 

 

 

 

 

 

 

 

 

 

 

2:. uﬁgﬁ; 4

. . 5.3:.“ no?! 43!

39.? hat ‘ame...
gums“

I... .
5331

.w.
. ﬂirtﬁz?

.
ﬁﬁvnﬁ. Hmﬁvaﬁr
"awry 3: “.1.

124%;

.1
at
- 1
Liz... E.
”43.53%.
Kata. .3

as ...,

I.
.3 : I.

Luau. .u

. .5 $5.63.

. v .31.. .

THESIS
/‘:

 

 

2' _ f

, .i. ‘1 LIBRARY

2‘0) Michigan State
University

 

 

This is to certify that the
dissertation entitled

CASE READY MEATS: AN OUTSOURCING DECISION FOR
FOOD RETAILERS

presented by

George Arthur Young

has been accepted towards fulfillment
of the requirements for the

PhD. degree in Agricultural Economics

 

 

WW. W

Wajor Professor’s Signature

Date

 

MSU is an afﬁnnative-action, equal-opportunity employer

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K‘IProj/Acc8Pres/CIRC/DateDue.indd

CASE-READY MEATS: AN OUTSOURCING DECISION FOR FOOD RETAILERS
By

George Arthur Young

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Agricultural Economics

2009

ABSTRACT
CASE-READY MEATS: AN OUTSOURCING DECISION FOR FOOD RETAILERS
By
George Arthur Young

Innovation in the way food products are processed and distributed through the
supply chain has been an important source of productivity growth in the US. and
economies worldwide. While process and product innovations have not ceased, the
uneven adoption of certain technologies and processes by retailers and consumers in
some product lines is curious.

The poultry industry long ago realized that case-ready chicken products added
value and increased proﬁts in its commodity product lines. These new products
combined improved production techniques, new processing and packaging technologies,
and a new retail marketing strategy in a way that completely revolutionized how retailers
stocked and sold poultry and how consumers used the product.

In comparison to the experience of chicken, case-ready beef has seen limited
success in the retail marketplace. The limited success of case-ready beef products is
surprising given several external demand factors favoring adoption of this advanced
technology of fresh beef products. The objective of this dissertation is to determine what
critical factors may affect fresh meat processors and retailers’ decision whether or not to
integrate case-ready beef products into their product lines.

Decision-makers in the fresh beef industry have for years contemplated and
attempted various strategies for increasing adoption of case-ready technology and this

research project seeks to assist the industry in those endeavors. To develop a better

understanding of the Operational concerns that help or hinder the successful introduction
of case-ready beef products into retail grocery stores, this project developed six research
questions to guide the research process. These questions were constructed by
interviewing key industry experts, by directly observing case-ready meat products as they
moved through the supply chain, and by conducting in-depth interviews with retail store
managers and distribution center managers from a limited assortment grocery store chain
with over 1,150 stores throughout the US. Also a thorough review of the Accounting,
Strategic Management and Transaction Cost Economics literature was conducted as they
relate to outsourcing and/or the “make or buy” decisions. From the six research
questions a total of twenty-four testable research propositions were developed that will
help identify the operational issues associated with the successful implementation of a
case-ready meat strategy.

The academic literature was in agreement about ﬁrms outsourcing products and
services. It is recommended that ﬁrms should outsource only those products that are not
core to the business. However, in the instance of retail meat processing and packaging,
the function is core to its business. This study focuses on the decision to convert its meat
operations to case-ready beef products, which is a decidedly different approach to
outsourcing. This research found that the adoption of case-ready meats is directly related
to maximizing the opportunities within the six operational issues and mitigating others.
An important observation is that central processing opportunities will become more
evident to industry decision-makers if and when formal alliances are formed whereby

retailers share point-of-sale data to ﬁrms throughout the supply chain.

ACKNOWLEDGEMENTS

This dissertation is dedicated to my Lord and Savior, Jesus Christ, from whom all
blessings ﬂow. I saw His hand working in my life from the ﬁrst day of graduate school
classes. While walking across campus, I saw the Gideons handing out books of the New
Testament, Psalms, and Proverbs to all takers. I accepted the book, walked a few steps,
and then was asked, “Do you know Jesus Christ and have you accepted Him as your
Savior?” When I responded, “Yes,” they smiled and said “If you ever need extra advice
or encouraging words, just Open this book. All of the answers you really need can be
found within those pages.” They were so right!

I soon found myself searching that little green book for ‘advice’ and the following
verses became my solace and inspiration as the intensity of my doctoral studies and the
size of our young family both increased:

Romans 12: 2 “And do not be conformed to this world, but be transformed by the
renewing of your mind, that you may prove what is that good and acceptable and perfect
will of God. Romans 12: 11 “Not lagging in diligence, fervent in spirit, serving the
Lord.”

Romans 5: 3-5 “And not only that, but we also glory in tribulations, knowing that
tribulations produces perseverance; and perseverance produces character; and character,
hope. Now hope does not disappoint, because the love of God has been poured out in our
hearts by the Holy Spirit who was given to us.”

John 15: 7-8 “ If you abide in Me, and My words abide in you, you will ask what you
desire, and it shall be done for you. By this My Father is glorified, that you bear much
fruit; so you will be My disciples.”

Psalms 32: 8 “I will instruct you and teach you in the way you should go, I will guide you
with My eye.”

iv

Reliance on these verses, daily Bible and devotional reading, and attending a
church truly grounded in Scripture not only deepened my relationship with Jesus Christ,
but also strengthened my determination to persevere the rigors of the Agricultural
Economics department at Michigan State University in order to complete my Ph.D. and
pursue the career path the Lord has laid out before me.

Laura, my wife and best friend, thank you for supporting my efforts to pursue the
PhD. Without your love and understanding this dream would not have been realized and
your sacriﬁce can never be quantiﬁed I thank the Lord daily for blessing my life with
you.

Clayton, Cole, and Cooper, our three precious sens, thank you for your patience
and trust. It is an honor to be the father of Godly young men who never cease to amaze
me. You’ve blessed my life in ways I never knew were possible. I know you have fond
memories of our life at Michigan State, but it was not been'an easy road. I thank you,
from the bottom of my heart, for your many sacriﬁces and understanding smiles given on
behalf of your graduate school father.

Herb Haygood, thank you for allowing Laura and me to be your ‘Michigan
Parents’. You were our surprise child and we thank the Lord for crossing our paths.

Thank you Dr. Jack Allen for mentoring me and treating me as your colleague
while ‘Team Teaching’ undergraduate and MBA classes in the Business School. Your
access to world-class Marketers and ability to bring them into the classroom was
incredible. It was a great privilege and honor to have been taught by you. Thank you for
the opportunity to meet and learn from your best friends, industry leaders in Food

Marketing. I will never forget those classes and conversations. Your teaching style and

rapport with students is reﬂected in my classes and student relationships of today. This
has not only helped me succeed in the classroom, but my students have greatly beneﬁted
from your tutelage as well. Many thanks for lending me your stream of research as the
subject of this dissertation and for your assistance in writing, and ultimately being
awarded, the grant. The funding of the grant was directly linked to their respect for you
and their regard for your topical research agenda.

Thank you Dr. Stephen Harsh for helping me complete the dissertation. Your
insights, suggestions, and direction were priceless and prescient. Thank you for always
having your door open, the lights on, and answering your phone for a student in need of
help and direction. I truly hope you know how many graduate students’ lives you have
directed and inspired toward success. Your efforts are greatly appreciated and now can
be seen in our students’ lives due to the nurturing and prodding we received from you.

Thank you Dr. Ricks for showing me the importance and richness of managerial
implications found within Agricultural Economics. Your constant friendship and
willingness to participate on my thesis committee were invaluable, thank you.

Thank you Dr. Robb Clarke for sharing your knowledge of food packaging and
the use of resources and facilities at the School of Packaging for this dissertation. Your
help with the initial meeting of industry leaders and MSU faculty was key towards
starting this project.

Thank you Mike Schilli for sharing your insights into case-ready meats based
upon your lifetime in the meat industry. Your extra efforts and corporate contacts were

instrumental in every facet of this project.

vi

Thank you Bob Harris for giving me access to your franchise stores, several
members of your management team, and access to you personally for the many questions
I had about the retail grocery store industry and case-ready meats. This project could not
have begun, nor ﬁnished, without your help and expertise.

Thank you Dr. Alan Schmid for taking an interest in my studies and research
endeavors, and in my personal well-being. I believe I speak for all my colleagues in
saying that a comment or note of afﬁrmation from you has always served as great
encouragement that we might just be on the right track after all.

Thank you to my dear colleagues from Michigan State University, Will Shields,
Iva Petrova, and Gregg Hadley, for you assistance in the classroom and in research
endeavors, for your willingness to be a sounding board for academic ideas, and most of
all for your enduring friendship that has far surpassed anything my humble words can
express. ’

Thank you as well, to my dear friends Todd and Krinda Power whom God so
graciously placed in my life while at Arizona State University. Your steady
encouragement to trust in God -always- has served this friend well and eternally.

I cannot end without acknowledging my mother who, during the 1950’s and ‘60’s, raised
her three sons alone. Her enduring commitment to us, her ceaseless efforts towards
excellence and professionalism, her steadfast spirit of adventure, her uncanny
determination to ‘live well’ on an incredibly limited budget, and her straighﬁforward
ability to ‘tell you how the cow eats the cabbage’ were the perfect training ground for
graduate school... and for life. I dedicate all that is implied in the ‘successful completion

of the doctoral degree’ to my dear mother, Mary Alice.

vii

TABLE OF CONTENTS

List Of Tables ..................................................................................................................... xi
List of Figures .................................................................................................................. xiv
Key to Abbreviations ......................................................................................................... xv
CHAPTER 1 — INTRODUCTION ...................................................................................... 1
1.1 Study Organization .................................................................................................. 1
1.2 Retailer and Consumer Acceptance of Case-Ready Chicken .................................. 4
1.3 Limited Success of Case-Ready Beef ...................................................................... 6
1.4 Importance of New Packaging and Processing Techniques .................................... 8
1.5 Food Safety Concerns .............................................................................................. 8
1.6 Shifts in the Structure of the Retail Grocery Store Indusz .................................. 11
1.7 Disadvantages Of Case-Ready Products ................................................................. 12
1.8 Research Issues ...................................................................................................... 14
1.9 Layout of Subsequent Chapters ............................................................................. 15
CHAPTER 2 — LITERATURE REVIEW ......................................................................... 16
2.1 Outsourcing Introduction ....................................................................................... 16
2.2 Early Outsourcing Decisions ................................................................................. 17
2.3 Principal Outsourcing Approaches ........................................................................ 20
2.4 Transaction Costs Approach to Outsourcing ......................................................... 22
2.5 Strategic Management Approach to Outsourcing .................................................. 31
2.6 Core Competencies ................................................................................................ 33
2.7 The World is Flat Method of Outsourcing ............................................. '._. .............. 37
2.7.1 Implications of Outsourcing Methodology for Case-Ready Meats ............... 44
2.8 Concluding Remarks for Outsourcing ................................................................... 45

CHAPTER 3 — RESEARCH DESIGN: STRATEGIES, PROPOSITIONS AND

3.]

- 3.2

3.3
3.4
3.5
3.6
3.7
3.8
3.9
3.10

METHODOLOGIES ................................................................................. 46
Overview of Research Design ............................................................................... 46
Research Categories ............................................................................................... 46
Selecting the Appropriate Research Strategy ........................................................ 50
Case Study Research Strategy ............................................................................... 52
Appropriate Uses of the Case Study Research Strategy ........................................ 53
Strengths of Building Theory from Case Studies .................................................. 54
Weaknesses of Building Theory from Case Studies .............................................. 55
Case Study Research Design ................................................................................. 56
Case Study as the Preferred Research Technique .................................................. 57
Case Study Beginnings: Key Informant and Industry Executive Interviews ........ 57

viii

3.11 Research Questions ................................................................................................ 61

3.12 Supply Chain Analysis of Case-Ready Meats ....................................................... 65
3.13 Questionnaire Design ............................................................................................. 70
3.13.1 Unit of Analysis ......................................................................................... 71
3.13.2 Respondents to the Questionnaire .............................................................. 73
CHAPTER 4 — DATA ANALYSIS: CROSS-TABULATIONS, MULTIPLE
REGRESSION AND BINARY LOGISTIC REGRESSION .................... 76
4. 1 Introduction ............................................................................................................ 76
4.1.1 Cross-Tabulations of the Research Propositions ........................................... 79
4.1.2 Analysis of General Case—Ready Product Operational Issues ....................... 80
4.2 Cross Tabulation Research Propositions: Case-Ready Product Issues .................. 89
4.2.1 Results of Packaging Operational Issues ....................................................... 91
4.2.2 Handling Operational Issues Data Analysis .................................................. 96
4.2.3 Data Analysis Operational Issues Associated with Shrink ............................ 99
4.2.4 Data Analysis of Food Safety Operational Issues ........................................ 103
4.2.5 Data Analysis of Workrnanship Operational Issues .................................... 107
4.3 Multiple Regression Analysis .............................................................................. 112
4.3.1 Multiple Regression Model: Food Safety Operational Issues ..................... 113
4.3.2 Independent Variables in the Food Safety Regression Model ..................... 1 14
4.3.3 Regression Results for Food Safety ............................................................. 115
4.3.4 Multiple Regression Model: Workrnanship Operational Issues .................. 116
4.3.5 Independent Variables in the Workrnanship Regression Model .................. 116
4.3.6 Regression Results: Workrnanship .............................................................. 1 17
4.3.7 Multiple Regression Model: Shrink Operational Issues .............................. 119
4.3.8 Independent Variables in the Shrink Regression Model .............................. 119
4.3.9 Regression Results for Shrink ...................................................................... 120
4.3.10 Multiple Regression Model: Packaging Operational Issues .................... 121
4.3.11 Independent Variables in the Packaging Regression Model .................... 122
4.3.12 Regression Results for Packaging ............................................................ 124
4.3.13 Multiple Regression Model: Handling Operational Issues ...................... 124
4.3.14 Independent Variables in the Handling Regression Model ..................... 125
4.3.15 Regression Results for Handling ............................................................. 126
4.3.16 Multiple Regression Model: All Operational Issues ................................ 127
4.3.17 Regression Results for All Operational Issues ......................................... 128
4.4 Econometric Analysis of Case-Ready Meat Data Set .......................................... 129
4.4.1 Logistic Regression ...................................................................................... 132
4.4.2 Logistic Regression Coefﬁcients ................................................................. 134
4.4.3 Evaluating the Performance of Logistic Regression Models ....................... 135
4.4.4 Binary Logistic Regression Relative to Case-Ready Meat Adoption .......... 140
4.4.5 Binary Logistic Regression Base Model 1: Ground Beef ............................ 140
4.4.6 Binary Logistic Regressions Model 1: First Run Results ............................ 142
4.4.7 Binary Logistic Regression Model 1: Second Run ...................................... 145
4.4.8 Binary Logistic Regression Model 1: Final Run ......................................... 147

4-4.9 Model 1: Multicollinearity Diagnostics ....................................................... 149

4.4.10 Assessment of Coefﬁcients in Logit Model I: Predicted/Observed Signs of
Coefﬁcients .............................................................................................. 152
4.4.11 Binary Logistics Regression Model 1: Managerial Implications ............ 155
4.4.12 Summary Conclusions for Model I ......................................................... 159
4.4.13 Binary Logistic Regression Model 2 Adoption Case-Ready Whole Muscle
Beef Products ........................................................................................... 160
4.4.14 Binary Logistic Regression Model 2: First Run ...................................... 161
4.4.15 Binary Logistic Regression Model 2: Second Run .................................. I64
4.4.16 Binary Logistic Regression Model 2: Final Run ..................................... 166
4.4.17 Binary Logistic Regression Model 2: Multicollinearity Diagnostics ...... 170
4.4.18 Binary Logistic Regression Model 2: Predicted/Observed Signs of
Coefﬁcients .............................................................................................. 172
4.4.19 Binary Logistic Regression Model 2: Managerial Implications .............. 174
CHAPTER 5 — SUMMARY AND CONCLUSIONS ..................................................... 178
5.1 Study Objectives .................................................................................................. 178
5.1.1 Purpose of Study .......................................................................................... 179
5.1.2 Data Sources and Methodology ................................................................... 180
5.1.3 General Conclusions Drawn from the Questionnaire .................................. 182
5.1.4 Major Findings from the Cross Tabulations Analysis ................................. 182
5.1.5 Multiple Regression ..................................................................................... 188
5.1.6 Binary Logistic Regression Model 1 ........................................................... 190
5.1.7 Binary Logistic Regression Model 2 ........................................................... 194
5.2 Observations ........................................................................................................ 197
5.2.1 Walmart’s Conversion to a Case-Ready Program ....................................... 199
5.3 Implications and Concluding Remarks ................................................................ 202
5.3.1 Managerial Implications .......... -. ................................................................... 202
5.3.2 Relationships with Customers ...................................................................... 204
5.3.3 Point of Sale Information Coordination ....................................................... 204
5.4 Suggestions for Future Research ......................................................................... 205
Appendix A ...................................................................................................................... 206
Bibliography .................................................................................................................... 2 1 8

LIST OF TABLES

Table 1: Costs Related to Selected Food Borne Pathogens, 2004 .................................... 10
Table 2 Organizations Engaging in Major Outsourcing Endeavors ................................ 20
Table 3: Research Strategies Choice Matrix ..................................................................... 51
Table 4: Geography of Questionnaires Returned by Corporate Stores ............................ 77
Table 5: Geography of Questionnaires Returned by Licensed Stores ............................. 78
Table 6: Case-Ready Meat Products Sold by Store Type. ............................................... 82
Table 7: Product Attributes Ranked by Store Type. ........................................................ 83
Table 8: Price Sensitivity of Fresh Meat Products by Store Type. .................................. 84
Table 9: Key Success Factors of Case-Ready Meat Products by Store Type. ................. 85
Table 10: Key Success Factors Related to Proﬁtability .................................................... 85
Table 11: Employee Friendly Case-Ready Meat Alternatives by Store Type. ................ 86
Table 12: Manager Friendly Case-Ready Meat Alternatives by Store Type. .................. 86
Table 13: Case-Ready Product Issues by Retail Store Type. ........................................... 88
Table 14: Case-Ready Meat Management Operational Issues by Store Type ................. 90
Table 15: Ground Beef Packaging Operational Issues by Retail Store Type. ................. 93
Table 16: Packaging Operational Issues by Retail Store Type. ....................................... 95
Table 17: Handling Operational Issues by Retail Store Type. ......................................... 98
Table 18: Handling (Stocking) Case-Ready Meat Cases by Retail Store Type. ............. 99
Table 19: Shrink by Retail Store Type. ......................................................................... 102
Table 20: Shrink Management as a Key Success Factor by Retail Store Type. ............ 103
Table 21: Food Safety and Product Traceability Factors by Retail Store Type. ........... 104
Table 22: Food Safety Operational Issues of Case-Ready by Retail Store Type. ......... 107

xi

Table 23: Workmanship Operational Issues by Retail Store Type. ................................ 109

Table 24: Additional Workrnanship Operational Issues by Retail Store Type. ............. 1 11
Table 25: Summary of Data: Food Safety Regression ................................................... 113
Table 26: Regression Results: Food Safety ................................................................... 115
Table 27: Summary Of Data: Workrnanship Regression ............................................... 117
Table 28: Workrnanship Regression Results ................................................................. 118
Table 29: Summary of Data: Shrink Multiple Regression ............................................. 119
Table 30: Regression Results: Shrink ............................................................................ 121
Table 31: Summary of Data: Packaging Regression ..................................................... 122
Table 32: Packaging Regression Results ....................................................................... 124
Table 33: Summary of Data: Handling Operational Issues ........................................... 126
Table 34: Handling Regression Results ......................................................................... 127
Table 35: Summary of Data: All Operational Issues Regression .................................. 128
Table 36: All Operational Issues Regression Results .................................................... 129
Table 37: Logit 1 Model: Adopted Case-Ready Ground Beef ....................................... 143
Table 38: Classiﬁcation Table (ab) Model I First Run Without Explanatory Variables 144
Table 39: Classiﬁcation Table (a) Model I First Run With Explanatory Variables ...... 145
Table 40: Binary Logistic Regression Model 1 Second Run: Store Adopts Case-Ready

Table 41:

Table 42:

Table 43:

Table 44:

Ground Beef .................................................................................................. 146

Classiﬁcation Table (ab) Logit Model 1 Second Run Without Explanatory
Variables ....................................................................................................... 147

Classiﬁcation Table (a) Logit Model 1 Second Run With Explanatory
Variables ....................................................................................................... 147

Binary Logistic Regression Model 1 Final Run: Store Adopts Case-Ready
Ground Beef .................................................................................................. 148

Classiﬁcation Table (ab) Logit Model I Final Run Without Explanatory
Variables, Intercept Only .............................................................................. 149

xii

Table 45:
Table 46:

Table 47:

Table 48:

Table 49:
Table 50:

Table 51:

Table 52:
Table 53:
Table 54:

Table 55:

Table 56:

Table 57:

Table 58

Classiﬁcation Table (a) Logit Model I With Explanatory Variables ............. 149
Multicollinearity Diagnostics for Logit Model la Ground Beef ................... 151

Binary Logistic Regression Model 1 Adoptions of Case-Ready Ground Beef
Products: Interpreting the Sign and Direction of the Predictor Variables ....153

Binary Logistic Regression Model 2 Base Model: Adoption Case-Ready
Whole Muscle Beef Products ........................................................................ 162

Classiﬁcation Table (ab) Logit Model 2 First Run Without Predictors ........ 163
Classiﬁcation Table (a) Logit Model 2 First Run With Predictors ............... 163

Binary Logistic Regression Model 2 Second Run Model: Adoption Case-
Ready Whole Muscle Beef Products ............................................................ 165

Classiﬁcation Table (ab) Logit Model 2 Second Run Without Predictors 166
Classiﬁcation Table (a) Logit Model 2 Second Run With Predictors ........... 166

Binary Logistic Regression Model 2 Final Run Model: Adoption Case-Ready
Whole Muscle Beef Products ........................................................................ 168

Classiﬁcation Table (ab) Logit Model 2 Adoption of Case-Ready Whole ...170

Classiﬁcation Table (a) Logit Model 2 Adoption-of Case-Ready Whole
Muscle Beef Products F inal Run With Predictors ........................................ 170

Multicollinearity Diagnostics for Logit Model 2a Whole Muscle Beef ........ 171

: Binary Logistic Regression Model 2 Adoption of Case-Ready Whole Muscle

Beef Products: Interpreting the Sign and Direction of the Explanatory
Variables ....................................................................................................... 1 73

xiii

LIST OF FIGURES

Figure 1: Cognitive Map of Core Competencies ............................................................. 36

Figure 2: Conceptual Model for Outsourcing a Core Product ....................................... 199

xiv

KEY TO ABBREVIATIONS

HACCP ...................... Hazardous Analysis of Critical Control Points
IT ................................ Information Technology

TCE ............................ Transaction-Cost Economics

RFID .......................... Radio Frequency Identiﬁcation

DC .............................. Distribution Center

POS ............................ Point of Sale

DSD ............................ Direct Store Delivery

FIFO ........................... First-In, First-Out

XV

CHAPTER 1— INTRODUCTION

1.1 Study Organization

Innovation in the way food products are distributed in the supply chain has been
an important source of productivity growth in the US. and economies worldwide.
Visionaries such as Clarence Birdseye and Sam Walton, by seeking to ﬁll important
unfulﬁlled market needs, radically changed the way that Americans purchase and
consume food. While this process of innovation continues unabated, the uneven adoption
of case-ready meat’s proven technologies and processes by some sectors of the food
industry is concerning; and is the motivation for this study. We trust the study will
contribute to enhanced market performance.

The objective of this research is to determine those critical factors that may affect
fresh meat processors’ and retailers’ decisions whether or not to integrate case-ready beef
products into their product lines. An important aspect of this objective is to obtain a
greater understanding of how retail store managers perceive consumer preferences for
case-ready beef products. In so doing there are several lessons to be derived from the
experience of the poultry industry as it adopted case-ready processing and packaging.

This research identiﬁes six operational issues that are associated with the
implementation of a successful case-ready beef strategy. A more complete understanding
of these issues is critical if the indusz is to adopt these new fresh beef products and
improve the performance of retail grocery stores. In addition, case-ready beef product
and supply chain management information from this project should prove valuable to

beef cattle producers, fresh beef processors and retail grocery ﬁrms.

The research questions and propositions were derived from six operational issues
that retail grocery store managers, processors and distribution center personnel need to
resolve as they consider adopting case-ready beef. The ﬁrst stage of the project involved
identifying issues from the pertinent literature and from direct ﬁeld observations of fresh
beef products (including case-ready) moving through the supply chain. Initial results
were then conﬁrmed by means of extensive consultation with industry experts that
formed the basis for a questionnaire survey of 220 retail grocery store managers of the
Save-A-Lot retail company.

The operational issues identiﬁed are grouped into six broad categories of issues
which together compose the framework of the research project:

1. Food Safety Operational Issues—Do customers and store managers

recognize the safety issues associated with and between both case-ready and in-

store processed products? How might a food borne illness outbreak associated
with fresh beef affect all store sales and meat sales?

2. Workrnanship Operational Issues—The supply of qualiﬁed meat

department personnel who can process meat and manage the inventory effectively

is shrinking and their wage costs are increasing. What are the long run
implications of this phenomenon on meat sales? How might the differences

between the case-ready products and products processed in-stores affect sales?

3. Shrink Operational Issues—What are the respondents (store managers)
viewpoints of shrink"? What are the advantages and disadvantages between the
two product choices as they relate to shrink issues?

4. Packaging Operational Issues—These issues center on consumers’
knowledge. Do they know or can consumers discern the differences and
attributes between the two product choices by looking at the packaging?

5. Handling Operational Issues—Do store managers know the advantages and
disadvantages between these two types of products and the effects on their
proﬁtability and customer satisfaction? What are the key handling issues for both
product types regarding keeping meat cases fully stocked during typical peak
shopping times between 4 PM and 7 PM?

6. Case-Ready Product Operational Issues—The assumptions of costs and
beneﬁts of case-ready products on store proﬁtability and customer satisfaction.

What are the key operational issues as they relate to the quality of the product?

In addition to addressing the above issues using qualitative and quantitative
assessment techniques, the study also includes in-depth analysis of how they affect the
decision to outsource a core grocery store item. When retailers are contemplating
adopting case-ready beef products they are in reality negotiating with processors to
outsource or produce fresh beef products that normally in the past are core to a retail
grocery store’s product mix. Strategic management theory is adamant that ﬁrms not

outsource items that are core to their business line. But important changes in markets,

 

1 There are several key issues related to shrink: customer/employee theft, incorrect inventory,
workmanship, and out-of-date product.

technology, ﬁrm linkages, etc. can result in substantial changes overtime. While this is
the traditional view of outsourcing, there is a new view or role for outsourcing taking
place. The world’s largest retailer Walmart, does not view outsourcing products such as
case-ready meats through the lens of fresh meat products being core or non-core but
whether a given decision to outsource improves its overall business model. Situations
such as case-ready meat products need to be researched differently, using new

perspectives and methods, which addressed in this study as well.

1.2 Retailer and Consumer Acceptance of Case-Ready Chicken

The poultry industry long ago realized that case-ready chicken products had the
potential of transforming a commodity product line into a valued-added product line that
would result in increased proﬁt margins. Case-ready chicken programs utilized new and
improved production techniques, processing and packaging technologies, and innovative
retail marketing strategies in ways that transformed how retailers ordered, stocked and
merchandised poultry. Consequently, case-ready poultry dramatically altered how
consumers purchased and prepared chicken in their homes. The new marketing strategy
involved segmentation of the market for whole fresh chickens, thus providing consumers
with product choices that better reﬂected their preferences and needs. Moreover, case-
ready chickens were simpler to prepare in the home. Early case-ready chicken product
innovations included separating and packaging breast, leg, thigh and wings, satisfying
customers increasing expectations for safer and more convenient poultry products.

Prior to the introduction of case-ready methods, the market for poultry products
was strictly a commodity whereby whole broilers were displayed in meat counters on

trays of ice. While this format was efﬁcient for processors and retailers, it largely

ignored the latent consumer preferences for poultry in more convenience forms that held
the potential for enhanced economic rewards associated with a marketplace
differentiation strategy.

Case-ready chicken products added tremendous value to commodity poultry '
products and contributed to its surge in popularity over the past 50 years. Consumers
beneﬁted in two important ways. First and foremost, they no longer had to cut up the
whole bird prior to cooking and decide how to use less desirable parts. Secondly,
consumers could also choose the types of poultry meat they most preferred, and these
packages could either be prepared or placed in the freezer for later use, with no apparent
decline in quality.

Grocery retailers also beneﬁted from case-ready chicken products. Prices they
received for these value added products increased. Also, food safety was enhanced as a
result of eliminating cross-contamination of poultry with other-ﬂesh meats. For example,
retailers no longer stored fresh poultry in exposed wire cages in walk-in coolers.2

Long term success of the transformation of case-ready chicken was a result of the
industry’s acknowledgement of the role producers and input suppliers have on the
ultimate success of these new products. As retail demand for the new case-ready poultry
products grew, processors began to standardize poultry breeds, feeds, and health products
to keep pace with demand. Soon, poultry producers and processors began raising broilers
under strict supply chain management arrangements, with each actor concentrating on
improving Operations in its respective ﬁeld of expertise. Processors expanded control

along the vertical supply chain by implementing exclusive contracts, formula pricing and

 

2 This storage practice was closely associated with cross contamination of other meat products that were
also stored in the coolers.

other captive supply production arrangements to improve product quality and
consistency. Further control of the vertical supply chain also allowed producers to lower
production costs, increase supplies, provide a uniform product, shorten cycle times, and
lessen the many uncertainties that can plague the production of commodity products
(Hayes et al, 1999; Sharland et a1, 2003).

Another important change in the poultry industry associated with case-ready
chicken was that broilers became more uniform in size, allowing processors to introduce
mechanized slaughtering, cleaning, cutting, and portioning. Mechanizing these stages
contributed to processors’ proﬁts by increasing capacity utilization, which also improved
the consistency of the ﬁnished products. The most signiﬁcant result was lowered labor
costs.

In the early 1960’s, over 80 percent of broiler production was dedicated to whole
fryers. By 2007, whole fryers accounted for 18 percent of production. In addition to fresh
case-ready chicken products, new value-added products such as breaded strips, chicken.
patties, and nuggets have been introduced as the industry moved even further from its

commodity orientation (Martinez and Stewart, 2003).

1.3 Limited Success of Case-Ready Beef

In comparison to the experience of chicken, the adoption of case-ready beef has
seen limited success. In 2007, approximately 25 percent of all retail meats were sold as
case-ready and only one major retail chain, Walmart, had committed to marketing 100
percent case-ready meats for all of its fresh meat offerings (Major, 2007). Many industry

experts had expected that case-ready beef, with its inherent advantages for processors,

retailers and consumers, would have followed the same successful path case-ready
chicken pursued over ﬁfty years ago (Barkema et al, 2001).

The limited success of case-ready beef products is surprising given the existence
of several favorable external demand factors such as consumers’ tastes and preferences
for good tasting and consistent fresh meat products, food safety issues,
governmental/legal/environmental issues surrounding the production and processing of
beef cattle, and new processing/packaging technologies favoring adoption of new
processing and packaging methods. Paradoxically, these demand factors are similar to
the ones the poultry industry faced when they considered the adoption of case-ready
programs.

Moreover, consumers have a heightened concern and awareness about the safety
of beef products due to highly publicized outbreaks of E. coli 0157:H7 in 1993 caused by
tainted hamburger meat and other publicized food safety catastrophes (Barkema et al,
2001). In addition, the supply of qualiﬁed meat department personnel is dwindling and
the wages paid for qualiﬁed meat joumeymen are on the rise, driving cost of maintaining
meat cutters employed in the retail setting to prohibitive levels (Keith, 2004; Salvage,
2003 and Brody, 2002).

In contrast with many other retail grocers that have introduced case-ready meats
with limited success, Walmart has adopted a case-ready meat program for all poultry,
lamb, beef and pork products for all of its retail store formats (Sam’s Club, a wholesale
store format, excepted). Several large retailers have succeeded with a partial adoption of
case ready products, typically limited to ground beef and whole muscle pork items (Excel

Meats, 2003).

1.4 Importance of New Packaging and Processing Techniques

Fresh beef and pork processors have recently gained the ability to successfully
produce case-ready products by means of technological developments in packaging and
processing, including new processing technologies that automate processing in new
Hazard Analysis Critical Control Points (HACCP) certiﬁed “clean rooms”. New
packaging technologies include ways of encapsulating the fresh meat products with a
modiﬁed atmosphere ensuring signiﬁcantly extended shelf life. For many years, fresh
beef has been centrally processed and packaged using conventional plastic ﬁlms and trays
providing a shelf life of no more than two or three days which is insufﬁcient for fresh
beef products to move through the supply chain and be sufﬁciently fresh for consumer to
merchandise to consumers. Case-ready beef products now can be centrally processed
within HACCP certiﬁed clean rooms, packaged and shipped throughout the United States
with anticipated shelf lives of 16-21 days", providing sufﬁcient time to flow through the
supply chain and into consumer’s homes before the expiration date (Bardic, 2003; Dolan,
2001; Bjerklie, 2001; Brody, 2002; Demetrakakes, 2001; Porter, 2002; Galosich, 1999;

Keith, 2002; Barry, 2002; Galosich, 2000; Young, 2001; Tosh, 1999; Bennett, 1995).

1.5 Food Safety Concerns
Despite the enormous negatives of food home illness to beef processors and
retailers, the industry has not adopted case-ready programs which are generally

acknowledged as the safer method of meat distribution. It is clear that each incident of

 

3 Case-ready products packaged using “Low-Oxygen” technologies can provide these extended shelf life
periods.

food borne illness associated with fresh meat not only harms the reputation and
proﬁtability of the companies directly involved, but also of the entire meat industry.

Even with the beneﬁcial attributes of case-ready meats with respect to lessening
the incidents of unsafe meats, given the realities of the meat industry, food safety is and
always will be a major concern for processors, livestock producers and grocery retailers.
The consequences for “getting it wrong” are enormous. The Center for Disease Control
and Prevention estimates that one in three Americans in (Buzby, 2001) became ill or had
serious health conditions related to a food borne illness. Many others are hospitalized
and several die each year (Mead et al, 1999).

On the demand side, consumers have heightened concerns about the safety of beef
products due to highly publicized outbreaks of E. coli 0157 :H7. In September 2007,
Topps Meat Company of New Jersey and the USDA recalled 21.7 million pounds of
ground beef containing E. coli 0157:H7. The tainted ground beef caused one fatality and
35 people to contract severe food poisoning. The total ground beef recalled due to E. coli
0157:H7 in 2007 was over 30 million pounds. This outbreak and other publicity of food
borne illnesses associated with beef and product recalls has likely accentuated
consumers’ lack of trust in the safety of supply of beef.

In 2004, the USDA’s Economic Research Service (ERS) estimated that ﬁve
speciﬁc bacterial pathogens caused severe human illnesses that cost the US. economy
$6.9 billion per year. As seen in Table 1, the ﬁve food borne pathogens cited by the ERS
are Campylobacter (Guillain-Barré syndrome), Salmonella, and Escherichia coli 0157:H7

(hemolytic uremic syndrome), Escherichia coli non-0157:H7 STEC, and Listeria

Monocytogenes (congenital and newborn infections resulting in chronic disability or
impairment) (Ollinger et al, 2004).

Table 1: Costs Related to Selected Food Borne Pathogens, 2004

 

Pathogen Cases Number of Deaths Costs
Hospitalizations (S Billions)
Campylobacter 1,963,141 10,53 9 99 $1.2
Salmonella 1,341,873 15,608 553 2.4
E.coli 0157:H7 62,458 1,843 52 0.7
Ecoli non-0157 31,229 921 26 0.3
Listeria 2,493 2,298 499 2.3
Total 3,401,194 31,209 1,229 $6.9

Source: Ollinger et al, 2004

Beef cattle producers and processors are particularly concerned with Ecoli 0157
because it can be introduced anywhere along the supply chain, its unparalleled ability to
survive most environments, and its toxicity as a pathogen. Ecoli 0157 is of particularly
insidious because it is commonly found in the nunen stomachs and intestines of beef and
dairy cattle.

Cross-contamination of fresh meat can occur along any stage of the supply chain
if proper hygiene procedures are not adhered to. In the past, researchers believed that
freezing fresh beef would kill many types of bacteria. However, only cooking at high
temperatures has demonstrated to be effective in killing the E.coli 0157 pathogen. At any
rate, both techniques negatively affect the taste that consumers desire, making proper
food handling throughout the supply chain all the more important (Pennington, 2000).
Meat processors have taken these facts about food safety and incorporated them into their
business model, speciﬁcally by processing fresh case-ready beef products in HAACP

certiﬁed clean rooms.

10

1.6 Shifts in the Structure of the Retail Grocery Store Industry

Food marketing in the retail grocery store venue has seen dramatic changes over
the past twenty years, the most signiﬁcant of which was the successﬁrl entry of general
merchandisers such as Walmart, K-Mart and Target Stores into traditional food retailing.4
Current market leader Walmart quickly reached the position of number one grocery
retailer in the US. just a few years after entering the grocery business in 1989. This
ascent was greatly aided by adapting their already highly efﬁcient purchasing and supply
chain structures and practices to accommodate the unique challenges of perishable
grocery items like case-ready beef products. Key to Walmart’s success is their efﬁcient
nationwide system of 99 distribution centers, 34 of these which are dedicated to the
distribution of grocery products. Grocery store industry analysts agree Walmart could
not sell as much product as they do without a very sophisticated distribution network.

The commitment of Wahnart to sell 100 percent of all fresh meat products case-
ready demonstrates that moving to case-ready products is likely to be proﬁtable, but only
for those retailers with a robust supply chain. A robust supply chain in a retail setting
was must accommodate not only dry goods but they must be able to accept highly
perishable ﬂesh and frozen products as well. This is particular true of accepting the
temperature sensitive case-ready products at ever stop along the supply chain. The good
news from Walmart"s perspective is the ﬁrm believes that switching to case-ready will
allow it to better serve its customer base by providing a safe and ﬂesh product that ﬁt

seamlessly within their existing supply chain competency.

 

’ It should be noted that shifts in the structure were also led by signiﬁcant changes in consumer’s
preference for food, which is not covered in this research project.

11

The impact of Walmart’s commitment to case ready products is signiﬁcant in an
industry where the concentration ratio of the top four (CR4) food retailers is 34 percent
and 72 percent in the largest 100 US. cities. (King et al, 2001, Barkema, 2001) In
2006/2007 Wal-Mart operated 2,176 supercenters and 574 Sam’s Club stores with over
$232 billion in total sales, versus $455 billion for the industry as a whole in 2006.
Kroger, the next largest retailer grossed over $66 billion in sales ﬂom 4,276 retail stores.
Many other large grocery chain stores believe that ﬂesh case-ready meat processors may
sell to them and others, but they will cater to Walmart and not other retailers. The fear of
other retailers centers around the sheer size and industry dominance Wahnart exerts on

the industry.

1.7 Disadvantages of Case-Ready Products

The total costs savings ﬂom adopting to case-ready beef products are not readily
recognized by retailers employing the industry standard cost of goods sold methodology.
Despite signiﬁcantly lower costs of production for case-ready beef products, the switch to
a case-ready format can be difﬁcult to justify. The literature and industry experts cite
difficulty in assigning costs savings to meat products produced under HACCP guidelines,
products that have extended shelf life, and that result in fewer employee injuries (Miller,
2001 and 2002). Many large supermarket chains cannot switch to case-ready because of
labor agreements with their employees that protect speciﬁc job classes, meat cutters in
particular (Petrak, 2007 and Wagar, 2005). Wal-Mart, which does not have unionized
employees, was not constrained by labor agreement issues.

