—-
an analysis of the standard score measures2 on innovative-
ness, education, age, conflict given, cooperation given,
conflict received, and cooperation received for 35 channels.

In analysis 5.1, although 150 persons completed the
Q-sort section of the research instrument, the Control Data

M

lThe analysis of the data which relate to Hypothesis
5.1 will be referred to as 'analysis 5.1,' while the one
which relates to hypothesis 6.1 will be called 'analysis 6.1.‘

2These data were converted to standard scores so
that the channels may be compared on the basis of their re—
sponses to each of several variables. In raw form the data
On each of these variables were expressed in units which
differed, such as dollars, pounds, number of chickens, and
so on, and hence were non-comparable.

210

 

 

 

 

 

—_—i

211

3600 computer at Michigan State University has a capacity
restriction of one hundred persons in a varimax factor an-
alysis rotation. To conform to this restriction, the per—
sons selected for analysis were those who participated in
only one channel, i.e., all channels are represented in the
final analysis, but persons who were named in several chan-
nels were eliminated to allow the discussion and analysis
of non—related channels.

Once the data for the two analyses were ready for
processing, the computer factor analysis routine performed
the following steps:3 (Step (a) involves persons, while
step (b) involves channels.)

Step 1 (a) In analysis 5.1, a matrix of intercorrelations
was formed by correlating every person's sort of the state-
ments with every other person's sort of the statements.

(b) In analysis 6.1, a matrix of intercorrelations
was formed by correlating the standard score of each channel
with the standard score of every other channel on each de-
pendent variable.

Step 2 (a) In analysis 5.1, this matrix of intercorrelations
was submitted to factor analysis using persons as variables
and statements as observations. A principal axis solution
was obtained. This was submitted to varimax rotation which
produced orthogonal factors. On this basis a factor repre—
sents a grouping of persons around a common pattern of sort-
ing the items. Hence, a factor represents a type of person.

M

3Much of the following discussion is based on a de—
scription of the WRAP—II computer routine programmed by Mr.
Albert Talbott of—tmept. of Communication, Michigan State
University. This description was published as: Malcom S.
MacLean, Thomas Danbury, and Albert Talbott, "Civil Defense
Belief Patterns: Technical Summary," a mimeographed Working
Paper, Dept. of Communication, Michigan State University,
East Lansing, Mich., 1962, pp. 19—30.

   

 

 

 

 

 

I?—
212

(b) In analysis 6.1, a matrix of intercorrelations
was submitted to factor analysis using channels as variables
and dependent variables such as age and innovativeness as
observations. A principal axis solution was obtained. This
was submitted to varimax rotation which produced orthogonal
factors. On this basis a factor represents a grouping of
channels around a common pattern of standard score measures
on dependent variables. Hence, a factor represents a type
of channel.

The factor analytic model, in effect constructed
hypothetical types of persons based on the way the actual
people sorted the items (or hypothetical types of channels
based on the actual measures of several dependent variables.)
The coefficients or loadings on the rotated factor solutions
may be viewed as each person's (or channel's) correlation
with each of these hypothetical types. Persons (or channels)
were assigned to the type that they were most like, that is,
they were assigned to the factor on which they had the high—
est loading.

Criteria for Selecting a Factor
Solution for Further Analysis
The varimax rotation of the principal axis factor

solution yielded a series of factor solutions. The princi—

Pal axis solution was first split into two factor groupings,

 

then three, four, and so on until the procedure was stopped
when a factor was encountered on which fewer than three per—
sons (or channels) had their highest loading. Any one of
these factor solutions might have been chosen for analysis.
To determine which factor solution should be chosen for an—

alYSiS, each factor in each factor solution was analyzed

 

 

 

 

 

 

EII:_________________________________I
213
to determine how homogeneous the members of that factor were
in terms of: (a) vertical integration, (b) channel type,
and (c) channel level.
With respect to each of these criteria, homogeneity
was determined by the relative amount of variance that oc—
‘curred within groups in relation to that which occurred be—

tween groups. An 'ideal' factor solution for analysis would

 

be one in which each factor is homogeneous with respect to
each criterion, and therefore dissimilar from other factors.
For example, imagine that as a result of the varimax rota—
tion, the following three factor solution emerged as one

of several solutions;

Table F.l

AN EXAMPLE OF PERFECTLY HOMOGENEOUS FACTORS

 

 

 

 

 

 

 

 

Factors
One Two Three
Vertical Integration low medium high
Channel types A B C
Channel levels bird and producers processor-
feed sup— distributors
pliers and retailers

 

This three factor solution contains homogeneous fac—
tors--each factor contains none of the types of people that
are contained in the other two factors. A factor would be—
come less homogeneous to the degree that it contained types

Of Persons who are contained in other factors. It should

   

 

 

 

7

214

be noted that the probability of the occurrence of perfectly
homogeneous factors is extremely low. In the real world,
most variables are imperfectly operationalized, most meas-
uring instruments are imprecise, and responses are subject

to a myriad of biases. Thus, the selection of a factor ro-
tation for further analysis is most often a matter of select—

ing the one which contains the most homogeneous factors from

 

a set of factor solutions which contain rather heterogeneous

groupings.