Another hurdle to overcome is consumers’ perception, with some justiﬁcation,

that ﬂesh meat processed on—site has higher quality than case-ready. The two largest club

12

stores, Sam’s Club and Costco, both sell ﬂesh choice graded meat that is processed and
packaged on-site. Currently, most case-ready beef products are of lower quality usually
are select grade. Select grades of beef cannot compete against the higher quality choice
grade for tenderness, marbling, and ﬂavor. At the 2007 American Meat Institute (AMI)
Meat Conference, the quality of ﬂesh meat products was cited as a major reason
underlying consumers’ choice of shopping venues (Petrak, 2007 and Major, 2007). For
this reason, retailers may hesitate to take a chance with a product where any gains may be
Offset by a loss of business, particularly to larger competitors. Walmart has chosen to
keep their meat cases full 24/7 with HAACP certiﬁed case-ready meat products in lieu of
processing a higher grade meat at each store within the entire chain.

The beneﬁts of the longer shelf life of case-ready can be offset by the cost of
potential diminished consmner perception of the quality of the meat. The “modiﬁed
atmosphere” packaging that creates a long shelf life does not resemble the packaging of
meat traditionally produced in the store. Case-ready packaging has a different color tray,
a tray that is taller to create the large “headspace” required to accommodate the modiﬁed
atmosphere gases, and uses a different looking ﬁlm or overwrap. Together, these new
components are different enough that customers realize that the meat is not produced on-
site, causing concerns about ﬂeshness and quality (Enis, 2007 and Major, 2007).

The principal reason case-ready beef programs has not experienced higher
adoption rates is that the per-unit costs of case-ready are higher for both retailers and
consumers (Major, 2006). Processors charge retailers more per unit for case-ready
products because. the limited market for case-ready meats requires them to maintain their

traditional processing lines, preventing processors ﬂom beneﬁting ﬂom economies of

13

scale. Retailers face a related problem in that not all types of whole muscle products are
available in the case-ready format, forcing them to choose between losing business due to
product lines that do not match their customers or maintaining their on-site processing

capacity, but at a lower utilization rate and higher unit cost.

1.8 Research Issues
The objective of this dissertation is to determine what critical factors may affect
ﬂesh meat processors and retailers’ decision whether or not to integrate case-ready beef
products into their product lines. In order to meet this objective, it is important to better
understand the operational issues that help or hinder the successful introduction of case-
ready beef products into retail grocery stores. Decision makers in the ﬂesh beef industry
have for years contemplated and attempted various strategies for increasing adoption of
case-ready technology. This research project seeks to assist the industry in those
endeavors.
Sub-Objectives of the Project
0 Of interest to this research project are the reasons why case-ready red meat
products have not been accepted as readily by retail supermarket ﬁrms and
consumers as case-ready broilers have been. Case-ready poultry programs
have grown steadily over the past thirty years in food stores, growth which has
encouraged many product innovations and provided a broad line of related
convenience products. These case-ready programs have resulted in numerous
beneﬁcial outcomes, including improved food safety, efﬁcient stocking, and

highly functional packaging.

14

o In contrast, the beef industry is eager to accelerate the change to what many in
the industry think someday will be the acceptance of case-ready beef. It is
important to gain better understanding how to achieve the same kind of
beneﬁcial results for the beef industry that have accrued to the progressive
poultry industry.

0 The question this research addresses is why, despite the many apparent
advantages of case ready beef products to retailers and consumers, as well as
the success of Walmart, consumers and retailers have been slow to adopt these
products. Fresh meat sales represent an important share of total store sales,
yet relatively little research has been conducted on relevant meat industry
Operations. Because the trade and academic literature does not provide
comprehensive answers to this central question, there is an apparent need for
this research project. It is expected that retailers, processors and packaging

companies will beneﬁt the most from this research project.

1.9 Layout of Subsequent Chapters

The dissertation is divided into ﬁve chapters. Chapter 2 provides the industry
setting of case-ready beef and its new packaging technologies and contains a literature
review related to the project. Chapter 3 details the survey design, and the survey
instrument used to address the research objectives. Chapter 4 provides data analysis
results. Chapter 5 provides a summary of the research ﬁndings, conclusions and

implications for further research.

15

CHAPTER 2 - LITERATURE REVIEW

2.1 Outsourcing Introduction

Successfully managed ﬁrms continually seek ways to lower their operating costs
while maintaining or improving the integrity and quality of their products. As part of this
process, ﬁrms often must decide whether or not to outsource production of
components/services or even the entire product or service. While a desire to lower costs
is the primary reason ﬁrms engage in outsourcing, potential other goals include;
improving access to state-of-the-art technology, reducing product-to-market cycle times,
removing low priority work in order to concentrate on core competencies, capturing the
beneﬁts of economies of scale and seeking performance and/or quality improvement
(Kakabadse, 2006; Gay and Essinger, 2000 and Quinn, 1999).

Decisions to outsource confront all industries. Neither Apple nor Dell computer
companies, for example, manufacture the components in their computers. These ﬁrms
instead rely on a sophisticated international network of ﬁrms to produce and assemble the
requisite electronic components—allowing them to concentrate their expertise in
marketing and supply chain management of the ﬁnished products. Even the Ford Motor
Company, whose River Rouge industrial complex was the world’s ﬁrst completely
vertically integrated automotive plant, now outsources production of a majority of the
automotive components in its cars and trucks. Ford and General Motors both anticipate
the 25 percent per year expansion of their outsourcing agreements with China to continue
until 2010 (F aircloth, 2007).

The strategic management literature has traditionally supported the precept that

only a ﬁrrn’s “non-core” products or services should be considered for outsourcing, while

16

“core” items should always remain within the ﬁrm (Prahalad and Hamel, 1990; Porter,
1985; Quinn and Hilrner 1994)

Outsourcing decisions can be viewed ﬂom either a purely cost perspective or one
that also considers other important strategic factors in major part due to greater
globalization of industry. Economic integration has progressed, leading to a more
complex version of outsourcing. Yet the overall goal of keeping ﬁrms competitive
remains. New models of outsourcing are based upon several key points such as the
availability of high speed data networks in many more countries (including many of the
poorest countries), expansion of economies around the globe driven by even more hyper-
competitive and Open markets in the United States, and a world where international
relationships are more deﬁned by trading relationships than military actions. It is
appropriate to reference Thomas Friedman, who has described this phenomenon in his
book “The World is Flat.” And it is in this context outsourcing is helping to drive this

world-wide phenomena.

2.2 Early Outsourcing Decisions

The 1960’s saw the ﬁrst signiﬁcant outsourcing of business support functions that
had traditionally been performed in house. These included purchasing or “timesharing”
computer times, payroll, inventory control, and accounts payable/receivable.
Outsourcing these support functions was viewed as a cost-cutting measure because
management was able to reduce the number of employees on the payroll (Fill and Visser,

2000). In the early 1960’s Frito Lay and Blue Cross Blue Shield outsourced all of their

 

5 In the 1960’s several computer ﬁrms SIA, Baric, and Compute] owned large and expensive mainframe
computers. These ﬁrms outsourced computer time (Gay and Essinger, 2000).

17

data processing or Information Services (IS) to Electronic Data Systems (EDS). These
early outsourcing contracts were very cost effective for the ﬁrms and those providing the
outsourced service (Lacity and Hirschheim, 1993).

During the 1970’s, large manufacturing ﬁrms struggling with dismal rates of
return responded to Wall Street’s lower stock valuations by outsourcing a large
proportion of non-core products/services (Bettis et al., 1992; Mclvor, 1997 and Probert,
1996). These same ﬁrms also began outsourcing facilities management tasks such as
building maintenance, foodservice, engineering, and security. Outsourcing was seen as a
viable strategic response to lower stock prices and helped ﬁrms reduce costs in non-core
departments, thus allowing them to become more competitive and proﬁtable (Hirschheirn
and Lacity, 2000).

The global recession of the early 19805 provided fertile ground for the strategic
decision to outSource. Outsourcing became an immediate response to the need for
corporate restructuring during these lean years, and allowed ﬁrms to produce the same
goods and services and maintain their core businesses while lowering costs. Internal
support services such as Information Technology (IT), payroll, billing, custodial, and
foodservices became especially popular outsourcing targets.

In 1988, Eastman Kodak outsourced the majority of its IT operations, ushering in
a new outsourcing wave for this all important support function that required unusual
amounts of human and physical capital not directly related to production (Earl, 1996,
Lacity and Hirschheim, 1993, Linder, et al., 2002, Lonsdale, 2001). Kodak’s outsourcing
decision allowed it to concentrate its managerial expertise, internal activities, and

resources toward core activities and helped it avoid maintaining specialized skills in-

18

house not core to its business. This function was a logical choice to be outsourced since
many of the IT costs were predicted to increase rapidly and would have required
signiﬁcant amounts of new capital to remain technologically current (Linder, et al.,
2002). Kodak’s successful move to outsourcing caused quite a stir among companies that
were spending considerable resources on keeping the IT function in-house. Soon other
major corporations were following this new trend. Chase Manhattan Bank, British
Petroleum, General Dynamics, JP Morgan, McDonnell Douglas, Xerox and Continental
Airlines were among those outsourcing to third party IT providers (Hirschheim and
Lacity, 2000; DiRomualdo and Gurbaxani, 1998).

As the trend to outsource IT grew, IT quickly became viewed as a commodity,
and demand surged for independent third party vendors whose core competencies were
maintaining and expanding state-of-the-art technology and human resources for these all
important areas of IT. These specialized IT ﬁrms offered economies of scale cost-
reductions and less risk than the pioneers in the 1980’s had faced. Fueling this trend was
the proliferation of third-party logistics (3PL) outsourcing ﬁrms and think tanks devoted
to the discipline of outsourcing. These included Electronic Data Systems (EDS),
Outsourcing Institute, Deloitte-Touche, and Accenture (Lacity and Hirschheim, 1993).

Presently, more than 90 percent of US. companies outsource at least one activity
or function according to the International Data Corporation (IDC), an international
research organization (Outsourcing Institute, 2003; Gay and Essinger, 2000). Table 2
lists the top organizations engaging in major outsourcing activities according to a 2006

report ﬂom the Outsourcing Institute.

19

Table 2 Organizations Engaging in Major Outsourcing Endeavors

 

Allied Signal Ford Motor Company Mobil Corporation
American Airlines GAP, Inc. Morgan Stanley
American Express General Electric Motorola

Amoco Corporation General Motors NIKE

Apple Computers Goldman Sachs Owen Corning
Compaq Computers Johnson Controls Procter & Gamble
Delta Airlines Kellogg Company Quaker Oats Company
US. Dept of Defense Merrill Lynch Sharp Electronics
Dow Chemical MetLife Sprint

DuPont Microsoft Boeing

Source: Outsourcing Institutes’ 2006 Outsourcing Audit

Companies currently outsourcing business functions indicate that reducing and
controlling costs are important (54 percent) as is improving the focus of the company (55
percent). The ability to ﬂee resources for other activities (38 percent), gain access to
world-class capabilities (36 percent), and to gain resources not available to the ﬁrm

internally (25 percent) were also identiﬁed as important (Outsourcing Institute, 2006).

2.3 Principal Outsourcing Approaches

Several disciplines have contributed to the theory of outsourcing including
accounting, economics and strategic management. From the retailer’s perspective,
adopting a case-ready meat strategy is outsourcing the production of at least a portion of
their ﬂesh meat. The most important tangible costs a retailer must consider when
deciding whether or not to outsource are related to cost minimization. The literature
refers to this as the “accounting approach” (Culliton, 1942; Gross, 1966; Cooper, 1990;
Gambino, 1980 and Ellis, 1992).

The other consideration many retailers consider is categorized as being strategic
in nature, and the pertinent literature cites this stream of literature as belonging to the
transaction costs approach (Coase, 1937; Williamson, 1975 / 1985; van Hoek, 2000;

Lonsdale, 2001; Poppo and Zenger, 1998; Logan, 2000; Park, 2000; Klass, 1999; Slater,

20

2000; Ellram and Carr, 1994; Maltz, 1992; Sislian and Satir, 2000; Maltz and Ellram,
I997; Prahalad and Hamel, 1990; Welch and Nayak, 1992; Venkatesan, 1992; Mclvor et
al, 1997; Probert, I997; Schoemaker, 1992; and Alexander, 1996).

The earliest published research on how manufacturing ﬁrms should consider
whether to outsource non-core products was conducted by Culliton (1942). Culliton’s
work was based strictly upon the fum concentrating on minimizing costs. He suggests
ﬁrms should consider using a complete cost-analysis the alternatives and chose the
cheaper alternative. While Culliton believes the decision criteria for choosing
products/processes to outsource should be solely based upon the goal of lowering costs,
his research suggests that ﬁrms making outsourcing decisions ought to balance its
decision by considering the relevant Strategic consequences as well.

The standard cost approach to outsourcing is driven by the critical assumption that
ﬁrms engaging in minimizing costs are actually increasing proﬁts based upon static
revenue sales for the ﬁrm (Maltz and Ellram, 1997). This cost approach requires only
accounting information for analyzing potential products that may be outsourced
(Jennings, 2002). According to Gardiner and Blackstone (1991) the accounting
information is fairly standard and will take into consideration ﬁxed costs, such as general
and administrative overhead and research and development, and variable costs, such as
materials, direct labor, and variable overhead costs.

While. the cost approach to outsourcing sounds straightforward, accurately
documenting and allocating these costs can be a daunting task and many ﬁrms have
inadequate costing systems that have not kept up with technology and ﬁrm expansion

(Y con and Naadimuthu, 1994; Ford et al., 1993; Mclvor, 1997; Tayles and Drury, 2001;

21

Giffr, 1993). Gambino (1980) and Williamson (1975 and 1981) both stipulate that cost
factors are the most important factors of determining whether to outsource. Williamson
states his point very succinctly; “economizing is more fundamental to the ﬁrm than
strategy.” In contrast, Culliton (1942) warns potential outsourcing ﬁrms that growing via
the cost approach only will hinder ﬁrms’ ability to grow strategically.

According to Dr. Richard Levin, University of North Carolina professor who
spent three years analyzing the accounting costs associated with retailers adopting case-
ready meat products. Levin was convinced retail grocers were not fully accounting for
all of the costs associated with processing meats in-stores. Levin thought retailers did not
have a ﬁrll accounting of several costs when considering case-ready meats. They include:
shrink, cost of processing waste, out-of-date product and sales lost due to out of stock
items. This lack of accurate accounting information indicates that many retailers may be

underestimating the beneﬁts of adopting a case-ready format (Hodgins, 2005 ).

2.4 Transaction Costs Approach to Outsourcing

The cost and strategic streams of outsourcing research both credit Ronald Coase
(1937) for ﬁrst discussing the phenomena of outsourcing and applying the tenets of
economics toward the topic. Coase concluded outsourcing raised three important and
questions: 1) Can ﬁrms minimize production costs through organization techniques? 2)
Why is all production not carried out by one large ﬁrm? And 3) are there other costs to
using price mechanisms? His seminal work, The Nature of the Firm (1937) provided
valuable insights into the above questions.

Coase began his analysis in The Firm by conSidering the circumstances under

which ﬁrms grow. He pondered the research question of: How ﬁrms should grow and

22

whether ﬁrms should hire new personnel or contract the labor out for a particular task?
This is the classic outsourcing question. Coase believed that as ﬁrms become larger there
exist decreasing returns to the entrepreneur function. Also, they increase their revenues,
increased overhead in the cost of production will in turn decrease returns. Coase viewed
these two costs as countervailing to the ﬁrm (Welch et al, 1992; Mclvor et al, 1997).

From Coase’s perspective, the willingness to outsource may be a function of the
ﬁrm’s size. As related to retail grocers, the question is: Should the retailer continue to
employ meat cutters in-house or outsource their function to case-ready meat processors?
According to Coase, as transactions increase with ﬁrm expansion, the entrepreneur often
fails to use the factors of production to their optimal use. In addition, the price of one or
more factors of production may rise because smaller, specialized ﬁrms often have
competitive advantages over the larger ﬁrm for certain factors. Expansion costs will
increase until they become equal with outsourcing the same transaction in the
marketplace or creating another ﬁrm to accomplish production via vertical integration of
ﬁrms. He maintained that the size of the ﬁrm is directly correlated with ﬁnding an
optimal balance with all costs, and ﬁrms mitigating those costs will tend to increase in
size. As these costs increase, the proclivity for ﬁrms to outsource will increase as well.
In theory then, retail grocers’ growth depends in part upon their success in outsourcing
products and processes to more efficient and specialized ﬁrms for those factors.

Coase classiﬁed transaction costs into three main areas: costs associated with
information, costs of negotiating, and costs of enforcement. According to Coase, new
ﬁrms will form when existing ﬁrms try to avoid these costs internally or to economize on

transaction costs (Logan, 2000). In Coase’s language, transaction costs can be deﬁned

23

further as those costs related to ﬁrms carrying out exchange that do not go through a price
mechanism, yet they are costs that are incurred in the transaction (Hobbs, 1996). These
often include food safety, inventory control and employee management. In a retail
grocery setting, in-store meat processing success depends substantially upon the skill of
the meat cutters and their ability to match inventories with customer demand Food
safety plays an important role as well. Case-ready processors are usually able to employ
more sophisticated and stringent food safety processing protocols than a retail grocer.
Outsourcing allows retail grocers to minimize their transaction costs by contracting with
one specialized ﬁrm that will be in a better position to evaluate and mitigate these
“hidden” costs.

Oliver Williamson in his seminal work Markets and Hierarchies (1975), states
that the theory of transaction-cost economics (TCE) is the conceptual basis for
outsourcing. In 1975 Williamson revived Coase’s earlier work on transaction costs and
attempted to determine the optimal business form a ﬁrm would take in the marketplace
based on the type of transaction costs they faced, with a key assumption that ﬁrms are
constantly engaged in mitigating transaction costs. Williamson believed the inherent
properties of a transaction should determine the most efﬁcient governance structure the
ﬁrm should take to mitigate transaction costs. His governance structures lie along a
continuum with market-based solutions at one end and hierarchical or strategic alliances
at the other end (Williamson, 1975/ 1985; Mclvor, 2000 and Hobbs, 1996).

Outsourcing transactions, according to Williamson (1975, 1985), have four key
factors or assumptions that underpin TCE theory. They include opportunism, bounded

rationality, asset speciﬁcity, and informational asymmetries. A central point of TCE

24

theory relates to the properties of transaction costs. The types and amounts of transaction
costs will determine where on the continuum a ﬁrm will have its governance structure.
Centrally processed case-ready meat processing requires highly specialized facilities and
assets, skilled and carefully trained employees that are monitored for compliance with
HAACP certiﬁcation. This suggests that often outsourcing is a more efficient model for
retailers to employ.

Another way Williamson (1975, 1985) complemented Coase’s work was by
positing that ﬁrms behave “opportunistically” in outsourcing relationships. Williamson
deﬁnes opportunism as self-interest with “guile.” He proposes that where few suppliers
of products or services exist in a given market, the potential for opportunistic behavior
will also co-exist. Globerrnan and Vining (2004) describe opportunistic outsourcing
behavior occurs when one party changes the agreed terms of a transaction to its favor.
This does not imply that all parties will act opportunistically, but the threat is present in
the marketplace. The risk of opportunistic behavior on the part of suppliers is greatest
when there are few ﬁrms for a buyer to choose ﬂom. Opportunistic behavior on the part
of suppliers is when they demand higher prices than agreed upon because of the reality of
few suppliers for the buyer to rely on (Hobbs, 1996).

With the complexity and prevalence of opportunistic behavior, costly contracting
is resorted to in order to lessen exploitation form both parties. An example in the retail
grocery indusz occurs when there are a few suppliers of case-ready meat products for
retailers to choose ﬂom. This increases the likelihood that a supplier or all suppliers will
exploit the lack of competition and behave opportunistically. This requires carefully

monitored contracts or the formation of strategic alliances.

25

Bounded rationality according to Williamson (1975, 1985) identiﬁes the cognitive
ability of humans to have and maintain rational boundaries. Williamson assumes humans
will act rationally up to those boundaries or to the limits of their ability. Mclvor (2000)
agrees with the physical limitations aspects of bounded rationally as “the rationality of
human behavior is limited by the ability of the actor to process information.” Bounded
rationality assumes that people intend to make rational decisions, but their capacity to
accurately evaluate all of the possible decisions is physically limited or constricted by the
enormous size of the choice set. It can also be interpreted as imposing limits on the
decision maker’s cognitive ability, which ultimately limits their rationality (Rindﬂeisch,
1997). Retail store managers, whose duties include monitoring many different classes of
employee, may not have the time of skill set to adequately monitor activities in their meat
department. This can provide a reason for outsourcing as a superior alternative.

Williamson’s deﬁnition of asset speciﬁcity relates to the signiﬁcant investments
in assets for a particular business or exchange, and those assets have a limited ability to
be converted into other uses for creating cash ﬂow and are non-trivial in size and scope.
The limiting ability to be converted into other uses presupposes these types of assets
cannot be redeployable to any other activities (Bienstock and Mentzer, 1999). Therefore,
asset speciﬁc investments possess a negligible residual value as well. An example is the
specialized facilities required by mandatory HACCP food safety protocols may exceed
the capacity of regional or smaller retailer grocery stores. I

The relevance of asset speciﬁcity in outsourcing case-ready meat products is
signiﬁcant and important. Retail grocers may be interested in outsourcing ﬂesh meat

production to an independent processor of case-ready meat products, but this activity

26

requires high asset-speciﬁc capital equipment expenditures to package the case-ready
products and would have severally limiting redeployable uses. It may prove difficult to
replace a supplier that has invested heavily in assets to fulﬁll the outsourcing of case-
ready meat production, because few processors may be willing to incur the requisite costs
for limited and speculative uses. Interviews with a very large meat processor conﬁrmed
these assertions.

The mere existence of factors of production that possess high levels of asset
speciﬁcity could mean there is a higher risk of opportunistic behavior ﬂom a supplier
using those speciﬁc assists. Mitigating asset speciﬁcity effectively will require both
parties to develop a trusting relationship and/or a formal alliance. Williamson (1985)
argues for this course of action so parties to the contracts become locked into the
relationship and the subsequent details for planning and cash ﬂow purposes.

Bounded rationality in concert with asset speciﬁcity creates an atmosphere that
hampers optimal decision making. This uncertainty is a fundamental supposition in TCE
theory. Williamson (1975, 1985) believed the source of uncertainty is found in the
bounded rationality of contracting in the market. He points out that the uncertainty in
bounded rationality makes it impossible to deal successfully with the market complexities
via contingent claims contracts. Market complexities are not only complicated to
envision, they are impossible to adjudicate ex ante. A ﬂequent premise in Williamson’s
(1985) research is that transactions subjected to the ex post opportunism will beneﬁt if
and only if safeguards can be created ex ante. Examples of these types of safeguards are

performance incentives and superior governance structure (Slater and Spencer, 2000).

27

Slater and Spencer argue (2000) argue that Williamson’s (1985) recognition of
uncertainty is conditional upon the complexity and number of possibilities in the
marketplace and the ﬁnite limits of cognitive ability humans possess to deal with every
potential marketplace scenario. They believe that ﬁrms can further exacerbate
transaction costs by operating in the presence of bounded rationality and uncertainty.

Mclvor (2000) states the main sources of uncertainty in transactions are centered
on the high levels of ambiguity found in the performance metrics. The presence of this
type of uncertainty implies the necessity of integrating safeguards or performance metrics
for all parties to adhere to. Transaction-cost uncertainty tries to protect against
opportunistic use of asymmetrical knowledge and information. These types of
uncertainties are of particular interest when knowledge and information can be used to
the detriment or loss of one party (V emimmen et al, 2000). It is not surprising that ﬁrms
will often choose to internalize these types of uncertainties for mitigation purposes
(Burke, 1998).

The amount of uncertainty that a ﬁrm faces has the potential to affect the
transaction costs associated with the different governance structmes. This uncertainty is
directly dependent on the speciﬁc business environment associated with the ﬁrm. Klass
et al (1999) found that ﬁrms reporting to be manufacturers operated in relatively stable
business environments and faced very little uncertainty. Manufacturers have and exercise
more control, thus controlling more factors of uncertainty. In contrast, ﬁrms in the
information systems industry reported they faced highly uncertain and unstable

environments due to the ongoing changes in technology requirements.

28

TCE theory states that ﬁrms facing high levels of uncertainty are more likely to
make ﬂequent labor adjustments and contracts that are not comprehensive in order to
remain ﬂexible when markets change. Firms facing stafﬁng and labor decreases caused
by outsourcing tend to forgo training and hiring investments to keep costs at a minimum
Firms outsourcing due to high levels of uncertainty will continue to incur costs associated
with terminating management and employees. F irms that are suppliers of products and
services (outsourcing ﬁrms) can more quickly adapt to changing labor needs versus ﬁrms
that make products in-house (Klass et al, 1999).

Inﬂequency as a transaction cost, according to Mclvor (2000) is one that is
seldom undertaken, produced or incurred The difﬁculties associated with inﬂequency
are bound to the number of transactions. With relatively low asset speciﬁcity and with
uncertainty combined with transactions that are fairly ﬂequent, these types of combined
transactions will be governed by market structures.

According to the TCE literature, retail grocery stores will outsource when the
total costs of production and all transaction costs are lower than producing ﬂesh beef and
pork products in-house. These same results were seen in Walker and Weber’s (1984)
seminal work using structural equation modeling (SEM) of predictors of make-or-buy
decisions.

TCE relaxes the requirement of perfect or full information assumptions found in
neoclassical economic theory and posits that many business transactions are deﬁned by
their incomplete and uncertain information asymmetries. Opportunism by suppliers also
includes asymmetrical information that will distort vital information ﬂom reaching

parties wishing to deal with suppliers. Asymmetrical information is present when

29

although some information exists for both buyers and suppliers; additional information
exists that is available only to suppliers. The asymmetry comes to ﬂuition when the one-
sided information is available only to certain parties, and they exploit that condition or
situation

Opportunistic use of asymmetrical information occurs in two ways. First is ex
ante, where information is hidden or not disclosed prior to a transaction. Examples of ex
ante occur when suppliers know of problems, but buyers cannot distinguish the problem,
and thus they adversely select the problem product or one-sided conditions. Contract law
and stipulations are used to mitigate this type of behavior. Ex post informational
asymmetries take place after the tranSaction has been completed and because of hidden
actions or characteristics of the product. An example of ex post asymmetrical
information occurs when suppliers knowingly conceal information concerning product
defects, any product recalls or discounts.

Mclvor (2000) aptly summarizes TCE as the fundamental theoretical foundation
for purchasing/buying departments to use to systematically analyze the relevant factors,
other than cost, to determine the boundaries of the ﬁrm and the efﬁcient form the ﬁrm
will take.

In the past TCE theory has been used as descriptive theory to help predict choices
made by fums about which governance structure should be chosen. TCE posits that
when ﬁrms outsource they are depending on market contracting as the chosen governance
structure. When ﬁrms depend on making the product internally they are relying on
organizational hierarchy as the preferred governance structure. Either way ﬁrms will

attempt to choose the governance structure that will minimize their total transaction costs.

30

Reducing transaction costs along the supply chain is fundamental to improving
efﬁciency, net income, and gaining competitive advantage. Reduced transaction costs
occur when suppliers and buyers form strategic alliances for cooperating and exchanging
important supply chain information.

TCE, with its focus on the boundaries of a ﬁrm and governance structures,
provides a plausible and helpful conceptual framework for examining the outsourcing of
ﬂesh meat products to an external processor. Outsourcing ﬂesh meat production
transaction costs include high uncertainty levels, unique asset speciﬁcity of producing
ﬂesh meat, technological change, and opportunities to develop economies of scale

production by the processor which can also beneﬁt the grocery retailer.

2.5 Strategic Management Approach to Outsourcing

Another theoretical approach to outsourcing is the strategic management method.
While this approach incorporates components of the cost approach into its methodology,
the strategic implications of outsourcing decisions are also considered (Canez et.al.,
2000; Prahalad, 1990; Porter, 1981; Probert, 1997; Welch, 1992; Venkatesan, 1992; and
Winkleman, 1999). Considering signiﬁcant changes in any business model requires
Operators to understand how those changes may affect the overall ﬁrm’s strategy, ﬂom
both the short and long term perspectives.

Porter (1981) views strategy or the strategic approach to management as
“management’s need to transform the daily chaos of events and decisions into an orderly
way of sizing up the ﬁrm’s position in the environment.” His method of transforming or
improving the ﬁrm is accomplished by implementing strategies that exploit a ﬁrm’s

strengths that are derived ﬂom internal opportunities, while accounting for the external

31

factors facing the ﬁrm. The strategic management literature stresses that the ﬁrm’s
endeavors should create customer value and competitive advantage. Porter (1980 and
1985) directly linked competitive advantage to a ﬁrm’s set of core competencies (Barney,
1991). The transaction costs theory in this thesis research project includes the
investments required to keep up with processing ﬂesh beef and pork (asset speciﬁcity),
the ability to develop economies of scale in case-ready meat production, and the levels of
uncertainty as they pertain to food safety and liability issues. These are all reasons why
retailers would consider outsourcing the production of ﬂesh meats and the purchase of
case-ready products.

Prahalad and Harnel’s (1990) Harvard Business Review (HBR) article
synthesized a new way of thinking about strategy that centered on the importance of a
ﬁrm’s set of core competencies. Prior to this article, most strategy debates revolved
around increasing market share and vertically integrating the ﬁrm. Prahalad and Hamel
(1990) deﬁne core competencies as combinations of production capabilities, skills and
technologies inherent to a ﬁrm’s product line. More precisely, they are a ﬁrm’s set of
core competencies which allows them to produce products that are technologically
advanced and superior to the competition. The classic examples of strategic core
competencies cited in their HBR article include Honda’s engine and transmission, Sony’s
miniaturization, Canon’s imaging/optics/micro processing.

Prahalad and Hamel’s (1990) work fully articulated the strategy behind
outsourcing. Their premise is built upon a ﬁrm’s ability to compete under tremendous
global competition. A growth strategy under these new circumstances will center on a

ﬁrm’s ability to “exploit” their core competencies and outsource all non-core activities

32

(V enkatesan, 1992; Mclvor, 1997; Quinn, 1994; and Kakabadse, 2000). It is postulated
that a growth strategy under these new circumstances will in reality fundamentally
change the way the ﬁrm does business, and ultimately lead to a competitive advantage in
the marketplace. It should be noted the strategy management scholars view outsourcing
decisions should be made at the senior management level, not ﬂom purchasing agents,
because of these types of decisions can fundamentally change a ﬁrm’s business model.
Therefore, these decisions are “strategic” in nature not tactical.

The strategy side of outsourcing is based upon the conventional wisdom that all
activities that are core to the ﬁrm should be maintained in-house and to outsource the
non-core activities. The literature consistently mentioned the strategy behind
successfully outsourcing non-core items was designed to allow ﬁrms the time and
resources to properly focus upon on the core business activities within the business
model, as a strategy to stay more competitive (Barney, 1990 and 1991; Bettis, 1992;
Quinn and Hilmer, 1994; Porter, 1980, 1981 and 1985; Welch and Nayak, 1992;
Venkatesan, 1992; Mclvor,l997; Mclvor, 2000; Probert, 1997; Lacity, I995;
Schoemaker, 1992; Alexander, 1996; Heikkila, 2002 and Kakabadse, 2005). Kakabadse

(2005) called the focus on a core strategy a sort of “de-integration” of the ﬁrm.

2.6 Core Competencies

According to Prahalad and Hamel (1990), core competencies are a function of a
fum’s intangible assets that deliver unique value to its customers, and cannot be easily
reproduced by the competition. There are three tests for identifying core competencies:
1) the core competence provides potential access to a wide variety of ’ markets; 2) core

competence should add noticeable beneﬁts to the ﬁnal product; and, 3) core

33

competencies should be difficult for a competitor to duplicate. Prahalad and Hamel are
adamant in connecting a ﬁrm’s set of core competencies with their long term ability to
continually produce new products, compete proﬁtably over the long term, and achieve a
lasting competitive advantage in the marketplace.

Prahalad and Hamel (1990) further explain that core competencies centered on
adding beneﬁts are usually the byproduct of combining new technology, new inputs and
production skills to improve a ﬁrm’s product offerings. They also note that core
competencies are not physical assets, because in time these assets can be readily
replicated by the competition and over time become obsolete (Mclvor, 2000 and Fleury,
2003)

Lonsdale and Cox (1997) clarify that core competencies should not be deﬁned
simply as those things the ﬁrm does “well.” Hodges (2003) puts it rather succinctly; “if
the process or activity didn’t create competitive advantage or a clear market
differentiation, it is likely not a core competency.”

Alexander and Young (1996) have deﬁned core competencies as the core
activities a ﬁrm should not outsource. These are: activities traditionally performed
internally with long standing precedent; activities critical to business performance;
activities creating current or potential competitive advantage, and activities that will drive
the future growth, innovation or rejuvenation of the ﬁrm.

Venkatesan (1992) does not refer to the terms core competencies in the decision
criteria for outsourcing. Instead he maintains the decision to outsource products and
services should be in accordance with three simple principles: focus on those components

that are critical to the product and what the ﬁrm is good at producing; outsource

34

components where suppliers maintain the competitive advantage, and possess scale and
cost econonries, and strong performance incentives; use outsourcing as a means to
generate employee commitment towards improving product and manufacturing
performance. Venkatesan’s viewpoints about outsourcing are a close ﬁt with the
subtleties of the grocery industry and with the outsourcing of ﬂesh meat products.

Schoemaker’s (1992) deﬁnition of core competencies is that they must be unique,
important, controllable, durable and able to generate excess proﬁts. Schoemaker was
adamant about the outsourcing decisions must produce short term. proﬁts. Jennings
(1992) points out that ownership of core competencies is second to having or possessing
the beneﬁts associated with your products and the integration must appear to be seamless
to customers, a more long term approach. Traditionally these items have been considered
“core” to a retail grocery store: fresh meat processed in-stores, ﬂesh produce and ﬂesh
bakery items. Fresh produce and bakery items are now largely outsourced or exist as
high margin niche products. Red meat is one of the few remaining grocery items still
processed in stores.

Mclvor (1997) suggests deﬁning what is core to a ﬁrm ﬂom the perspective of not
only the customers, but the market and industry. This approach is consistent with the
strategic management and the supply chain management literature’s value chain approach
to earning competitive advantage. The value chain approach of core competencies was
discussed in Quinn and-Hilmer (1994), where Ford Motor Company discovered tier one
and tier two level suppliers possessed greater product knowledge, technology levels,
speciﬁc research/development investments and speciﬁc human resources capable of

producing products and services than did Ford. "Therefore, Ford learned to partner with

35

suppliers. The supply chain management and logistics literature refer to these value
chain improvements as “best in class”. As the value chain contributes best in class
products/services/components, to the ﬁnished product the ﬁrm must learn to manage and
become best in class at their other core competencies, not just outsourcing.

Torkkeli and Tuominen (2002) developed a simple cognitive map (Figure 1) of
the importance of core competencies towards adding value to the ﬁrm’s end products.
The casual relationship should not focus on every competence resulting in a competitive
advantage and every competitive advantage being related to core competencies. The map
simply means the core competence should lead the ﬁrm to perform better in the
marketplace than its competitors (competitive advantage), and the competitive advantage
should lead to superior products as deﬁned by the ﬁrrn’s customers (added value), with
the net result yielding improved customer satisfaction with the ﬁrrn’s products/services.

Figure 1 Cognitive Map of Core Competencies

Core Competence ——> Competitive Advantage —> Added Value

Source: Torkkeli, 2002

Torkkeli and Tuominen (2002) have also added to the literature by developing a
multi-step criteria that core competencies should adhere to if a ﬁrm is to successfully
incorporate the competencies into a ﬁrm’s strategy. Core competencies must remain
essential to corporate survival in both the short and long term, be invisible or seamless to
competitors and be difficult to imitate, along with a combination of skills, resources and

processes. Core competencies are a capability by which the ﬁrm can sustain its business

36

model over time. They are essential to the implementation of the strategic vision of the

ﬁrm.

2.7 The World is Flat Method of Outsourcing

Another approach to outsourcing that is relevant for this thesis is ﬂom the
perspective of Thomas Friedman’s critically acclaimed book The World is F (at: A Brief
History of the T wenty-F irst Century (2005). Friedman made his observations about
outsourcing while traveling in India when he realized the global playing ﬁeld has been
leveled or “ﬂattened”. As the author visited several high tech research centers, he
concluded the “earth was really ﬂat” because new digital technologies have enabled ﬁrms
ﬂom around the world to gain access to global markets and a global workforce. It then
occurred to Friedman that this phenomenon is not only taking place in India, but
throughout the whole world. These new breakthroughs have created a great many
opportunities for ﬁrms to collaborate and compete in real time with any ﬁrm, “any place”
in the world, on equal footing, competing for the production of goods and services.
Much of the knowledge and production pools ﬂom around the world are now connected
through new digital technologies and ready to be used by almost anyone.

According to Friedman the world is now ﬂat, because of ten events that occurred
in the 19905 and culminated in about 2000. The ﬁrst two “ﬂatteners” are historical in
. nature, beginning with the monumental collapse of the Berlin wall on November 9, 1989.
When the wall was brought down it represented not only the victory of ﬂeedom over
Oppression, it also signaled a win for Capitalism over Communism. This event is
remarkable for regions of the world. namely USSR, Eastern Europe and China that have

rarely experienced market and personal ﬂeedoms.

37

Another historical event was the day Marc Andreessen’s Netscape went public on
August 9, 1995. This signaled the exact day the world became truly connected with each
other, and ﬁrms ﬂom around the world could now become transformed into a global
community. This day also is seen as the beginning of the world wide “dot com” boom
where billions of dollars of investment capital were seeking to invest into this new
economy. The new economy was soon referred to as the intemet economy, and large
capital investments were shortly ﬂowing into global communication inﬂastructure and
networks anticipating high returns. Instead these new investments yielded unusually high
stock prices like Netscape was enjoying. In reality this environment proved to be very
volatile for the new intemet economy, because there was a lot of investment capital
seeking too few functioning business models. While many of the early dot com ﬁrms
went bankrupt and some merged with other ﬁrms, in their wake was left a very
sophisticated and complex communication inﬁastructure and networks.

A third “ﬂattener” is the efﬁcient use of the newly developed inﬂastructure and
networks to maximize work ﬂow within ﬁrms. Work ﬂowing in this new infrastructure
became “interoperable” with many other systems. An example in Friedman’s book is
where the marketing department is connected with key suppliers, the suppliers are
connected with their manufacturers, and the manufacturers are connected with their input
suppliers. When the work ﬂows are interoperable in an integrated supply chain it has a
seamless interconnected communication system allowing information to ﬂow where it
needs to ﬂow, uninterrupted. He describes the workﬂow before it became seamless as
the marketing department wOuld run Microsoft software and the inventory department

would run Novell, information did not ﬂow as easily as it could. The seamless travel of

38

information or work ﬂow required a new data description language coupled with new
intemet language standards. New software developments designed to improve work ﬂow
began to ﬂourish under the strictly enforced language standards.

Open source code was Friedman’s fourth “ﬂattener”. Open source codes were not
developed nor maintained from the protected position of Intellectual Property (IP) rights
but, rather ﬂom a community perspective. The open source movement is a collaboration
of software writers that begin their assembly from the bottom up versus a top down
approach. The top down approach is standard when hierarchies are directed to be
incorporated. Open source codes are the underlying programming language or
instructions that allow software applications and different computer systems and devices
to not only communicate with each other but to display web pages, exchange emails all in
a standard form. While all IP software developers guard their source code ﬂom any
duplication, open source code developers readily make theirs available online for
anyone’s use, with no charge. This new model challenged software and hardware
developers to be creative and to bring their new products to market cheaper and quicker
than past IP protected business models.