Procedures for Describing Each Type

Once a factor solution was selected, it was analyzed
to determine how the types were similar. In analysis 5.1
the types were described in terms of attitudinal response
patterns, whereas in analysis 6.1, types were described in
terms of dependent variables. To provide these descriptions
the computer routine did the following:

Step 1 (a) In analysis 5.1, each pattern of sorting the
items associated with each factor or type of person was
estimated. This was done by weighting each item response

of each of the persons most highly associated with a given
factor by the degree to which they were loaded on that fac-
tor. The higher a person's loading on the factor, the greater
was the weight. These weighted responses were summed across
each item separately. This produced an item array of weighted
responses for each factor in the rotated factor analysis
solution selected. The arrays of weighted responses were

then converted to Z-scores.

(b) In analysis 6.1, each pattern of standard scores
on the dependent variables associated with each factor or
tYPe of channel was estimated. This was done by weighting
each score of each variable of each of the channels most

M

4Ibid., p. 23.

 

 

 

 

 

 

Ii—
215

highly associated with a given factor by the degree to which
they were loaded on that factor. The higher a channel's
loading on the factor, the greater was the weight. These
weighted scores were summed across each variable separately.
This produced a variable array of weighted scores for each
factor in the rotated factor analysis solution selected.
The arrays of weighted scores were then converted to Z-scores.
Associated with each of the hypothetical types con—
structed by the factor analytic model is a pattern of sort-
ing the items (or a pattern of scores on the variables).
Estimation of these patterns, in effect, is estimation of the
hypothetical sorts (or scores on the variables) of the hypo-
thetical types constructed. To do this, a weighted average
of the sorts of the people (or scores on the variables for
the channels) most highly associated with a given type or
factor was calculated. The more a person's sort (or a chan-
nel's standard score on a variable) was like the hypothetical
type, the more weight it received in the average. The spe—

cific weight given was as follows:

factor loading

 

weight = _ 2
l — factor loading

Step 2 (a) In analysis 5.1, the arrays of item Z—scores
for each factor, called factor arrays, were ordered from
strongest agreement to strongest disagreement for each fac—
tor. This provides a hierarchy of agreement for each factor
or type of person.

 

(b) In analysis 6.1, the arrays of dependent vari—
able Z-scores for each factor, called factor arrays, were
ordered from highest standard score to lowest standard score
for each factor. This provides a hierarchy of the relative
importance with which each variable describes each factor
or type of channel.

EDGE 3 (a) In analysis 5.1, in order to determine what
differentiates one factor from all others, those statements

 

 

 

 

————i '
216
which each factor agreed or disagreed with more than any
other factor were isolated.

(b) In analysis 6.1, in order to determine what
differentiates one factor from all others, those dependent
variables which described a particular factor more than other
factors were isolated.

The attitudinal and behavioral descriptions of each

factor type which emerged from this analysis are discussed

in Chapter 4, "Findings of the Study."

 

‘h—

S and 61m this step, the statements at each extreme

0f the array are identified, and employed to describe each
factor.

 

 

 

 

 

APPENDIX G

PROCEDURE FOR CHOOSING AMONG FACTOR SOLUTIONS

The varimax procedure used in analyses 5.1 and 6.11
produced a series of factor solutions, as discussed in 527
pgndix F. The purpose of this appendix is to explain how
a factor solution was chosen for analyses 5.1 and 6.1. This
appendix is organized into the following four sections:
(1) dimensions by which each factor was described——integra-
tion, channel type, and channel level, (2) criteria for
selecting a factor solution for further analysis, (3) fac—
tor solution chosen in analysis 5.1, and (4) factor solu—

tion chosen in analysis 6.1.

Dimensions by Which Each Raptor Is Described

 

1. Integration

The degree to which the persons (or channels) in
each factor vary in terms of vertical integration is sum—
marized by the following three integration measures:

(a) Integration range—-the distance between the
largest and smallest integration value. This provides a
relatively crude means of describing the extent to which
alternate factors differ. It does not indicate the degree

“—

1The analysis of the data which relate to hypothesis
5.1 will be referred to as 'analysis 5.1' while the one which
relates to hypothesis 6.1 will be called 'analysis 6.1.'

217

 

 

 

 

 

 

L7—1
218

to which integration scores tend to 'group' at some point
or points in the distribution.

(b) Median integration value--this central value
in the distribution provides a means of determining the ex-
tent to which the average scores of alternate factors differ.

(c) The inter-quartile range: Q3 — Ql—-this
value describes the integration values between which the
middle fifty percent of the respondents are contained. Its
width indicates the degree to which the middle integration
scores are grouped. The narrower the interquartile range,
the more homogeneous on integration is the factor. Its ab—
solute values indicate whether these scores are low, medium,
or high on integration.

2. Channel Type

To distinguish among the structural types of chan—
nels included in analyses 5.1 and 6.1, the following descrip-
tions of each are provided;

Channel Type A: Integration range: 7—39. Channel
type A represents independent producers who are associated
with similarly independent channel members. It contains
producers who purchase their inputs independently and sell
their eggs to independent processors who in turn sell to

retail stores. This is the least integrated of the channel
types included.

 

Channel Type B: Integration range; 20-23. The
persons engaged in channel type B participate in a quality
control program of a cooperative. Although there is no
aQreement on the part of the producers engaged in this pro-
gram to link them with the feed supplier or the processor
for a long period of time, they do receive a premium price
for their eggs provided that the eggs meet certain quality
Standards. Channel type B is relatively non-integrated.

 

 

 

 

 

Ii—_’

219

Channel Type C: Integration range: 30—32. Chan-
nel type C employs a sixty—day written agreement to link
producers to various cooperative feed mills and to two
cooperative-owned processing plants. Channel type C is
relatively non-integrated.

Channel Type D: Integration range: 38—61. Chan-
nel type D is formed on the basis of production contracts
between a number of producers and a feed milling corporation.
The eggs from the producers in this type of channel go through
several processors to several different retailers. This
channel is in the medium range of vertical integration.

Channel Type E: Integration range: 38—49. Chan-
nel type B is composed of several contract channels. The
persons in these channels by the large are linked by contracts
which exist between one feed company and only one or a few
producers. In all cases these contractors send their eggs
to one or more processors who in turn sell to several re—
tailers.

Channel Type F: Integration range: 52—57. In
channel type F egg production contracts link one feed mill
and several egg producers. The eggs from these producers
flow to one processor only and this processor sells to one
retailer only.

Channel Type G: Integration range: 54-65. This
contract channel links a cooperative and several producers.
The cooperative owns two pr0cessing plants and most of the
eggs flow to two retail chains.

Channel Type H: Integration range: 54—100. The
persons in channel H run owner-integrated operations. These
persons have their own feed mills and may or may not supply
their own pullets. They may or may not process their own
eggs but in no cases do they retail their eggs.

 

Channel Type S: Integration range: 56-100. The
persons in this channel type are located in the South. They
have either contract operations or they are engaged in owner—
integrated operations. These are channel types G and H which
are located in the South.

3. Channel Level

The data in analyses 5.1 and 6.1 came from persons

at all channel levels. Therefore the degree to which the

persons in a particular factor type belong to one or only

 

 

 

 

 

220

a few of these five channel levels, is a measure of the
homogeneity of that factor.
Criteria for Selecting a Factor
Solution for Further Analysis

The varimax rotation of the principal axis solution
yielded a series of factor solutions, any of which could
have been chosen for further analysis. As discussed in
Appendix F, to determine which factor solution was to be
analyzed, the relative homogeneity of each factor in each
factor was subjectively estimated on the basis of three cri—
teria: vertical integration, channel type, and channel level.
The factor solution chosen for analysis was the one which
contained factors which were the most homogeneous in terms
of each criterion.

The homogeneity of the actual factor solutions chosen

for analyses 5.1 and 6.1 is discussed below.

EggtggySolution Chosgpifor Analysis 5.1

The seventh of the eleven factor solutions was chosen
for analysis because the seven factors which it contained
were relatively more homogeneous than those contained in
the other factor solutions. Table G.l summarizes these seven

factors in terms of the three criteria.

Explanation of Table G.l. The seven factors are

listed in Table G.l in ascending order of vertical integra—

tion, based on their median integration scores. The Q1 and

Q3 integration values tend to roughly approximate an

 

 

 

 

221

 

 

.m m m m H I H
H *m H H H l l
H N *v N l I I
m H m H .m m N
W N I I H m N
... N I l H *0 H
I m H I H I ...HV
N N H H N I
m d v I H N N
o m m H .e H N
...m m *m ...m H *5 *d
H *m N H H H H
v m I I m d H
O.m¢ m.Nm O.mH o.mH O.Hm m.hH m.mm
O.Nm m.©N O.H¢ m.h¢ m.mN m.Om m.HN
O.mm O.mb 0.0m O.mm m.bn o.mv O.mm
hm om mm O.Hm m.h¢ mm mm
OOlefl OOHIHN flmln omlhm mmlb flmlb mmION
a m m a m m H

hocwswmuw Dwmbmmhm mmDMUHccH*

<11IIJUQLQI'LILDCEU)

mmfluemflmmum
Hm>wq Hmccmcu .m

hmHHmbmmno
MObSbHMbmelhOmmmooumno
umUDconmum

umHHQQSm cuflmnw
nmflHdQDm commnH
meUCQSmmnm

H®>md chcmrU .N

Ha I ma

Ho

.3

cmflcwE coebmuombCH
mmcmn COHDmumwaH

COHbmummbcH .H

 