Number ﬁve on the list is the outsourcing of various products and services to
global vendors and suppliers. This ﬂattener is arguably Friedman’s least developed, but
combined with other ﬂatteners he makes a compelling case for outsourcing as another
cause for the earth being ﬂat. Friedman postulated that the ability to move certain
processes and services offshore has created a worldwide demand for quality employees

that can produce high quality products and services cheaply. Friedman’s book develops

39

several outsourcing examples which are decidedly global in nature and stress the
importance of developing collaborations, versus outsourcing for lowering costs.

Number six is offshoring, which is decidedly different than outsourcing products
and services to ﬁrms in other countries. Offshoring is moving the entire production
facilities offshore, anticipating generous savings for labor and inputs. Friedman notes
many ﬁrms will relocate offshore to save on production costs and to also build a
marketing presence in the new host country. New offshoring business models that
include (1) the ability to maintain market share in the United States, (2) lower production
costs signiﬁcantly and (3) enter new foreign markets where competition may be
nonexistent can provide a formidable and often proﬁtable strategy.

Supply chaining is number seven in Friedman’s list of “ﬂatteners”. He says
supply chain management practices have revolutionized a ﬁrm’s ability to efﬁciently
move and monitor product inventory as it ﬂows through a supply chain. Friedman’s view
of supply chain management is related to the practice’s ability to gain efﬁciencies and
improve production. This view is also consistent with the supply chain management
literature. His book cites Walmart’s use of the latest Radio Frequency Identiﬁcation
(RFID) technology to track pallets as they move ﬂom the manufacturer’s plant to the
Distribution Center (DC) and ultimately into retail stores. Walmart has also utilized the
RF 1D technologies to improve product demand modeling throughout the entire chain and
to pass that information downstream to the distribution center and then back to the
manufacturer.

Insourcing is Friedman’s number eight “world ﬂattener”. Insourcing is deﬁned

by contrasting the vertically integrated ﬁrm with ﬁrms that contract with outside ﬁrms to

40

outsource any number of production and service functions. Typical examples cited are
(1) the supply chain management and logistics management used by large ﬁrms that are
performed within the ﬁrm as opposed (2) to smaller ﬁrms employing a tactic of hiring
outside ﬁrms to work inside the ﬁrm to gain efﬁciencies. A key point is that in this case
outside ﬁrms work with the ﬁrm. Friedman relates insourcing to creating value within
the ﬁrm by collaborating horizontally, not vertically. The insourcing relationship is about
yielding control and information through relationships, and developing each other’s
strengths and efﬁciencies. These relationships do not resemble the standard vendor-
customer relationships of the past, which were adversarial by design.

“Flattener” number nine is termed “In-Forming”, which relates to the
development of various search engines such as Yahoo, Google, Safari, and MSN which
were all created for the public’s use and designed to be destinations for gathering
information. In the early days of Yahoo, users were surprised at the amount of relevant
information that was available just be entering a few search terms and then wait as Yahoo
searched the entire web. Today we expect to ﬁnd what we are looking for, instantly and
not be awed with our new found results, compared to the early days of searching on
Yahoo when people were in awe when the engine produced the desired results, time after
time. One of the founders of Google, Sergey Brin, gives credit to their early success to
the fact that people have a pent up demand for information, a desire to be “in-formed.”
These search engines ﬂattened the world by allowing everyone to be connected without
regard to geography, and competition amongst ﬁrms was not based upon asymmetrical
information monopolies found in the marketplace before search engines provided almost

perfect information.

41

The last “ﬂatteners” are the “steroids,” which are any devices that help or enhance
a ﬁrm’s ability to communicate and better compete in the global arena of products and
services. Friedman states there are several factors that have allowed the communication
steroids to ﬂourish. They include the almost daily improvements in: computational and
storage capability, along with the new universal wireless technology. All of these
advancements, matched with the changes in hardware technology, have dramatically
improved our ability to communicate professionally and personally. The world becomes
“very ﬂat” very quickly when all of the “ﬂatteners” start to get “turbocharged” by all of
the steroids.

Friedman hypothesizes a “ﬂat” world creates useful and creative solutions when
there is convergence of the ten or more ﬂatteners as they work together with the goal of
sharing of knowledge and work processes, in real time. This convergence creates a new
playing ﬁeld, one that is “flat”. When the world goes “ﬂat”, all businesses should ask .
how they will adapt, how will the walls come down to Operate differently? All ﬁrms
working in a ﬂat world will experience the reality of new and different partners
possessing different ideas and new processes, along with new habits for building
horizontal collaborations.

Friedman challenges outsourcing relationships to take on a new context, one that
is organized horizontally. Horizontal collaborations are decidedly different ﬂom the
traditional vertical relationship in outsourcing; they possess an almost organic
atmosphere. Horizontal in Friedman’s vernacular refers to the multiple forms of
collaboration, and power is not wielded by size or IP rights. A necessary ingredient in

two way collaborations is when power is taken out of the equation. When there is a true

42

collaboration in a ﬂat world outsourcing relationship, all parties are destined to work
together and share information. This arrangement helps ﬁrms become more efﬁcient and
traditional hierarchies naturally will become ﬂatter. These types of collaboration allow
small ﬁrms to act big, big players are allowed to act small and everyone has an incentive
to work on their own and contribute to the collaboration.

Outsourcing in a ﬂat world morphs the inner strengths of the collaboration not to
build walls with supply chain partners, but to ﬂatten them and increase one’s ability to
add value. Like all business relationship strains will develop between partners, and it is
suggested ﬁrms should ﬁrst reﬂect or look inward and then outward for solutions.
History has shown hyper competition can turn value added products into commodities.
The ﬂat world suggests a way of dealing with increased competition should be through
the development of collaborations and alliances. Flat world outsourcing means
collaborating faster, and with your partners in order to grow and gain market share.

Friedman ﬁrrther states that ﬁrms should not outsource in a ﬂat world to solely
reduce costs, because cutting costs alone creates relationships built upon one partner
making less money to keep the relationship viable. A one-sided outcome is the opposite
of a win-win relationship. More importantly, simply lowering costs takes the focus away
ﬂom the value consumer’s purchase and may hinder any new product innovation.
Product innovation thrives in collaborations and strategic alliances where information is
readily provided to improve product quality and production processes. A collaboration
mindset is a necessary prerequisite for ﬁrms to become ﬂat and interested in developing

new markets where customers pull the product through the supply chain versus

43

manufacturers/processors pushing the product through the system. Firms should instead

outsource to grow the ﬁrm, satisfy their customers and improve their business model.

2.7.1 Implications of Outsourcing Methodology for Case-Ready Meats

Friedman’s outsourcing methodology has several signiﬁcant implications for
retailers contemplating outsourcing their case-ready meat products. Here are the relevant
sections of Friedman’s outsourcing methodology as they relate to case-ready meats: (l)
The advent of the net and its ability to link processors with retailers was key towards both
ﬁrms maximizing the work ﬂow between ﬁrms. Before these advancements were
implemented and accepted successful case-ready meat programs were problematic,
because there simply was too much data and information required between ﬁrms.
Conventional methods of sharing data were too inefﬁcient. (2) Case-ready meats require
specialty processing/packaging equipment and materials; the intemet facilitates the
interaction between supply chain partners and the ability to stay current with new
offerings being introduced around the world. (3) Supply chaining is a vital ingredient for
case-ready meat products to move seamlessly ﬂom beef cattle producers to consumer’s
reﬂigerators. Supply chaining perceives outsourcing as “pulling” product through a
supply chain versus, pushing it through. Pushing product through system begins and
ends with the end user and connects all the relevant partners and stops of the supply chain
together. (4) The ﬁnal components for Freidman’s outsourcing methodology are the
steroids. Retailers must manage inventory moving through their supply chain and the use
of wireless technologies for ordering and managing inventories are a critical component
for successfully adopting a case-ready meat outsourcing strategy. Without wireless

technologies the speed and accuracy of inventory management breaks down, making an

44

outsourcing strategy for case-ready meats nearly impossible. The steroids allow meat
department personnel the necessary time to ﬁnally market these new products and make

more connections and relationships with customers.

2.8 Concluding Remarks for Outsourcing

There is substantial evidence that the centrally processed and packaged case-ready
beef is a complete product: no more processing or labor needs to be applied, and it is
ready for the consumer. Another important point is true as well; grocery stores cannot
easily and efficiently replicate the necessary HACCP inﬂastructure for processing and
packaging in each and every retail grocery store.

Processors have developed case-ready beef products that are safe and non-labor
intensive product. When retailers outsource ﬂesh beef products and adopt case-ready
beef products they are now able to concentrate on marketing products, building customer
relationships and improving their business model. Retailers outsourcing meat production
in a ﬂat world can build relationships with and unite beef cattle producers, meat
processors and retailers together with one common goal of providing an excellent center

of the plate protein for consumers around the world.

45

CHAPTER 3— RESEARCH DESIGN: STRATEGIES, PROPOSITIONS AND
METHODOLOGIES

3.1 Overview of Research Design

This chapter discusses the strategies and methodologies used to determine those
critical factors that affect retailers’ decision-making regarding outsourcing conventional
in-store meat cutting and packaging to case-ready beef programs offered by meat
processors. Choosing the correct research design must be appropriate for the topic of this
thesis project, and a central theme and goal of this chapter. Section 3.2 describes the
range of research categories available to the researcher; 3.3 describes the process of
choosing the most appropriate strategy. Section 3.4 describes the merits of the case study
as a research strategy; and section 3.5 discusses the use of case studies in this research.
Sections 3.6 through 3.8 explain the strengths and weaknesses of the research design and
the reasons for selecting the case study approach. Section 3.9 will explain how the case
study was used in this research project.

Finally, section 3.10 presents the six basic research questions and twenty-four
research propositions. The last two sections discusses the analysis of the supply chain for
case-ready beef; and an explanation of the questionnaire and how it was used to survey a

sample of retail store managers of the Save-A- Lot company.

3.2 Research Categories

Scientiﬁc research can be separated into ﬁve basic categories: ﬁeld experiments,
laboratory experiments, ﬁeld studies, case studies and survey research (Yin, 2003,
Churchill, 1999, Cook and Campbell, 1979 and Kerlinger, 1986). Each category

possesses differing logic, techniques of data collection and ways to analyze the data; and

46

each has its own sets of advantages and disadvantages (Kerlinger, 1986). These ﬁve
research categories are also associated with experimental or non-experimental research;
or whether the data is collected in the ﬁeld or in a laboratory setting (Cook and Campbell,
1979)

Experimental research is recognized by a simple causal hypothesis: if X then Y. It
is designed to measure or manipulate X and record the results of Y. In Marketing and
Supply Chain Management disciplines, experimental research is referred to as causal
research, where the research design is concerned with determining cause-and-eﬂect
relationships (Churchill, 1999).

At a minimum, all experimental research involves a treatment, a unit of analysis,
an outcome. measurement, random assignment/selection", and some form of inferring
outcomes associated with the treatment. The researcher manipulates the controls of one
or more variables, called a treatment, and then assesses their effects on the dependent
variable (Cook and Campbell, 1979). This research technique has traditionally been
reserved for use in the “hard sciences” because of its ability to replicate the experiment
for veriﬁcation purposes. Recently, market researchers have begun to use experimental
design, especially in the advertising discipline (Churchill, 1999; Hair er a1, 1998).

Non-experimental research takes place when the researcher does not have control
of the independent variables because direct control is not possible and random
assignment is not an option. Kerlinger notes that non-experimental research has three

glaring weaknesses: the incapability of incorporating independent control variables; the

 

6 Without random assignment or selection the experiment is a “quasi-experiment.”

47

incapability of randomizing the variables; and a substantial risk of improper
interpretation of the results (Kerlinger, 1976).

The most important differentiation between experimental research and non-
experimental research is the ability to incorporate independent control variables into the
research. Control variables are used to screen out unconnected forces that may infer
specious inferences and is also related to the ability with which units receive treatments at
a particular time. Control is related not only to the independent variables, but also to the
research setting, where levels of treatment can be strictly applied and used in concert with
other control variables. Therefore, control is used to mitigate issues of validity (Cook
and Campbell, 1979 and Kerlinger, 1986).

Kerlinger (1976) deﬁnes ﬁeld experiments “as a research study in a realistic
situation in which one or more independent variables are manipulated by the
experimenter under as carefully controlled conditions as the situation will permit.” The
difference between laboratory experiments and ﬁeld experiments is relegated to control;
experiments enjoy a greater degree of control than ﬁeld experiments.

The weaknesses of ﬁeld experiments are related to issues of control. They do not
have the same level of control as laboratory experiments. While independent variables
can be controlled, the reality of extraneous effects found in the ﬁeld can affect the
research environment sufﬁciently to question the results; and it is extremely unlikely that
conditions in the ﬁeld can be adequately replicated as they can in laboratory experiments
(Kerlinger, 1976).

The strengths of ﬁeld experiments are their ability to study large processes,

complex issues, and settings that cannot be replicated in a laboratory. Field experiments

48

are ideally suited to research questions about phenomena and theory, which beneﬁt from
the greater levels of ﬂexibility (Kerlinger, 1976).

Kerlinger (1976) deﬁnes laboratory experiments “as research study in which the
variance of all or nearly all of the possible inﬂuential independent variables not pertinent
to the immediate problem of the investigation is kept at a minimum. This is done by
isolating the research in a controlled physical situation and by manipulating one or more
independent variables under rigorously specified, operationalized, and controlled
conditions.” Kerlinger’s deﬁnition clearly implies that lab experiments have a high
degree of control of the conditions and variables.

Field studies are non-experimental research that focuses on the interactions of
values, perceptions, behaviors, outcomes, attitudes, and relationships between dependent
and independent variables. Katz and Kahn (197 8) postulate that the majority of ﬁeld
studies are centered on hypothesis testing and exploratory research. Exploratory research
endeavors to study and explain relationships among variables, gain insights, establish
priorities for ﬁrture research, clarify concepts, and develop hypotheses and testing (Cook
and Campbell, 1986; Kerlinger, 1976; Churchill, 1999; and Selltiz, eta], 1976).

Case study research seeks to establish a ﬁamework for researching complex
phenomena. While this method allows a researcher to retain the important characteristics
and the real-life events related to researching a phenomenon, some argue that laboratory
and ﬁeld research are more rigorous in advancing knowledge (Yin, 2003 and Churchill,
1986)

Survey research, the most common method of collecting primary data in the

social sciences, usually relies on the use of questionnaires as the instrument for collecting

49

primary data (Churchill, 1999 and Hair et al, 1999). Surveys are conducted using
differing methods depending upon the research budget, target audience, and research
questions. Survey techniques have several different methods available for use. They
include; telephone interviews, personal (face-to-face) interviews, self-administered
questionnaires, or a combination of methods. Personal interviews have a reputation of as
one of the most effective ways of obtaining detailed/in-depth demographic information
and it allows for posing extensive and probing questions. A self-adrninistered mail
survey is the most cost effective survey method (Converse and Presser, 1986).

Survey research is considered to be an effective technique in obtaining a wide
variety of data and is viewed as the most economical of all collection techniques when all
factors associated with the research project are taken into account (Kerlinger, 1976 and
Churchill, 1999). The primary data collected using this technique can be accurate and
within the sample error if properly drawn samples are used.

There are several criticisms associated with using survey research techniques as a
means of gathering primary data, they include; a lack of depth in the information
gathered and low response rates. Another disadvantage of survey research is having to

sacriﬁce the ability of posing follow-up questions.

3.3 Selecting the Appr0priate Research Strategy

Research design literature suggests the type of research strategy chosen should
depend on the research questions to be asked; the amount of control the researcher
possesses or has the ability to maintain over behavioral events to be studied, and ﬁnally,

the focus or interaction amongst the contemporary events of the research project. Yin

50

(2003) developed a matrix to compare and contrast the differing research strategy
choices, summarized in Table 3.

Table 3: Research Strategies Choice Matrix

 

Strategy Form of Research Requires Control Focuses on
Questions Over Behavioral Contemporary
Events? Events?
Experiments How, Why Yes Yes
Surveys/Questionnaires Who, What, Where, No Yes
How Many, How
Much
Archival Analysis Who, What, Where, No Yes/No
How Many, How
Much
History How, Why No No
Case Study How, Why No Yes

Source: Yin, 2003

Yin (2003) notes a common misunderstanding researchers encounter is the
different research strategies should be arranged in a hierarchical order. Yin’s notion of
hierarchical order for research strategies surmises that research strategies should follow a
prescribed order or “hierarchy”, where this order negates any beneﬁts from not following
the prescribed order based upon what phase the research was in, either exploratory,
descriptive, or casual (explanatory). Yin suggests a more suitable vision of the different
strategies would be a “pluralistic one”, where each strategy could be used in all three
phases—exploratory, descriptive, or causal or explanatory or portions of each (Yin,
2003)

Following Yin’s lead this research project began with an exploratory phase which
included a series of interviews and observations7 which were conducted with meat
industry executives representing both fresh meat processors and a Limited Assortment

retailer. After the exploratory phase the research project began the explanatory phase of

 

7 In the observation phase the research focused on fresh meat products moving through the retailer’s supply
chain, beginning at the retail level where personnel took an inventory of fresh meat products in storage and
displayed in meat cases.

51

the project by describing the results of the interviews and observations in a questionnaire.
The questionnaire was the primary means of collecting data to be analyzed and

interpreted by the project.

3.4 Case Study Research Strategy

Case studies are appropriate for exploratory research projects that ask “how” and
“why” questions about phenomena where the setting is difficult to recreate outside its
natural setting and the variables are difficult to quantify. Case studies are also a useful
tool when used for the development of preliminary tools to be used in more structured
analysis such as surveys and experiments (Eisenhardt, 1989), (Chetty, 1996), (Rowley,
2002). Yin, (2003) also suggests the constructs and variables may be too numerous to
quantify and account for, making experiments inappropriate (Ghauri, 2002). Eisenhardt,
(1989) states case study research can be used in theory building. ,

Many research studies of phenomena occurring in the business arena cannot be
undertaken if they are removed from their real-life environment; as it is not feasible to
attempt to recreate them in an artiﬁcial setting. Eisenhardt (1988) states the boundaries
between the phenomena and the environment in which they occur are not clear. In
particular a case study approach should be considered when the purpose of the study is to
gain a better understanding of the dynamics surrounding a phenomena within a unique
setting and in which multiple sources of evidence are used (Yin, 2003) and (Eisenhardt,
1988). An experimental research strategy would deliberately separate phenomena from
its context or surroundings in order to highlight a few variables under controlled

circumstances (Yen, 2003), (Churchill, 1999), and (Ghauri, 2002).

52

3.5 Appropriate Uses of the Case Study Research Strategy

Yin (2003), Churchill (1999) and Ghauri (2002) postulate the use of case-study
research as technique to begin to build theory. The literature is replete with examples of
research that yield generalizable, universal claims gained from case study research that
are theory building projects. Burgleman (1983), Mintzberg & Hughes (1985), Eisenhardt
& Bourgeois (1988), and Pettigrew (1988) provide examples of research problems that
relate to strategic management. Burgleman (1983) studied the management of new
ventures, Mintzberg & McHugh (1985) looked into the strategy formulation, Eisenhardt
& Bourgeois (1988) researched a problem focused on strategic decision-making for ﬁrms
in high velocity environments (retail grocery stores are an example of high velocity) and
Pettigrew (1988) focused on the strategic problem of how to remain competitive and deal
with strategic change.

While the data sources for each projects mention above differed, they all
incorporated some use of archival information, observation techniques were also
employed and person to person interviews were also used Only Eisenhardt & Bourgeois
(1988) incorporated the use of questionnaires into their data collection (Eisenhardt,
1989)

Data analysis is at the heart of creating or building theory from case studies.
Mintzberg & McHugh undertook the examination of 2,500 movies, analyzing movie
revenues, the subject of each ﬁlm, ﬁlm sponsorship, and many other criteria. Mintzberg
& McHugh also used tabular graphs and displays about each case, a monumental
assortment of data transformed into useful information. Data analysis is required in order

to observe any unique patterns that may ﬂow form the data and the use of cross-

53

tabulation of the data set should permit generalizable patterns to emerge (Eisenhardt,
1989)

Eisenhardt (1989) identiﬁes three analytical tactics to employ in analyzing the
data set. One way is to identify group similarities coupled with the intergroup
differences. In this research project an example of this tactic would be to analyze all
store similarities and compare and contrast among the intergroup differences of corporate
and franchise stores. The goal of the ﬁrst step is to search for patterns and attempt to
match with intergroup or within group cross-tabulations. A second is to select pairs of
cases and then list the similarities and differences among and between pairs. This
method Eisenhardt (1989) states is to ferret out the subtle similarities and differences.
The second is a useful exercise in breaking up what may be simple observations into a
more sophisticated comprehension of the data.

A third approach is to divide the data by sources and observe any trends between
the different data sources. Again, the researcher should seek to identify any patterns,
cross-patterns, agreeing data, and disagreeing data. When disagreeing data is evident,
stating the obvious results is the beginning of further analysis and postulating trends.
Further analysis employing this tactic would be to separate the data into subgroups and
cross-tabulate the “mini-groups” among the others, looking for and exposing interesting

patterns Eisenhardt, (1989).

3 -6 Strengths of Building Theory from Case Studies
The likelihood of generating new theory from case studies is found in producing
novel theory from creative insights found within the data analysis. Creative insights are

revealed from the agreeing and contradictory results found in the data and reconciling the

54

 

results across observations and among differing types of analysis. New theory emerging
ﬁom this type of research can be tested with new constructs and hypotheses generated for
a second-generation research project. Another reason for ﬁrst employing the case study
technique is that the new theory emerging from the analysis is closely associated with the
consistent evidence found within the data analysis (Yin, 2003). Engaging in a case study
that focuses on a phenomenon occurring a business setting before collecting data with a
questionnaire has many beneﬁts, and those beneﬁts can add value to the entire project.
The beneﬁts include: a deeper and personal knowledge of the environment where the
phenomenon occurs, an opportunity to gain insights from the ﬁeld (not just from the
literature review), an opportunity to witness the phenomenon operating in a supply chain
setting and ﬁnally developing a survey instrument based upon the results of case study

and the literature review could produce a more robust instrument to analyze.

3.7 Weaknesses of Building Theory from Case Studies
The most signiﬁcant weakness associated with theory-building from case studies
is the overwhelming amount of data and the enticement of building theory that attempts
to encapsulate all of the data results. This weakness is caused in part by failure to
carefully assess the most important relationships from the redundant and/or superﬂuous

relationships (Eisenhardt, 1989).

Another weakness of building theory from case studies is that the research may
not be capable of generalizing the theory and/or the theory describes an idiosyncratic
phenomenon. Examples of moderate idiosyncratic theory-building are on Burgelman’s
( l 983) work, which produced a process model that linked multiple levels within

oI‘ganizations. Mintzberg & McHugh (1985) developed strategy-making themes for the

55

Canadian Film Board Eisenhardt & Bourgeois (1988) developed a mid-range theory
linking power, politics, and ﬁrm performance. Harris & Sutton (1986) conceptualized a
framework about the functions of parting ceremonies for displaced members. All of
these theories are novel, testable, and valid; they are theories about and endemic to

speciﬁc phenomenon and provide worthwhile understanding of theory building.

3.8 Case Study Research Design

A research design outlines the components of the research and the proposed plan
for examining and analyzing the data collected. A well thought out research design
problem directs the researcher to the following: the method of investigation, observations
to study, and how many observations should be made, suggestions towards which
respondents to target, how to ask the questions, suggestions towards which analytical
tools should be used, and likely outcomes. A research design is basically the structure or
framework developed to guide the collection of and analysis of the information gathered
during the collection phase. It can also be useﬁrl in the development of an entire plan for
linking and guiding the conceptual research problem into relevant and practical empirical
research.

Yin (2003) stipulates ﬁve essential components for designing a case study
research project. They include: a study’s research questions, its research propositions (if
any), its unit(s) of analysis, the logic linking the data to the propositions, and the criteria
for interpreting the ﬁndings. This thesis research project has followed the case study

rubric and methodologies espoused by Yin and Eisenhardt.

56

3.9 Case Study as the Preferred Research Technique
This thesis research project followed the case study rubric and methodologies espoused
by Yin and Eisenhardt in light of the fact that the case study research strategy focuses on
understanding a phenomenon and the dynamics present within a single non-replicable
setting when the boundaries between the phenomenon and context are not clearly evident
(Yin, 1993 and Eisenhardt, 1988). Case studies offer the researcher a unique and
rigorous method to measure and record the behavior of phenomena.

In light of the research project’s goals, objectives, and the numerous research
propositions, the use of case study research strategy for documenting the operational
challenges and concerns of case ready beef products constitutes a plausible use of the

technique.

3.10 Case Study Beginnings: Key Informant and Industry Executive Interviews

From its inception, this researcher was granted remarkable access to operators in
the meat industry with knowledge of and experience in both conventional and case-ready
fresh meat programs. Moreover, several Michigan State University faculty members
served as resources. Executives and faculty who participated in the study include the
Executive Vice President, Hormel Foods Corporation; Vice President, Cargill Meats
Solutions; Chairman of the Board, Alliance Foods; Senior Manager-Perishables, Save-A-
Lot Foods ; Perishable Managers Alliance, Foods; Distribution Center Manager, Save-A-
Lot Foods; Marketing and Supply Chain Faculty and Meat Science faculty. This group
of key informants provided invaluable insights about the issues facing the case-ready
meat industry, in a forum that was conducive for exchanging ideas and suggestions.

These individuals were of great beneﬁt to the case study project because of their

57

knowledge of the issues, industry, and consumers. This access to industry practice was
useful in successfully completing the qualitative study of meat industry practice and
decision-nmking processes. It provided insights which enabled the creation of a process
map of case-ready meat products that are distributed through the supply chain—from
meat processors, distribution centers and retail stores. This resource team conﬁrmed the
operational issues to be studied, and they provided advice and counsel in the development
of the questionnaire. Moreover, they provide valuable general industry knowledge for
the researcher. Executives from Save-A-Lot Corporation provided the researcher access
to retail store managers (respondents for the survey) located throughout the country, thus
assuring a high response rate in the survey used in the study.

This high level of collaboration among these supply chain members was thought
necessary given the unique biological characteristics of the fresh meat case-ready
products and the signiﬁcant new investments of the technology employed by all parties.
With unfettered access to key industry companies, the case study was able to effectively
study the case-ready meat phenomena in its distinctive environment.

The Cargill Meats Solutions executive explained that a key component of their
efforts to market case-ready meats to retailers was going to be based upon an exhaustive
Activity Based Costing (ABC) analysis showing that when all costs for case-ready meats
are taken into consideration, case-ready meat programs product will be acknowledged as
being “cost effective”. The ABC takes into account, among other things, the true cost
retailers incur when they experience out-of-stocks in the meat case as a result of the
difficulty in maintaining adequate production in the store’s meat department. Research

conducted by Cargill revealed that most grocery stores experience out-of-stocks during

58

peak shopping times of the day between (4PM and 7PM), which coincides with meat
department personnel cleaning up the processing and packaging area at the end of their
shift Therefore, Cargill was convinced the adoption of case-ready meats would
accelerate when retailers were able to take all of the costs into consideration.

Another cost factor for retailers to consider is the “shrink”. Shrink is deﬁned as
any dollar amount that is less than full-expected retail price for any and all fresh beef
products. Most retailers have difﬁculty in quantifying shrink because accounting for it
requires rigorous and time consuming accounting procedures. Indeed, merely counting
shrink creates additional costs when time is lost ﬁlling out the reports. Several of the
features and beneﬁts of case-ready meats help to reduce shrinkage and reduce overall
costs in the supply chain.

The Executive Vice President of Hormel Foods Corporation was supportive of
Cargill, pointing out, that case-ready programs potentially provide improved economic
performance by processors, retailers—and beneﬁts consumers, as well. Hormel is
constantly researching ways of identifying retail customers who can create a positive cash
ﬂow for Hormel and ﬁnding way to reward those customers, a win for Hormel and win
for the retailer. Hormel believes it is essential to promote the adoption of case-ready
programs by identifying companies that envision the mutuality of beneﬁts of connecting
the entire supply chain, including cattle producers, feeders, processors distribution
centers and retailers—as well as including the ultimate consumers in the value chain.

Since the supply chain for beef is not integrated from processors to retailers, as is

the case of the poultry industry, essential data moving forward and back though the

supply chain is lacking. In the absence of this data, Cargill is processing cattle according

59

to economic dermnd analysis techniques and Point of Sale (POS) data. Currently, daily
demand curves are analyzed that drive processing models for plants, and for cattle
purchasing to be processed. Although demand modeling is useful in calculating how
many beef cattle to process; it does not reﬂect consumer’s taste and preferences.
Demand modeling is an example of pushing product through a supply chain. It is not as
effective as marketing programs that are designed to more accurately meet the needs of
customers in the supply chain and marketplace. Marketing effectiveness requires
qualifying and quantifying all of the factors that shift demand. Firms that market
products programs are have greater opportunities to obtain greater proﬁts for the products
that are differentiated in the minds of the consumers.

The Hormel executive added that effective marketing requires retailers to become
more connected with its Shopper’s wants and needs; and sharing this information to
processors who can take this into account as it develops products and programs.
Information of this kind can be the basis for improving products, demand analysis and
more accurate processing information. There is a great need for this kind of
collaboration: it is the basis for creating true win-win relationships.

Hormel is interested in learning more about the decision-making process of
retailers which will provided an explanation of why case-ready meats are not being
adopted as rapidly as forecasted.

The Chairman of Alliance Foods is concerned that case-ready meats might meet
the needs of the larger chain stores (Wahnart and Kroger) but that the smaller chain

stores might be ignored if there is a wholesale shift to all stores converting to a case-

ready meat strategy. An example of this concern is related to the situation that an

60

average Kroger store orders $70,000 in fresh meat and a smaller store only orders
$15,000, yet both stores will need case-ready meat products, they have different volumes,
and different customers demanding different product at different times. When stores
process fresh meat, each store becomes an expert in what products their particular store
needs and when they need. it This reality may be causing smaller chain stores to resist
adopting case-ready meats and not to give up their autonomy in processing product when
they need it.

Alliance Foods executive suggested that with the more precise information
throughout the supply chain, processors could Direct Store Delivery (DSD) fresh case-
ready meats to individual stores, thus negating the cost for product to be warehouse at a
distribution center. This is how many large volume products are distributed to grocery
stores, thus reducing costs and improving efﬁciency. Such an approach to meat
distribution has the potential to be managed by the processor supplying the stores,
especially as the trust level between the supplier and retailer is fostered

Direct Store Delivery (DSD) of case-ready meats would have a minimum amount
of new infrastructure changes required to properly implement this strategy. Most stores
have sufﬁcient reﬁigerator space to accommodate current inventory levels, which would
negate additional any infrastructure investment. However, a potential obstacle would be
the need to train meat department personnel to manage this new business model. These

points eventually became a central theme for the questionnaire.

3.1] Research Questions
To develop a better understanding of the operational concerns that help or hinder

the successful adoption of case-ready beef programs into retail grocery stores, the project

61

developed six research questions to guide the research process. These questions were
developed based upon several informational components. These include (1) interviewing
key industry experts during a one day research seminar held at Michigan State
University, (2) by directly observing case-ready meat products as they moved through the
supply chain, and (3) by conducting in-depth interviews with retail store managers and
distribution center managers from SAL grocery store chain with over 1,150 stores
throughout the US.

From each research of the six research questions twenty-four testable research
propositions were developed to identify the operational issues associated with the
successful implementation of, or lack thereof of a case-ready meat strategy.

The SAL stores have two speciﬁc types of stores corporate stores and licensed
stores. Corporate stores are stores that are managed by the parent company. Licensed
stores are managed by individuals and corporations who have obtained a franchise. All
stores, regardless of ownership receive all of their grocery products from the parent
company via a SOphisticated distribution system.

The six main research questions are:

RQl What are the differences between corporate and licensed stores in their
perceptions of the operational concerns associated with case-ready products?

RQ; What differences exist between corporate and licensed8 stores in their perception
of the operational concerns associated with food safety?;

RQ3 What are the differences between corporate and licensed stores in their
perceptions of the operational concerns associated with workmanship?;

RQ4 What are the differences between corporate and licensed stores in their
perceptionsof the operational concerns associated with shrink?;

 

3 . . . .
Licensed stores are srrrular to franchise stores.

62

RQs What are the differences between corporate and licensed stores in their
perceptions of the operational concerns associated with packaging?;

RQg What are the differences between corporate and licensed stores in their
perceptions of the operational concerns associated with handling?;

The twenty-four research propositions listed below were derived ﬁom the six
operational concerns and underlie the survey instrument design. Each research
proposition purports to have this null hypothesis: 110;: There is no signiﬁcant difference
between corporate and licensed stores with respect to...

”lemma, mam, There are several key success factors that all ﬁ‘esh meat
products must possess in order to be considered viable in a retail grocery store format.
These attributes include: product appearance, shrink management, proﬁtability, shelf-life,
food safety, out-of-stocks, product consistency, SKU varieties, and product traceability.
Rch.....,.,...gy product. Managers assume that case-ready meats are more expensive. RP3Cm.
"My mam, Managers feel that case-ready meats are a more employee friendly program
than in-store processed meats. ”40,..de products Managers know how their customers
perceive fresh meats with respect to color, servings per package, and package price.
RP5(:,..,.......3y prod-ct- Managers would not have a problem training a traditional meat
department manager to manage all perishables, not just meats. RPmmmdy mam,
Managers have a poor perception of the relative costs between case-ready and non-case
ready products.

RP'II-‘ood “my Store managers believe case-ready beef and pork products are safer
than products processed in stores. RPgFood "my An outbreak of food borne illness
associated with fresh meat purchased at either corporate and franchise stores would be

catastrophic to store sales. Rngood “my Store managers’ perception is that customers

63

who do not recognize the safety of both products would respond by purchasing more
fresh beef and pork products if they believed those products were the safest available.

RPmmmmp Store managers are ﬁnding it more difficult to hire qualiﬁed meat
cutters at competitive wages. RPuw-ormmp- Implementing a case-ready meat program
will demand additional skills such as forecasting daily and weekly demand, along with
proper inventory management and less meat processing workmanship skills.
RanWP Case-ready beef and pork products compared to in-store produced
products are more cost effective. RP13w°rkmump Case-ready meats will have lower stock
outs and therefore maintain a higher rate of customer retention.

RPMSW Store managers do not believe there is a signiﬁcant difference in shrink
levels when selecting fresh meat product options. RPlssm Store managers do not
believe there is a signiﬁcant difference between fresh meat product options as they relate
to managing the other operational concerns and managing shrink levels. RPmsmk Store
managers are able to distinguish the top three sources of shrink. RPnsmk Store
managers do not believe there is a signiﬁcant difference between fresh meat product
options in formulating strategies for controlling and addressing shrink.

RPlgpmgmg Case-ready meat products that require “blooming” cause operational
issues as a result of preparing that product to be displayed in the meat case. RPlgp.¢k.ghg
Case-ready meat packing is considered “attractive.” Ronp.ck.g..g Store managers prefer
the case-ready beef that is packaged in low oxygen, low headspace. Rlep.chg.ng Store
managers believe the case-ready beef packaging alternative of high oxygen/high

headspace is a detriment to fresh meat sales.

64

M223...“ Store managers typically ﬁnd it difﬁcult to keep their ﬂesh meat
cases fully stocked between 4PM and 7PM and case-ready meats can resolve this
problem. RPZSHaldling Managers believe a key factor towards increasing the shelf life of
ﬂesh meats is holding and maintaining proper temperatures (e. g. rapidly transporting
ﬂesh meat products ﬂom the receiving area into the walk-in coolers in a timely manner)
furthermore case-ready meats are at a disadvantage in this area. RPunndung Case-ready
meats are easier to manage for proper rotation and to avoid out- of-date product and help
maintain pr0per adequate inventory levels.

The research questions and propositions were the basis for developing the survey

instrument, which is discussed later in this chapter.

3.12 Supply Chain Analysis of Case-Ready Meats

The ﬂesh meat supply chain case study component of this research was initiated
to docmnent the operational issues case ready meats face as they move through the
supply chain. This process begins with the retail grocer ordering the product and ends
when the products are selected by the consumer and put in their shopping basket. This
phase of the research project focused on analyzing the relevant activities, events and
processes through which case-ready meat must pass within the ﬂesh meat supply chain.
This research was also instrumental in designing, testing and data collection of the survey
instrument using the Yin (2003), Eisenhardt (1989), and Creswell’s (2005/ 1994) research
design ﬂameworks.

Prior to conducting the supply chain analysis, phone interviews of top executives,
mid-level executives and store managers were collected to get an overall sense of the

industry, product knowledge, and to document any concerns they had about the research,

65

case-ready meat products, the state of the industry, and the role meat products play in
overall customer satisfaction. Next, a series of visits to retail grocery stores and
distribution centers (DC) throughout three Midwestern states were undertaken to collect
data in person, to observe product movement and the facilities’ operations. These visits
corroborated the information gathered in the earlier phone interviews.

Observing product transactions and movements through the supply chain was
important to identify weaknesses in the supply chain; and it was vitally important in
developing a process map. During the store and DC visits the procedures retailers took
before they ordered products ﬂom the processor were observed and how and when the
order was placed. A follow-up visit to the same facilities was undertaken to observe
ﬂesh product arriving at the DCs and the third party logistics personnel in charge of
driving the product ﬂom the centrally located meat processing plant to the DC were also
interviewed. Of particular note were the turnaround times, ﬂom when the product was
ﬁrst ordered until that particular order arrived at the DC. The average turnaround time
recorded was approximately 12 days.

The next phase of the research was to map product moving through the supply
chain, beginning with product arriving at the DC. Hunt (1999) describes process
mapping a supply chain as “product moving through a series of steps designed to
ultimately produce a product or service.” The stops along the supply chain should be
viewed as value chains, denoted by the contribution each stop adds to the ultimate value
of the product. Denoting where value occurs is the highest concern in producing a

process map. Supply chain management activities are typically tasked with creating

66

value at each stop along the chain or the stop is eliminated or changed. Hunt (1999)
further states the beneﬁts to using process mapping below.
Beneﬁts to process mapping:

0 Conﬁrmation of value being added at each stop;

0 Reductions in product and service costs;

0 Fewer system integration failures;

- Uniformly better process understanding;

0 Improvement in overall business enterprise operations and performance.

Hunt (1999) found that ﬁrms can better understand their processes or systems
better by creating process maps that show all of the activities associated in producing
goods and services. Process maps should distinguish the functions a system is designed
to perform; and confum it is accomplishing those functions. It is essential that the
process map reveals what is actually being accomplished, not what the system is intended
to accomplish.

The process mapping literature suggested ﬁrst to observe product/services
ﬂowing through the supply chain noting cycle times, temperature readings at each stop,
personnel/equipment engaged in moving product, any guidelines or metrics in place
associated with each stop and noting the inventory procedures in place (Darnelio, 1996),
(Lientz, 2002), (Burgess, 1998), (Biazzo, 2002), (Bond, 1999), (Collins, 1997),
(Gourishankar, 2003), (Denton, 1995), (Hunt, 1999), (Fulscher, 1999), (Mason, 1997),
(Keller, 1999) and (Anjard, 1998). To accomplish this phase, two DC were

independently observed in preparation of the process map.

67

The results of the observation phase were compiled and a draft of the process map
was constructed. A conference with corporate management was held to discuss the
preliminary results and resolve any inconsistencies with the process map. Discussing the
preliminary results was necessary to ensure an accurate representation was being
provided in the process map. A conference with corporate management was held to
discuss the preliminary results of process mapping and resolve minor inconsistencies in
order to ensure an accurate representation. The process map is in Appendix A.

One of the most signiﬁcant advantages of case-ready meat packaging is
dramatically increased shelf life compared with conventionally packaged meats in retail
stores. However, it is critically important to maintain the temperature at 40 degrees
Fahrenheit or lower. Understanding that the product was vulnerable to temperature
variations along the supply chain, a “Cold-Chain Vulnerability Index” was developed to
analyze the temperature at each critical point.