COHbmnombCH MO umpuo OCHochm< CH mmd>B nowumh

hbflwcwooeom
Monomm mo MHnmanU

 

H.m mHmwd<Z< ZOHBDHOm mOBU<m Zm>mm .WBHMZMOOZOI

 

H.D mHQMB

mOBU<m m0 mmmbm<mz

 

 

 

 

 

7

222

integration continuum. The increase in integration that
occurs between factors one and seven may also be seen by
noting that the most frequent channel type categories (which
are marked with an asterisk) tend to lie along a diagonal
which runs from upper left to lower right. This corresponds
with the increase in vertical integration that occurs in
both: (a) the channel types, moving from type A to type S;
and (b) factor types, moving from factor one to factor seven.

The factors in the seven factor solution were rela—
tively more homogeneous on all three criteria than were those
contained in any of the other ten factor solutions. Its
factors were relatively more homogeneous than those con-
tained in the two, three, four, and five factor solutions
where the factors tended to be quite similar to one another
in terms of integration range, median, interquartile range,
and channel type.

Little additional explanatory power would have been
gained by using the eight, nine, ten, or eleven factor solu-
tions because these contained the seven relatively homogene-
ous factors given in solution seven, pips numerically small
factors which were basically like one of the original seven
in terms of the three criteria. In other words, using fac—
tor solutions eight through eleven would have contributed
little in the way of additional information, and would have

given up part of the simplifying power that is the principal

strength of factor analysis.

 

 

 

 

 

7

223

A parallel procedure was employed to select the fac—
tor solution for analysis 6.1. This process is explained

below.

Factor Solution Chosen for Analysis 6.1

 

The fourth of the seven possible factor solutions

was chosen for analysis. The four factors which it contained

 

were relatively more homogeneous than those contained in the
other factor solutions. Table G.2 summarizes these four

factors in terms of the criterion of homogeneity.

Explanation of Table G.2. The four factors in Table

 

G.2 are listed in approximate order of integration, based
on the median integration score of all the channels in each
factor, irrespective of the sign of the factor loading.

The term 'approximate' is used to caution the reader that
these factors could not be ordered in terms of integration
as easily as could those in analysis 5.1, in which each of
the factor loadings of the persons who loaded most strongly
on a factor was pOSitive. Some of these loadings in analy—
sis 6.1 are positive while others were negative.

The factor loadings may be viewed as each channel's
correlation with each factor type. The sign attached to
each factor loading may be viewed in the same manner as the
Sign attached to a correlation coefficient——it indicates
the direction of association. To illustrate, suppose that

channel 034 has a positive loading of .92 on factor one,

 

 

 

 

 

 

 

 

224

 

 

 

 

 

 

 

 

 

 

dm HO I m0
Ho Ho
mm mo
m.hm we CMHUTZ
OOHImm OOHIOm OOH Om mmcmm
COHpmuOmbCH .H
mWCHUmOH nowomm
®>Hbmm®z LDHB mHmccmCU .m
3 HN om Ho I mo
m om Om HO
Om he om mo
Om He mm OH GMHomS
Hons Hmlmm emnmm mmum magma
COHpmumw#CH .H
mOCHUMOH nonumm
m>Hmeom Lsz mHmccmru .<
NwflmcmmoEom homumm mo MHumbHuU
w m m H
COHbmnmecH mo
umpuo mmeonudm< CH mmdwe nebumm

 

 

H.m WHmwd<z< .ZOHHDHOW mOBU<m MDOh

aNBHmZmOOZOI mOEU<h m0 mmmbm<m2

 

N.0 OHQMB

 

 

 

 

—7

225

whereas channel 021 has a negative loading of .95 on factor
one. Suppose further that conflict is the dimension which
characterizes the channels in factor one. The difference

in signs of the two factor loadings means that the two chan—
nels are opposite one another on conflict——if channel 034
has high conflict, then channel 021 has low conflict.

In Table G.2, the criterion of homogeneity’ is ap—
plied to: (a) channels with positive factor loadings, and
(b) channels with negative factor loadings.

Better information on factor homogeneity is yielded
from analysis of those channels in each factor which have
positive factor loadings. In this case, the groups are
arrayed in ascending order of integration median.

The factors contained in this four factor solution
were relatively more homogeneous on integration than were
those contained in any of the other six solutions. Factor
solutions one, two, and three were not selected for analysis
because the factors contained in each of them were relatively
alike in terms of integration. Factor solutions five, six,
and seven were not chosen because they contained the same
factors given in solution four, plus much smaller factors
which were like one of the original four in terms of inte-

gration and channel type.

h“..—

‘Only one of the three criteria of homogeneity are
relevant to this discussion. This criterion is median in—
tegration.

 

 

 

 

 

 

APPENDIX H

DESCRIPTION OF THE OPERATION OF THE
"MULTIPLE REGRESSION DELETE“ COMPUTER ROUTINE

The data used to test hypotheses 4.1 and 4.2 were
analyzed with the aid of a multiple regression stepwise de—
letion computer routine called pspgp.

The authors of this program describe its operation
as follows:

In stepwise deletion, an initial least squares equa—
tion is obtained using all of the independent variables.
One variable is then deleted from the equation and a
new least squares estimated. A second variable is
deleted and the least squares equation recalculated.

The procedure continues until a variable selected as

a candidate for deletion meets one_or more stopping
criteria. (at page 1)

. . . The variable selected for deletion at any step
will be the variable which will be missed the least,
that is, when the least squares equation is calculated
for the next step, a greater part of the variation

in the dependent variable can be accounted for by the
remaining independent variables than if any other in—
dependent variables . . . had been deleted. (at page 2)

1
As the variable which was missed the least was dropped
from the regression equation, the difference in the resulting
R2 was noted. These differences in R2 at each step in the
deletion process represent the percent of variance explained

in the criterion variable by adding each predictor variable

to the prediction team.

M

1Mary E. Rafter and William L. Ruble, Stepwise De-
letion of Variables from a Least Squares Equation LSDEL

Routine), STAT Series Description No. 8, Agricultural Ex—

periment Station, MichiganSState UniversitY9 East Lansing,
Michigan, January, 1968.

226

 

 

 

 

 

 

 

BIBLIOGRAPHY

 

 

 

 

BIBLIOGRAPHY

Alderson, Wroe. Dynamic Marketing Behavior. Homewood,
Illinois: Richard D. Irwin, 1965.

. Marketing Behavior and Executive Action. Home-
wood, Illinois: Richard D. Irwin, 1957.

Assael, Henry. "The Political Role of Trade Associations
in Distributive Conflict Resolution," Journal of
Marketin , Vol. 32. Chicago: American Marketing
Association, April, 1968.

Baker, Ralph L. Coordinated Egg Production and Marketing
in the North Central United States. I. Kinds of
Programs. North Central Research Publication 160.
Wooster, Ohio: Ohio Agricultural Research and De—
velopment Center. Research Bulletin 973, July, 1965.

. Integrating Egg Production and Marketing. Wash—
ington, D.C.: U.S. Department of Agriculture, Mar—
keting Research Division. Agricultural Marketing

Service. Marketing Research Report No. 332, June,
1959.

Baligh, Helmy H. "A Theoretical Framework for Channel Choice,“
in P. D. Bennet (ed.), Economic Growthlnggpetition,
and World Market. Chicago: American Marketing As-
sociation, 1965. Reprinted in Bruce E. Mallen, Egg
Marketing Channgl, A Conceptual Viewpoint. New York:
John Wiley and Sons, Inc., 1967.

—~—_____a and Leon E. Richartz. Vertical Market Structures.
BOston: Allyn and Bacon, Inc., 1967.

Berg, Thomas L. "Designing the Distribution System,“ in
William D. Stevens (ed.), The Social Responsibilities
of Marketing. Chicago: The American Marketing As—
sociation, 1962.

BOWersox, Donald J. "Changing Channels in the Physical Dis—
tribution of Finished Goods," in Peter D. Bennett
(ed.), Proceedings of the Fall Confeggpggflgiygyg
American Marketing Association, September, 1965.
Chicago: American Marketing Association. Reprinted
in George Fisk and Donald F. Dixon, Theories for

Marketing Systems Analysis. New York: Harper and
ROW, 1967.

 

228

I_H—

 

 

 

 

 

 

 

229

. "Physical Distribution Development, Current Status
and Potential," Journal of Marketing. Chicago:
American Marketing Association, Vol. 33, January,
1969.

, and McCarthy, E. J. "Strategic Planning of Ver-
tical Marketing Systems.“ A mimeographed paper de—
livered at the Channels Seminar sponsored by the
American Marketing Association Task Force on Market-
ing Methodology. November 6-8, 1968. Evanston,
Illinois.

Breyer, Ralph F. Quantitative Systemic Analysis and Control:
Study No. l-—Channel and Channel Group Costing.
Philadelphia: Ralph F. Breyer, 1949.

. "Some Observations on “Structural' Formation and
Growth of Marketing Channels,“ in Revis Cos, Wroe
Alderson, and Stanley J. Shapiro (eds.), Theor in
Marketing (second series). Homewood, Illinois:
Richard D. Irwin, Inc., 1964.