Case-ready meat products are particularly vulnerable to factors in the cold chain
including time, temperature and handling that may lead to product degradation. The
index assigns a numeric value to potential risks at each critical control point. The index
is a ranking ﬂom 1-10, 10 being the greatest risk or potential for degradation. The
process map also cataloged all operational issues associated with product moving through
supply chain.

Further analysis of the information gathered in the process mapping revealed that
case-ready meat products were arriving ﬂom the processor at the DCs consistently in
_ excellent condition; staged to enter the retailer’s supply chain. The mapping analysis

demonstrated that the quality of case—ready products was maintained throughout the

68

supply chain. Much of this success is due to DC conﬁgurations speciﬁcally designed for
efﬁcient movement of product into the appropriate storage areas. DCs were also
designed to maintain refrigerated foods at a speciﬁc temperature on the receiving docks
before being moved into refrigerated storage. Many older DCs do not have such
advanced facilities; and when case-ready programs are expanded, renovations would be
necessary.

Observations of case-ready products shipped ﬂom DCs and arriving at retail
stores throughout the Midwest revealed that they were properly staged, inventoried, and
stored prior to stocking the retail meat cases. Each store had metrics in place to facilitate
these ﬁmctions. It should be noted these metrics were developed by the Save-A-Lot
senior perishable managers in collaboration with technical product advice provided by its
supplying meat processor. The artifacts gathered in this phase of the case study further
substantiated the sustained high quality of case-ready meats distributed through the
supply chain. Additional insights gained from the mapping process led to the decision to
conduct additional interviews of SAL management and hourly personnel.

Several important observations were made during the process mapping phase.
There were no signiﬁcant delays in product movement along the cold chain. This was
true for both boxed beef9 and case-ready products. This ﬁnding is a testament to the
dedication and professionalism and collaboration of the meat processor and retailer’s
team members. The health and well being of their customers depends upon the

adherence of food safety metrics designed in its systems. This ﬁnding contradicts the

 

9 Boxed beef refers to subprimal beef products (about 40-451bs per box) vacuum packaged in heavy duty
bags and shipped in sturdy corrugated boxes to grocery stores. These are the products that butchers process,
portion and package.

69

conventional wisdom among case-ready meat processors that the retail personnel were

mishandling the product.

3.13 Questionnaire Design

Each question was vetted and pre-tested for relevance by several industry
executives. Faculty members and a graduate student in the Marketing and Supply Chain
Department reviewed the questionnaire for content and validity. For statistical relevance,
Michigan State University’s Statistical Consulting Service (SCS) reviewed the survey"s
questions and possible answers. The SCS provided invaluable insights and suggestions
toward improving the questionnaire’s format, Likert scale designs, and consistency. The
order of the questions was consistent with other survey designs in that it assumed
respondents review the questionnaire prior to beginning (Churchill, 1999) and (Creswell,
1994)

The physical layout of the questionnaire was designed for ease of completion. All
questions were preceded with explicit instructions that were highlighted so they could be
differentiated from the answer set. Instructions were designed to explain to respondents
precisely how to answer each question. These steps were important in maintaining
survey instrument reliability and valid content. Respondents were instructed to review
the entire questionnaire in advance of answering (Churchill, 1999) and (Creswell, 1994).

The questionnaire consisted of 10 one-sided pages and was estimated to take
about 30 minutes to complete. A cover page was attached to the questionnaire describing
the research project’s challenges and opportunities, and explaining why the respondents’
answers/comments would be essential to the success of the project. The cover page also

showed the project’s university affiliation.

70

The questionnaire packet also included an “Informed Consent Form,” which
Michigan State University requires of all research involving human subjects and/or data
collected from living human subjects (including preexisting data). This is subject to
University Committee on Research Involving Human Subjects (UCRIHS) review and
consent prior to the questionnaire being sent and completed by respondents. The project
fulﬁlled UCRIHS’s protocol; and permission was granted. The consent form was
included with the questionnaire; and was signed and dated before the respondent
completed the questionnaire. One hundred ninety three retail managers of SAL managers

completed the required UCRIHS forms and responded to the questionnaire.

3.13.1 Unit of Analysis
According to Yin (2003), the unit of analysis should be deﬁned as the critical factor
within the case study. Often times the nature of the research questions will pose
problems in deﬁning the unit of analysis. When this occurs Yin suggests that researchers
should consider defaulting back to the phenomenon being studied for deﬁning a clear and
concise connection ﬂom the phenomenon to the original research question(s). It should
be noted that if a prior case study research has been conducted the unit of analysis should
follow the original research’s unit of analysis.

In keeping with Yin’s suggested methodology for choosing a unit of analysis and,
given there is no prior research to draw ﬂom, this research project will present its
analysis following the research questions which tie to the six operational issues. For this
research project the unit of analysis will be the six operational issues of: Food Safety

Operational Issues; Workmanship Operational Issues; Shrink Operational Issues;

71

Packaging Operational Issues; Handling Operational Issues; and Case-Ready Product
Operational Issues.

The questionnaire began with a series of demographic questions designed not to
inﬂuence answers to the speciﬁc attribute questions of the propositions. By asking
speciﬁc attribute questions ﬁrst, the attitude of respondents may be affected Churchill
(1999) and Hair et a1 (1999). Therefore, demographic and general questions were
positioned ﬁrst in an “inverted funnel sequence” in the hope of gaining respondents’
conﬁdence and to calm possible anxiety associated with completing an academic
questionnaire.

Next, the self-administered questionnaire addressed the six categories of
operational issues discovered in the case study and process map phase of the research.
Market research has shown that respondents rarely complete questions asking for
personal opinions. One plausible explanation is that respondents may be afraid of
making grammar and spelling mistakes. However, the project required the inclusion of
speciﬁc open-ended “follow-up” questions to be a necessary part of the questionnaire.

With marketing research literature warning against the use of open-ended
questions, one technique employed in this questionnaire was to obtain answers to
important open-ended questions by asking a pertinent “yes” or “no” question and then
asking a related open-ended question. The point of this technique was to obtain the
valuable “yes” or “no” data and then allow the respondent an opportunity to further
expound on the subject. The literature emphasized the importance of consistently coding

the results of open-ended questions and look for patterns.

72

The questionnaire was comprised of a total of 87 questions. Of these, 15
questions were open-ended and required a written response. The remaining 72 questions
were yes-no questions, ranking questions, and Likert scale questions. All Likert scale
questions utilized a scale beginning with 1 and ending with 5 and remained constant
throughout the questionnaire.

To keep the Likert scale constant for respondents, the ﬁrst choice on the scale was
1 and always represented the lowest value. The Likert questions had three different
scales for responses, ranging ﬂom 1 being “not satisﬁed” to 5 being “very satisﬁed.” On
a second scale, 1 represented “strongly agree” and 5 “strongly disagree.” In the last

Likert scale, 1 represented a “big concern” and 5 “not a concern.”

3.13.2 Respondents to the Questionnaire

The target population was determined to be 220 store managers within one chain
of a nationally recognized, Limited Assortment grocery store. The number of stores to
survey was determined jointly with the project and senior corporate management of SAL.
The 220 stores were represented equally by 110 corporate stores and 110 ﬂanchise stores
throughout the United States. Each store had a manager who was personally responsible
for the success of the store. This individual was identiﬁed as the respondent to the
questionnaire.

The national chain is comprised only of corporate and franchise stores. Corporate
stores are owned and operated by the corporation. Franchise stores operate under a
licensing agreement with the corporation. Both stores appear identical and operated in

similar ways, offering the same products and services.

73

To ensure the respondents had a working knowledge of case-ready meats, the
project worked directly with corporate management in identifying 220 stores that met the
criteria. The stores were evenly divided into geographically diverse groups. To further
ensure success in obtaining a high completion rate, corporate management communicated
essential aspects of the study with corporate and ﬂanchise Division Managers10 and
requested their support in the study. Division managers received the questionnaires and
distributed them to their district managers, who in turn personally distributed the
questionnaires to the individual store managers, thus demonstrating commitment to this
important research project.

It was important to obtain survey data from respondents who possessed some
knowledge of case-ready meat programs, which resulted in avoiding random sampling of
respondents. A downside of not randomly sampling the entire population is the presence
of sampling bias found in non-probability sampling. Nevertheless, it was decided that the
information ﬂom the survey would offer valuable insights and outweighed the value of
probability sampling.

Convenience sampling is a recognized sampling technique used in exploratory
research when the project is interested in gaining insights represented as approximations.
Convenience sampling is related to non-probability sampling. This is a method of
selecting samples based on their availability and the judgments of the researcher that the
samples will be representative. In probability sampling all respondents in a known
population has a mathematical probability of being chosen for the sample. Likewise, a

key component of non-probability sampling is the reality that a known segment of the

 

'0 Division Managers are in charge of several District Managers, and District Managers are in charge of
several retail stores.

74

population has no chance of being part of the sample (Hair et al, 1999) and (Churchill,
1999)

A further consideration in the use of convenience sampling was the fact expressed
by executive management that not all store managers within the SAL chain have
experience with case-ready meats. It was therefore deemed logical that store managers
who completed the questionnaire possess at least some knowledge of case-ready meat
operations so that they were able to make informed responses to the questionnaire
concerning the relevant operational issues. Corporate executives deemed this
requirement was met if the store sold the product either currently or in the past. District
managers had extensive knowledge of each store manager’s current and past expertise,
which was inﬂuential in the decision to choose convenience sampling.

It was determined that the validity of the results was directly related to the
respondents’ ability to understand the speciﬁc case-ready questions; and it seems certain
that the respondents possessed sufﬁcient knowledge of case-ready meat products and its

operational issues.

75

CHAPTER 4— DATA ANALYSIS: CROSS-TABULATIONS, MULTIPLE
REGRESSION AND BINARY LOGISTIC REGRESSION

4.1 Introduction

Chapter 4 discusses the results of the questionnaire data analysis. Because of the
range and ﬂexibility of the data collected from the questionnaires three statistical
techniques were employed to analyze the data set: cross tabulations, multiple regression
and binary logistic regression.

Respondents to the self-administered convenience survey completed and returned
193 of 220 questionnaires, yielding an overall 88 percent response rate. Of the 193
usable responses, corporate store managers completed and returned 99 percent of their
questionnaires (n=109). Licensed store managers returned 76 percent of their
questionnaires (n=84). The following section summarizes the respondents by category
and geographical location.

A total of 109 corporate store questionnaires were returned ﬂom eight different
geographic areas11 throughout the United States. As Table 4 indicates, the corporate
stores are predominantly located in states with large urban centers. The Southeast
division returned the largest percentage of questionnaires at 27.5 percent, followed by the
Pennsylvania division. The Chicago division has the lowest percentage of responses with

5.5 percent.

 

” The titles for the geographic areas come from the corporate headquarters of the Limited Assortment retail
grocery store chain.

76

Table 4: Geography of Questionnaires Returned by Corporate Stores

 

Corporate Store Division Corporate Stores
N %

Southeast Division 30 27.5
Florida and Georgia

Pennsylvania Division 19 17.4

Northeast Division 15 l 3 .8
Massachusetts, New York,
New Jersey

Mid-Atlantic Division 12 l 1.0
Maryland, Virginia

Ohio Division 1 1 10.0
Canton and Cleveland

MidWest Division 9 8.2
Tennessee, Mississippi,
Louisiana

California Division 7 6.4
California and Arizona

Chicago, Illinois Division 6 5.5

Totals 109 100%

 

Seen below in Table 5 the licensed store divisions are organized a bit differently.
The Midwest division for licensed stores represents a more traditional deﬁnition of the
Midwest states, and the Southeast division for licensed stores is a much larger division
since it includes Mississippi, Louisiana and Tennessee. Licensed stores were given a
total of 110 questionnaires and they returned a total of 76.3 percent. Table 5 indicates
that the Midwest division returned the largest percentage of questionnaires at 46.4
percent followed by the Southeast division which returned 21.4 percent. The smallest

return was ﬂom the Mid-Atlantic division with 3.6 percent.

77

Table 5: Geography of Questionnaires Returned by Licensed Stores

 

Licensed Store Division Licensed Stores
N %
Midwest Division 39 46.4
Michigan, Ohio, Indiana, Wisconsin, and Iowa
Southeast Division 18 21.4

Florida, Mississippi, Louisiana
Tennessee, Kentucky

Northeast Division 12 14.3
Pennsylvania
New York

Southwest Division 1 2 1 4 .3
Texas
Oklahoma
Missouri

Mid-Atlantic Division 3 3.6
Maryland

Totals 84 100%

 

Comparing Table 4 and Table 5 a greater number of corporate stores completed
and returned questionnaires than the licensed stores. When it was determined that more
of the licensed stores were choosing not to participate, corporate management contacted
the respective division and district managers with a reminder about the missing
questionnaires (n=26) to effect a higher response rate. The reminder did not yield any
more completed questionnaires. The project is deeply indebted to corporate management
for their follow-up efforts.

According to the literature, the self-administered convenience survey response
rates enjoyed by the project were exceptionally high, and well above expected rates of
similar instruments (Selltiz et al, 1976; Cook and Campbell, 1979; Yue and Cooper, 1983
and Ghauri et al, 2005). There are several plausible explanations for the high response
rates. First, the corporate store managers maintain and foster a direct and open
relationship with division and district managers. This point was apparent to the store
managers (respondents) when they observed corporate management recruiting division

and district managers to distribute the questionnaires throughout the entire retail chain.

78

Another possible explanation for high response rates is that corporate management was
very effective in identifying stores that have a working lmowledge of case-ready meats
and are interested in the topic. These respondents may have deemed it worthwhile to
spend their time completing the questionnaire. While the project enjoyed an astounding
99 percent response rate for corporate stores, the licensed stores had a high response rate

of 76 percent. The response rate for both types of stores was 87 percent.

4.1.1 Cross-Tabulations of the Research Propositions

The questionnaire contained a series of ranking and Likert scale (interval)
questions to address the twenty-four research propositions regarding the adoption of a
case-ready meat strategy. The ranking questions asked respondents to rank their choices.
The Likert scale questions sought an opinion ﬂom a range of interval choices on a scale
of 1-2-3-4-5. The Pearson chi-square test statistic is appropriate for this data analysis,
because the possible answers are simple counts that fall into speciﬁc category. The
Pearson chi-square test allows for determining whether the observed counts in cells are
different ﬂom expected counts. The Pearson chi-square test statistic assumes a discrete
distribution rather than a normal distribution, but does not reveal the nature of any
relationship. The statistic is states that a relationship exists or does not exist (Hair et al,
2000).

For the cross tabulation analysis. a binary dependent variable was created for the
193 returned questionnaires. With corporate SAL stores coded “0” (n=109) and licensed
SAL stores coded “1” (n=84). The hypothesis to be tested in the cross tabulations and
chi-square test statistic was that corporate stores and licensed stores respondto the Likert

scale questions in the same way at the .05 probability level. Therefore, if the hypothesis

79

can be rejected by the Pearson chi-square test statistic, then statistical evidence exists that
two stores did not answer in the same way and the research could draw inferences about
the results between store types. If the chi-square test statistic is not signiﬁcant at the
critical alpha value of .05 (test statistic > than .05), we would then fail to reject the
hypothesis and conclude that there is no signiﬁcant difference between corporate stores

and licensed stores.

4.1.2 Analysis of General Case-Ready Product Operational Issues

This section of the data analysis is interested in the questions designed to elicit
responses about both types of ﬂesh meat products being sold in both formats (case-ready
and products processed in-stores). The research propositions for case-ready products
(CRP) are:

Rle There are several key success factors that all ﬂesh meat products

must possess in order to be considered viable in a retail grocery store

format. These attributes include: product appearance, shrink management,

proﬁtability, shelf-life, food safety, out-of-stocks, product consistency,

SKU varieties, and product traceability.

RPM-RP Managers assume that case-ready meats are more expensive.

RPW Managers feel that case—ready meats are a more employee

ﬁiendly program than in-store processed meats.

RP4.CRP Managers know how their customers perceive fresh meats with

respect to color, servings per package, and package price.

RP5_CRp Managers would not have a problem training a traditional meat 7

department manager to manage all perishables, not just meats.

80

Rpm Managers have a poor perception of the relative costs between

case-ready and non-case ready products.

This section also discusses the responses concerning speciﬁc attributes of case-
ready products, and What case-ready product key success factors the alternative is able to
produce for stores that market case-ready meat products. Questions in the survey
pertaining to product operational issues include Q2, Q3, Q5, Q6, Q10, Q13, Q17, Q19-
24, and Q31.

Table 6 displays the questions and responses as to which CRP store managers are
marketing at their respective stores. The corporate stores sold 100 percent case-ready
ground beef and 40.5 percent of the licensed stores sold the product. The percentage of
licensed stores selling case-ready ground beef may be higher than what actually occurs
because the question should have stipulated that case-ready ground beef does not include
Chubs”.

The sample of respondents selling case-ready beef cuts was predictable and
correlated to store type. A total of 90.8 percent (n=99) of corporate stores sold case-
ready beef cuts and the majority of licensed stores 88.1 percent (n=74) did not. Stores
that sell case-ready pork products had a different distribution. Eighty-nine percent
(n=97) of the corporate stores sold case-ready pork products, and 8.3 percent of the stores
did not. Three stores cited the question as not applicable. The licensed stores had an
interesting response, note with 42.9 percent of the stores selling case-ready pork cuts.

Why this response rate is so high deserves further investigation.

 

'2 Chubs are tubes of ground beef similar that come in different weights and lean-to-fat ratios. Many bulk
pork sausage products are sold in 802 and 16 oz Chubs.

81

Table 6: Case-Ready Meat Products Sold by Store Type.

 

 

 

Question #2 Responses Corporate Licensed
N % N %
Does your store sell case— 0=No 0 0 47 56.0
ready ground beef? 1= Yes 109 100 34 40.5
2=N/A 0 0 3 3.6
Does your store sell case- 0=No 7 6.4 74 88.1
ready beef cuts? l= Yes 99 90.8 9 10.7
=N/A 3 2.8 l 1.2
Does your store sell case- 0=No 9 8.3 46 54.8
ready pork cuts? 1= Yes 97 89.0 36 42.9
2=N/A 3 2.8 2 2.4

 

Respondents were asked several questions about speciﬁc case-ready meat
products for sale in their individual stores. The questionnaire investigated store sales of
the three major categories of fresh meats: ground beef, whole muscle beef cuts (e.g.
steaks, roasts, etc.), and whole muscle pork cuts (e.g. steaks, roasts, etc.).

Sales of ground beef as a category is important in the product mix and contributes
signiﬁcantly; to proﬁt for all retail grocers. According to the Food Marketing Institute’s
Meat & Poultry Fact Sheet, sales of ground beef products as a single category of meat
products accounts for 42 percent (27.7 pounds per capita annually) of all meat sales. This
is a close second to the sales of beef cuts at 44 percent (29.3 pounds per capita annually).
With that in mind respondents were asked if the store grinds beef and/or pork products.

Question 3 asked respondents if their store grinds beef and/or pork products. A
total of 99.1 percent of the corporate stores and 27.4 percent of licensed stores do not
grind products. These results are perplexing in light of the responses to question 2 about
selling case-ready ground beef. The number of licensed stores not grinding beef/pork
products yet selling case-ready ground beef does not add up to a plausible ﬁgure (all
stores sell these products, and they either grind the products or purchase case-ready). A
partial explanation could be that some licensed stores sell only chub or bulk ground beef

products.

82

Question 5 (see Table 7) was designed to identify and quantify speciﬁc attributes
of case-ready meat products. The attributes are color, servings per package, and package
price. Respondents ﬂom both types of stores thought that the number one attribute of
ﬂesh meatis price, followed by color of the meat, and then servings per package. These
results are logical given the store format offers limited assortrnents of products for sale
and deep discount pricing for all products.

Table 7: Product Attributes Ranked by Store Type.

 

 

 

Question QS Responses Corporate Licensed
N '/o N '/o

How would your customers rank l= First Preference 28 25.7 34 40.5
the color attribute of case-ready =Second Preference 60 55.0 23 40.5
fresh beef and pork products? 3=Third Preference 18 16.5 18 16.7

N/A 3 2.8 2 2.4
How would your customers rank l= First Preference 4 3.7 4 4.8
the servings per package attribute =Second Preference 25 22.9 17 20.2
of case-ready ﬂesh beef and pork 3=Third Preference 77 70.6 61 72.6
products? N/A 3 2.8 2 2.4
How would your customers rank l= First Preference 74 67.9 44 52.4
the package price attribute of 2=Second Preference 21 ‘ 19.3 31 36.9
case-ready ﬂesh beef and pork . 3=Third Preference 11 10.1 7 8.3
products? N/A 3 2.8 2 2.4

 

Question number 6 was designed as a corollary question of question 5 and
explores key demographic traits for Limited Assortment Store customers. Most of the
customers who shop at Limited Assortment Stores seek low prices. This question sought
to discern which of the three product offerings was deemed the most “price sensitive.”
Corporate and licensed store respondents were consistent in their responses that ground
beef was the most sensitive, followed by beef cuts (see Table 8). The least sensitive was

pork cuts.

83

Table 8: Price Sensitivity of Fresh Meat Products by Store Type.

 

 

 

Question 6 Responses Corporate Licensed
N % N %
Rank how sensitive your 1= Most Sensitive 50 45.9 18 21.4
customers are to the prices per 2=Moderately Sensitive 28 25.7 47 56.0
pound charged for fresh whole 3=Least Sensitive 20 18.3 16 19.0
muscle beef products. N/A 1 1 10.1 3 3.6
Rank how sensitive your 1= Most Sensitive 42 38.5 62 73.8
customers are to the prices per 2=Moderately Sensitive 32 29.4 13 15.5
pound charged for ﬂesh ground 3=Least Sensitive 25 22.9 6 7 .1
beef products. N/A 10 9.2 3 3 .6
Rank how sensitive your 1: Most Sensitive 6 5.5 1 1.2
customers are to the prices per 2=Moderately Sensitive 39 35.8 21 25.0
pound charged for fresh pork 3=Least Sensitive 53 48.6 59 70.2
cuts and grinds products. N/A 1 1 10.1 3 3.6

 

Questions 10, 13, 17 were designed to better understand how respondents viewed
the key success factors linked to product appearance, shelf life, and Store Keeping
Units13 (SKU). These questions are similar and in nature and will be discussed together.
The results ﬂom Table 9 below show that both store types prefer the appearance of ﬂesh
meat products produced in-store by 61.6 percent. Case-ready product appearance was
chosen by 24.2 percent of corporate and licensed stores. The composition of that ﬁgure
masks the fact that only 8.6 percent of licensed stores viewed case-ready product
appearance as superior to in-store. The opposite pattern is observed when respondents
were asked the question regarding shelf life. Over 64 percent of the respondents ﬂom
both store types chose case-ready as the superior product when considering shelf life as a
key success factor. Question 17 was used to determine respondents’ opinions on which
product offered the most SKU varieties. The results conﬁrmed what was learned ﬂom
the case study namely, case-ready products had a limited offering, but the results were far
ﬂom being a consensus. Over 52 percent of all respondents felt that meat produced in-

store offered more SKU varieties. However, 25.9 percent of the respondents thought

 

‘3 The Food Marketing Institute (FMI) defines SKU as a number that identiﬁes each separate brand, size,
ﬂavor, color or pack of a product.

84

there was no difference. The licensed stores believing that there was no difference in
SKU varieties were higher at 33 percent.

Table 9: Key Success Factors of Case-Ready Meat Products by Store Type.

 

 

 

Key Success Factor Responses Corporate Licensed
N '/o N '/o
1= In-store 52 47.7 65 80.2
Q10. Product Appearance 2=Case-ready 39 35.8 7 8.6
3=No difference 18 16.5 9 11.1
0=N/A 0 0 3 3.6
1= In-store 17 15.6 29 35.8
Q13. Shelf Life 2=Case-ready 78 71.6 45 55.6
3=No difference 14 12.8 7 8.6
0=N/A 0 0 3 3.6
l= In-store 52 47.7 48 60.0
Q17. SKU Varieties 2=Case-ready 35 32.1 5 6.3
3=No difference 22 20.2 27 33.8
0=N/A 0 0 4 4.8

 

Question 12 was concerned with the highly important proﬁtability “key success
factor.” It asked the respondents which ﬂesh meat alternative they deemed more
proﬁtable. Aggregating the results ﬂom both store types yields an interesting result.
Table 10 below indicates that 57 percent of all respondents thought that meat produced
in-store was more proﬁtable. It should be noted that 17.9 percent of licensed stores felt
that the case-ready meat alternative was more proﬁtable, an important point with respect
of acceptance of case-ready meats by licensed stores.

Table 10: Key Success Factors Related to Proﬁtability

 

Key Success Factor: Responses Corporate Licensed
N % N %
l= In-store 47 43.1 63 75.0
Q 12. Proﬁtability 2=Case-ready 50 45 .9 15 17.9
3=No difference 12 _ 11.0 3 3.6
0=N/A 0 0 3 3.6

 

Table 11 highlights the results for Question 28, which is the question of
“employee ﬁiendliness.” Over 78 percent thought case-ready meat alternative was

friendly toward employees.

85

Table 11: Employee Friendly Case—Ready Meat Alternatives by Store Type.
Corporate Licensed

 

(hrestion #28 Responses N % N %

Are case-ready meat alternatives 0=No 10 9.2 24 28.6

“employee ﬁiendly?” l= Yes 99 90.8 41 48.8
2=N/A 0 0 19 22.6

 

Table 12 shows the responses to the question about how “manager ﬁiendly” a
case-ready meat alternative is in the marketplace. The results coincide with other
questions asked in the questionnaire concerned with the managerial implications of a
case-ready meat implementation. Manager ﬁiendly products are closely related to high
levels of customer loyalty and customer loyalty in grocery store industry is closely linked
with higher per customer sales. In addition, many industry experts believe customer
loyalty is earned through direct interaction between store personnel and shoppers.
Grocers that enjoy high rates of customer loyalty have in the past earned positive
economic proﬁts. Whole Foods, Wegrnans, and Ukrops are three companies that have
maximized their customer’s store loyalty relationship by creating a positive experience in
the meat and produce departments. These results regarding “manager ﬁiendliness” are
very positive for case-ready meats for both types of stores, but less so for the licensed
stores.

Table 12: Manager Friendly Case-Ready Meat Alternatives by Store Type.

 

Question #30 Responses Corporate Licensed
N % N '/o
Are case-ready meat 0=No 4 3.7 18 21.4
alternatives “manager 1: Yes 105 96.3 39 46.4
friendly?” 2=N/A ' 0 0 27 32. l

 

Several other speciﬁc case-ready product Likert scale questions were posed to
further understand respondents’ views of speciﬁc case-ready product offerings. Table 13
below presents these results. For these questions it is important to focus at how licensed

stores viewed case-ready product offerings since the corporate stores were fairly

86

unanimous in their responses. These results are related to the strength of having the case
study provide insights for formulating research propositions some of which that ran
counter to what some of the research and trade journals were espousing about case-ready
product offerings.

As indicated in Table 13 over 60.3 percent of all respondents agree or strongly
agree that case-ready ground beef products are available in the lean-to-fat ratios their
store customers prefer, which was counter to what was revealed in the interview process.
Having the correct ratios is instrumental towards providing customers with a product they
desire at a competitive price. Similar results are reported for case-ready ground beef
being available in the pack sizes (weight per package) their customers demand.

While the news is good for the availability of ground beef products, it is less so
for whole muscle beef products. Only 34.2 percent of all respondents agree or strongly
agree case-ready whole muscle beef products are available in the cuts their customers
prefer. A partial explanation for this low ﬁgure coincides with several interviews of meat
managers where they felt the strength of processing meat in-stores is the ﬂexibility in

meeting customer preferences on a location by location basis.

87

Table 13: Case-Ready Product Issues by Retail Store Type.

 

 

 

 

Case-Ready Meat Product Responses Corporate Licensed
Issues N % N “/o
1= Strongly Disagree 0 0 4 4.8
Q19. Case-ready ground beef 2 7 6.4 7 8.3
products are available in the 3 26 23.9 23 27.4
lean-fat ratios my customers 4 52 47.7 22 26.2
prefer. 5= Strongly Agree 24 22.0 18 21.4
N/A 0 0 10 l 1.9
l= Strongly Disagree 1 .9 13 15.5
Q20. Case-ready whole muscle 2 24 22.0 15 17.9
beef products are available in 3 49 45.0 26 31.0
the cuts my customers prefer. 4 25 22.9 14 16.7
5= Strongly Agree 20 18.3 7 8.3
N/A 3 2.8 15 17.9
1= Strongly Disagree 1 .9 7 8.3
Q22. Case-ready ground beef 2 7 6.4 12 14.3
products are available in the 3 22 20.2 16 19.0
pack sizes my customers prefer. 4 51 46.8 26 31.0
5= Strongly Agree 28 25.7 9 10.7
N/A 0 0 14 16.7
1= Strongly Disagree 1 .9 12 14.3
Q22. Case-ready whole muscle 2 9 8.3 17 20.2
beef products are available in 3 43 39.4 21 25
the pack sizes my customers 4 41 37.6 13 15.5
prefer. 5= Strongly Agree 12 11 4 4.8

N/A 3 2.8 17 20.2

The results for question 22 in Table 13 above shows only 36.3 percent of
respondents agree or strongly agree that case-ready whole muscle beef cuts are available
in the pack sizes their customers prefer. This result is particularly important given the
observations and discussions with store customers during the case study phase. They
indicated that their choice to purchase any center of the plate protein is dependent upon
how many servings were in the pack (“pack size”). This viewpoint also concurs with
interviews of meat department personnel and store executives. They felt that the pack

size was more strongly inﬂuenced by the number of portions than the package weight.

88

4.2 Cross Tabulation Research Propositions: Case-Ready Product Issues

This section focuses on the Likert scale questions (Q72-Q76) that addressed
speciﬁc operational issues for the case-ready meat products and stated in four of the
research propositions (RPz, RP3, RPS, and RP6). These questions endeavored to shed
light upon these speciﬁc issues: assumptions about the costs of case-ready products being
more expensive than other products (RPz), case-ready meats are considered an “employee
ﬂiendly”(RP3), the ability of meat department manager to manage all perishables, not
only ﬂesh meats (RP5), and managers have a low perception of the relative costs between
case-ready and in-store processed meats (RP6).

Table 14 summarizes the respondent’s answers to questions 72-76. Questions 73
and 74 were not signiﬁcant at the critical alpha level and therefore we fail to reject the
null hypothesis. The results for questions 73 and 74 indicate that corporate and licensed
store answers for each scale contain approximately the same percentage.

Question 72 asked a fundamental question about adopting a case-ready meat
scenario. Speciﬁcally, is it difﬁcult to hire qualiﬁed meat cutters? Several journal and
trade articles (Keith, 2004; Salvage, 2003; Brody, 2002 and Hodgins, 2002) have
commented about the ever-increasing difficulty of hiring and retaining qualiﬁed meat
cutters. Table 14 demonstrates that most respondents view this as a difficulty, as seen by
the majority of responses are above the median.

Question 75 asked if case-ready products were as cost-effective as meat produced
in-store. The results varied by store type. Corporate stores viewed the case-ready ‘

product as cost effective, licensed stores disagreed.

89

Question 76 asked if the increasing cost of workers’ compensation related to
hiring and paying meat cutters was becoming cost-prohibitive. While many respondents
ﬂom both types of stores answered not applicable, most responses ranged ﬂom agree to
strongly agree that workers compensation was becoming cost-prohibitive.

Table 14: Case-Ready Meat Management Operational Issues by Store Type

 

 

 

 

 

Case-Ready Meat Operational Likert Scale Responses Corporate Licensed
Issues: Likert Scale Questions N % N %
1= Strongly Disagree 9 8.3 8 9.5
Q72. It has become increasingly 2 10 9.2 10 11.9
more difﬁcult to ﬁnd qualiﬁed 3 19 17.4 9 10.7
meat cutters for my store. 4 13 11.9 17 20.2
5= Strongly Agree 18 16.5 31 36.9
N/A 40 36.7 9 10.7
Q73. A traditional meat 1: Strongly Disagree 6 5.5 5 6.0
department manager and/or meat 2 7 6.4 7 8.3
cutter could be trained to manage 3 1 1 10.1 9 10.7
all perishables not just meats. 4 33 30.3 29 34.5
FAIL TO ACCEPT 5= Strongly Agree 42 38.5 27 32.1
N/A 10 9.2 7 8.3
Q74, Implementing a case-ready 1= Strongly Disagree 2 1.8 2 2.4
meat program will shift meat 2 1 .9 5 6.0
mgmt. from production activities 3 21 19.3 14 16.7
to forecasting/inventory 4 43 39.4 35 41 .7
management. 5= Strongly Agree 28 25.7 25 29.8
FAIL TO ACCEPT N/A 14 12.8 3 3.6
1= Strongly Disagree 5 4.6 19 22.6
Q75. Case-ready beef and pork 2 7 6-4 19 22.6
products compared to in-store 3 29 26.6 22 26.2
produced meat products are more 4 37 33.9 11 13 .1
cost effective. 5= Strongly Agree 22 20.2 8 9.5
N/A 9 8.3 5 6.0
1= Strongly Disagree 2 1.8 5 6.0
Q76. Cost of workers’ 2 4 3.7 9 10.7
compensation for employees that 3 35 32.1 35 41.7
cut meat is becoming prohibitive. 4 32 29.4 14 16.7
5= Strongly Agree 18 16.5 9 10.7
N/A 18 16.5 12 14.3

 

There are two questions that showed no differences between corporate and
licensed stores. These questions asked managers about possible future scenarios for meat
department personnel when using a case-ready meat strategy. Both store types
overwhelmingly agreed or strongly agreed that meat department personnel could be

trained to manage the entire perishables category of meat and produce. Processors were

90

enthusiastic about the marketing expertise current meat department personnel could
provide customers on a daily basis if their primary duties were to manage the category

versus processing meats.

4.2.1 Results of Packaging Operational Issues

The four packaging research propositions include:

Rlex Case-ready meat products that require “blooming” cause

operational issues as a result of preparing that product to be displayed in

the meat case.

RP19.p]( Case-ready meat packing is considered “attractive.”

RPmpK Store managers prefer the case-ready beef that is packaged in low

oxygen, low headspace.

RP11-PK Store managers believe the case-ready beef packaging alternative

of high oxygen/high headspace is a detriment to ﬂesh meat sales.

Twelve Likert scale questions concerning the key packaging issues were drafted
ﬂom the proceedings of the March 19 videoconference at Michigan State University and
after completing the case study with industry executives, retail Distribution Center (DC)
and retail grocery stores in Michigan, Illinois, Wisconsin and Indiana. The packaging
questions were intended to gather information about several aspects of case-ready meat
packaging found within the four research propositions.

Nine of the twelve questions found in Table 15 and Table 16 ﬁt into two
categories: (1) Is the integrity of the product maintained in the case-ready pack? (2)
Which packaging alternative is more visually attractive? The last three packaging

questions address three completely different topics including the time required for the

91

product to bloom”. Question 69 asked manager’s viewpoint about their customers’
opinions of case-ready packaging alternatives and effectiveness of case-ready meat trays.

Three packaging questions were centered on the “very different looking” high
oxygen/high headspace packaging alternative, which is different ﬂom products processed
and packaged in-stores. Meat processors preferred the high headspace pack for a time
because of the “protection” the high headspace provided during transit and storage in the
retailer’s storage cooler prior to being stocked in the meat cases. During several
interviews store managers suggested that many of their customers felt that the high
headspace package was too different ﬂom standard products and would have problems
being accepted in the marketplace. The customers reactions were based merely on how
decidedly different this new pack appeared ﬂom what they normally purchased.
Knowing the perceptions store managers have regarding packaging alternatives was
deemed important.

All of the ﬁve questions in Table 15 had signiﬁcant Pearson chi-square p-values
at the .05 probability level. These ﬁve questions indicate one obvious positive result.
For question 26, which relates to the packaging of case-ready ground beef containing a
very high lean-to-fat (85/15) ratio product, 56.5 percent of all respondents agree or
strongly agree this packaging displayed a “cherry red” color for the meat. This result is
promising for the adoption of case-ready programs because this is the color meat experts
and retailers associate with “ﬂeshness.” One partial explanation for 85/15 lean to fat

ratio rating higher is that the 70/30 product does not contain signiﬁcantly higher lean

 

M Case-ready products are shipped in boxes from the meat processor, within the boxes product is
overwrapped with oxygen barrier bags. This bag must be opened to allow oxygen in the air to “bloom” the
case-ready products. Once the products are bloomed they will have a “cherry red” color and be ready for
the meat cases.

92

meat which contains more oxymyoglobin. Question 27 with the even higher 90/10 ratio
scored lower than the 85/ 15, and higher than the 70/30. An explanation is that many SAL
stores do not sell the more expensive 90/10 ground beef.

Table 15: Ground Beef Packaging Operational Issues by Retail Store Type.

 

 

 

 

 

Case-Ready Packaging Responses Corporate Licensed
Questions N “/o N %
Q25 (Packaging Environment) l= Strongly Disagree 2 1.8 12 14.3
Case-ready ground beef (not 2 12 11.0 8 9.5
chubs) with a 75/25 lean—fat 3 39 35.8 18 21.4
ratio have a “cherry red” color. 4 42 38.5 22 26.2
5== Strongly Agree 12 11.0 5 6.0
N/A 2 1.8 19 22.6
Q26 (Packaging Environment) 1= Strongly Disagree 0 0 8 9.5
Case-ready ground beef (not 2 5 4.6 7 8.3
chubs) with a 85/ 15 lean—fat 3 27 24.8 17 20.2
ratio have a “cherry red” color. 4 59 54.1 27 32.1
5= Strongly Agree 17 15.6 6 7.1
N/A 1 .9 19 22.6
Q27 (Packaging Environment) 1: Strongly Disagree 2 1.8 8 9.5
Case-ready ground beef (not 2 2 1.8 6 7.1
chubs) with a 90 / 10 lean—fat 3 22 20.2 14 16.7
ratio have a “cherry red” color. 4 44 40.4 19 22.6
5= Strongly Agree 18 16.5 4 4.8
N/A 21 19.3 33 39.3
Q62 (Packaging Alternative) 1= Strongly Disagree 5 4.6 3 .6
Low-Oxygen / Low Head Space 2 10 - 9.2 2 2.4
packaging of case-ready ground 3 43 39.4 25 29.8
Eggs not a detriment to sales. 4 38 34.9 28 33.3
5= Strongly Agree 10 9.2 10 11.9
N/A 3 2.8 16 19.0
Q63 (Packaging Alternative) l= Strongly Disagree 3 2.8 7 8.3
High-Oxygen/ High Head Space 2 21 19.3 13 15.5
packaging of case-ready gound 3 48 44.0 36 42.9
be_efis not a detriment to sales. 4 23 21.1 8 9.5
5= Strongly Agree 12 11.0 3 3.6
N/A 2 1.8 17 20.2

 

The results of the two current packaging alternatives available to processors low-
oxygen/low-head space and the high-oxygen/high-head space are shown in Table 15.
The packaging alternatives for ground beef products, the low-oxygen alternative had 44.6
percent of all respondents agree or strongly agree is not a detriment to sales. Question 63
explored the high/oxygen high/head space packaging alternative for ground beef: the

results showed this packaging alternative is not a detriment to sales with only 23.8

93

percent of all respondents agree or strongly agree it is not. The low-oxygen/low head
space for ground beef'5 had a stronger following among respondents.

All twelve questions in Table 16 had signiﬁcant Pearson chi-square p-values.
Forty one percent of the respondents agree or strongly agree the low/oxygen
low/headspace packaging alternative was not a detriment to sales. These results conﬂict
with the case study, interviews and the literature review: that is they indicated the low
oxygen alternative was the better alternative. In retrospect, these two questions could
have been worded differently to better address which alternative was superior for
marketing purposes to their respective customer base.

Question 68, indicated that 57.5 percent of all respondents agree or strongly agree
that blooming case-ready meats causes operational problems. The results suggest that
processors need to constantly adjust the time and metrics to successfully bloom product.