Bucklin, Louis P. "The Economic Structure of Channels of
Distribution," in Martin L. Bell (ed.), Marketin
A Maturing Discipline. Chicago: American Market-
ing Association, 1960.

"The Locus of Channel Control, " in Robert L. King
(ed. ), Marketing and tpgiNew Science of Planning.
1968 Fall Conference Proceedings of the American
Marketing Association. Chicago: American Market-
ing Association, 1968.

. "Postponement, Speculation, and the Structure

of Distribution Channels," Journal of Marketing Re—
search. Chicago: American Marketing Association,
Vol. 2. February, 1965.

. A Theory of Distribution Channgl Structure.
Institute of Business and Economic Research, Gradu—
ate School of Business Administration, University
of California, Berkeley, 1966.

Craig, David R., and Werner K. Gabler. "Competitive Stru le
99
for Market Control," The Annals of the American Acad—
emy of Political and Social Science. May, 1940.

Davidson, William R. "Innovation in Distribution," in M. S.
Moyer and R. E. Vosburgh (eds.), Proceedings of the
June, 1967, Conference of The American Marketing

Association. Chicago: American Marketing Associa-
tion.

 

I—H—

 

 

 

 

 

 

230

Dickinson, Roger, Jr. "Channel Management by Large Retail-
ers," in Robert L. King (ed.), Marketing End the New
Science of Planning. 1968 Fall Conference Proceed—
ings of the American Marketing Association. Chicago:
American Marketing Association, 1968.

Dixon, Donald F. "The Emergence of Marketing Systems," in
Economic and Business Bulletin, School of Business
Administration, Temple University, Vol. 17, No. 4
(June, 1965). Reprinted in Theories for Marketing

S stems Anal sis. George Fisk and Donald F. Dixon
(eds.). New York: Harper and Row, 1967.
Dodd, Stuart Carter, and Thomas R. Gerbrick. "Word Scales

for Degrees of Opinion," Language and Speech, Vol.
3, Part 1. January—March, 1960.

Edwards, Allen L. Techniques of Attitugg Scale Construction.
New York: Appleton-Century-Crofts, Inc., 1957.

"Egg Industry Interview with Richard L. Pockman," Egg Indus—
try. August, 1968.

Fisk, George. Mgrketing Systgms: An Introductory Analysis.
New York: Harper and Row, 1967.

Gallimore, William W. Contracting and Othgpplntegration
Arrangements in thg Turkey Industry. U.S. Depart—
ment of Agriculture, Economic Research Service.
Marketing Economics Division. Marketing Research
Report No. 754. November, 1965.

 

Granbois, Donald H., and Ronald P. Willett. "Patterns of
Conflicting Perceptions Among Channel Members," in
George L. Smith (ed.), Reflections on Progress in
Marketing. American Marketing Association Proceed—
ings. Chicago: American Marketing Association,
Dec., 1964.

Guilford, J. P. Psychometric Methods. New York: McGraw—
Hill Book Company, Inc., 1954.

Hollander, Stanley C. "Channel Management by Small Retail-
ers," in Robert L. King (ed.), Marketing and the
New Scigpge of Planning. Chicago: American Market—
ing Association, 1968.

Hughes, C. David. "Developing Marketing Strategy Through
Multiple Regression," Journal of Marketing Research.
Chicago: American Marketing Assoc1ation, Vol. 3.
November, 1966.

 

 

 

 

 

 

231

King, Charles W. "Adoption and Diffusion Research in Mar-
keting: An Overview," in Raymond L. Hass (ed.),
Science,Technology, gpd Marketing. 1966 Fall Pro—
ceedings of the American Marketing Association.
Chicago: American Marketing Association, 1966.

. "Systems Analysis at the Public-Private Market-
ing Frontier," Journal of Marketing. Chicago: Amer-
ican Marketing Association, Vol. 33, January, 1969.

Kotler, Philip. Marketing Management. Englewood Cliffs,
New Jersey: Prentice-Hall, Inc., 1967.

Larzelere, Henry E. Current Styles and Shapes in Egg Con-
tracting——II. Agricultural Economics Report. Dept.
of Agricultural Economics, Michigan State University,
East Lansing, Michigan, Feb., 1966.

. The Egg Industry in Transition. Agricultural
Economics Report No. 79. Dept. of Agricultural
Economics, Michigan State University, East Lansing,
Michigan, March, 1967.

Lazer, William, and Eugene J. Kelley. "The Systems Approach
to Marketing," in William Lazer and Eugene J. Kelley,
Managerial Marketing: Perspectives and Viewpoints.
Homewood, Illinois: Richard D. Irwin, Inc. Revised
edition, 1962. '

’Lewis, Edwin H. "Channel Management by Wholesalers," in
Robert L. King (ed.), Marketing and the New Science
of Planning. 1968 Fall Conference Proceedings of
the American Marketing Association. Chicago: Amer—
ican Marketing Association, 1968.

. Mgrketing Channels: Structure and Strategy. New
York: McGraw-Hill Book Co., 1968.

 

Little, Robert W. "Power and Leadership in Marketing Chan—
nels." A mimeographed paper delivered at the Chan-
nels Seminar sponsored by the American Marketing
Association Task Force on Marketing Methodology.
November 6-8, 1968, Evanston, Illinois.

MacLean, Malcom S., Thomas Danbury, and Albert Talbott.
"Civil Defense Belief Patterns: Technical Summary."
A mimeographed working paper published by the Dept.
of Communication, Michigan State University, East
Lansing, Michigan, 1962.

 

 

 

 

 

 

 

7

232

McCammon, Bert C., Jr. “Alternative Explanations of Insti—
tutional Change and Channel Evolution," in Stephen A.
Greyser (ed.), Toward Scigptific Marketing, Proceed—
ings of the Winter Conference of the American Mar-
keting Association. December, 1963. Chicago: Amer—
ican Marketing Association. Reprinted in Bruce E.
Mallen (ed.), The Marketing Channel: A Conceptual
Viewpoint. New York: John Wiley and Sons, 1967.

, and Albert D. Bates. "The Emergence and Growth
of Contractually Integrated Channels in the American
Economy," in P. D. Bennett (ed.), Economic Growth
Competition and World Markets. Chicago: American
Marketing Association, 1965. Reprinted in Bruce E.
Mallen (ed.), The Marketing Channel: A Conceptual
Viewpoint. New York: John Wiley and Sons, Inc.,
1967.

 

, and Robert W. Little. "Marketing Channels: An—
alytical Systems and Approaches," in George Schwartz

(ed.), Science in Marketing. New York: John Wiley
and Sons, Inc., 1965.

McCarthy, E. Jerome. Basic Marketing: A Manggerial Approach.
Third Edition. Homewood, Illinois: Richard D. Irwin,
Inc., 1968.

, and Donald J. Bowersox. "Strategic Planning of
Vertical Marketing Systems." A mimeographed paper
delivered at the Channels Seminar sponsored by the
American Marketing Association Task Force on Market—
ing Methodology. November 6—8, 1968. Evanston,
Illinois.

 

, and Robert J. Williams. "Simulation of Production-
Marketing Channels," in Raymond M. Hass (ed.), £9;—
enceL Technology,L and Marketing. 1966 Fall Confer-
ence Proceedings of the American Marketing Associa—
tion. Chicago: American Marketing Association,
1966.

 

 

McClure, Peter J., and John K. Ryans, Jr. "Differences
Between Retailers and Consumers' Perceptions,"
Journal of Marketing Research. Chicago: American
Marketing Association, Vol. 5, Feb., 1968.

McNemar, Quinn. Psychological Statistics, 3rd Ed. New York:
John Wiley and Sons, Inc., 1962.

 

 

 

 

 

 

 

Ii—

233

McVey, Philip. "Are Channels of Distribution What the Text—
books Say?" Journal of Marketing. Chicago: Amer—
ican Marketing Association, Vol. 24, January, 1960.
Reprinted in Wm. Lazer and Eugene J. Kelley, Mpg:
ggerial Marketing: Perspectives and Viewpoints,
Revised Ed. Homewood, Illinois: Richard D. Irwin,
Inc., 1962.

Mallen, Bruce. "Conflict and Cooperation in Marketing Chan-
nels," in L. Geo. Smith (ed.), Reflectiqps on Progress
in Marketing. Chicago: American Marketing Associa—
tion. Reprinted in Bruce E. Mallen (ed.), The Mar—
keting Chanpgl, A Conceptual Viewpoint. New York:
John Wiley and Sons, Inc., 1967.

 

Mighell, Ronald L., and Lawrence A. Jones. Vertical Coor—
dination in Agriculture. U.S. Dept. of Agriculture.
Economic Research Service. Farm Economics Division.
Agricultural Economic Report No. 19, February, 1963.

 

Miller, James G. "Living Systems: Basic Concepts," Behav-
ioral Science, Vol. 10, No. 3, July, 1965.

Niles, Edwin E., Robert J. Williams, and R. L. Kohls. Coor—
dinated Egg Production and Marketing in the North
Central States II. Characteristics and Attitudes
of Independent and Contggpt Egg Producers. North
Central Regional Research Publication No. 162.
Lafayette, Indiana: Purdue University Agricultural
Experiment Station Research Bulletin No. 795, May,
1965.

 

 

Organization and Competition in the Poultry and Egg Indus—

tries. Washington, D.C.: National Commission on
Food Marketing. Technical Study No. 2, June, 1966.