Question 69 asked respondents if their customers” could tell the difference
between case-ready meat packaging and in-store processed packaging. Fifty nine percent
of all respondents agree or strongly agree that their customers could tell the difference.
This result ran counter to the literature review and the case study results. One plausible
explanation is that many SAL customers also shop at Walmart which only sells case-
ready meats.

Responses to Question 70, also conﬂicts with prior research and interviews. New

black case-ready trays were considered a more attractive tray than the standard white tray

 

’5 It should be noted that the results for whole muscle meat cuts for the two packaging alternatives had the
same results as ground beef; the low-oxygen alternative had more agreement than the high-oxygen
alternative.

94

for in-store processed meats. Only 36.3 percent of the respondents agree or strongly
agree that the trays are attractive.

Table 16: Packaging Operational Issues by Retail Store Type.

 

 

 

 

 

 

 

Case-Ready Packaging Responses Corporate Licensed
Questions N % N '/o
Q64 (Packaging Alternative) l= Strongly Disagree 3 2.8 4 4.8
Low-Oxygen / Low Head Space 2 7 6.4 6 7.1
packaging of case-ready mg! 3 45 41.3 26 31.0
§u_ts is not a detriment to sales. 4 43 39.4 19 22.6
5= Strongly Agree 9 8.3 9 10.7
N/A 2 1.8 20 23.8
Q65 (Packaging Alternative) 1: Strongly Disagree 4 3.7 6 7.1
High-Oxygen/ High Head Space 2 17 15.6 19 22.6
packaging of case-ready bcgf 3 50 45.9 32 38.1
_<_:u_ts is not a detriment to sales. 4 26 23.9 6 7.1
5: Strongly Agree 10 9.2 2 2.4
N/A 2 1.8 19 22.6
Q66 (Packaging Alternative) l= Strongly Disagree 1 .9 7 8.3
Low-Oxygen / Low Head Space 2 7 6.4 12 14.3
packaging of case-ready po_rk 3 22 20.2 16 19.0
cuts & Ends is not a detriment 4 51 46.8 26 31.0
to sales. 5: Strongly Agree 28 25.7 9 10.7
N/A 0 0 14 16.7
Q67 (Packaging Alternative) l= Strongly Disagree 1 .9 12 14.3
High-Oxygen / High Head Space 2 9 . 8.3 17 20.2
packaging of case-ready Mk 3 43 39.4 21 25.0
guts Q grinds is not a detriment 4 41 37.6 13 15.5
to sales. 5= Strongly Agree 12 11.0 4 4.8
N/A 3 2.8 17 20.2
Q68 (Miscellaneous) l= Strongly Disagree 1 .9 6 7.1
Case-ready meat products that 2 6 5.5 7 8.3
require “blooming” cause 3 20 18.3 27 32.1
operational problems. 4 62 56.9 29 34.5
5= Strongly Agree 1 .9 19 22.6
N/A 0 0 8 9.5
Q69 (Miscellaneous) 1: Strongly Disagree 1 .9 7 8.3
My customers know the 2 7 6.4 12 14.3
difference between case-ready 3 22 20.2 16 19.0
packaging and in-store 4 51 46.8 26 31.0
packaging. 5= Strongly Agree 28 25.7 9 10.7
N/A 0 0 14 16.7
Q70 (Miscellaneous) l= Strongly Disagree 1 .9 12 14.3
Case-ready beef and pork 2 9 8.3 17 20.2
products are packaged in 3 43 39.4 21 25.0
attractive trays. 4 41 37.6 13 15.5
5= Strongly Agree 12 11.0 4 4.8
N/A 3 2.8 17 20.2

 

95

4.2.2 Handling Operational Issues Data Analysis

Several meat industry research articles (Brody, 2004, 2003 and 2002) had
concluded that case-ready meat products were not a success because of inadequate
handling problems along the supply chain, beginning with the case-ready meat products
arriving ﬂom the processor at the distribution center. The videoconference with industry
experts contradicted those assertions. When conducting the case study, poor handling
procedures were not observed. Moreover, follow up questions of supply chain personnel
disclosed the handling problems had not been recently witnessed They indicated that the
handling issues in the past were attributed to the older case-ready packaging and not the
handling metrics along the supply chain. The fragile nature of case-ready products in
earlier product offerings was still being referenced even after those valid concerns were
addressed and remedied by the packaging industry in conjunction with the meat
processing industry.

The case study ﬁrrther revealed that case-ready meat processors have developed
and disseminated metrics or best management practices for the proper handling of case-
ready meat products to distribution centers and retailers. This step dramatically improved
performance and fulﬁlled expected shelf-life dates. Retailers and distribution centers
have made signiﬁcant capital improvements to upgrade shipping and handling facilities,
which have improved the performance of case-ready products.

The case study revealed several handling issues pertinent to implementing a
successful case-ready meat strategy and those issues were the theme of the three handling

research propositions. The research propositions for handling are:

96

Rqum Store managers typically ﬁnd it difﬁcult to keep their ﬂesh meat

cases fully stocked between 4PM and 7PM and case-ready meats can

resolve this problem.

Rqum Managers believe a key factor towards increasing the shelf life of

ﬂesh meats is holding and maintaining proper temperatures (e.g. rapidly

transporting ﬂesh meat products ﬂom the receiving area into the walk-in

coolers in a timely manner) furthermore case-ready meats are at a

disadvantage in this area.

Rqum Case-ready meats are easier to manage for proper rotation and to

avoid out-of-date product and help maintain proper adequate inventory

levels.

Table 17 summarizes the ﬁndings of the Likert scale handling questions 49
through 55. These questions all had signiﬁcant chi-square results except question 50.
These questions address the underlying importance of factors that relate to the proper
handling that case-ready meats. Managers of both corporate and licensed stores were
satisﬁed with the way product is being handled. These results, like the ﬁndings ﬂom the
video conference contradicted the meat industry journal articles about inadequate

handling metrics along the supply chain have lead to poor adoption of case-ready meats.

97

Table 17: Handling Operational Issues by Retail Store Type.

 

 

 

 

 

 

 

Case-Ready Handling Responses Corporate Licensed
Operational Issues N % N %
1=Notatall 2 1.8 3 3.6
Q49. How satisﬁed are you with 2 7 6.4 14 16.7
the way your store accounts for 3 32 29.4 1 1 13.1
every ﬂesh meat package not sold 4 45 41.3 33 39.3
at full retail? 5: Very satisﬁed 22 20.2 21 25.0
N/A 1 .9 2 2.4
1= Not at all 0 0 0 0
Q50. How satisﬁed are you with 2 2 1.8 0 0
the way your store stocks meat 3 16 14.7 7 8.3
cases to ensure FIFO methods are 4 47 43.1 39 46.4
followed? 5= Very satisﬁed 43 39.4 38 45.2
FAIL TO REJECT p = .353 N/A 1 .9 0 0
1= Not at all 0 0 0 0
Q51. How satisﬁed are you with 2 2 1.8 4 4.8
the way your store maintains 3 20 18.3 7 8.3
adequate inventory levels? 4 65 59.6 39 46.4
5= Very satisﬁed 21 19.3 34 40.5
N/A 1 .9 0 0
1: Not at all 0 0 0 0
Q52. How satisﬁed are you with 2 1 1.2 1 1.2
the way your store manages ﬂesh 3 19 17.4 10 11.9
beef and pork inventory in the 4 53 48.6 28 33.3
walk-in cooler? 5= Very satisﬁed 34 31.2 44 52.4
N/A 2 1.8 l 1.2
l= Not at all 0 . 0 0 0
Q53. How satisﬁed are you with 2 6 5.5 l 1.2
the way your store maintains 3. 24 22.0 14 16.7
adequate inventory levels? 4 60 55.0 42 50.0
5: Very satisﬁed 17 15.6 27 32.1
N/A 2 1.8 0 6.0
1= Not at all 0 0 0 0
Q54. How satisﬁed are you with 2 5 4.6 2 2.4
the way your store prepares for 3 22 22.0 8 9.5
monthly meat specials? 4 56 51.4 34 40.5
5: Very satisﬁed 25 22.9 40 47.6
N/A 1 .9 0 0
l= Not at all 0 0 I 1.2
Q55. How satisﬁed are you with 2 5 4.6 6 7.1
the way your store prepares for 3 39 20.2 7 8.3
increased sales experience 4 42 51.4 31 36.9
associated with special meat 5= Very satisﬁed 19 22.9 40 35.7
Jromotions? ' N/A 4 3.7 9 10.7

 

Table 18 shows the results of Question 4, which asked which times of the day,
were difﬁcult to keep the meat cases stocked. This question was asked because case-
ready meat products do not need to be processed and packaged on premise; rather, they
only require pricing and stocking in the meat display case. This is a critical point because

98

case-ready beef products require a minimum twenty minute bloom period. All stores
chose 4:00 PM as the most difﬁcult time to keep the meat case stocked. It is important to
note that 10 AM-l PM and 4 PM-7 PM periods were the most difﬁcult times to keep the
cases stocked The Q4 results are consistent with information gained ﬂom interviews
with meat cutters. The time between 4 PM and 7 PM can be the busiest times of the day

for grocers, which coincides when meat department personnel end production activities

 

for the day.

Table 18: Handling (Stocking) Case-Ready Meat Cases by Retail Store Type.
Case-Ready Handling Responses Corporate Licensed
Qgestions Prob Not Prob Prob Not Prob
Q4. When is it difﬁcult 9 AM-lO AM 108 1 74 10
to maintain adequately 10 AM-l PM 94 15 72 12
stocked meat cases? 1-4 PM 74 35 70 14

4 PM—7 PM 42 67 36 48
7 PM-until closing 94 15 58 26

 

4.2.3 Data Analysis Operational Issues Associated with Shrink

Fresh meat shrink is deﬁned as any dollar amount that is less than full-expected
retail price. Causes of shrink are of major concern for meat processors. They were
adamant about beneﬁts of case-ready meat programs: Numerous Activity Based Costing
research projects revealed the costs of case-ready meat products was not higher compared
with in-store processed meats when all in-store costs of processing meats are taken into
account. Moreover, it was learned that many retailers underestimate shrink in its
accounting systems.

There are four research propositions associated with shrink:

99

RPM.“ Store managers do not believe there is a signiﬁcant difference in

shrink levels when selecting ﬂesh meat product options.

RPng Store managers do not believe there is a signiﬁcant difference

between ﬂesh meat product options as they relate to managing the other

operational concerns and managing shrink levels.

RPmsx Store managers are able to distinguish the top three sources of

shrink.

RP17.SK Store managers do not believe there is a signiﬁcant difference

between ﬂesh meat product Options in formulating strategies for

controlling and addressing shrink.

The cross-tabulations unveiled several key points about the sources of shrink and
the associated relationships with either processing meats in-store or a case-ready meat
program Prior research discovered seven key sources of shrink associated with ﬂesh
meats. They include customer theft, employee theft, defective packaging, inadequate
workmanship, incorrect inventory issues, out-of-date product, and, for case-ready
products, the shrink associated with non-bloom packages of meat.

The seven issues listed are the basis for the shrink questions. Table 19 describes
the results. Over 79 percent of respondents thought customer theft was a signiﬁcant
source of shrink. There are decidedly different scores to this category of shrink between
stores. Corporate stores recorded 41.3 percent of their responses as a “big concern”
compared to ﬂanchise stores recording 28.6 percent. The only other question of shrink
that had signiﬁcant results was the question of workmanship. Corporate stores indicated

workmanship issues are as an important concern of shrink. More than 79 percent of store

100

managers thought that workmanship was a source of shrink Whereas, licensed stores
only 45 percent thought workmanship was a source of shrink with only 12 percent of that
score above the median.

These results show a clear difference between store types and how they view
workmanship as a source of shrink. The case study revealed that some case—ready meat
products were being discounted and discarded due to inconsistent and/or poor
workmanship issues coming ﬂom the processor. These results conﬁrm the case study
ﬁndings about workmanship of case-ready meat products being a key source of shrink.
This is particularly important as corporate stores market 100 percent case-ready meat
products.

Shrink resulting ﬂom employee theft (Question 45) reveals several important
points. First, close to 75 percent of both store types rated employee theﬁ in the range of 3
to 5. This outcome would substantiate managers concern that they do not have an
internal theft problem; however, research conducted by the supermarket industry
contradicts this ﬁnding. An industry report conducted by the National Supermarket
Research Group and sponsored by the National Grocers Association entitled 2003 —2004
Shrink Survey found that the largest single source of shrink in grocery stores is ﬂom
employees. Employee theﬁ has increased signiﬁcantly the past ﬁve years; however,
controlling employee theft in a retail setting is difﬁcult and costly in terms of
apprehending and prosecution. Radio Frequency Identiﬁcation (RF ID) products have a
low cost tag that could be implanted into each case-ready package tray that accounts for

the products movement along a supply chain (including aspects of inventory/product

101

traceability); and can be used to sound an alarm if the product moves outside the store

without payment

Table 19: Shrink by Retail Store Type.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Operational Issues Associated Responses Conrorate Licensed
with Shrink N '/o N °/o
1= Big Concern 45 41.3 24 28.6
2 22 20.2 15 17.9
Q39. Customer theﬂ 3 24 22.0 23 27.4
FAILED TO REJECT 4 17 15.6 18 21.4
5= Not a Concern 0 0 4 4.8
N/A 1 0.9 0 0
1= Big Concern 10 9.2 10 11.9
2 30 27.5 10 l 1.9
Q40. Defective and damaged 3 40 36.7 26 31.0
packaging 4 23 21.1 30 35.7
5= Not a Concern 5 4.6 7 8.3
N/A 1 0.9 1 1.2
1= Big Concern 15 13.8 2 2.4
2 40 36.7 12 14.3
Q41. Workrnanship issues 3 31 28.4 24 28.6
4 21 19.3 27 32.1
5= Not a Concern 1 0.9 19 22.6
N/A 1 0.9 0 0
1: Big Concern 7 6.4 4 4.8
2 14 12.8 14 16.7
Q42. Incorrect inventory issues 3 37 33.9 17 20.2
4 43 39.4 27 32.1
5= Not a Concern 7 6.4 22 26.2
N/A 1 0.9 0 0
1: Big Concern 18 16.5 12 14.3
2 33 30.3 15 17.9
Q43. Non-bloom issues/ground 3 34 31.2 1 1 13.1
beef 4 14 12.8 13 15.5
5: Not a Concern 9 8.3 12 14.3
N/A 1 0.9 21 25.0
1=Big Concern 19 17.4 11 13.1
2 39 35.8 15 17.9
Q44. Non-bloom issues/beef cuts 3 33 30.3 9 10.7
4 14 12.8 9 10.7
5= Not a Concern 3 2.8 13 15.5
N/A 1 0.9 27 32.1
l= Big Concern 12 11.0 8 9.5
2 14 12.8 13 15.5
Q45. Employee theft 3 36 33.0 25 29.8
4 39 35.8 26 31.0
5= Not a Concern 7 6.4 12 14.3
N/A 1 0.9 0 0

 

 

Table 20 indicates that when comparing ﬂesh meat processed in-stores with case-
ready meat products, case-ready meats were superior in managing shrink, with case-ready

102

earning 48.7 percent compared to in-store with 37.3 percent. Managers of both store
types clearly indicated that case-ready meats possessed characteristics conducive to
managing shrink. This is an important ﬁnding for ﬁuther adoption efforts.

Table 20: Shrink Management as a Key Success Factor by Retail Store Type.

 

Key C-R Success Factors Responses Corporate Licensed
N % N '/o
l= In-Store 27 24.8 45 55.6
Shrink management 2=Case-Ready 65 59.6 29 35.8
3=No Difference 17 15.6 7 8.6

 

4.2.4 Data Analysis of Food Safety Operational Issues

The next section discusses the results of the food safety survey questions derived
ﬂom these three research propositions:

RP7.ps Store managers believe case-ready beef and pork products are safer

than products processed in stores.

RPus An outbreak of food borne illness associated with ﬂesh meat

purchased at either corporate and franchise stores would be catastrophic to

store sales.

RP9.ps Store managers’ perception is that customers who do not recognize

the safety of both products would respond by purchasing more ﬂesh beef

and pork products if they believed those products were the safest

available.

The role of food safety towards expanding the adoption rates of case-ready meats
was unclear during the video conference and case study phase of the thesis project. It
should be noted that food safety was not discounted in any way, but any speciﬁc role the

issue played in furthering adoption rates was unclear. Food safety is and always will be a

103

major concern for cattle producers, fresh meat processors, grocery retailers and
consumers.

Table 21 reveals the results of two very important food safety questions.
Question 14 investigates which alternative was perceived as being “safer.” The
combined results for store types indicates that the case-ready alternative 80.3 percent as
the “safest” meat product. Even the licensed stores selected case-ready as the safest
alternative. Since these stores sell in-store processed meats this is a promising result for
case-ready meats.

Question 18 explored the topic product traceability. There is a difference between
how respondents answered this question (see Table 21). Corporate stores believe case-
ready meat was a key success factor at 84.4 percent whereas only 67 .9 percent of the
licensed stores agreed Together, 77.4 percent of respondents thought that product
traceability as a key success factor were associated with the case-ready meat alternative.
The results found in question 14, show another positive attribute for case-ready meats.

Table 21: Food Safety and Product Traceability Factors by Retail Store Type.
Key C-R Success Factors Responses Corporate Licensed
N % N %
l= In-Store 2 1.8 15 18.5

 

Q14. Food safety 2=Case-Ready 101 92.7 54 66.7
. =No Difference 6 5.5 15 18.5

0=N/A 0 0 3 3.6

1= In-Store 5 4.6 15 18.5

Q18. Product traceability 2=Case-Ready 92 84.4 55 67.9
3=No Difference 12 11.0 1 1 13.6

0=N/A 0 0 3 3.6

 

Table 21 addresses food safety as a “key success factor” ﬂom only two factors;
overall food safety and product traceability. While the food safety issues have been
addressed earlier, product traceability is new component within the food safety category.

There is a substantial literature on the subject of fresh meat food safety and how it can be

104

improved and product traceability will be an option that helps ensure a safer supply of
ﬂesh meats in the long run. Product traceability is a component of acquiring more
information as it relates to exactly where in the supply chain the food safety incident
occurred. Further upsides to tracing product along a supply chain is the potential for
gathering more data at stops along the supply chain (including customers) and providing
information to all to improve product quality and safety.

In Table 22 speciﬁc questions concerning retail meat sales and problems
associated with food safety are analyzed. Only two questions were not signiﬁcant in the
chi square test at the .05 probability level, question 57 and question 59. A further review
of the data for those two questions reveals respondents ﬂom corporate and licensed stores
supplied similar responses.

1 Question 57 was interested in obtaining manager’s viewpoints on a possible food
safety issue occurring at their store and how that would affect overall sales. Over 91
percent of all respondents indicated that the occurrence would cause overall sales to fall
dramatically.

Question 58 asked about a food safety incident and the consequences the incident
would have only on meat sales. Over 70 percent of corporate store managers thought the
incident would jeopardize just meat sales. Close to 26 percent answered at the median
and above to ‘strongly agree,” which suggests that the incident would affect all sales.
While the licensed stores’ answers were more evenly distributed across the scale, these
were at and below the median, suggesting an incident would harm mainly meat sales.
Several explanations for these results include the lack of competition for customers

shopping at the SAL stores (there is little competition amongst grocery store retailers

105

located in these areas) and customers may not associate a meat food safety issue other
food purchases.

Question 59 explored the volume of meat sales associated with perceived safety
of the product. Over 91 percent of all respondents scored at the median and above,
revealing the respondents agreed their customers would indeed purchase more ﬂesh meat
products if they believed the products were the safest available. The percentages for each
interval in the Likert scale were roughly the same percentages for both store types and
thus not pass the chi square test. Managers’ perceptions of safe meat products towards
customers purchasing more ﬂesh meat are an important ﬁnding and consistent with the
literature review.

The important question in the survey addressing whether case-ready meats are the
safest available was covered in question 60. Over 92 percent of corporate and licensed
store managers indicated case-ready meats are the safest possible alternative. Corporate
stores “strongly agreeing” to the proposition were 45.9 percent of their total sample.
Licensed stores responded less enthusiastically, and the chi-square was signiﬁcant. The
difference between stores was found in the ranking above the median.

The last question in this series attempted to gather speciﬁc data as to whether the
respondents believe case-ready meat products are the safest. Corporate stores at the 98
percent level at the median and above felt they were the safest products. Licensed stores
were not as enthusiastic of the proposition. While over 70 percent were at or above the
median, there was 63 percent of the sample population at or below the median,
suggesting these store managers are not as convinced that case-ready products are the.

safest.

106

Table 22: Food Safety Operational Issues of Case-Ready by Retail Store Type.

 

 

 

 

 

Operational Issues Associated Responses Corporate Licensed

with Food Safety N % %
Q57. A food borne illness 1= Strongly Disagree 1 .9 l 1.2
outbreak associated with fresh 2 3 2.8 9 10.7
meat purchased at my store 3 9 8.3 12 14.3
would be catastrophic to 4 18 16.5 16 19.0
overall store sales. 5: Strongly Agree 77 70.6 45 53.6
FAIL TO ACCEPT P=.115 N/A 1 .9 l 1.2
l= Strongly Disagree 45 41.3 21 25.0
Q58. A food borne illness 2 21 19.3 25 29.8
outbreak associated with fresh 3 13 l 1.9 15 17.9
meat purchased at my store 4 9 8.3 14 16.7
would be catastrophic to meat 5= Strongly Agree 20 18.3 7 8.3
sales only. N/A 1 .9 2 2.4
Q59. My customers would l= Strongly Disagree 4 3.7 2 2.4
respond by purchasing more 2 7 6.4 8 9.5
ﬂesh beef and pork products if 3 22 20.2 20 23.8
they believed those products 4 39 35.8 22 26.2
were the safest available. 5= Strongly Agree 36 33.0 30 35.7
FAIL TO ACCEPT P: 660 N/A 1 .9 2 2.4
l= Strongly Disagree 1 .9 7 8.3
Q60. Case-ready beef and pork 2 l .9 8 9.5
products are “safer” than beef 3 12 1 1.0 30 35.7
and pork products processed in- 4 43 39.4 18 21.4
store. 5= Strongly Agree 50 45.9 16 19.0
N/A 2 1.8 5 6.0
l= Strongly Disagree 1 . .9 9 10.7
Q61. Case-ready meat products 2 l .9 11 13.1
are the safest meats available 3 24 22.0 33 39.3
4 49 45.0 13 15.5
5: Strongly Agree 32 29.4 13 15.5
N/A 2 1.8 5 6.0

 

4.2.5 Data Analysis of Workrnanship Operational Issues

The presence or absence of workmanship in meat products can be found in
several areas: well trimmed fat, the positioning of the meat product in the tray, uniformity
among all other packages, neatness in wrapping, and proper positioning of the label. The
workmanship operational questions in the survey were derived ﬂom these research

propositions:

107

RPuth Store managers are ﬁnding it more difﬁcult to hire qualiﬁed meat

cutters at competitive wages.

RPu.wx Implementing a case-ready meat program will demand additional

skills such as forecasting daily and weekly demand, along with proper

inventory management and less meat processing workmanship skills.

RP12.wx Case-ready beef and pork products compared to in—store

produced products are more cost effective.

RP13.WK Case-ready meats will have lower stock outs and therefore

maintain a higher rate of customer retention.

The workmanship category had several issues that were addressed in the research
propositions. The ﬁrst and obvious issue relates to the ﬁnished products consumers
purchase and the importance workmanship has on those consumer decisions. During the
case study retail executives and managers stressed the quality (not necessary the highest
possrble quality of meat) of meat products as being very important and closely related to
consumers choices of center of the plate proteins. Quality has a major impact on the
ﬁnished meat products that consumers purchase and quality of the product was directly
related to the “workmanship” involved in processing. While quality is implied for all
consumer products (and protected by consumer laws and regulations) the perception of
quality for ﬂesh beef products is on display in each and every package of beef. These
lessons on quality were ﬁrst learned by retailers in consumer satisfaction surveys that
revealed the ﬁrst criteria consumers use to judge an entire grocery store is the produce
department, followed by the meat department. After these results were discovered,

retailers around the country began to dramatically improve the quality of the produce

108

section by upgrading the produce departments and changing the size and layout by
adding new ﬁxtures, and using continuous replenishment strategies. It may be argued
that meat departments are more difficult to enhance; however, research has demonstrated
the importance of making efforts to do so. The ﬁrst two operational issues associated
with the quality of workmanship were addressed in product appearance and product
consistency. Table 23 reveals both corporate and licensed stores rated fresh meat
processed in-stores to have better product appearance. Over 47 percent corporate stores
rated in-store product as being the better alternative. This statistic is revealing because
corporate stores market mainly case-ready meat products. This result is stipulating
respondents view in-store processed meats as possessing a higher level of workmanship.

Choosing which ﬂesh meat alternative was superior in product consistency
followed the option each store is selling. Only twenty-two percent of case-ready
managers thought meat processed in-store was more conSistent than case-ready meat
products. Whereas, twenty-six percent of the licensed stores thought case-ready was the
superior alternative in product consistency. Clearly, the workmanship issues residing
with the case-ready meat program alternative need to be addressed in any efforts towards
improving adoption rates.

Table 23: Workrnanship Operational Issues by Retail Store Type.

 

 

Key C-R Success Factors Responses Corporate Licensed
N % N %
1= In-Store 52 47.7 65 77.4
Product Appearance 2=Case-Ready 39 35.8 7 8.3
3=No Difference 18 16.5 9 10.7
1= In-Store ' 24 22.0 43 51.1
Product Consistency 2=Case-Ready 64 58.7 21 25.1
3=No Difference 21 19.3 17 20.2
0=n/a 0 0 3 3.6

 

The Likert scale questions for workmanship can be found in Table 24 (questions

72, 73 and 74). Question 72 addressed the potential problem for stores that continue to
109

cut and process meats in-stores, ﬁnding a consistent/reliable source of labor. The other
two questions were developed to address the concern of how to manage a meat
department if it was no longer necessary to employ meat cutters. One key point
discovered in the case study on this subject was that the meat department requires capable
managers. The case study revealed that current meat department personnel would need
varying levels of management training in order to succeed in these new management
positions. However, meat operations experience would facilitate their preparation for
effectively managing the department.

The results for question 72 revealed that 28.5 percent agreed there is perceived
difﬁculty in ﬁnding qualiﬁed meat cutters. Many corporate store managers chose not to
answer because they do not hire meat cutters. Over 57 percent of licensed store managers
agreed it was becoming more difﬁcult in attracting qualiﬁed meat cutters. It was
anticipated that the results for this question would show an even higher agreement;
nevertheless, this result paralleled what was discovered in the literature review.

Table 24 highlights the results of question 73 that asked respondents about meat
department personnel being retrained to manage all perishables, in addition to ﬂesh
meats. Over 68 percent of managers ﬂom corporate stores thought this approach could
be achieved. Licensed store managers had similar viewpoints with over 67 percent
agreeing. The results could provide valuable insights when implementing and managing
a case-ready meat strategy.

Finally, Table 24 addresses the question of meat department personnel shifting
responsibilities towards forecasting and inventory management which is an essential

aspect of managing case-ready meat programs. This question was included in the study

110

as a result of meat processors’ expressed interest in better understanding demand for
ﬂesh meat at the store level. Having trained personnel dedicated to the important task of
forecasting and corroborating micro demand models for inventory would greatly impact
both retailer and processor. Properly executed, this new position could begin to bring
consumer’s tastes and preferences into micro-demand models. Processors expressed the
need for more consumer input for the products they produce especially, case-ready meats.
These results demonstrate both store types either agree or strongly agree that current
personnel could accomplish these new forecasting and inventory management tasks; as
well as possibly managing the entire perishables (meat and produce) category.

Table 24: Additional Workrnanship Operational Issues by Retail Store Type.

 

 

 

Case-Ready Workmanship Responses Corporate Licensed
N % N %

1= Strongly Disagree 9 8.3 8 9.5
Q72. It has become 2 10 9.2 10 11.9
increasingly more difficult to 3 19 17.4 9 10.7
ﬁnd qualiﬁed meat cutters 4 13 l 1.9 17 20.2
for my store. 5: Strongly Agree .18 16.5 31 36.9
N/A 40 36.7 9 10.7
Q73. A traditional meat 1: Strongly Disagree 6 5.5 5 6.0
department manager and or 2 7 6.4 7 8.4
meat cutter could be trained 3 l 1 10.1 9 10.8
to manage all perishables not 4 33 30.3 29 34.9
just meats. 5= Strongly Agree 42 38.5 27 32.5
N/A 10 92 6 7.2
Q74. Implementing a case- ]: Strongly Disagree 2 1.8 2 2.4
ready meat program will shift 2 1 .9 5 6.0
meat management ﬂom 3 21 19.3 14 16.9
production activities to 4 43 39.4 35 41.7
forecasting and inventory 5: Strongly Agree 28 25.7 25 29.8
mgmt of ﬂesh meats. N/A 14 12.8 3 3.6

 

lll

4.3 Multiple Regression Analysis

Six regression models were developed, one for each operational issue (except the
case-readyl6 meats operational issue) and one model exploring a combination of the
operational issues. The dependent and independent variables for each model were chosen
ﬂom the research propositions which were based upon the interviews with retail and meat
processing industry executives, visits to retail Distribution Centers, grocery stores
throughout Michigan, Illinois Indiana and Wisconsin and the literature review. Both the
dependent and independent variables are based on Likert scale questions.

The use of Likert scale data as a dependent variable has its detractors and
advocates. Likert scale data is considered interval data, other deﬁnitions for Likert scale
include categorical and ordinal. To use likert scale data as a dependent variable in
multiple regression a key assumption is made about the intervals of the Likert scale. The
key assumption is there is an equal distance between points in the scale. For example in
this research project a ﬁve point Likert scale is used, where the implications of the scale
are:

1=Strongly Disagree 2=Disagree 3=Neutra1 4=Agree 5=Strongly Agree
From the above Likert scale the differences between each interval in the scale is the same
and equal to 1 and therefore sufﬁciently continuous between intervals.

Several research articles employing the multiple regression technique have used
Likert scale data as both dependent and independent variables (Graham and Nafukho,

2006; Toffalo and Pederson, 2005; Champion and Leach, 1989). These articles cite the

importance of using Likert scale data as the dependent variable for its capability of

 

'6 It was determined this operational issue was signiﬁcantly covered by the other regression models.

112

capturing the “richness” found within the Likert scale. Researchers who have performed
due diligence in draﬂing research propositions can yield one seminal question for the
purpose of “capturing” the essence of the topic of interest. For this project six multiple
regression models were draﬂed and each model has a seminal Likert scale question that
incorporates the key dimensions and perceptions of this topic collapsed into one interval

scale dependent variable. The next sections will cover these multiple regression models.

4.3.1 Multiple Regression Model: Food Safety Operational Issues

The dependent variable for this regression is found in Table 25 and it was derived
ﬂom several literature review articles hypothesizing case-ready meats were demonstrably
safer than the ﬂesh meat processed in-stores (Major, 2007 and Brody, 2005). Store
managers and industry executives also conﬁrmed the realities of processing meats in-
stores over time has proven to be more difﬁcult to maintain when zero tolerances for
safety is required.

Table 25: Summary of Data: Food Safety Regression

 

Variable Description Mean Standard
Deviation

Dependent Variable:

Q61 Case-ready meat products are the safest meats available. 3.76 1.285

1=Strongly Disagree to 5=Strongly Agree

Independent Variables:

Q57 A food borne illness outbreak associated with ﬂesh meat purchased at my

store would be catastrophic to overall store sales. 4.33 1.071

1=Strongly Disagree to 5=Strongly Agree

Q58 A food borne illness outbreak associated with ﬂesh meat purchased at my

store would be catastrophic to meat sales only. 2.43 1.453
1=Strongly Disagree to 5=Strongly Agree

Q59 My customers would respond by purchasing more fresh beef products if

they believed those products were the safest available 3.8] 1.171
1=Strongly Disagree to 5=Strongly Agree

Q72 It has become increasingly more difficult to ﬁnd qualiﬁed meat cutters for

my store. 2.62 1.946
1=Strongly Disagree to 5=Strongly Agree
Q2 Store Type (Dummy Variable) .44 .497

 

For N=193 Observations

113

4.3.2 Independent Variables in the Food Safety Regression Model

The explanatory variables can be seen in Table 25 above, where the ﬁrst two
variables were drafted in response to the second research proposition concerning the
impact of food borne illness associated with meat purchased ﬂom the respondent’s store.
The two concerns of these variables relate to lost sales of the entire grocery store and lost
sales within the meat department. Many fast food restaurants associated with customers
contracting food home illnesses have experienced a near loss of the entire brand for the
chain. Consumers have a difficult time separating the food borne illness ﬂom other menu
items in restaurants/ fast food establishments, whereas retail grocery stores have been able
to isolate and contain any losses to the meat department, while maintaining a majority of
revenue ﬂom other categories within the grocery store.

The third explanatory variable is related to the third research proposition about
consumer’s demand for safer meat products. From a marketing perspective case-ready
meats has an important feature; namely, product processed under the “safest” possible
protocols and conditions. And, the question relates to potential sales increases resulting
ﬂom this safety feature. From the fairly high mean score for this question it appears store
managers agree with the proposition.

Many store managers began their retail careers in the meat department; working
their way into management ranks. Store managers observe that pool of talent to draw
ﬂom for future meat cutters continues to dwindle; and that it is becoming more difﬁcult
to ﬁnd a steady source of meat cutters to process meats in-stores (Keith, 2006). The

dwindling pools of talent with more meat department personnel seeking higher paying

114

management positions has caused retail grocery executives and top management to

contemplate the long term disadvantages associated with in-store processing.

4.3.3 Regression Results for Food Safety
Table 26 shows the regression results for the food safety model, including the
variables of interest, standardized coefﬁcients, standard errors, t—statistics and P-values.

Table 26: Regression Results: Food Safety

 

Variable Estimated Standard t- P- Expected
Coefﬁcient Error Statistic Value Srg’ n

Intercept 1.842 .435 4.238 .000

Q57 Foodbome illness all sales .267 .080 3.330 .001 Positive

Q58 Foodbome illness to just meat .139 .057 2.451 .015 Positive

sales

Q59 More sales if meat products were .215 .071 3.022 .003 Positive

safer

Q72 Difficulty ﬁnding cutters .011 .043 .257 .798 Positive

Q2 Store Type (Dummy Variable) -.374 .169 -5.715 .000 Positive

 

F=l6.051 (p-value=.000) [8:300 Adjusted 18:.282

The overall statistical signiﬁcance of this model is acceptable. Within the
ANOVA results the F-stat is 16.051 which was statistically signiﬁcant below the .01
level (p-value=.000), where the hypothesis of the independent variables having zero
explanatory power or no effect over the dependent variable is rejected. In fact, three out
of four variables were signiﬁcant, including the dummy variable for store type. All
variables were signiﬁcant at the 1 percent level, except the variable for ﬁnding a steady
supply of meat cutters. The R2 (Coefficient of Determination) for this model was a
respectable 30 percent and adjusted R2 was 28.2 percent, where the independent variables
explained 30 percent of the variability of case-ready meats being considered the safest
meats available.

The results of this regression provide a better understanding of the food safety

factors resulting in case-ready programs being viewed the safest meat alternative. The

115

signs for two of the independent variables associated with food borne illness make sense
ﬂom a manager’s perspective. If managers believe the more effective case-ready is
relative to product safety as compared to in-store processing, the more likely they are to
accept that alternative.

The negative sign for the dummy variable along with the signiﬁcant p-value
means the licensed stores view case-ready meat products are “safer” than in-store

processed meats.

4.3.4 Multiple Regression Model: Workrnanship Operational Issues

The dependent variable of interest for this regression model centers on the
difﬁculty of recruiting steady supplies of meat cutters to process meat in-stores. Meat
cutters are directly responsible for the quality of the end product and all of the tangible

and intangible beneﬁts associated with their efforts and resources at their disposal.

4.3.5 Independent Variables in the Workrnanship Regression Model

The explanatory variables of interest for this regression model can be viewed in
Table 27. The ﬁrst explanatory variable concerns case-ready meat processors and
grocery store executives believing meat department personnel could with the requisite
training manage the entire perishables category.

The reality of it being difﬁcult to ﬁnd meat cutters is especially problematic for
corporate store senior management as it pertains to producing high quality meat products
on a daily basis. A partial explanation is the rising costs of workers compensation, the
second independent variable of interest.

Also impacting the dependent variable are the tangible and intangible shrink

losses associated with workmanship issues. In retailing to document and manage the

116

tangible and intangible factors of shrink is a daunting task, a few examples will be
provided. The tangible factors include waste in processing/cutting meat products ﬂom
subprimals into individually portioned and packaged products. An example of the
intangibles associated with workmanship of meat products are inconsistent portions and
defective packages. These products typically are not sold before the “sell by date” has
expired, become a shrink statistic and harm the image of the store.

Table 27: Summary of Data: Workrnanship Regression

 

Variable Description Mean Standard
Deviation

Dependent Variable:

Q72 It has become increasingly more difficult to ﬁnd qualiﬁed meat cutters 2.63 1.951

for my store.

1=Strongly Disagree to 5=Strongly Agree

Independent Variables:

Q73 A traditional meat department manager and / or meat cutter could be

trained to manage all perishables not just meats. 3.60 1.569

1=Strongly Disagree to 5=Strongly Agree

Q76 Cost of workers compensation for employees that cut meat is becoming

cost prohibitive. 2.92 1.552
1=Strongly Disagree to 5=Strongly Agree

Q41 Workmanship as a source of shrink ,

1=Big Concern to 5=Not a Concern 2.99 1.160
Q2 Store ije (Dummy Variable) .43 .497

 

For N=l 93 Observations

4.3.6 Regression Results: Workmanship

Reviewing Table 28 below reveals the overall statistical signiﬁcance of this
model is acceptable and checking the ANOVA results the F-stat is 22.856 and it was
statistically signiﬁcant below the .01 level (p-value=.000), where the hypothesis of the
independent variables having zero explanatdry power or no effect over the dependent
variable is rejected. In fact, three out of four variables were signiﬁcant, including the
dummy variable for store type. All variables were signiﬁcant at the 1 percent level,
except the variable for about shrink associated with workmanship. While this variable is
not within the desired less than 5 percent category for being signiﬁcant, it is a respectable

ll7

6.8 percent. The R2 (Coefﬁcient of Determination) for this model is respectable at 32.8
percent and adjusted R2 is 31.4 percent.

The results of this regression are positive for developing a better understanding of
the factors associated with workmanship. All of the signs for the independent variables
associated with workmanship make sense from a manager’s perspective and are
consistent with the literature review.

Finally, given the research propositions for workmanship, this model shows
variables of interest that are signiﬁcant towards explaining the dependent variable along
with a respectable R2 makes an encouraging interpretation of this model.

Table 28: Workmanship Regression Results

 

Variable Estimated Standard t-Statistic P-Value Expected
Coefﬁcient Error SigE

Intercept .229 ..461 .497 .62 1

Independent Variables

Q73 Meat cutters manage .336 .078 4.309 .000 Positive

perishables

Q76 Cost of workers. Comp. .384 .079 4.846 .000 Positive

Q41 Workrnanship -.208 .113 -1.834 .068 Negative

Q2 Store type (Dummy Variable) .404 .264 6.006 .000 Positive

 

=22.856 (p-value=.000) R2=.328 Adjusted R3=.314

118

4.3.7 Multiple Regression Model: Shrink Operational Issues

Table 29: Summary of Data: Shrink Multiple Regression

 

Variable Description Mean Standard
Deviation

Dependent Variable:

Q49 How satisﬁed are you with the way your store records (accounts for) 3.64 1.114

every fresh meat package not sold at full retail?
1=Not at all to 5=Very satisﬁed
Independent Variables:
Q7 How satisﬁed are you with the way your store veriﬁes that fresh meat
deliveries from the Distribution Center (DC) are within acceptable temperature 3.79 .925
ranges?
1=N0t at all to 5= Very satisﬁed
Q8 How satisﬁed are you with the way your store transfers fresh meat products
from the receiving area to the walk-in coolers? 4.42 .71 l
1=Not at all to 5=Very satisﬁed
Q9 How satisﬁed are you with the way your store writes the delivery date on
each fresh meat box received from the Distribution Center? 4.13 1.055
l=Not at all to 5= Very satisﬁed
Q50 How satisﬁed are you with the way your store stocks fresh meat cases to
ensure that First-In-First-Out (FIFO) methods are followed? 4.26 .774
1=N0t at all to 5= Very satisﬁed
Q53 How satisﬁed are you with the way your store maintains adequate fresh

meat case levels of fresh beef? 3.92 .856
1=Not at all to 5= Very satisﬁed
Q39 Customer theft (source of shrink) 2.30 1.205
Q40 Defective and damaged packaging (source of shrink) 2.95 1.115
Q41 Workrnanship (source of shrink) 2.99 1.157
Q42 Incorrect inventory issues (source of shrink) 3.39 1.113
Q43 Non-bloom issues associated with case-ready-ground beef (source of 2.46 1.479
shrink)
Q44 Non-bloom issues associated with case-ready beef cuts (source of shrink) 2.26 1.456
Q45 Employee theft (source of shrink) 3.17 1.144
Q46 Out of date product (source of shrink) 2.99 1.170
£2 Store Type (Dummy Variable) .44 .497

 

For N=193 Observations

4.3.8 Independent Variables in the Shrink Regression Model

There are several explanatory variables for this regression model shown in Table
29. These independent variables fall within three distinct categories of shrink. The ﬁrst
relates to product’s arrival at the shipping dock (questions 7-8-9). This category is about
discerning proper procedures for the meat: either case-ready or boxed beef moving from

the receiving dock to the walk-in storage coolers. The second category concerns several

119

of fresh meat’s inventory issues (questions 50 and 53); and the last is about speciﬁc

sources of shrink (questions 39-46).

4.3.9 Regression Results for Shrink

Table 30 shows the regression results for the shrink model, including the variables
of interest, standardized coefficients, standard errors, t-statistics and P-value.

The overall statistical signiﬁcance of this model is acceptable. Within the
ANOVA results the F-stat is 3.756 and it was statistically signiﬁcant which is well below
the .01 level (p-value=.000), where the hypothesis of the independent variables having
zero explanatory power or no effect over the dependent variable is rejected. Only one of
the explanatory variables was signiﬁcant below the 5 percent value. This was the
variable relates to inventory arriving to a store and the concept of First-in-First-Out
(FIFO). One other variable concerning satisfaction about keeping inventory levels of the
retail meat cases at satisfactory levels. This variable had a signiﬁcance level of 5.4
percent, barely above the 5 percent acceptable range. The R2 (Coefﬁcient of
Determination) for this model was only 22.8 percent and adjusted R2 was 16.7 percent.

The results of this regression were disappointing given what was suggested by the
literature and the involvement of stores and DC from four states were involved in the
development of these independent variables. There is some consolation in knowing that
the two signiﬁcant variables were identiﬁed as being critical towards proactively
minimizing shrink and is a critical factor in maximizing the shelf life of ﬂesh meat, either
in-store processed or case-ready. Inventory management of all fresh meat products

begins with the theory of FIFO. The second independent variable that related to

120

maximizing store sales and minimizing shrink by keeping the meat cases adequately
stocked at all times.

Table 30: Regression Results: Shrink

 

 

Variable Estimated Standard t-Statistic P- Expected
Coefﬁcient Error Value Srg' n

Intercept .196 .632 .310 .757
Independent Variables
Q7 Veriﬁes delivery date .064 .092 .698 .486 Negative
Q8 Transfers meat into coolers -.027 .123 -.216 .829 Positive
Q9 Writes delivery dates .097 .082 1.183 .238 Positive
Q50 Satisﬁed with FIFO .458 .115 3.989 .000 Positive
Q53 Satisﬁed meat case levels .195 .100 1.940 .054 Positive

' Q39 Customer theﬁ (source of shrink) .094 .074 1.277 .203 Positive
Q40 Defective and damaged -.056 .082 -.685 .494 Positive
packaging (source of shrink)
Q41 Workrnanship (source of shrink) -.024 .088 .274 .785 Negative
Q42 Incorrect inventory (source of .018 .086 .212 .832 Positive
shrink)
Q43 Non-bloom grd beef (source of .054 .083 .644 .521 Positive
shrink)
Q44 Non-bloom beef cuts (source of .001 .083 .010 .992 Negative
shrink)
Q45 Employee theﬁ (source of shrink) -.031 .084 -.373 .710 Positive
Q46 Out of date product (source of .070 .080 .873 .384 Positive
shrink)
Q2 Store type (Dummy Variable) -.083 .194 -.955 .341 Positive

F=3.756 (p-value=.000) R3=.228 Adjusted 18:. 167

4.3.10 Multiple Regression Model: Packaging Operational Issues

The dependent variable for packaging operational issues for this regression is
found in Table 31 which is based upon a literature review of the case-ready packaging
alternative. Store managers and industry executives were mixed in their assessment of
the packaging options of case-ready meats. They stated that their opinions were based
upon years of producing and marketing in—store meats. The packaging dependent
variable encapsulates the research propositions related to “case-ready meat

attractiveness.”

121

Table 31: Summary of Data: Packaging Regression

 

Variable Description Mean Standard
Deviation

Dependent Variable:

Q70 Case-ready beef products are packaged in attractive trays. 3.40 1.148

1=Strongly Disagree to 5=Strongly Agree

Independent Variables:

Q63 High-oxygen/high head space packaging of case-ready ground

beef is not a detriment to sales. 2.76 1.285

1=Strongly Disagree to 5=Stronglv Agree

Q65 High-oxygen/high head space packaging of case-ready beef

cuts is not a detriment to sales. 2.69 1.289
1=Strongly Disagree to 5=Str0ngly Agree

Q68 Case-ready meat products that require blooming cause

operational problems. 2.5 l l .21 5
l=Strongly Disagree to 5=Strongly Agree

Q69 My customers know the difference between case-ready

packaging and in-store packaging. 3 .20 1.309
1=Strongly Disagree to 5 =Strongly Agree
Q2 Store Type (Dummy Variable) .43 .497

 

For N=l 93 Observations

4.3.11 Independent Variables in the Packaging Regression Model

The explanatory variables are shown in Table 32. From interviewing store
managers and DC management, these independent variables were identiﬁed as relating to
the research propositions for packaging. These four questions (variables) were repeated
throughout visits to stores and DC when discussing case-ready meat packaging.

The ﬁrst two questions were drafted in response to the packaging research
proposition about store managers and DC believing case-ready meat packaging using
high oxygen/ high head space is detriment to fresh meat sales. These variables related to
meat processors who at the time required this extra head space to successfully “modify”
the packaging atmosphere to obtain the additional days of shelf life. From their
perspective, this type of packaging alternative was too different from the packaging of in-
store meats and would encounter customer resistance. Offsetting these negatives is the
beneﬁt of the extra headspace to protect the meat from being unnecessarily bruised in

shipping, while being stocked in meat cases and handled by consumers. If the meat was

122

bruised in any way, degradation would begin and the extra shelf life would not be
beneﬁcial.

The blooming variable was in response to store managers and DC management
experiencing certain case-ready meat products needing time to bloom before the package
was ready to be displayed. These times were non-negotiable for certain products,
especially the beef products. The blooming phase was required to bring the product ﬁ'om
a dormant stage where the product was a purple color (resulting ﬁom a complete lack of
oxygen) to a bright cherry red color (in the presence of oxygen). Many retailers shared
negative experiences of not “blooming” the case-ready beef products in time to keep
empty meat cases fully stocked.

Retailers and processors alike value customers’ opinions of packaging. Store
managers who are in contact with customers on daily basis realize those customers are
able to identify where meat is packaged by the type of packaging displayed.

The third explanatory variable relates to the third research proposition concerning
consumers’ demand for safer meat products. From a marketing perspective retailers on
case-ready programs would like to market the attributes of the product being processed
under the “safest” possible protocols and conditions in order to maximize sales. From the
fairly high mean score for this question it looks like store managers agree with the

proposition within the question.

123

Table 32: Packaging Regression Results

 

Variable Estimated Std t- P- Expected
Coefﬁcient Error Statistic Value Sig!
Intercept 2.882 .289 9.984 .000 Positive
Q63 High oxy grd beef .181 .115 1.576 .117 Negative
Q65 High oxy beef cuts -.093 .119 -.784 .434 Negative
Q68 Blooming problems .092 .071 1.298 .196 Positive
Q69 Cust. know pkg .091 .068 1.350 .179 Positive
Q2 Store type (Durrrrny Variable) ~.603 .184 ~3.274 .001 Positive

 

F=5.039 (p-value=.000) R2=.119 Adjusted R2=.096

4.3.12 Regression Results for Packaging

Table 32 above shows the regression results for the packaging model. The overall
statistical signiﬁcance of this model is not acceptable. Within the ANOVA results the F-
stat is 5.039 and it was statistically signiﬁcant below the .01 level (p-value=.000), where
the hypothesis of the independent variables having zero explanatory power or no effect
over the dependent variable is rejected. The problem with this model is that all of the
explanatory variables are not signiﬁcant at the 5 percent level. The R2 (Coefﬁcient of
Determination) for this model was a very low 11.9 percent and the adjusted R2 is even

lower at 9.6 percent. Overall, there are inconclusive results regarding packaging options.

4.3.13 Multiple Regression Model: Handling Operational Issues

The dependent variable for this regression is found in Table 33 and it is designed
as a proxy for all four research propositions, where the term handling is deﬁned as the
human interaction/treatment of meat products.

Meat department personnel and grocery store management are in full agreement
that amply stocked meat display cases throughout the day is the main goal. A daily
challenge for stores‘that process meat in stores is keeping the cases full. Because of the
multi-step processes it takes to achieve this goal, one of the biggest draws to case-ready

meats is the ability of any store personnel to keep the cases full, 24/7. In stores where

124

they process meat, if a particular item is sold out and the butcher/meat cutter has left the
department, slots in the meat cases will remain empty until the next day, with consequent
lost sales. Moreover, many customers do not have substitutes for particular cuts and

grinds of fresh beef and may elect to shop elsewhere.

4.3.14 Independent Variables in the Handling Regression Model

The explanatory variables for the handling regression model are shown in
Table 33. The ﬁrst three variables are concerned with handling the product after it is
delivered to the store. These questions are particularly pertinent to stores that sell case-
ready beef where temperature tolerances are much stricter than for boxed beef. The next
three variables are concerned with standard operating procedures for handling fresh meat
to ensure correct inventory practices are followed These handling procedures are
necessary to maximize shelf life and manage shrink. The last two variables are related to

maximizing fresh meat sales during the ﬁrst of the month.

125

Table 33: Summary of Data: Handling Operational Issues

Variable Description Mean Standard
Deviation

Dependent Variable:
Q53 How satisﬁed are you with the way your store maintains adequate fresh 3.92 .856
meat case levels.

1=Not at all to 5= Very satisﬁed
Independent Variables:
Q7 How satisﬁed are you with the way your store veriﬁes that fresh meat
deliveries from the Distribution Center (DC) are within acceptable temperature 379 .925
ranges?

1=Not at all to 5= Very satisﬁed
Q8 How satisﬁed are you with the way your store transfers fresh meat products
from the receiving area to the walk-in coolers? 4.42 .71 1
1=Not at all to 5= Very satisﬁed
Q9 How satisﬁed are you with the way your store writes the delivery date on
each fresh meat box received from the Distribution Center? 4.13 1.055
1=Not at all to 5= Very satisﬁed
Q50 How satisﬁed are you with the way your store stocks fresh meat cases to
ensure that First-In-First-Out (FIFO) methods are followed? 4.26 .774
1=Not at all to 5= Very satisﬁed
Q51How satisﬁed are you with the way your store maintains adequate
inventory levels of fresh beef products stored in walk-in coolers 4.06 .795
1=Not at all to 5=Very satisﬁed
Q52 How satisﬁed are you with the way your store manages fresh beef
inventory to ensure the oldest product in the walk-in cooler is stacked in a way 4.17 .905
it is the next product to be picked for restocking.

1=Not at all to 5=Very satisﬁed ,
054 How satisﬁed are you with the way your store prepares for the increase
sales experienced during the ﬁrst days of each month. 4.09 .846
1=Not at all to 5= V erv satisﬁed
Q55 How satisﬁed are you with the way your store prepares for increased sales

experienced with special meat promotions. 3.62 1.306
1=Not at all to 5= Very satisﬁed
Q2 Store Type (Dummy Variable) .44 .497

 

For N=193 Observations

4.3.15 Regression Results for Handling

Table 34 shows the regression results for the handling model.

126

Table 34: Handling Regression Results

 

 

Variable Estimated Standard t-Statistic P-Value Expected
Coefﬁcient Error ﬂgn
Intercept .484 .364 1.331 .185
Q7 Veriﬁes temp .019 .056 .333 .740 Negative
Q8 Transfers meat to walk-in .042 .074 .551 .575 Positive
Q9 Writes date -.013 .050 -.256 .798 Positive
Q50 Satisﬁed FIFO -.065 .082 -.796 .427 Positive
Q51 Satisﬁed inventory levels .123 .084 1.464 .145 Positive
Q52 Stock walk-in cooler .335 .069 4.843 .000 Positive
Q54 Store prepare FOM .335 .075 4.476 .000 Positive
Q55 Satisﬁed prep for sales .055 .040 1.383 .168 Positive
Q2 Store type (Dummy Variable) .057 .103 .949 .344 Positive

 

F=19.300 (p—value=.000) R’=.487 Adjusted R’=.462

The overall statistical signiﬁcance of this model is acceptable. Within the
ANOVA results the F-stat is 19.300 and it was statistically signiﬁcant below the .01 level
(p-value=.000), where the hypothesis of the independent variables having zero
explanatory power or no effect over the dependent variable is rejected. However, only
two variables in the handling regression were signiﬁcant. These are the variables about
satisfaction of handling fresh meat products to ensure the walk-in storage room is
organized for employees to pull the oldest product ﬁrst and the last variable that relates to
how well stores prepare for increased sales at the ﬁrst of the month.

The R2 (Coefﬁcient of Determination) for this model was 48.7 percent and the
adjusted R2 was 46.2 percent. The results of this regression are positive for developing a
better understanding of two key factors related to handling of fresh meats and the

satisfaction stores have towards maintaining adequate fresh meat case levels.

4.3.16 Multiple Regression Model: All Operational Issues
The ﬁnal regression model was developed from a combination of research
propositions. The dependent variable for this model tests the hypothesis that case-ready

beef products compared to “in-store” produced meat products are more cost effective (see

127

Table 35). The independent variables of interest and their respective research
propositions are listed below.

Table 35: Summary of Data: All Operational Issues Regression

Variable Description . Mean Standard
Deviation

Dependent Variable:

Q75 Case-ready beef products compared to “in-store” produced meat products 2.96 1.471

are more cost effective.

1=Strongly Disagree to 5=Strongly Agree

Independent Variables:

Q22 Case-ready ground beef products are available in the pack sizes my

customers prefer. 3.38 1.395

1=Strongly Disagree to 5=Strongly Agree

Research Proposition: Case-Ready Issues

Q42 Incorrect inventory issues are a source shrink.

1=Big Concern to 5=Not a Concern 3.39 1.113

Research Proposition: Shrink Issues

Q53 How satisﬁed are you with the way your store maintains adequate fresh

meat case levels of fresh beef product? 3.92 .856

1=Not at all to 5= Very satisﬁed

Research Proposition: Handling Issues

Q60 Case-ready beef products are safer than beef products processed in-stores.

l=Strongly Disagree to 5=Strongly Agree 3.76 1.285

Research Proposition: Food Safety Issues

Q70 Case-ready beef products are packaged in attractive trays.

1=Strongly Disagree to 5=Strongly Agree 3.39 1.146

Research Proposition: Packaging Issues ‘

Q76 Cost of workers compensation for employees that cut meat is becoming

prohibitive 2 .91 1 .5 54

1=Strongly Disagree to 5=Strongly Agree

Research Proposition: Workrnanship Issues

Q2 Store Type (Dummy Variable) .44 .497

 

For N=193 Observations

4.3.17 Regression Results for All Operational Issues

The overall statistical signiﬁcance of this model is within the acceptable range.
The ANOVA results the F-stat is 14.953 and it was statistically signiﬁcant below the .01
level (p-value=.000), where the hypothesis of the independent variables having zero
explanatory power or no effect over the dependent variable is rejected. Indeed, all of the

variables of interest were signiﬁcant except the dummy variable for store type.

128

The R2 (Coefﬁcient of Determination) for this model was 36.1% and the adjusted
R2 was 33.7 percent, where the independent variables and one dummy variable explained
36.1 percent of the variability of case-ready meats being more cost effective.

The results for this regression are positive for developing a better understanding
of the operational factors that help explain case-ready meats being a cost effective
alternative for marketing fresh beef products. The signs for all of the independent
variables are what were predicted. Initially the sign for the shrink variable (Q42) was
determined to be positive. If incorrect inventory issues for fresh meat increase (this
would be negative) then case-ready programs would be considered more cost effective.

Table 36: All Operational Issues Regression Results

 

Variable Estimated Standard t-Statistic P-Value Expected
Coefﬁcient Error Sign
Intercept -.252 .596 -.423 .673
Q22 Grd beef pack sizes .190 .073 2.591 .010 Positive
Q428hrink inventory issues -.2 l 5 .081 -2.653 .009 Negative
Q53 Satisﬁed with meat case levels .314 .109 2.886 .004 Positive
Q60 C-r products safest .214 .083 . 2.582 .011 Positive
Q70 C-r meats attractive trays .222 .085 2.612 .010 Positive
Q76 Cost of workers. comp. .213 .060 3.525 .001 Positive
Q2 Store type (Dummy Variable) -.087 .209 -1.239 .217 Positive

 

F=l4.953 (p-value=.000) R-‘~=.361 Adjusted 2 =.337

The next section will discuss the results from the econometric analysis of the data

SCI.

4.4 Econometric Analysis of Case-Ready Meat Data Set

A further analysis of the case-ready data set is warranted given the ﬂexibility of
the research design. Logistic regression is an econometric technique that is e useful in
illuminating the decisions retail grocery store managers face about expanding the
adoption of case-ready meats. It is particularly useful in further explaining the presence

or absence of unique characteristics in a binary dependent variable for a set of

129

explanatory variables. The binary dependent variables in this research project are
designed to capture the richness and inherent power of the explanatory independent
variables as they relate to the adoption of ﬂesh case-ready meat programs. Explanatory
independent variables in logistic regression have speciﬁc data requirements; the data
must be either interval or categorical. A central theme for the survey questions in this
project were designed around acquiring responses based upon a ﬁve point Likert scale
where data points are considered interval data.

The nature of binary dependent variables causes statistical and methodological
diﬁ'rculties and opportunities. In particular, predicted probabilities ﬁom a standard
Ordinary Least Squares (OLS) regression may exceed either the lower limits which is (0)
or the upper limits (1). Given the desire of the project to employ binary dependent
variables along with Likert scale explanatory/independent variables, the literature
suggests using two alternatives, logistic and probit regression analysis (Maddala, 1992;
Kennedy, 1993; Kmenta, 1986 and Gujarati, 1995).

Maddala (1992) suggests research projects possessing these types of binary
dependent variables are closely associated with using linear and non-linear probability
models where the dependent variable is binary and the variable takes the usual value of 1
or 0 with the option of including many explanatory independent variables. Both logistic
and probit regression analysis techniques are plausible choices. The literature (Maddala,
1992; Kmenta, 1986; and Gujarati, 1995) suggests using either technique where the data
analysis is concerned with dependent variables that offer two distinct alternatives (binary

dependent variables).

130

Kennedy (1993); Kmenta (1986) and Gujarati (1995) offer insights into choosing
between logistic regression and probit regression. The probability difficulties of binary
dependent variables are overcome in both the logistic and probit functions “S” shaped
distribution curve. The “S” shaped curves, while somewhat linear in the middle; they are
curved at both ends. The logistic regression model produces fairly similar regression
lines to the probit model; with the logistic regression having “fatter” tails in the logistical
distribution. It should be noted the differences in these curves is negligible. The fatter
. tails in the logistic regression model are particularly useful when the data is clustered in
the tails and the probit distribution curves “approach the axes more quickly”. The probit
regression tail is “skinnier” and is considered more appropriate for outliers in the data and
looking for differences (Kmenta, 1986). More importantly to this research project when
logistic regression is compared to probit regression, is that the area in the tails in the
logistic regression may yield larger probability estimates, not as close to 0 when the Z is
small; and possibly smaller probability estimates as the Z gets larger.

Another reason for choosing logistic regression is the ability to underscore the
relationships and strengths amongst the explanatory independent variables of interest
with the binary dependent variables. Kennedy (1993) discusses speciﬁc use of logistic
regression is for analyzing the classic “make or buy” decisions facing many business
leaders, similar to this project. Hoetker (2007) further cites the “appropriate” use of
logistic regression when a ﬁrm is deciding on making or buying a component
(outsourcing). The project is similar to Kennedy and Hoetker’s make or buy research
where the retail stores surveyed either sold case-ready meat products (buy) or otherwise

(make). Given a review of the relevant literature this research project will use the logistic

131

regression technique. The next sections will cover the logistic regression analysis of the

data set.

4.4.1 Logistic Regression

Bliss (1935) is credited with ﬁrst popularizing the applied use of the probit
regression in the mid-19303 in the ﬁeld of biology to research the dosage/response
relationships of biological organisms. He was interested in knowing how much
insecticide (dosage) needed to be applied in order to kill the predation. The eradication
of the predation would be coded (1) or if the predation lived the value would be (0) (Lai
and Chan, 2004).

Following Bliss’s successful introduction of probit analysis researchers in the
social sciences also adopted the technique (Aldrich and Nelson, 1984). In 1944 Joseph
Berkson developed logistic regression, the ancillary technique for Bliss’s probit
regression. Overtime the use of both techniques has ﬂourished. According to Hoetker
(2007), the use of probit and logistic regression models in the strategic management
journals alone has grown exponentially from early 1980’s at which time the techniques
represented only 10 percent of all published articles.

The decision to use a logistic regression is further based upon the ability of the
technique to explore the behavioral choice or an event classiﬁcation, such as adoption of
case-ready meat programs. A better understanding of the binary choice to adopt case-
ready meats is a central goal of this research project. It is generally accepted a priori,
researchers do not know the underlying relationships between variables and possible

complexity. Therefore, a better understanding is attainable by using this technique.

132

Agri-food marketing decisions about brand choices and product selections or non-
selections are considered examples of behavioral choice events.

The logistic regression model differs signiﬁcantly from the OLS regression, and
offers a powerful tool for analyzing binary dependent variables with explanatory
. independent variables derived from interval data.

Kennedy (1993) outlines the classic make or buy logistic regression model below,
where the observable binary dependent variable speciﬁes that the probability a store sells

(buys) case-ready meats is:
e)“3

prob (buy) = W

This in turn implies that the probability of not buying (make) is

prob (make) = 1 — prob (buy) = W

The likelihood function is formed as
exiB 1

i1+exi5 , 1+3sz
1

L:

Where i refer to stores who buy fresh case-ready beef and j refers to those make
(process fresh beef in stores).

Kennedy (1993) further explains that maximizing this likelihood with respect to [3
produces the maximum likelihood (MLE) of B. Where the MLE -is not the most
“probable” value of B, but the value of B that maximizes the probability of drawing the
sample actually obtained from the data set. For the nth individual, then, the probability of

buying fresh case-ready meats is estimated as

133

eanMLE

 

1 + eanWlLE

For the expression given above, the logit model implies that

prob(buy) __

 

 

__ exB
prob(make)
so that log likelihood (LL) is
[ prob(buy) _
prob(make) — X8

4.4.2 Logistic Regression Coefﬁcients

There are two distinct logistic regression models that will be developed The two
models attempt to predict or further explain the presence or absence of case-ready meat
adoption based upon the explanatory independent variables (predictors) found within the
survey questionnaire instrument. The presence or absence of any relationship between
the binary dependent variable and the explanatory variables is based upon the statistical
signiﬁcance of the Homer-Lemeshow chi square test statistic for the entire model. The
analysis will also determine the coeﬁicient’s signiﬁcance levels and in which direction
they move. Beny (2007) stipulates social scientists in the past have used binary
dependent variables with logistic and probit regression research and those efforts were
restricted to interpreting the “sign and signiﬁcance of the maximum likelihood estimates
for the coefﬁcients.” Plausible presence of interaction can be inferred by signiﬁcance

levels and the sign between the binary dependent variable and the explanatory variables.

134

This research follows the same path for inference amongst the variables of interest. It
should be noted, if the coefﬁcients are not signiﬁcant we reject the hypothesis of

interaction of the independent variable with the dependent variable.

4.4.3 Evaluating the Performance of Logistic Regression Models

There are ﬁve recognized statistics that evaluate the performance of the binary
logistic regression model: 1) the model chi square, 2) Wald statistic, 3) Hosmer-
Lemeshow model chi square, 4) percent correct predictions (classiﬁcation table) and 5)
two pseudo R2 statistics (Cox & Snell and Nagelkerke).

The ﬁrst statistic produced in logistic regression using SPSS is Omnibus Tests of
Model Coefficients which is a chi square statistical test concerned with how well the
overall model performs (Garson, 2008). In particular this chi square test reveals the
interrelationship between the dependent and the explanatory variables, ﬁrst a chi square
is performed from the perspective of an intercept only model (titled Block O) and the
ﬁnal omnibus chi square test is performed on a second model that incorporates all of the
explanatory independent variables (Block 1). These separate tests are designed to
conﬁrm the existence or presence of relationships between variables. As a check of the
signiﬁcant levels of the variables a chi square test statistic will reveal any effects the
explanatory variables have on the binary dependent variable. A signiﬁcant chi square test
statistic implies there is an “adequate ﬁt of the data to the logistic regression model”
(Garson, 2008). The null hypothesis for this chi square implies that when the p-value is
signiﬁcant the test denotes the model is considered statistically signiﬁcant, and the data

ﬁts the model.

135

The second statistical test is the Wald test statistic which is designed to test the
signiﬁcance level of the model coefficients. To determine the individual importance of
each explanatory variable a Wald chi square test is conducted. This statistic is gathered
by taking the ratio of the logistic coefﬁcient [3 to the variable’s standard error, then

squaring the product, yielding the Wald chi square statistic.

 

B coefficient )2

Wald Test Statistlc = (SE coefficient

The Wald statistic and its uses are similar to how a t-value in an OLS regression is
used and interpreted The statistic shows if the coefﬁcient is meaningfully different from
zero: If so, it can be concluded the explanatory variable is making a signiﬁcant
contribution to the binary dependent variable. As the above equation notes, SE5 are an
important part of the statistic and it is suggested by Menard (2002) that large SE5 will
produce non signiﬁcant coefﬁcients. The Wald is also instrumental in ascertaining the
signs for each explanatory/independent variable coefﬁcient.

The third. statistic is the Hosmer-Lemeshow (2000) test which is considered a
form of a goodness of ﬁt test designed just for logistic regression. It tests the assumption
that the data was generated by the model and is considered a more robust test than a
traditional chi square (Garson, 2008). The Hosmer-Lemeshow (I-I-L) test is computed
from a chi-square using the observed and expected frequencies to test the “fit” of the
logistic regression model. It should be noted that a “well-ﬁtting” logistic regression
model should show non-signiﬁcance in the H-L p-value. The non-signiﬁcant p-value
illustrates the model’s prediction is not signiﬁcantly different from the observed values
and, therefore, the model’s coefﬁcient’s estimates ﬁt the data at an acceptable level. If

the p-value is <.05 we reject the null hypothesis that there is no difference between the

136

observed and model predicted proportions. This logistic regression chi square test
statistic if signiﬁcant does not imply the degree of variance explained of the dependent
variable, only signiﬁcance (Garson, 2008 and Menard, 2002).

The fourth statistic is the percent correct predictions. ' This statistic can be found
within the Classiﬁcation Table, and is deemed the “practical” results from running the
logistic regression model. The percent correct predictions originate from a simple two by
two matrix which aggregates the correct and incorrect estimates based upon both the
explanatory and the constant. The matrix assigns the binary probabilities of Os or Is for
each cell and compares those to the actual; and the matrix yields the percent correct
tables along with an overall calculated score. The statistics within the classiﬁcation table
postulates the estimated p is greater than or equal to .50 when a store sells (in this project
buys the product from a processor) and not occur otherwise (make case-ready beef
products). Logistic regression makes an assumption about that the two rows in the matrix
having a different percent correct, if they were the same the model has homogeneity
(Garson, 2008). According to Menard (2002) complete homogeneity in the classiﬁcation
table is an “unacceptable” solution, and heterogeneity in the percent correct predictions is
the acceptable norm. According to the results found within the classiﬁcation table a good
model that offers predictive success will have a higher percent correct when the
explanatory/independent variables are included (Menard, 2002 and Garson, 2008).

The ﬁnal test statistics for judging or assessing model performance are the pseudo
R2 test statistics. Hoetker (2007) and Garson (2008) both suggest predicting whether a
logistic regression model has “goodness of ﬁt” is problematic and is decidedly different

than in OLS regression analysis. In OLS, R2 is the classic test of how well the model ﬁts

137

the data, or goodness of ﬁt. In logistic regression, Hoetker (2007) and Menard (2002)
both cite numerous researchers incorrectly reporting “pseudo-” statistics as a predictor of
how well their logistic regression model performed. The downside to relying on the
pseudo R2 is the test statistic must report which pseudo R2 they are reporting, and without
that speciﬁc information the interpretation and reliance on the models ﬁt is precarious.
Most importantly the pseudo R2 does not correspond to the percent of variance in the
model that is explained and does not have any interpretable meaning (Hoetker, 2007).
SPSS reports two pseudo R2 statistics, Cox and Snell and Nagelkerke.

Garson (2008) explains the Cox and Snell R2 for logistic regression was
developed to replicate the traditional R2 for OSL. To try and replicate the traditional R2
Cox and Snell based their “pseudo R2” on the log likelihood function of the ﬁnal model
as compared to the log likelihood of the baseline model”. Noting the maximum value is
less than 1, this presented problems with interpretations and it created an opportunity for
the development of another R2 statistic. ’

—2LLnuu]2/“

C &5 UP d R2: —[
ox ne seu o 1 -2LLk

Veall (1996) states that the pseudo R2 measures cannot be used to measure a
model’s goodness of ﬁt when the dependent variables are binary in value. Amemiya
(1981) offers a contrarian view to the use pseudo R2 for goodness of ﬁt where he cites
using the R2 in the probit situation will approximate the model’s goodness of ﬁt and
Dhrymes (1986) suggest using the pseudo R“!2 by incorporating the F statistic for testing

the explanatory power of the coefficients.

 

‘7 Notation for all Pseudo R2 models: LL= Log Likelihood, null = model with predictors, k = model without
predictors, just the intercept.

138

Nagelkerke centered his analysis on the weakness of the R2 of Cox & Snell which
could not reach 1.0. He made the necessary updates and modiﬁcations to ensure the R2
could range ﬁom 0—1.0. According to Garson (2008) Nagelkerke’s R2 “divides Cox and
Snell’s R2 by its maximum in order to achieve a measure that ranges from 0-1.0.
Therefore Nagelkerke’s R2 will normally be higher than the Cox and Snell measure but
will tend to run lower than the corresponding OLS R2. The Nagelkerke R2 is considered

the most reported “pseudo” R2.”

—2LLnuu 2’“
d R, _ 1‘1 —21.Lk 1
Nagelkerke Pseu o — 1_(2LLnu11)2/n

Menard (2002) and Veall (1996) cite another example of a logistic regression
pseudo R2 test statistic, the McFadden Pseudo R2. The McFadden R2 is also known as the
likelihood ratio index because the ratio must fall between 0 and 1. The resulting R2 are
typically smaller in magnitude than the previous three pseudo R2 and these lower R2 ratios
typically demonstrate the model that includes the predictors is a better ﬁt than the model
with just the intercept. While the McFadden pseudo R2 results tend to be smaller as
compared to other pseudo R2 results. McFadden R2 values that range between .20 percet
and .40 percent are considered very acceptable. Below is McFadden’s pseudo R2.

I-'I-'nu11

McFadden's Pseudo R2 = 1 —
LI.k

Given the conﬂicting statistical assessment for determining the goodness of ﬁt for
logistic regression models, Hoetker (2007) offers researchers some intuitive advice on
reporting how well the logit model ﬁts the data set. His suggestion is to provide the
reader the exact proportion of correct predictions the model made along with each of the

Pseudo R2 ﬁgures reported in the SPSS Logistic Regression output. While this method is
139

not unanimous, it is one method of showing the model’s goodness of ﬁt. This project

will only report the Cox & Snell and the Nagelkerke pseudo R2 test statistics.

4.4.4 Binary Logistic Regression Relative to Case-Ready Meat Adoption

Of primary interest for the binary logistic regression analysis will be why fresh
case-ready programs have been slower in their adoption rates. This analysis was
designed to help identify the qualitative traits or explanatory variables that may affect a
retail grocer’s propensity to adopt the fresh case-ready meat alternatives. There are two
primary proteins that make up case-ready meat products; ground beef and whole muscle
cuts”. Two binary logistic regression models were designed with these two primary
product categories in mind. The two binary logistic regression models will now be
referred to as Model 1 and Model 2. Model 1 will focus on adoption of case-ready
ground beef and Model 2 on whole muscle meat products. Both models will examine

hypotheses about the qualitative traits or characteristics that impact adoption rates.

4.4.5 Binary Logistic Regression Base Model 1: Ground Beef

The binary dependent variable in Model 1 centers on the propensity of the two
retail store types to choose to adOpt case-ready ground beef products. Ground beef is one
of the larger selling categories purchased in grocery stores today (Maj or, 2008) and thus a
very important product in a retailer’s fresh meat product mix. Ground beef is a category
of products that consumers remember the price per pound for when shopping for their
weekly products. It should be noted that the average household shops at three to four

stores every week for their entire list of grocery products, including health and beauty

 

‘8 Whole muscle cuts include all of the various roasts and steak products.

140

items. Many retailers including SAL use the price of hamburger as a loss leader to attract
consumers to purchase their entire array of grocery products. The binary dependent
variable in Model 1 represents a discrete choice of adopting case-ready ground beef
products from a set of mutually exclusive choices. The differentiation between store
types was deemed important to this project. Because of the initial importance of knowing
whether the respondents sold these types of case-ready meats and the variable being
binary, no changes in the coding of the dependent variables was required. The binary
dependent variable for the Model 1 was constructed from crbeeﬁ- adopted case-ready
ground beef] =Yes and otherwise=0. Of the 193 stores in the survey, 143 have adopted
the fresh case-ready ground beef alternative, 50 stores have chosen otherwise.

As mentioned earlier, an important ingredient of logistic regression is the
technique’s rigid data requirements. The data for the dependent variable must be binary
and the data for all explanatory variables is limited to either interval data (Likert scale) or
categorical data. The binary dependent and explanatory variables for both models
conform to these requirements.

Model I started with a base model that incorporated the explanatory variables or
predictors in the data set that relate to further adoption of case-ready ground beef. These
variables were based upon the retail store and DC visits, interviews with industry/store

executives and literature review. The base Model 1 has for the dependent variable:
adopted case-ready ground beef = f (veriﬁes temp ranges of delivered meat, writes

delivery dates on boxes, c-r has the lean/ratios stores require, customer theft source of
shrink, defective packaging source of shrink, workmanship issues source of shrink, non-

bloom issues of ground beef source of shrink, satisfaction level of the accounting of meat

141

not sold at full retail, satisfaction of FIFO of meat in walk-ins, satisfaction meats are
stacked in walk-in, satisfaction store keeps meat case inventories, satisfaction store preps
for ﬁrst of the month sales, satisfaction store preps for specials, more sales if meat
products were thought to be safer, c-r meats are safer than in-store processed meats,
blooming of c-r meats is a problem, c-r products use attractive trays, difficult to ﬁnd
qualiﬁed meat cutters, meat cutters could be trained to manage the perishables category,
meat managers could forecast demand and manage versus cut meat, and c—r products are

most cost effective than in-store processed meats )19

4.4.6 Binary Logistic Regressions Model 1: First Run Results

Running a binary logistic regression on the base model revealed that 15 of 22
explanatory variables in the model were statistically insigniﬁcant (see Table 37 below).
While few variables were statistically signiﬁcant, the base model had several promising
results, including a signiﬁcant Omnibus Test for the data ﬁtting the model, seven
signiﬁcant coefﬁcients (Wald chi square), and a signiﬁcant Homer-Lemeshow (H-L)
model chi square test. These results showed the underlying strength of the base model.
Coupled with very good pseudo R2 results, the Cox & Snell was 46 percent and the

Nagelkerke was 68 percent.

 

'9 Logit 1 base model explanatory variables as listed in the questionnaire: q7, q9, ql9, q39, q40, q41, q43,
q46, q49, q50, q52, q53, q54, q55, q59, q60, q68, q70, q72, q73, q74, and q75.

142

Table 37: Logit 1 Model: Adopted Case-Ready Ground Beef

 

 

 

All Stores
Parameter Estimate SE Sig
Veriﬁes temps upon delivery (Q7) -0_68 0.35 *
Delivery dates are written on each meat pkg (Q9) 0.89 0.32 ***
C-R has grd beef lean/fat ratios cust. demand (Q19) 0,49 024 H
Shrink: Customer Theﬁ (Q39) -0.63 0.25
Shrink: Defective pkg (Q40) -0.42 0.32
Shrink: Workmanship (Q41) -0.33 0.32
Shrink: Non-bloom (Q43) Ground Beef 0.10 0.20
Shrink: Out of Date (Q46) 0.03 0.30
Accounting for Pkg Not Sold at Full Retail (Q49) 0_ 30 0,29
Satisﬁed with FIFO Procedures Walk-1n (Q 50) -0.28 0.70
Satisﬁed storage of c-r meats in walk-in (Q52) 0,37 0.55
Satisﬁed Meat Case Inventory Levels (Q53) -0,31 051
Satisﬁed Store Preps First-of-Month Sales (Q54) —0,89 057
Satisﬁed Store Preps for Meat Specials (Q55) .036 0.31
More Sales if Meat Products were Safer (Q59) 0.22 0.28
C-R Meats Safer than In-Store Processed (Q60) 0.72 0.30 **
“Blooming” of C-R meats is a problem (Q68) -0.74 0.30 **
C-R Meat Packages Use Attractive Trays (Q70) 0,85 0, 33 **
Difficult to Find Qualiﬁed Meat Cutters (Q72) .064 0,22 **"‘
Meat Cutters Manage Perishables Category (Q73) 0,22 0.24
With C—R Meat Mgrs. Forecast/Inventory (Q74) -0.32 0.26
C-R Products Are More Cost Effective (Q75) 0.22 0.30
Intercept 2.95 3.23
Onmibus Tests of Model Coefﬁcients (Chi Square) 119,801 22df ***
Hosmer—Lemeshow Model Chi-Square” 12.33 8df H"

Pseudo R2 for Base Model:
Cox and Snell: 46% Nagelkerke: 68%
Significant Codes:****p>.05 ***p<.01 **p<.05 *p<.10
Number of Observations 193
Given that many variables were not signiﬁcant, a separate diagnostic test for
multicollinearity was conducted which revealed many of the explanatory variables had
multicollinearity problems.

A ﬁnal statistical test is the “classiﬁcation table” which was conducted for the

ﬁrst run of the base model (see Table 38). The test ﬁrst includes only the intercept and

 

2° NOTE: According to Menard (2002) a Hosmer-Lemeshow model chi-square statistic registering non-
signiﬁcant concludes the model adequately ﬁts the data being analyzed.