Palamountain, Joseph C., Jr. The Politics of Distribution.
Cambridge, Massachusetts: Harvard University Press,
1955.

Pedersen, John R., and Fred L. Faber. Major Marketing Chan-
nels for Shgll Eggs in 18 Metropolitan Areas. Wash-
ington, D.C.: U.S. Dept. of Agriculture. Marketing
Economics Division. Economic Research Service.
ERS—219, February, 1965.

Rafter, Mary E., and William L. Ruble. Stepwise Deletion
of Variables from a Leggt Squares Equation. (LSDEL
Routine) Description, STAT Series No. 8. Agricul-
tural Experiment Station, Michigan State University,
East Lansing, Michigan, January, 1968.

   

 

 

 

234

Revzan, David A. "Marketing Organization Through the Chan—
nel," Wholesaling in Marketing Organization. New
York: John Wiley and Sons, Inc., 1961. Reprinted
in Bruce E. Mallen (ed.), The Marketing Channel, A
Conceptual Viewpoint. New York: John Wiley and
Sons, Inc., 1967.

 

Ridgway, Valentine F. "Administration of Manufacturer—
Dealer Systems," Administrative Science Quarterly,
Vol. 1, March, 1957.

Rock, James S., and L. B. Darrah. Survey_of Integrated Egg
Production, A. E. Res. 165. Ithaca, New York: Cor-
nell University Agricultural Experiment Station,
Feb., 1965.

Rogers, Everett M. Bibliography on the Diffusion of Innova—
tions. Diffusion of Innovations Research Report
No. 6. East Lansing, Michigan: Dept. of Communi—
cation, College of Communication Arts, Michigan State
University, July, 1967.

. "Categorizing the Adopters of Agricultural Prac—
tices," Rural Sociolo , Vol. 23, No. 4, December,
1958.

. Diffusion of Innovations. New York: The Free
Press, 1962.

, and L. Edna Rogers. "A Methodological Analysis
of Adoption Scales," Rural Sociolo , Vol. 26, No.
4, December, 1961.

Roy, Ewell P. "The Broiler Chicken Industry," in John R.
Moore and Richard G. Walsh (eds.), Market Structure
of the Agricultural Industries. Ames, Iowa: The
Iowa State University Press, 1966.

 

Ruble, William L., Donald F. Kiel, and Mary E. Rafter. 93l—
culation of Least Squares (Regression) Problems on
the LS Routine. STAT Series Description No. 7.
Agricultural Experiment Station, Michigan State
University, East Lansing, Michigan, April, 1968.

Schwartz, George. "Breyer Study: Quantitative Systemic
Analysis and Control," in Developggnt of Marketing
Theory. Cincinnati, Ohio: Southwestern Publishing
Company, 1963. Reprinted in Bruce E. Mallen (ed.),
Eflé Marketing Channel, A Conceptual Viewpoint. New
York: John Wiley and Sons, Inc., 1967.

 

 

 

 

 

 

 

235

Siegal, Sidney. Nonparametric Statistics. New York:
McGraw—Hill Book Co., Inc., 1956.

Stasch, Stanley P. "The Stability of Channel Systems:
Two Dynamic Models," in Science, Technology; and
Marketin , edited by Raymond M. Haas. Proceedings
of the 1966 Fall Conference of the American Market—
ing Association, 1966.

Staudt, Thomas A., and Donald A. Taylor. A Managerial In—
troduction to Marketing. Englewood Cliffs, New
Jersey: Prentice-Hall Publishing Co., 1965.

 

Stephenson, William. The Study of thavior. Chicago: The
University of Chicago Press, 1953.

 

Stern, Louis W., and Ronald H. Gorman. "Conflict in Distri-
bution Channels: An Exploration," Chapter 10 in
Louis W. Stern (ed.), Distribution Channels: Be—
havioral Dimgpsions. Boston: Houghton Mifflin Com—
pany, 1969.

. Distribution Channels: Behavioral Dimensions.
Boston: Houghton Mifflin Company, 1969.

 

, and Jay W. Brown. "Distribution Channels: A
Social Systems Approach," in Louis W. Stern (ed.),
Distribution Changgls: Behavioral Dimensions. New
York: Houghton Mifflin Company, 1969.

 

Stigler, George J. "The Division of Labor Is Limited by
the Extent of the Market," Journal of Political
Economy, June, 1951.

Sturdivant, Frederick D. "Determinants of Vertical Inte—
gration in Channel Systems," in Science, Technology
and Marketing, 1966 Fall Conference Proceedings of
American Marketing Association. Chicago: American
Marketing Association, 1966.

 

Wittreich, Warren J. "Misunderstanding the Retailer,"
Harvard Busipgss Review, May-June, 1962.

 

Zober, Martin. Marketing Management. New York: John
Wiley and Sons, Inc., 1964.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

IIIIIII’I

51390

I
'I
l
[I
[I