143

another test is conducted that includes the explanatory variables. The second step of
including the independent variables is performed to determine if the variables added
explanatory power to the model (see Table 39). The classiﬁcation table below is
particularly useful; especially knowing that the logistic regression does not assume the
relationships between the dependent and explanatory variables is linear. The technique
for this statistic is testing for overall predictive success of the entire model, highlighting
the percent correct and incorrect for a particular binary dependent variable model.

Table 39 shows the results of the classiﬁcation table begin with the null model
without explanatory variables, which did very well, at 74.5 percent. When the
explanatory variables were added, the model added signiﬁcantly to the predictive nature
of the model. Menard (2002) suggests when the explanatory variables are added the
overall percentage predicted should increase. As Table 38 and Table 39 suggest the
overall accuracy of Model 1 for the ﬁrst run indicates the model’s ability to predict
successfully is 91.1 percent, an excellent result, especially for a base model.

Table 38: Classiﬁcation Table (ab) Model I First Run Without Explanatory

 

Variables
Adoption
Has Adopted Percentage
Otherwise C-R Correct
Step Adoption Otherwise o
0 OR Ground Beef 0 50 0 /0
Has Adopted C -R 0 143 100.0%
(a) Intercept only model. Overall 745%

(b) The cut value is .50

144

Table 39: Classiﬁcation Table (a) Model I First Run With Explanatory Variables

 

Adoption
Has Adopted Percentage
Otherwise C-R Correct
ft“? 0mm“ 37 12 75.5%
Has Adopted C -R 5 138 96.5%
(a) The cut value is .50 Overall 91.1%

 

4.4.7 Binary Logistic Regression Model 1: Second Run

At this stage of the binary logistic regression analysis Menard (2002) suggests
eliminating variables and rerunning the model. Therefore, a second run for Model 1 was
prepared by dropping seven variables with p-values greater than 35 percent which
produced much better results and much cleaner multicollinearity diagnostics (see Table
40). The Omnibus Tests of Model Coefficients (Chi Square) and Homer-Lemeshow
Model chi square test statistics were both signiﬁcant. Of particular interest associated
with this run were the nine signiﬁcant explanatory variables revealed by the Wald chi
square p—values and the continued consistent results of the pseudo R2 ﬁgures. The
pseudo R2 results for this run includes a Cox & Snell R2 was 46 percent and the

Nagelkerke R2 was 67 percent.

145

Table 40: Binary Logistic Regression Model 1 Second Run: Store Adopts Case-

 

 

 

Ready Ground Beef

All Stores
Parameter Estimate SE Sig
Veriﬁes temps upon delivery (Q7) -O.64 0.32 *
Delivery dates are written on each meat pkg (Q9) 0,86 0,32 ***
C-R has grd beef lean/fat ratios cust. demand (Q19) 0,44 0,22 **
Shrink: Defective pkg (Q40) -0.42 0.27
Shrink: Workrnanship (Q41) —0.37 0.30
Accounting for Pkg Not Sold at Full Retail (Q49) 05 3 025 **
Satisﬁed with FIFO Procedures Walk—In (Q 50) .022 0,49
Satisﬁed Store Preps First-of-Month Sales (Q54) 4,086 0,54 **
Satisﬁed Store Preps for Meat Specials (Q55) -0,43 0.32
More Sales if Meat Products were Safer (Q59) 0,24 0,26
C-R Meats Safer than In-Store Processed (Q60) 079 0,26 ***
“Blooming” of C-R meats is a problem (Q68) -067 0,27 **
C-R Meat Packages Use Attractive Trays (Q70) 0,92 0,30 ***
Difﬁcult to Find Qualiﬁed Meat Cutters (Q72) -O.62 0.22 ***
With C-R Meat Mgrs. F orecast/Inventory (Q74) —0_17 0,22
Intercept

3.88 2.83
Orrmibus Tests of Model Coefﬁcients (Chi Square) 118.21 15df "*
Hosmer-Lemeshow Model Chi-Square 7.067 8df H"
Pseudo R2 for Base Model:

Cox and Snell: 46% Nagelkerke: 67%

Significant Codes:****p>.05 ***p<.01 **p<.05 *p<10
Number of Observations 193

 

The classiﬁcation table results for the second run of Model 1 were again
excellent. Starting with the null model of only the intercept the results (see Table 41) for
the overall percentage of the second run was 74.5 percent. After adding the explanatory
variables in the second run the classiﬁcation model’s overall percentage increased to 90.7
percent (see Table 42). This is important towards implying the Logit l model’s ability of

further explaining the factors of adoption.

146

Table 41: Classiﬁcation Table (ab) Logit Model I Second Run Without Explanatory

 

Variables
Adoption
Has Adopted Percentage
Otherwise C-R Correct
Step Adoption Otherwise o
0 OR Ground Beef 0 50 0 /0
Has Adopted OR 0 143 100.0%
(a) Intercept only model. Overall 745%

(b) The cut value is .50

Table 42: Classiﬁcation Table (a) Logit Model I Second Run With Explanatory

 

Variables
Adoption
Has Adopted Percentage
Otherwise C-R Correct
Step Adoption Otherwrse 39 11 78.0%
Has Adopted C-R 7 136 95.1%
(a) The cut value is .50 Overall 90.7%

 

4.4.8 Binary Logistic Regression Model 1: Final Run

Again, following Menard’s (2002) procedure for conducting logistic regression
analysis, the next step in analyzing Model 1 was to drop the highly insigniﬁcant variables
and re-rrm Model 1 again. The results in dropping four explanatory variables for the ﬁnal
run of Model 1 continued the trend of producing excellent statistical results for the entire
model.

The results for the ﬁnal run of Model 1 can be viewed below in Table 43
beginning with the Omnibus chi square test producing highly signiﬁcant p-values for the
explanatory variables in the model to predict the binary dependent variable. Garson
(2008) suggest a signiﬁcant result here corresponds to the data “adequately” ﬁtting the
proposed model.

The Wald chi square statistics showed ten out of eleven explanatory variables

with signiﬁcant p-values. The only insigniﬁcant variable in the ﬁnal run was related to

147

workmanship issues of case-ready meats as a source of shrink. The number of
explanatory variables in the ﬁnal model also meets the suggested ratio of observations to
variables for logistic regression, should be a minimum of 10. For this model the ratio is
193 observations to 11 variables which give a 17.5 ratio.

The Homer-Lemeshow Model chi square test statistic (6.632) for the entire model
is highly signiﬁcant, a very positive and encouraging result for the entire model because
this is the recommended statistic for judging the overall “ﬁt” of the binary logistic
regression model. The two pseudo R2 results for the ﬁnal model were very promising as
well, with the Cox & Snell R2 reporting at 45 percent and the Nagelkerke R2 reported a
signiﬁcantly higher test statistic of 65 percent.

Table 43: Binary Logistic Regression Model 1 Final Run: Store Adopts Case—Ready

 

 

Ground Beef
All Stores

Parameter Estimate SE Sig
Veriﬁes temps upon delivery (Q7) _ -0.64 0.30 *
Delivery dates are written on each meat pkg (Q9) 0.80 0.29 ***
C-R has grd beef lean/fat ratios cust. demand (Q19) 0.44 0.21 **
Shrink: Defective pkg (Q40) -0.48 0.25
Shrink: Workmanship (Q41) —0.39 0.27
Accounting for Pkg Not Sold at Full Retail (Q49) 0.43 0.23 *"‘
Satisﬁed Store Preps First-of-Month Sales (054) -1,41 0.45 **
C-R Meats Safer than In- Store Processed (Q60) 0,88 0,25 ***
“Blooming" of C-R meats is a problem (Q68) .066 024 **
C-R Meat Packages Use Attractive Trays (Q70) 0,76 0.26 ***
Difﬁcult to Find Qualiﬁed Meat Cutters (Q72) -0.60 0.18 ***
Omnibus Tests of Model Coefﬁcients (Chi Square) 113,88 lldf ***
Hosmer-Lemeshow Model Chi-Square 6_632 8df ****
Pseudo R2 for Base Model:

Cox and Snell: 45% Nagelkerke: 65%
Significant Codes:****p>.05 ***p<.0l **p<.05 *p<.10
Number of Observations 193

The classiﬁcation table results for the ﬁnal run of Model 1 were again excellent.

This is an important ﬁnding for Model 1, as these results help assess the performance of

148

the entire model, starting with the null model of only the intercept the results (see Table
44) for the overall percentage of the ﬁnal run was 74.1 percent. After adding the
explanatory variables the classiﬁcation model’s overall percentage jumped to 89.1
percent (see Table 45). This is important towards implying the Logit 1 model’s ability of
further explaining the factors of adoption.

Table 44: Classiﬁcation Table (ab) Logit Model I Final Run Without Explanatory

 

Variables, Intercept Only
Adoption
Has Adopted Percentage
Otherwise C-R Correct
Step Adoption Otherwise 0 50 0
0 OR Ground Beef ’
Has Adopted C-R 0 143 1000

(a) Intercept only model.

0
(b) The cut value is .50 Overall 74.1%

 

Table 45: Classiﬁcation Table (a) Logit Model I Final Run With Explanatory

 

Variables
Adoption
Has Adopted Percentage
Otherwise » C-R Correct
Step Adoption Otherwrse 38 12 76.0%
Has Adopted C -R 9 134 93.7%
(a) The cut value is .50 Overall 89.1%

 

4.4.9 Model 1: Multicollinearity Diagnostics

Garson (2008) and Menard (2002) describe multicollinearity in logistic regression
as one explanatory variable being a linear function of another explanatory variable.
Effects of multicollinearity include lack of coefficients being statistically signiﬁcant with
the overall model showing signiﬁcance and coefﬁcients not having the predicted sign.
When the coefficients have incorrect signs then the subsequent analysis and conclusions

are jeopardized. Therefore, inspecting Model 1 for multicollinearity is of particular
149

interest towards conﬁrming the reliability of the explanatory independent variables in the

ﬁnal model.

There is a quick diagnosis for checking for multicollinearity which can be found
in the logistic regression results; look for any standard errors with values greater than or
equal to 2.0. A more detailed and rigorous test for multicollinearity can also be
completed in an OLS regression of the logistic regression model. In these
multicollinearity diagnostics there are two statistics of import; the tolerance levels and the
variance inﬂation factor. Menard (2002) states the existence of multicollinearity with
explanatory variables when tolerance levels are less than or equal to 40 percent. The
tolerance levels are the percentage of variance in an explanatory variable that is not
explained by the other variables in the model. Variance inﬂation factor levels detecting
multicollinearity in logistic regression are at levels exceeding 2.5. The variance inﬂation
factor is a factor of the coefﬁcient of interest multiplied during the logistic regression due

to the levels of multicollinearity found within the logistic regression model.

The variance inﬂation factor is

WF : Tolerance

*VIF is always greater than or equal to 1.

For the quick diagnosis of multicollinearity in Model 1 the logistic regression
results revealed none of the explanatory variables in the SPSS output with high standard
errors at or above the 2.0 threshold. The next step was to construct an OLS regression for

Model 1 for the purpose of checking the accompanying diagnostic statistics for
150

multicollinearity. The diagnostics showed no signs of multicollinearity for any variables
for either the tolerance level or variance inﬂation factor. A closer review of Table 46
shows all of the tolerance levels for Model 1 to be at or above the prescribed levels (they
were all greater than 40 percent) for multicollinearity. The variance inﬂation factor for
each coefﬁcient showed excellent results as well. None of variance inﬂation factors were
greater than 2.5. We can conclude multicollinearity is not a problem for Model 1 after
conﬁrming in logistic regression no inﬂated values of the standard errors, and in the OLS
regression of Model 1 none of the tolerance levels or variance inﬂation factors were at

levels detecting problems with multicollinearity.

Table 46: Multicollinearity Diagnostics for Logit Model 1a Ground Beef

 

 

Collinearity Statisties*
Logistic Regression Model 1 Tolerance VIF
(Constant)
Veriﬁes temperature ranges Q7 1 0.85 1.18
Writes delivery dates Q9 0.85 1.18
Ground beef lean to fat ratios Q19 0.78 1.28
Shrink defective packaging Q40 0.76 1.32
Shrink workmanship issues Q41 0.73 1.38
How satisﬁed accounting of meat not sold at full retail Q49 0.79 1.27
How satisﬁed store is prepared for FOM Q54 0.73 1.36
C -R products safer than in-store processed Q60 0.72 1.38
Blooming c-r is a problem Q68 0.91 1.10
C-R products use attractive trays Q70 ' 0.76 1.31
More difﬁcult to ﬁnd meat cutters Q72 0.91 1.10

 

' Binary dependent variable: Adopted c-r ground beef products

*Note: Tolerances S 0.40 = Problems with multicollinearity VIF _>_ 2.50 = Problems with
multicollinearity

 

151

4.4.10 Assessment of Coefficients in Logit Model I: Predicted/Observed Signs of
Coefficients

The previous section the analysis demonstrated that Model 1 did not have
multicollinearity problems with the explanatory independent variables. This allows for a
further review of those key variables associated with expanding the adoption rates of
case-ready ground beef products. As shown earlier, the assessment of the explanatory
variables will be limited to interpreting the signs of each coefﬁcient and their associated
signiﬁcant levels (see Table 47).

Table 47 also displays the explanatory variable’s hypothesis for the predicted sign
for each coefﬁcient in the ﬁnal run of Model 1, and the subsequent results discovered in
the analysis. Note again this research was relying on the predicted and observed
coefficient sign as another level of interpretation for the relevant factors that help further
the adoption levels case-ready meat programs. This section in particular addresses the

signs of the coefﬁcients and how they relate to the adoption of ground beef products.

152

Table 47: Binary Logistic Regression Model 1 Adoptions of CaseReady Ground
Beef Products: Interpreting the Sign and Direction of the Predictor
Variables

 

Predictor Variable Hypothesis and Predicted Sig! / Direction Results
Q7. How satisﬁed are you with the way your store veriﬁes that fresh meat deliveries
ﬁ'om the Distribution Center (DC) are within acceptable temperature ranges?
o 1=Not at all to 5= Very Satisfied Passed
Ho2 During the case study portion of the project it was discovered that temperatures
of meat products were not being checked upon delivery. The sign was predicted to
be negative. '
Q9. How satisﬁed are you with the way your store writes the delivery date on each
fresh meat box received from the DC?
0 1=Not at all to 5= Very Satisﬁed
Ho; Research showed the shelf-life of c-r beef was maximized when the oldest Passed
product in the walk-in was the next product to go in the meat cases. The chances of
this metric being fulﬁlled were predicated on the delivery date being written on each
box. The sign was predicted to be positive.
Q19. Case-ready ground beef products are available in the lean — fat ratios my
customers prefer?
o I=Strongly Disagree to 5=Strongly Agree
Ho; Research efforts and interviews with processors hypothesized c-r products were Passed
being produced with the lean to fat ratios retailers preferred. When the lean to fat
ratios are available in the ratios retailers demand, further adoption of case-ready
meats is more likely. The sign was predicted to be positive.
Q40. Defective and damaged packaging is a source of shrink.
o 1=Big Concern to 5=Not a Concern
Ho; When defective packaging is a big concern for case-ready beef products, further Passed
adoption of case-ready meats is more likely. The sign was predicted to be negative.

Q41. Shrink: Workrnanship issues

0 1=Big Concern to 5=Not a Concern Variable
Ho : When workmanship issues are a big concern, further adoption of case-ready was not
meats is more likely. The sign was predicted to be negative. signiﬁcant

Q49. How satisﬁed are you with the way your store records (or accounts for) every
fresh meat package (product) not sold at full retail?
0 1=Not at all to 5= Very Satisﬁed FAILED
Ho; Research and interviews revealed most packages of meat not sold at full retail
were products with expired “use by” dates (shelf life). Case-ready meats have
decidedly longer shelf life and a negative sign would have encouraged further
adoption of case-ready meats. The was predicted to be negative.
Q54. How satisﬁed are you with the way your store prepares for the increased sales
experienced during the ﬁrst few days of each month?
0 1=Not at all to 5= Very Satisﬁed Passed
Ho; When managers are not satisﬁed with the way their store prepares for increased
sales during the ﬁrst few days of the month, further adoption of case-ready meats is
more likely. The sign was predicted to be negative.

153

Table 47 (cont’d)
Q60. Case-ready beef products are the safer than beef products processed in-stores?
o 1=Strongly Disagree to 5=Strongly Agree
Ho: 1f store managers think case-ready beef products are safer than iii-store Passed
processed meats, further adoption of case-ready meats is more likely. The sign was
predicted to be positive.
Q68. Case-ready meat products that require “blooming” cause operational

problems.
a 1=Strongly Disagree to 5==Strongly Agree Passed
Ho: If managers strongly disagree with blooming causing operational issues, further
adoption of case-ready meats is more likely. The sign was predicted to be negative.
Q70. Case-ready beef products are packaged in attractive trays.
o 1=Strongly Disagree to 5=Strongly Agree Passed
Ho: If case-ready beef products are packaged in attractive trays, further adoption of
ease-ready meats is more likely. The sign was predicted to be positive.
Q72. It has become increasingly more difﬁcult to ﬁnd qualiﬁed meat cutters for my
store.
0 1=Strongly Disagree to 5=Strongly Agree FAILED“
Ho: If store managers agree it is more difficult to ﬁnd qualiﬁed meat cutters, ﬁrrther
adoption of case-ready meats is rrrore likely. The sign should be positive.
* Could be a selection bias problem for some respondents because they don’t use
meat cutters.

The explanatory/predictor variable about case-ready ground beef products being
available in the lean to fat ratios customers prefer passed the predicted and observed
analysis. If store managers believe the availability of ground beef pack sizes they may be
more likely to adopt case-ready meats. Price points and other supply and demand issues
are based upon the lean to fat ratios of ground beef. Without the presence of the correct
lean to fat ratios further adoption would be resisted.

The explanatory variable associated with the shrink category concerned with
defective and damaged packaging passed the predicted and observed test with its negative
sign. Prior to the questionnaire being deve10ped earlier research endeavors suggested
that case-ready packaging for ground beef offerings had undergone signiﬁcant product
changes to overcome prior problems with defective and damaged packaging. Therefore,
as levels of this coefﬁcient (defective and damaged packaging) decrease, adoption levels

for case-ready ground beef are expected to increase.

The sign for the coefﬁcient related to case-ready beef products being safer than
product processed in stores was another very encouraging result for the project because a
positive sign for this coefﬁcient would encourage adoption. As the project began there
was substantial debate concerning food safety. Processors were adamant about its safety
and hopeful that managers viewed case-ready products as safer. This result from the
store managers was also consistent with the literature review, as well.

The coefﬁcient sign for the variable addressing case-ready products being
packaged in attractive trays is an important ﬁnding. Case-ready products are packaged in
completely different looking trays to accommodate the rigorous requirements associated
with the new technologies and to achieve increased shelf life. The positive sign for this
coefﬁcient is consistent with the research ﬁndings and for further adoption of the case-

ready programs.

4.4.11 Binary Logistics Regression Model 1: Managerial Implications

Table 47 synthesized the results for the predicted/observed coefficient signs and
their interpretation for the predictor variables for Model 1. This table is also a useful tool
for categorizing the signiﬁcant explanatory variables or predictors into three broad
categories related to the further adoption of case-ready ground beef programs. These
broad categories are in fact the major headings for the research’s six operational issues
that were discussed earlier in chapter four. The three adoption categories include: supply
chain metrics, product attributes for processors, retailers and consumers and management
tools, speciﬁc for limited assortment retailers to help them improve their business model.

The ﬁrst category, supply chain metrics, identiﬁes the key factors discovered in

the research that are conducive for ground beef to effectively and efﬁciently to move

155

through the supply chain. The importance of this category cannot be underestimated as
supply chain metrics in the past have been instrumental in moving perishable products
through a supply chain with maximum shelf life for both retailers and customers. Shelf
life is important to retailers as it reduces several sources of shrink, improves sales for
processors and proﬁt and ﬁnally provides consumers more product options and improves
satisfaction levels of the entire retail grocery store shopping experience. The retail
grocery store shopping experience has been viewed by industry analyst as key towards
improving gross store sales and increase customer loyalty. Below are the pertinent
factors.

Supply Chain Metrics Discovered in Model 1

0 Importance of monitoring temperatures of product delivered from the stores. This
could also be inferred for DC receiving product from processors.

0 Importance of writing the delivery date on each carton of fresh beef delivered to
stores. Recognizing and implementing this important step is deemed critical towards
maximizing inventory management and ensuring product as the maximum amount of
time to be sold for full retail price (no markdowns or write-offs).

o Defective and damaged packaging from the perspective of supply chain metrics
relates to the necessity of discovering where along the supply chain the damage was
incurred and make the necessary changes to improve stops along the supply chain and

mitigate future damaged products.

Importance of monitoring and reconciling products not sold at full retail. This metric
is related to properly and systematically quantifying product not getting through the
supply chain. Proper accounting should be the catalyst for improving the supply

chain.

The second adoption category is product attributes. Speciﬁc product attributes are

deemed important for all participants in the supply chain. Product attributes are directly

related to fulﬁlling demand Further explication and understanding of product attributes

is an important ﬁrst step in current demand scenarios and the ability to help with future

shifts of demand for case-ready ground beef products. Below are the pertinent factors.

Speciﬁc Product Attributes Demanded by All Supply Chain Participants

Case-ready ground beef products are available with the lean-to-fat ratios my store and
consumers demand. Retailers must have product in the “pack sizes” their customers
demand. Whole muscle beef products cut and packaged in stores use this
data/information to ensure their stores have the whole muscle pack sizes their
customers demand, especially for whole muscle meats. Interviews of store managers
revealed this important point.

Shrink associated with workmanship issues in a case-ready scenario is an opportunity
for managers to account for all products with “workmanship” issues before customers
view/buy the product. This information needs to ﬂow back downstream to processors
to help them constantly improve their offerings by isolating the source of the

workmanship issues.

157

New packaging technologies and creative packaging efforts have transformed the way
many food products are marketed.

Case-ready beef products being viewed as safer than products produced in-stores is
key attribute retailers and consumers demand. Currently when other fresh food
products are deemed safer this fact has allowed food marketers to differentiate those
products in the marketplace and are able to charge higher prices.

Case-ready products are packaged in attractive trays. Current packaging research
posits one way to differentiate products in the marketplace and in minds of consumers

is through new and innovative packaging alternatives.

The last adoption categories are the management tools to help further adoption

endeavors. Management tools as a category for further adoption of case-ready products is

related to the product’s ability to help management constantly improve their business

model. An underlying truth about management in the retail arena is the repetitive nature

of the tasks within the business model.

Management Tools

Case-ready meat’s ability to help managers maximize store sales during the ﬁrst few
days of each month. Limited assortment stores experience unusually high levels of
store sales during the ﬁrst few days of each month and products that can be seen as
less labor and maximize sales will be positively viewed by decision makers.

Case-ready meat’s ability to mitigate the need for ﬁnding future supplies of meat
cutters. Meat cutters have higher labor cost per unit as compared to other stOre

personnel and the supplies of qualiﬁed meat cutters are going down.

158

4.4.12 Summary Conclusions for Model 1

The logistic regression Model 1 objective is to provide a better understand of the
factors that will help further the adoption of case-ready ground beef programs. This is
especially important when the research is trying to predict the presence and subsequent
absence of explanatory variables pertinent to the further adoption of case-ready ground
beef programs. This was accomplished by the model.

Grocery retailers, meat department managers, and meat processors are very
knowledgeable about ground beef and whole muscle cuts. Coincidently, these two
categories of proteins represent a signiﬁcant portion of the entire meat category in sales
and they drive important marketing and product mix decisions play a key role as retailers
attempt to market the entire grocery store. Model 1 was employed to identify and assess
those relevant factors.

The overall ﬁt and quality of the entire model was acceptable as demonstrated by
the highly signiﬁcant Homer-Lemeshow goodness of ﬁt test. The ﬁt and quality of the
entire model was also seen in the respectable pseudo R2 results: Cox & Snell and the
Nagelkerke R25. The classiﬁcation table showed the combined model predicted correct
and incorrect estimates at 89.1 percent of the time. As for the strength of the explanatory
variables, the Wald chi square statistics revealed eleven out twelve explanatory
independent variables were signiﬁcant. These combined results shed light on the factors
of adoption to be considered by retailers and processors interested in further adoption of

case-ready ground beef programs.

4.4.13 Binary Logistic Regression Model 2 Adoption Case-Ready Whole Muscle
Beef Products

After the reviewing the results of the Model 1, another logistic regression model
was constructed to study case-ready whole muscle products. Without prior theory to
guide which predictor variables to include in the base logistic regression models as was
the case with Model 1, Model 2’s base model was developed ﬁ'om the results and
information gathered during visits to retail stores / Distribution Center, interviews with
industry experts and the literature review. The second logistic regression model studies
the factors that will help further adoption rates of the other important case-ready whole
muscle products“.

Model 2 includes several explanatory variables from the base model used in
Model 1 and adapted the other necessary components related to expanding the adoption

rates of whole muscle beef products. The dependent variable: adopted case-ready whole
muscle beef products = f (veriﬁes temp ranges of delivered meat, writes delivery dates on

boxes, c-r whole muscle products come in the “cuts” customers want, c-r whole muscle
products come in the pack sizes customers want, customer theft source of shrink,
defective packaging source of shrink, workmanship issues source of shrink, non-bloom
issues of whole muscle cuts source of shrink, out of date product source of shrink,
satisfaction level of the accounting of meat not sold at full retail, satisfaction of FIFO of
meat in walk-in storage rooms, satisfaction meats are stacked in walk-in, satisfaction
store keeps meat case inventories, satisfaction store preps for ﬁrst of the month sales,

satisfaction store preps for specials, more sales if meat products were thought to be safer,

 

2’ Whole muscle beef products have two broad categories: 1) roasts and 2) steaks.

160

c-r meats are safer than in-store processed meats, blooming of c-r meats is a problem, c-r
products use attractive trays, difficult to ﬁnd qualiﬁed meat cutters, meat cutters could be
trained to manage the perishables category, meat managers could forecast demand and
manage versus cut meat, and c-r products are most cost effective than in-store processed

meats).

4.4.14 Binary Logistic Regression Model 2: First Run

The results of running the initial Model 2 in SPSS using binary logistic
regression, the results showed 11 of 23 explanatory variables in the model were
statistically insigniﬁcant and the Homer-Lemeshow (H-L) model chi square was not
signiﬁcant. A complere display of the results for this the ﬁrst run on Model 2 can be

viewed below in Table 48.

161

Table 48: Binary Logistic Regression Model 2 Base Model: Adoption Case-Ready

 

 

 

Whole Muscle Beef Products
All Stores
Parameter Estimate SE Sig
Veriﬁes temps upon delivery (Q7) 0.18 0.32
Delivery dates are written on each meat pkg (Q9) 1.52 0.39 ***
C-R has the whole muscle cuts for my store (Q20) 0,56 0,39
C-R has the whole muscle packs for my store (Q23) 0,57 0.35 *
Shrink: Customer Theft (Q39) -0.24 0.26
Shrink: Defective pkg (Q40) 0.91 0.37 **
Shrink: Workrnanship (Q41) -1,11 0.38 ***
Shrink: Non-bloom whole muscle beef cuts (Q44) 0.88 0.29 ***
Shrink: Out of Date (Q46) -0.56 0.37
Accounting for Pkg Not Sold at Full Retail (Q49) 00? 0.35
Satisﬁed with FIFO Procedures Walk-1n (Q 50) .059 0,73
Satisﬁed storage of c-r meats in walk-in (Q52) -0.12 0.60
Satisﬁed Meat Case Inventory Levels (Q53) -0,33 0.55
Satisﬁed Store Preps First-of-Month Sales (Q54) -129 0.63 **
Satisﬁed Store Preps for Meat Specials (Q55) -0,63 0.35 *
More Sales if Meat Products were Safer (Q59) .01] 0, 32
C-R Meats Safer than In-Store Processed (Q60) 1,93 0.47 "*1
“Blooming” of GR meats is a problem (Q68) .10] 0,42 **
C-R Meat Packages Use Attractive Trays (Q70) 1,10 041 "It
Difﬁcult to Find Qualiﬁed Meat Cutters (Q72) -0.63 0.21 ***
Meat Cutters Manage Perishables Category (Q73) 02] 0.26
With C-R Meat Mgrs. Forecast/Inventory (Q74) .050 0.31
C-R Products Are More Cost Effective (Q75) 0,80 0,33 **
Intercept -5.37 2.37 **
Omnibus Tests of Model Coeﬁicients (Chi Square) 117,33 23df **
Hosmer-Lemeshow Model Chi-Square22 17.93 8df **"‘*
Pseudo R2 for Base Model:
Cox and Snell: 60% Nagelkerke: 81%
Significant Codes:****p>.05 ***p<.0l **p<.05 *p<.10
Number of Observations 193

 

The results for the ﬁrst run on Model 2 were similar to the initial run on Model 1
for ground beef, where many explanatory variables were non-signiﬁcant, but the overall
base model showed signs of having underlying strength. Model 2 did have several

interesting results; including a highly signiﬁcant Omnibus Tests of Model Coefﬁcients

 

22 NOTE: According to Menard (2002) a Hosmer-Lemeshow model chi-square statistic registering non-
signiﬁcant concludes the model adequately ﬁts the data being analyzed

162

coupled with excellent pseudo R2 results, the Cox & Snell R2 was 60 percent and the
Nagelkerke R2 was 81 percent.

Another promising result highlighting the underlying strength of the base model
can be found in the results of how well the Model 1 classiﬁed both dependent and
explanatory (predictor) variables (Menard, 2002). Menard stresses the importance of
realizing that “classiﬁcation models impose the constraint that the model should classify
as many cases into each category as are actually observed in each category. Further
noting, the proportion of cases observed should be at least the same proportion as
predicted. To the extent that a model fails to meet this criterion, it fails as a classiﬁcation
model” (Menard, 2002).

Table 49 shows the results of the classiﬁcation table without predictors at 56
percent and Table 50 shows when the explanatory variables are included in Model 2, the
model produces a respectable overall 92.2 percent. On the ﬁrst run the model did
extremely well.

Table 49: Classiﬁcation Table (ab) Logit Model 2 First Run Without Predictors

 

Adoption
Has Adopted Percentage
Otherwise C-R Correct
Step Adoption Otherwise o
0 GR Whole Muscle 0 85 0 /0
Has Adopted OR 0 108 100.0%

(a) Intercept only model.

1b) The cut value is .50 Overall 56.0%

 

Table 50: Classiﬁcation Table (a) Logit Model 2 First Run With Predictors

 

Adoption
Has Adopted Percentage
Otherwise C-R Correct
Step Adoption Otherwise
1 GR Whole Muscle 75 10 882%
Has Adopted C -R 5 103 95.4%
(a) The cut value is .50 Overall 92.2%

 

163

A separate diagnostic test for multicollinearity was conducted, revealing relatively
few explanatory variables; however, it had multicollinearity problems. Only 3
explanatory variables showed signs of multicollinearity, which were not consistent with
both the tolerance and the variance inﬂation factors: only one or the other. With these
promising results, another run of Model 2 was conducted which is discussed in the next

sections.

4.4.15 Binary Logistic Regression Model 2: Second Run

The second run of Model 2 began by dropping seven highly non-signiﬁcant
coefﬁcients from the base model with p-values ranging ﬁ'om 0.30-0.85. The explanatory
variables dropped include: Q7, Q49, Q50, Q52, Q53, Q59, and Q73.

The results of the second min are listed in Table 51 below and Model 2 was vastly
improved over the ﬁrst run. The second run had sixteen explanatory variables in the
model. The table below shows there are twelve signiﬁcant explanatory variables as
determined by the Wald chi square test statistics.

The Omnibus Tests of Model Coefﬁcients was again similar to the ﬁrst run with a
highly signiﬁcant chi square test statistic. The encouraging result between the ﬁrst and
second runs of Model 2 is found in the Homer-Lemeshow Model chi square with the
entire model chi square is now signiﬁcant after dropping seven explanatory variables
from the ﬁrst run to this the second run. The now signiﬁcant Homer-Lemeshow statistic
demonstrates that Model 2 adequately ﬁts the data set being analyzed.

The pseudo R2 results in this the second run were similar to the ﬁrst run with a
Cox & Snell R2 of 59 percent and the Nagelkerke R2 being 79 percent, showing good

results. In logistic regression the pseudo R2 results are to be interpreted as showing the

164

relative strength of the model’s association between the explanatory variables (predictors)
and the binary dependent variable. These pseudo R2 results in this second run show a
relatively strong association between the explanatory and dependent variable.

Table 51: Binary Logistic Regression Model 2 Second Run Model: Adoption Case-
Ready Whole Muscle Beef Products

 

 

 

All Stores
Parameter Estimate SE Sig
Delivery dates are written on each meat pkg (Q9) 1,43 0,42 ***
C-R has the whole muscle cuts for my store (Q20) 0,57 0.39
C-R has the whole muscle packs for my store (Q23) 0.54 0.35
Shrink: Customer Theﬁ (Q39) -0.20 0.23
Shrink: Defective pkg (Q40) 0.81 0.34 **
Shrink: Workmanship (Q41) -1.07 0.36 ***
Shrink: Non-bloom whole muscle beef cuts (Q44) 0.86 0.27 ***
Shrink: Out of Date (Q46) -0.60 0.35 *
Satisﬁed Store Preps First-of-Month Sales (Q54) .18] 0,50 ***
Satisﬁed Store Preps for Meat Specials (Q55) -0.59 0.32 "'
C-R Meats Safer than In- Store Processed (Q60) 1,87 0,39 ***
“Blooming” of C-R meats is a problem (Q68) .098 0,39 **
C~R Meat Packages Use Attractive Trays (Q70) 089 035 **
Difficult to Find Qualiﬁed Meat Cutters (Q72) 059 0.18 ***
With C-R Meat Mgrs. Forecast/1nventory(Q74) .039 0,25
C-R Products Are More Cost Effective (Q75) 0,71 0.30 **
Intercept -5.04 2.08 ”*
Omnibus Tests of Model Coeﬁicients (Chi Square) 172,94 16df ***
Hosmer-Lemeshow Model Chi-Square 9.52 8df **"‘*

Pseudo R2 for Base Model:
Cox and Snell: 59% Nagelkerke: 79%
Signiﬁcant Codes:****p>.05 ***p<.0l "mos *p<10
Number of Observations 193
Tables 52 and 53 below show the results of the classiﬁcation tables from the
second run of the model. The classiﬁcation table shows excellent results, with an overall
percentage of 91.7 percent. More importantly these two tables showed the model
improved from Step 0 with the baseline (constant only) logistic regression model of 56
percent. By adding the explanatory variables of Model 2 in Step lof the second run
revealed the explanatory variables of interest are interpreted as useful predictors of

adoption of case-ready whole muscle beef programs.

165

Table 52: Classiﬁcation Table (ab) Logit Model 2 Second Run Without Predictors

 

Adoption
Has Adopted Percentage
Otherwise C-R Correct
Step Adoption Otherwise
0 GR Whole Muscle 0 85 0%
Has Adopted OR 0 108 100,0%
(a) Intercept only model. Overall 56.0%

(b) The cut value is .50

Table 53: Classiﬁcation Table (a) Logit Model 2 Second Run With Predictors

 

Adoption
Has Adopted Percentage
Otherwise C-R Correct
Step Adoption Otherwise o
l C-R Whole Muscle 75 10 88'2 /°
Has Adopted C—R 6 102 94.4%
(a) The cut value is .50 Overall 91.7%

 

Another diagnostic test for multicollinearity was conducted on the variables from
the second run before running another model. It revealed only one explanatory variable
showing borderline multicollinearity problems. The tolerance level was 41 percent and.
the variance inﬂation factor was 2.45. With these promising results, a ﬁnal run of Model

2 was conducted which is discussed in the next sections.

4.4.16 Binary Logistic Regression Model 2: Final Run

The ﬁnal run of Model 2 began by dropping three non-signiﬁcant coefﬁcients
with p-values ranging ﬁom 0.12-0.39. The explanatory variables dropped include: Q20,
Q39, and Q74.

The results of the ﬁnal run for Model 2 are listed below in Table 54. The ﬁnal
run began with thirteen explanatory variables in the model and ﬁnished with eleven

signiﬁcant explanatory variables as determined by the Wald chi square test. The model

166

was improved from the second run, albeit slightly. One noticeable change in this model
was it had only two non-signiﬁcant explanatory variables. The ﬁrst non-signiﬁcant
variable was related to shrink associated with product that was out of date; the other
variable relates to the store’s ability to prepare for ﬂesh meat sold at special prices.

The Omnibus Tests of Model Coefficients was again similar to the last run with. a
highly signiﬁcant chi square test statistic. The Homer-Lemeshow Model chi square, for
the entire model was again signiﬁcant after dropping the three explanatory variables
postulating the ﬁnal run of Model 2 adequately ﬁts the data set being analyzed.

The pseudo R2 results were slightly less than the previous run with a Cox & Snell
R2 of 58 percent and the Nagelkerke R2 being 78 percent. R2 statistics in logistic
regression run noticeably lower than OLS regression, revealing the strength of these R2
results. Final conclusions from these R2 results are that there is a relatively strong

association between the explanatory variables and the dependent variable.

167

Table 54: Binary Logistic Regression Model 2 Final Run Model: Adoption Case-
Ready Whole Muscle Beef Products

 

 

 

All Stores
Parameter Estimate SE Sig
Delivery dates are written on each meat pkg (Q9) 1.17 0.37 ***
C-R has the whole muscle packs for my store (Q23) 0,92 0.26 ***
Shrink: Defective pkg (Q40) 0,63 0.30 **
Shrink: Workrnanship (Q41) -1.09 0.34 *"‘*
Shrink: Non-bloom whole muscle beef cuts (Q44) 0,74 0,24 ***
Shrink: Out of Date (Q46) ~0.45 0.31
Satisﬁed Store Preps First-of-Month Sales (Q54) -1_70 0,48 ***
Satisﬁed Store Preps for Meat Specials (Q55) -0,49 0.30
C-R Meats Safer than In-Store Processed (Q60) 1.63 0.35 ***
“Blooming” of C-R meats is a problem (Q68) -0,84 0.34 **
C-R Meat Packages Use Attractive Trays (Q70) 07 3 0.31 **
Difﬁcult to Find Qualiﬁed Meat Cutters (Q72) 06] 0.17 ***
C-R Products Are More Cost Effective (Q75) 0,53 0.25 **
Intercept -3 .99 1 .90 **
Omnibus Tests of Model Coefﬁcients (Chi Square) 167,45 13df ***
Hosmer-Lemeshow Model Chi-Square 10.59 8df *"'*"‘

Pseudo R3 for the Final Run of Model 2:
Cox and Snell: 58% Nagelkerke: 78%
Signiﬁcant Codes:****p>.05 ***p<.0l **p<.05 *p<10
Number of Observations 193
Out of the 13 explanatory variables in Model 2, there were only two with non-
signiﬁcant p-values (see Table 54 above). The ﬁrst failed variable was concerned with
shrink associated with non-bloom issues in whole muscle beef cuts. Research and
interviews with industry executives revealed that new processing and packaging
technologies had been implemented to assuage past non-bloom issues with whole muscle
beef cuts. Non-bloom issues the new technologies hoped to alleviate include: product
that would not fully bloom prior to being placed into the meat cases (there would be parts
of the product that would not turn “cherry red”), and the case-ready packages which were
packed in corrugated boxes and master bags requiring different blooming regimes and

times lacked a consistent method for blooming. The past reactions/viewpoints of the

respondents coupled with the results of the literature review about the blooming problems

168

of these products caused the predicted sign to be negative. One plausible explanation is
the research misinterpreted the predicted sign where a positive sign should have been
predicted, meaning the new packaging and processing technologies had improved the
blooming capabilities and “more” amounts of this variable would in reality increase the
adoption of center of the plate protein.

The last variable with a failed predicted sign was concerned with the difﬁculty of
ﬁnding qualiﬁed persons to cut and process fresh meat products in-stores. Prior research
again documented the problems retailers were encormtering in hiring qualiﬁed persons to
cut meat. A positive sign would have reinforced the results of the research. A partial
explanation for the negative sign centers on a selection bias problem for some
respondents. A signiﬁcant number of respondents do not use meat cutters to process their
fresh meats, and which could have caused the sign to be negative. Negative amounts of
this variable would cause adoption of the, whole muscle meats to increase, is a plausible
interpretation for the sign of this variable.

Tables 55 and Table 56 below show the results of the classiﬁcation tables from
the ﬁnal run of Model 2. Similar to the other classiﬁcation tables, results of this run
showed an improvement from Step 0 with the baseline (constant only) logistic regression
model of 56 percent to Step 1 where the explanatory variables were added. The
classiﬁcation table shows nearly identical results from the last run, with an overall
percentage of 90.7 percent. This improvement between Step 0 and Step 1 reveals the
explanatory variables of interest for Model 2 can be interpreted as useful predictors of
adoption of case-ready whole muscle beef products. This is a signiﬁcant and

encouraging result for the project.

169

Table 55: Classiﬁcation Table (ab) Logit Model 2 Adoption of Case-Ready Whole

 

Adoption
Has Adopted Percentage
Otherwise C-R Comet
Step Adoption Otherwise 0
0 OR Whole Muscle 0 85 -0 /o
(a) Intercept only mode]. Overall 56.0%

(13 The cut value is .50

Table 56: Classiﬁcation Table (a) Logit Model 2 Adoption of Case-Ready Whole
Muscle Beef Products Final Run With Predictors

 

Adoption
Has Adopted Percentage
Otherwise C-R Correct
Step Adoption Otherwise o
1 CR Whole Muscle 74 ll 87'1 /°
Has Adopted CR 7 101 93.5%
(a) The cut value is .50 Overall 90.7%

 

4.4.17 Binary Logistic Regression Model 2: Multicollinearity Diagnostics

The ﬁnal run in logistic regression has many signiﬁcant variables and a check for
multicollinearity is prescribed. First, none of the standard errors of the coefﬁcients in
Table 57 are considered large enough (greater than 2.0) to signal problems with
multicollinearity. Next, an OLS regression with the explanatory variables and the
dependent variable for Model 2 was constructed to check for multicollinearity. The
diagnostics showed no signs of multicollinearity for any variables for either the tolerance
level or variance inﬂation factor and all of the tolerance levels for the model are at or
above the prescribed levels (they were all greater than .40) for multicollinearity (see
Table 57 below). The variance inﬂation factor for each coefﬁcient showed excellent

results as well. None of the variance inﬂation factors were greater than 2.5. Therefore, it

170

is concluded that multicollinearity is not a problem for Model 2 having conﬁrmed the
logistic regression output ﬁles none of coefficients possessed inﬂated values of the
standard errors. And in a separate multicol linearity diagnostic test none of the tolerance
levels or variance inﬂation factors were at levels detecting problems with

multicollinearity.

Table 57: Multicollinearity Diagnostics for Logit Model 2a Whole Muscle Beef

 

 

Collinearity Statistics*

Imgit Model 2 Tolerance VIF
(Constant)

Writes delivery dates Q9 0.85 1.18
Whole muscle pack sizes Q23 0.70 1.44
Shrink defective packaging Q40 0.65 1.53
Shrink workmanship issues Q41 0.65 1.53
Shrink non-bloom beef cuts Q44 0.82 1.22
Shrink out-of-date Q46 0.74 1.35
How satisﬁed store is prepared for FOM Q54 0.73 1.36
How satisﬁed store is prepared for specials Q55 0.74 1.35
C-R products safer than in-store processed Q60 0.69 1.45
Blooming c-r is a problem Q68 0.88 1.13
C-R products use attractive trays Q70 0.71 1.42
More difficult to ﬁnd meat cutters Q72 0.84 1.20
C-R products more cost effective Q75 0.65 1.54

 

' ' Binary dependent variable: Adopted c-r ‘whole muscle beef products

*Note: Tolerances S .40 = Problems with Multicollinearity VIP 2 2.50 = Problems with
Multicollinearity

 

171

4.4.18 Binary Logistic Regression Model 2: Predicted/Observed Signs of
Coefﬁcients

As in Model 1, a summary of the signs for the explanatory variables for Model 2
is in order. A plausible check for the robustness of binary logistic regression models and
the explanatory power of the variables can be accomplished by analyzing the signs of
each explanatory variable. This analysis is limited to comparing the predicted sign for
each explanatory variable with the actual results of the binary logistic regression in
Model 2. The interpretation of the sign for each variable will be restricted to the
coefﬁcients that are statistical signiﬁcant as determined in the SPSS results of the Wald
chi square statistical test for each coefficient. See Table 58 (below) where each
explanatory variable, predicted sign direction and the actual results are reported

As the Table 58 below shows, the remaining nine variables have coefﬁcient signs
that are consistent with prior analysis and with observations discovered during the ﬁeld
research. These variables combined with the two variables that have plausible
explanations of their signs conﬁrm the “robustness” of this entire model, which further
identiﬁes those factors retailers and processors could employ towards gaining greater

adoption rates of fresh case-ready whole muscle meat programs.

172

Table 58: Binary Logistic Regression Model 2 Adoption of Case-Ready Whole
Muscle Beef Products: Interpreting the Sign and Direction of the
Explanatory Variables

 

Explanatory Variable/Hymthesis/Predicted itgp Direction Results
Q9. How satisﬁed are you with the way your store writes the delivery date on each
fresh meat box received from the DC?
0 1=Not at all to 5: Very Satisﬁed
Ho; Research showed the shelf-life of or beef was maximized when the oldest Passed
product in the walk-in was the next product to go in the meat cases. The chances of
this metric being fulﬁlled were predicated on the delivery date was written on the
boxes. The sign was predicted to be positive which would encourage further
adoption of c-r beef products.
Q23. Case-ready whole muscle beef products are available in the pack sizes my
customers prefer?
o 1=Strongly Disagree to 5=Strongly Agree
Ho; Research efforts and interviews with processors hypothesized c-r products were Passed
being produced with the pack sizes retailers preferred. When these whole muscle
products are available in the pack sizes retailers demand for their customers, further
adoption of case-ready meat products is more likely. The sign was predicted to be
posrtive.
Q40. Shrink: Associated with defective and damaged packaging.
o 1=Big Concern to 5=Not a Concern
Ho; Current designs for case-ready whole muscle meat products have changed
because defective packaging used to be a problem for these types of meat products. Passed
Processors took that information and developed new packaging for these products
and believed the packaging problems were no longer a problem. The sign was
predicted to be positive.
Q41. Shrink: Associated with workmanship issues.
0 1=Big Concern to 5=Not a Concern
Ho1 When workmanship issues are a big concern, adopting case-ready is more likely. Passed
The sign was predicted to be negative.
Q44. Shrink: Associated with non-bloom issues in whole muscle beef cuts.
0 1=Big Concern to 5=Not a Concern
Ho, When non-bloom issues related to shrink are a big concern, adopting case-ready FAILED
is more likely. The sign was predicted to be negative.
Q46. Shrink: Associated with out of date product Variable
o 1=Big Concern to 5=Not a Concern was not
Ho, When out of date product issues related to shrink are a big concern, adopting signiﬁcant
case-ready is more likely. The sign was predicted to be negative.
Q54. How satisﬁed are you with the way your store prepares for the increased sales
experienced during the ﬁrst few days of each month?
0 1=Not at all to 5=Very Satisﬁed Passed
Ho: When preparing for increased sales during the ﬁrst few days of the month are a
problem for managers and they are not satisﬁed with those efforts, adopting case-
ready is more likely. The sign was predicted to be negative.
Q55. How satisﬁed are you with the way your store prepares for the increased sales
experienced during the ﬁrst days of the month associated with special fresh meat
promotions? Variable
0 1=Not at all to 5=Very Satisﬁed was not
Ho: When preparing fresh meat promotions for increased sales during the ﬁrst few signiﬁcant
days of the month are a problem for managers and they are not satisﬁed with those
efforts, adopting case-ready is more likely. The sign was predicted to be negative.

173

Table 58 (cont’d)
Q60. Case-ready beef products are safer than beef products processed in-stores?
o 1=Strongly Disagree to 5=Strongly Agree
Ho: If store managers think case-ready beef products are safer than in-store Passed
processed meats they are more likely to adopt case-ready. The sign was predicted to
be positive.
Q68. Case-ready meat products that require “blooming” cause operational
problems.
0 1=Strongly Disagree to 5=Strongly Agree Passed
Ho: Ifmanagers strongly disagree with blooming causing operational issues adopting
case—ready is more likely. The sign was predicted to be negative.
Q70. Case-ready beef products are packaged in attractive trays.
o l=Strongly Disagree to 5=Strongly Agree Passed
Ho: If case-ready beef products are packaged in attractive trays more likely to adopt.
The sign was predicted to be positive.
Q72. It has become increasingly more difﬁcult to ﬁnd qualiﬁed meat cutters for my
store. Failed’
0 1=Strongly Disagree to 5=Strongly Agree
Ho: If store managers agree it is more difﬁcult to ﬁnd qualiﬁed meat cutters they
would be more likely to adopt case-ready meats. The sign was predicted to be
posrtive.
Q75. Case-ready beef products compared to “in-store” processed beef products are
more cost effective.
0 1=Strongly Disagree to 5=Strongly Agree Passed
Ho: Research and interviews discovered the prices charged for c—r products ceteris
paribus were signiﬁcantly higher than meat processed in-stores and the transaction
costs and other intangibles mitigated the higher costs. The project was hopeful the
sign would be positive.
* Could be a selection bias problem for some respondents, because they don’t use meat cutters.

4.4.19 Binary Logistic Regression Model 2: Managerial Implications

From the Table 58, which highlights the signs and interpretations for the predictor
variables in Model 2, it is possible to categorize the signiﬁcant explanatory variables or
predictors into three broad categories that yielded managerial implications for whole
muscle beef products. These categories were in fact the major headings for the research’s
six operational issues that were discussed earlier: The three adoption categories include:
1) supply chain metrics, 2) product attributes demanded by- processors, retailers and
consumers and 3) management tools, speciﬁc for Limited Assortment Retailers to help

them improve their business model.

174

Supply Chain Metrics (Model 2)

Importance of marking the delivery date on each carton of fresh beef delivered to
stores. Recognizing and implementing this important step is deemed critical towards
improving inventory management. It ensures products have maximum amount of
time to be sold for ﬁrll retail price (avoiding markdowns or write-offs).

Defective and damaged packaging from the perspective of supply chain metrics
relates to the necessity of discovering where along the supply chain the damage was
incurred and make the necessary changes to improve stops along the supply chain and
mitigate future damaged products.

“Blooming” case-ready meat products is a problem. Metrics designed by processors
and retailers are not internalized and institutionalized. Blooming case-ready meat

products takes time and planning. If these metrics are not followed product will not

get through the supply chain efficiently.

Speciﬁc Product Attributes Demanded by All Actors (Model 2)

Case-ready whole muscle beef products are available in the complete range of pack
sizes. Retailers strive to disiplay “pack sizes” that are in demand by their customers.
Production and sales of whole muscle beef products that are cut and packaged in
stores provides data/information to ensure their stores have the whole muscle pack

sizes that optimize sales. Interviews of store managers revealed this important point.

175

Shrink associated with workmanship issues in case-ready programs are an
opportunity for managers to account for all products with “workmanship” issues
before customers view/buy the product This information needs to get back to
processors to help them continually improve their offerings by isolating the source of
the workmanship problems. Packaging has been viewed by as marketing tool to help
retailers.

Case-ready beef products are viewed by store managers as being safer than products
produced in-stores; and it is clear that safety is an important feature to both retailers
and consumers. Currently when other fresh food products are deemed safer this fact
has allowed food marketers to differentiate those products in the marketplace and are
able to charge higher prices.

Case-ready products are packaged in attractive trays. Current packaging research
posits one way to differentiate products in the marketplace and in minds of consumers

is through new and innovative packaging alternatives.

Management Tools (Model 2)

Case-ready meat programs enable managers to maximize store sales during the ﬁrst
few days of each month at which time shoppers have greater income for food
shopping. Limited assortment stores experience unusually high levels of store sales
during the ﬁrst few days of each month Products that can be seen as less labor

intensive will be positively viewed by decision makers. (Similar to Model 1)

176

Case-ready meat programs mitigate the need for searching for skilled meat cutters.
Meat cutters have higher labor cost per unit as compared to other store personnel;
moreover, the availability of qualiﬁed meat cutters is declining (similar to Model 1)

Case-ready meat programs are viewed as more “cost effective” as compared to fresh
meats processed in-stores. This predictor was not signiﬁcant in Model 1; once all
costs, including transaction costs are included could make case-ready meat an
attractive alternative. Albeit for whole muscle meat products only, not ground beef.
This is consistent with research/industry interviews on how the production of ground
beef at a central point beneﬁts by utilizing meat trim that at store-level would be

regarded as waste/ considered shrink.

177

5.1

CHAPTER 5— SUMMARY AND CONCLUSIONS
Study Objectives

The objective of this research is to determine those operational issues that may

affect fresh meat processors’ and retailers’ decisions whether or not to integrate case-

ready program into their product lines.

To develop a better understanding of the

operational concerns six research questions were developed the research process by

means of: interviewing key industry experts dining a one day research seminar held at

Michigan State University, directly observing case-ready products as they moved through

the supply chain; and by conducting in-depth interviews with retail store managers and

distribution center managers from Save-A-Lot (SAL) grocery store chain with over 1,150

stores throughout the US.

RQI

RQz

RQs

RQ4

RQs

RQs

The six main research questions are:

What are the differences between corporate and licensed stores in their
perceptions of the operational concerns associated with case-ready products?

What differences exist between corporate and licensed23 stores in their perception

of the operational concerns associated with food safety?;

What are the differences between corporate and licensed stores
perceptions of the operational concerns associated with workmanship?;

What are the differences between corporate and licensed stores
perceptions of the operational concerns associated with shrink?;

What are the differences between corporate and licensed stores
perceptions of the operational concerns associated with packaging?;

What are the differences between corporate and licensed stores
perceptions of the operational concerns associated with handling?

 

‘7 . . . .
‘3 Licensed stores are srrnrlar to franchise stores.

178

in their

in their

in their

in their

5.1.1 Purpose of Study

Decision-makers in the fresh beef industry have for years contemplated and
attempted various strategies for increasing adoption of case-ready technology and this
research project seeks to assist the industry in those endeavors.

Sub-Objectives of the Project

0 Of interest to this research project are the reasons why red meats case-ready
programs have not been accepted as readily by the retail grocery store industry as
has been the case with the poultry industry. Case-ready broilers (chicken)
programs have enjoyed steady growth and industry acceptance over the past thirty
ﬁve years. Centralizing the cutting and packaging of broilers has encouraged
many new product innovations in the form of a broad line of convenience
products. Case-ready broiler programs have beneﬁted retailers and consumers in
several ways including: providing consumers with safer products, efficient
stocking, and highly functional packaging. Both retailers and consumers have
overwhelmingly accepted these developments.

0 In contrast, the beef industry is eager to accelerate the change to what many in the
industry think someday will be the acceptance of case-ready beef and would like
to better understand how to achieve the same kind of results that have accrued to
the far more progressive poultry industry.

0 The question this research addresses is why, despite the many apparent
advantages of case ready beef products to retailers and consumers retailers have
been relatively slow to adopt case-ready programs. Yet, for many years red meat

industry leaders have expressed their belief that eventually the logic of adopting

179

this technology will take place; especially now that Walmart (the world’s largest
retailer) has recently adopted case-ready programs. This issue is critically
important given the important share of store sales of red meats. Because the trade
and academic literature does not provide answers to this central question, there is
substantial need for this project. It is expected that Limited Assortment Store
retailers, processors and packaging companies are likely to beneﬁt ﬁom this

research project.

5.1.2 Data Sources and Methodology

The project began with a one-day seminar at Michigan State University with meat
industry experts and academicians to discuss the need for and appropriate approach to
researching the topic. Next, a case study was conducted to collect information and
artifacts about the current case-ready meat program in use at Save-A-Lot company (a
Limited Assortment Store) and to document the case-ready meat supply chain. Another
goal of the case study was to develop a relevant survey questionnaire based upon
corroborated documented operational issues observed in the process. Documentation of
the supply chain was accomplished by developing a process map of case-ready products
at each point in the supply chain.

Documenting the supply chain employed direct observation of the entire case-
ready process. Included were a meat processor, three Distribution Centers, and 12 SAL
retail grocery stores. Several hypotheses about certain operational issues were dispelled
during this phase. Initially, it had been assumed case-ready products arrived at
Distribution Centers ﬁ'om the meat processor in a degraded state, which worsened as it

was shipped and handled to the stores. To the contrary, it was observed that product

180

arrived at Distribution Centers as speciﬁed in the ordering system to the meat processors.
Another misperception that was dispelled was that standard operating procedures at the
Distribution Centers were not being followed, subsequently jeopardizing product quality
as it entered the retailer’s cold chain It was observed that state-of-the art facilities,
procedures and metrics were present at the Distribution Centers resulting in maximized
shelf life of all perishable products, including case-ready products.

To further understand the supply chain of case-ready products, a process map was
constructed to document each stop along the supply chain beginning with product
arriving from the meat processor and ending when the products are displayed for sale in
retailers’ display cases. It should be noted the process map begins with the product being
ordered ﬁ'om individual retailers. Documenting the process map was instrumental in
identifying the key operational issues.

The next phase of the research was to undertake a comprehensive literature
review of case-ready products from several perspectives, including the existing case-
ready models in use, packaging alternatives and food safety issues associated with all
fresh meats. A review of the Accounting and Strategic Management literature was
conducted relative to topics of outsourcing and/or “make-or-buy” management decisions.
A complete review of the Transaction Cost Economics literature was performed, paying
particular attention to the economic perspectives of outsourcing decisions. These
literature reviews resulted in better understanding of the supply chain and the operational

issues, as well as outsourcing decisions of functions considered core to a ﬁrm.

181

The questionnaire was developed based upon the research questions and
propositions that centered on the following operational issues: workmanship, shrink,
handling, case-ready product, packaging, and food safety.

The questionnaire also included several questions related to fresh meat sales and
costs. One of the cost related questions was asked respondents about their “perceptions”
of costs associated with adopting a case-ready strategy. Another question sought
retailers’ perceptions of retail price differences between case-ready products and in-store
packaged meats. During the data analysis phase it was apparent most of the respondents
did not provide useful data in response to these questions. Follow up interviews with
SAL managers revealed most respondents believed the information required to accurately

answer these questions was proprietary, which discouraged responses.

5.1.3 General Conclusions Drawn from the Questionnaire

Questionnaires were administered to 110 corporate stores and 110 licensed stores.
Of the 220 questionnaires, 193 were returned. Corporate stores returned 109 and licensed
stores returned 84, yielding an overall 87.7 percent response rate. Senior management
encouraged store managers to respond to the survey, which accounts for this high
response rate. The analysis of the ﬁrm-level data began with cross-tabulations of the
Likert scale and ranking scale questions, as well as a qualitative evaluation of several
open-ended questions. The data was also analyzed using multiple regression and binary

logistic regression.

5.1.4 Major Findings from the Cross Tabulations Analysis
Food Safety Propositions: The ﬁrst research proposition relative to food safety

showed 65.8 percent of respondents from both store types agreed or strongly agreed case-

182

ready meats were the safest alternative compared with in-store processed meats. This
result is impressive given the large number of respondents’ stores that do not sell case-
ready meats. The results concerning the second food safety proposition illustrated there
was no difference between store types that a foodbome illness would be catastrophic to
sales. Moreover, both store types agreed or strongly agreed with the statement. One of
the food safety research propositions hypothesized that if customers recognized that the
ﬂesh meat they were purchasing was the “safest” available, their customers would
respond by purchasing more from that store. Respondents agreed or strongly agreed their
customers would indeed purchase more fresh meats if they believed those products were
the safest. This is an important ﬁnding in support of adopting case-ready programs in as
much as case-ready products are processed and packaged under food safety protocols and
HACCP regulations.

Workrnanship Propositions: Several of the workmanship research propositions
pertained to the appearance of fresh meat products. Nearly sixty-two percent of all store
managers believed ﬁ'esh meat prepared in-stores were superior in appearance to case-
ready products. The case study and literature review revealed case-ready packaging
ﬁrms were convinced the low head space packaging (oxygen/low atmosphere packaging
technology) closely resembled in-store processed meat products, which enhances the
favorable “workmanship” impression of case-ready products.

Respondents were asked which alternative (case-ready or in-store processed) was
more consistent in workmanship. Nearly sixty-one responded there was no difference

between alternatives. The consistency of acceptable workmanship is an important

183

ﬁnding for case-ready meat processors that are concerned with its perceived lack of
positive product appearance.

Shrink Propositions: Shrink was deﬁned for respondents as any dollar amount
received that was less than full-expected retail price for meat products. Nearly ﬁfty
percent of all respondents reported that case-ready programs were superior in managing
shrink Several follow up questions were asked about speciﬁc sources of shrink,
including: customer and employee theft, defective packaging, workmanship issues, non-
bloom issues, and out-of-date product. All of these questions scored as a concern for
both store formats. The two biggest concerns for both store types were customer theft
and when case-ready ground beef products did not “bloom” to the traditional red color.

Another shrink-related proposition was concerned with respondent’s satisfaction
with how store records accounted for fresh meat packages not sold at full retail price.
Both store formats responded being satisﬁed However, these ﬁndings were inconsistent
with the literature review, meat executive interviews, the one-day seminar, and Dr.
Levin’s Activity Based Costing analysis of case-ready products.

Packaging Propositions: The packaging propositions were developed around three
key observations. The ﬁrst concerned case-ready products that require “blooming”
before they are ready to be stocked in the meat cases. The results of blooming question
from corporate store respondents documents that 46.8 percent disagreed or strongly
disagreed “blooming” case-ready products cause operational problems. This point is well
taken, because corporate store managers are familiar with this task and are in the position

of accurately commenting on this issue. For the licensed stores 54.8 percent disagreed or

184

strongly disagreed blooming was a problem. These ﬁndings are supportive of case-ready
advocates.

The second proposition focused on the attractiveness of case-ready packaging.
Corporate stores responded with 65.1 percent agreed or strongly agreed the packaging
was attractive compared with the licensed stores responded with 39.3 percent agreed or
strongly agreed. For licensed stores the percentages were larger when they were
questioned about the low headspace packages which closely resemble in-store processed
meats. This is another ﬁnding supportive of case-ready pr0ponents.

Finally, a research proposition inquired of store managers’ opinions regarding
case-ready packaging alternative of high oxygen, high headspace relative to being a
detriment to fresh meat sales. The corporate stores responded with 32.1 percent agreed or
strongly agreed with that statement, while the licensed stores were decidedly different,
with only 13.3 percent of respondents believing the high oxygen packaging for ground
beef was not detriment to sales. As expected, the low-oxygen low headspace for ground
beef and whole muscle cuts packaging alternative was preferred to high oxygen
alternative. This ﬁnding portends more products being packaged in a low oxygen/low
headspace tray and ﬁlm.

Handling Proposition: It should be noted that handling relates to the human
interaction/treatment of meat products. This proposition elicited both the tangible and
intangible aspects of handling. Case-ready products have been promoted as the best
means of maintaining ﬁesh meat product inventory levels meat display cases throughout
. all hours of store operation. The chi square test statistic was not signiﬁcant, revealing no

difference between store types in keeping meat cases serviced once meat cutters leave the

185

store (typically 4:00 PM) which is the acknowledged most challenging time to keep fresh
meat cases stocked. According to SAL management, the time ﬁ'om 4 PM until 7 PM
generally is the busiest time in most Limited Assortment Stores, and a time to maximize
store revenue by keeping the meat cases fully stocked

Proﬁtability Pr0position: The ﬁrst case-ready product research propositions were
interested in the proﬁtability of both alternatives. Just over 43 percent indicated they
believe that processing in-stores was more proﬁtable than case-ready programs, while
45.9 percent believed the case-ready program alternative was more proﬁtable. Licensed
stores overwhelmingly believe (75 percent) in—store processed meats were more
proﬁtable; and only 17 .9 percent responded that case-ready programs are more proﬁtable.
This is consistent with the interviews of industry executives/management and store
managers who reported the low proﬁtability image of case-ready programs are the
principal reason for slow adoption.

Fresh Meat Preference Proposition: The proposition relating to fresh meat
preferences were based upon several factors: color, servings per package and package
price. Managers of both store formats chose package price as the as being most
important to shoppers. Another question asked store managers to rate how sensitive their
customers would be to prices for several products. Ground beef was the most sensitive to
price, for both formats. Another set of propositions was interested in acquiring
information about speciﬁc case-ready attributes, including: product appearance, shelf-
life and item (SKU) variety. Surprisingly, 47.7 percent of corporate store managers, all

of which are operation with case-ready programs, believe in-store processed meats

186

possessed better appearance. Managers of licensed stores agreed (80.2 percent). This
result suggests further investigation to determine the basis of this ﬁnding.

Shelf Life Proposition: The proposition relative to shelf life is an important issue
and linked directly to sales and shrink for retailers across formats. Nearly 65 percent of
respondents believe case-ready programs are superior. SKU varieties are linked directly
to the amormt of product offerings a retailer can offer; and 52.9% percent of respondents
from both kinds of retailers believe that in-store processed meats offer more variety.
However, 25.9 percent of respondents believe there was no difference. One explanation
of this result is that both formats are Limited Assortment Stores which as a matter of
policy limits product assortment in all product categories.

Several product questions were designed around these important product
attributes: the lean-fat ratios, pack size, preferred cuts, and the availability of ground beef
products having sufficient lean-fat alternatives. Both store types agreed and strongly
agreed ground beef was available in the appropriate lean-to-fat ratios. Another
proposition questioned whether beef cuts being sold were those their customers preferred
In corporate stores, 68.3 percent of respondents answered from the median to strongly
disagree; while the licensed stores responded at 66.4 percent from the median down to
strongly disagree. These sharp differences corroborate meat processors’ views that more
beef cuts are required. Not only are more types needed, but in pack sizes that conform to
Limited Assortment Stores’ customers. It was also discovered that there is suitable
availability of case-ready ground beef in the desired pack sizes for store customers in

both store types. Meat processors commented during the interview process that case-

187

ready ground beef programs provide retailers choices of several pack sizes and lean-to-
fat ratios

Workrnanship Proposition: A key workmanship research proposition focused on
the additional skill sets meat department personnel would be required to have once
adopting a case-ready strategy. The cross tab showed 67.9 percent of all respondents
agreed or strongly agreed meat cutters could feasibly be trained to manage the entire
perishables category. The exact same results were recorded for the question concerning
shiﬁing meat department personnel from meat cutting to category forecasting and
inventory management (67 .9 percent of respondents agreed or strongly agreed). This is
considered an important ﬁnding as it pertains to managing the perishables department
(meats and produce).

An open ended question was posed concerning what case-ready products
customers would purchase which are not currently being offered. In the 16 respondents
the following products were cited in order: Premium steaks, extra lean ground beef

products, and premium-quality pork and beef roasts.

5.1.5 Multiple Regression

Six regression models were developed, one for each operational issue (except the
case-ready meats Operational issue) and one model exploring a combination of the
operational issues. Both the dependent and independent variables are based on Likert
scale questions.

The multiple regressions models had decidedly mixed results. Only the food
safety and workmanship models produced results containing both signiﬁcant independent

variables and decent R2 scores. Another strategy was employed at that point to employ a

188

dependent variable based upon the economics or cost effectiveness of case-ready
products, matched with a combination of independent variables constructed from all
operational issues. Indeed, all of the variables of interest were signiﬁcant, except the
dummy variable for store type. Each of the ﬁnal independent variables in the hybrid
model was directly related with the dependent variable, “case-ready products being the
more cost effective meat alternative.”

A difﬁculty in the project was quantifying these research propositions using
multiple regressions on their individual contributions with ﬁve separate dependent
variables. Two out of ﬁve regression models produced signiﬁcant results for their
respected research propositions. With the overall results being mixed, the project built a
hybrid regression model which was constructed from all six categories of research
propositions. All of the coefﬁcients in the hybrid model were signiﬁcant with correct
coefﬁcient signs.

The variables of interest for the hybrid model include: case-ready products have
the ground beef pack sizes stores demand, satisfaction levels with shrink inventory issues,
satisfaction with meat case inventory levels, case-ready products are the safest, case-
ready meats use attractive trays and the cost of worker’s compensation is becoming cost
prohibitive.

Three important conclusions can be drawn ﬁom the hybrid model: all of the six
operational issues were represented as independent variables in the model, all of the
independent variables were highly signiﬁcant, and the independent variables included in
the hybrid model were all topics of concern in the case study. The results for this

regression are positive for developing better understanding of the operational factors that

189

help to explain why case-ready programs are effective, which should encourage an

increase in adoption rates of case-ready programs.

5.1.6 Binary Logistic Regression Model 1

Because a better understanding of the binary choice to adopt case-ready programs
is a goal of this research project, it was decided to use a logistic regression technique to
identify the behavioral choice or an event classiﬁcation This method is appropriate
when the relationships between variables and possible complexity are unknown. These
binary choices relate to decisions made about product selections or non-selections of
case-ready programs.

The primary purpose of the binary logistic regression analysis was to better
understand why fresh case-ready products have experienced slow adoption. This analysis
was designed to help identify the qualitative traits or explanatory variables that may
affect a retail grocer’s propensity to adopt the fresh case-ready programs.

There are two primary proteins that make up the majority of the products sold in
retail grocery stores; ground beef and whole muscle cuts, which were the basis for two
binary logistic regression models. Model I focused on adoption of case-ready ground
beef and Model 2 on whole muscle meat products. Both models were examined using
hypotheses about the qualitative traits or characteristics that affect adoption decisions.

To help assess the performance of Model 1, a percent correct prediction statistic
was tabulated which can be found within the Classiﬁcation Table, and is deemed the
“practical” results from running the logistic regression model. The classiﬁcation table
statistic revealed the null model of only the intercept yielded 74.1 percent. After adding

the explanatory variables, the classiﬁcation model’s overall percentage increased to 89.1

190

percent, which is indicative of the Logit l model’s ability of further explaining the factors
of case-ready programs adoption decisions.

Fm‘ther analysis of the signiﬁcant explanatory variables or predictors for Model 1
were put into three broad categories related to the decision process of adopting case-
ready ground beef programs. These broad categories are the major headings for the
research’s six operational issues that were discussed in chapter four. The three adoption
categories relevant to Limited Assortment Store retailers in helping them improve their
business model include: supply chain metrics; product attributes for processors, retailers
and consumers; and management tools.

The ﬁrst category, supply chain metrics, identiﬁes the key factors discovered in
the research that are conducive for ground beef to effectively and efﬁciently move
through the supply chain and ultimately into retailers’ display cases. The importance of
this category cannot be overstated as supply chain metrics in the past have been
instrumental in distributing perishable products through a supply chain while achieving
maximum shelf life for both retailers and customers. Increased shelf life is an important
attribute for retailers, which mitigates several other sources of shrink, improves sales for
processors and proﬁt. Moreover, it provides consumers more product options and
improves satisfaction levels of the entire retail grocery store shopping experience

resulting in greater proﬁtability. Below are the pertinent factors.

191

Supply Chain Metrics Discovered in Model 1

Importance of monitoring temperatures of product delivered from the stores. This
could also be inferred for DC receiving product ﬁ'om processors.

Importance of writing the delivery date on each carton of fresh beef delivered to
stores. Recognizing and implementing this important step is deemed critical towards
maximizing inventory management and ensuring product has the maximum amount
of time to be sold for ﬁrll retail price (no markdowns or write-offs).

Defective and damaged packaging from the perspective of supply chain metrics
relates to the necessity of discovering where along the supply chain the damage was
incurred and make the necessary changes to improve stops along the supply chain and
mitigate future damaged products.

Importance of monitoring and reconciling products not sold at ﬁrll retail. This metric
is related to properly and systematically quantifying product not getting through the
supply chain. Proper accounting should be the catalyst for improving the supply

chain.

The second adoption category is product attributes which are important for all

participants in the supply chain: processors, retailers and consumers. Product attributes

are directly related to fulﬁlling shopper demand. A more complete understanding of

product attributes is an important step in constructing demand scenarios and the ability to

help with future shifts of demand for case-ready ground beef products. Below are the

pertinent factors.

192

Speciﬁc Product Attributes Demanded by all Participants in the Supply Chain

0 Case-ready ground beef products are available with the required lean-to-fat ratios.
Retailers must have product in the “pack sizes” their customers demand. Whole
muscle beef products cut and packaged in stores use this data/information to ensure
their stores have the whole muscle pack sizes their customers demand. Especially for
whole muscle meats. Interviews of store managers revealed this important point.

- Shrink associated with workmanship issues in a case-ready scenario is an opportunity
for managers to account for all products with “workmanship” issues before customers
view/buy the product. This information needs to get back to processors to help them
constantly improve their offerings by isolating the source of the workmanship issues.
Packaging has been viewed by as marketing tool to help retailers.

o Case-ready products being viewed as safer than products produced in-stores is key
attribute retailers and consumers both require. Currently when other fresh food
products are deemed safer this fact has allowed food marketers to differentiate those
products in the marketplace and are able to charge higher prices.

- Case-ready products are packaged in attractive trays. Current packaging research
posits one way to differentiate products in the marketplace and in minds of consumers

is through new and innovative packaging alternatives.

The last adoption categories are the management tools that help further adoption
rates. This category is related to the product’s ability to help management continuously
improve their business model. An underlying fact about many management tasks in the

Limited Assortment Store business is the repetitive nature of many of the tasks.

193

Management Tools

0 Case-ready’s ability to help managers maximize store sales during the ﬁrst few days
of each month. Limited assortment stores experience unusually high levels of store
sales during the ﬁrst few days of each month and products that can be seen as less
labor and maximize sales will be positively viewed by decision makers.

o Case-ready’s ability to mitigate the need for ﬁnding future supplies of meat cutters.
Meat cutters have higher labor cost per unit as compared to other store personnel and

the supplies of qualiﬁed meat cutters are going down.

5.1.7 Binary Logistic Regression Model 2

The second binary logistic regression model was interested in the factors that will
help further adoption rates of case-ready whole muscle meat products. There were 13
explanatory variables in Model 2, and only two of these variables had non-signiﬁcant p4
values. The failed variables were related to shrink associated with product being out of
date and satisfaction levels of how the store preps for meat specials.

The results of the percent correct predicted (classiﬁcation tables) displayed a 56
percent in Step 0 when only the dependent variable is included. When the explanatory
variables were added to the classiﬁcation table in Step 1 the results increased to 90.7
percent. The improvement between Step 0 and Step 1 reveals the explanatory variables
of interest for Model 2 can be interpreted as useﬁrl predictors of adoption of case-ready
whole muscle beef products. This is a very signiﬁcant and encouraging result for the

project.

194

The use of binary logistic regression again makes it possible to categorize the
signiﬁcant explanatory variables or predictors into three broad categories that yielded
managerial implications for whole muscle beef products. Similar to Model 1 the results
of Model 2 had three broad categories which were in fact the major headings for the
research’s six operational issues. The three adoption categories include: 1) supply chain
metrics, 2) product attributes demanded by- processors, retailers and consumers and 3)
management tools, speciﬁc for limited assortment retailers to help them improve their
business model. Here are the results of that analysis:

Supply Chain Metrics (Model 2)

0 Importance of writing the delivery date on each carton of fresh beef delivered to
stores. Recognizing and implementing this important step is deemed critical towards
maximizing inventory management and ensuring product as the maximum amount of
time to be sold for full retail price (no markdowns or write-offs).

o Defective and damaged packaging from the perspective of supply chain metrics
relates to the necessity of discovering where along the supply chain the damage was
incurred and make the necessary changes to improve stops along the supply chain and
mitigate ﬁrture damaged products.

0 “Blooming” case-ready products are a problem. Metrics designed by processors and
retailers are not internalized and institutionalized. Blooming case-ready products
takes time and planning. If these metrics are not followed product will not get through

the supply chain efﬁciently.

195

Speciﬁc Product Attributes Demanded by All Actors (Model 2)

Case-ready whole muscle beef products are available in the pack sizes stores and
consumers demand. Retailers must have product in the “pack sizes” their customers
demand. Whole muscle beef products cut and packaged in stores use this
data/information to ensure their stores have the whole muscle pack sizes their
customers demand, especially for whole muscle meats. Interviews of store managers
revealed this important point

Shrink associated with workmanship issues in a case-ready scenario is an opportunity
for managers to account for all products with “workmanship” issues before customers
view/buy the product. This information needs to get back to processors to help them
constantly improve their offerings by isolating the source of the workmanship issues.
Packaging has been viewed by as marketing tool to help retailers.

Case-ready products being viewed as safer than products produced in-stores is key
attribute retailers and consumers demand. Currently when other fresh food products
are deemed safer this fact has allowed food marketers to differentiate those products
in the marketplace and are able to charge higher prices.

Case-ready products are packaged in attractive trays. Current packaging research
posits one way to differentiate products in the marketplace and in minds of consumers

is through new and innovative packaging alternatives.

196

Management Tools (Model 2)

5.2

Case-ready’s ability to help managers maximize store sales during the ﬁrst few days
of each month. Limited assortment stores experience unusually high levels of store
sales during the ﬁrst few days of each month and products that can be seen as less
labor and maximize sales will be positively viewed by decision makers. (Similar to
Model 1)

Case-ready’s ability to mitigate the need for ﬁnding future supplies of meat cutters.
Meat cutters have higher labor cost per unit as compared to other store personnel and
the supplies of qualiﬁed meat cutters are going down. (Similar to Model 1)
Case-ready’s ability to be viewed as more “cost effective” as compared to fresh meats
processed in-stores. This predictor was not signiﬁcant in Model 1; once all costs,
including transaction costs are included could make case-ready an attractive
alternative. Albeit for whole muscle meat products only, not ground beef. This is
consistent with research/industry interviews on how the production of ground beef

can absorb or use more products deemed to be waste (considered shrink) into the

ﬁnished product.

Observations

A signiﬁcant portion of this research project was concerned with the

issues/factors related to increasing the adoption rates of case-ready products. From the

retailer’s perspective the decision to adopt a case-ready strategy is one that would require

them to outsource a core item of its operations. Given the literature is resolved in

suggesting ﬁrms outsource only those functions that are considered non-core to the ﬁrm,

this research project dealt with a contrary situation in as much as retail meat operations

197

clearly is core to its business model. However, this has presented the research with the
opportunity to break new ground

One of the goals of outsourcing any business function is to achieve a more
efﬁcient and proﬁtable company by allowing it to focus on what the ﬁrm does best:
presuming the functions are core competencies.

Friedman (2004) challenges outsourcing relationships to take on this new context,
one that is organized horizontally. Horizontal collaborations are decidedly different from
the traditional vertical relationship in outsourcing in the sense that they foster an
atmosphere where strategic alliances are sought. Figure 2 below is a conceptual model
for outsourcing core products. The key loop in the model is found in the selection of the
outsourcing partner. The selection is based upon the ﬁrm ﬁrst explicating or identifying
the issues through a systematic process and then analyzing the key operational issues
associated with the core product or service. The explicating and analyzing loop are the
basis for starting the collaborative relationship with the outsourcing partner. When there
is a true collaboration in a “ﬂat world”, outsourcing partners are encouraged to work
together in sharing information whereby both ﬁrms constantly improve their respective
businesses. Outsourcing meat operations to processors and adopting case-ready
programs using this model could facilitate Limited Assortment Store retailers to become

more efﬁcient and effective, thus improving its business model.

198

Figure 2: Conceptual Model for Outsourcing a Core Product

 

 

 

 

 

 

Increased A I er
Decision to Selection of Adoption Retailer
Retailer Outsource a :9 Outsourcing :1 Rates :1: With an
Core Product Partner Improved
(Successful Business Model
Outsourcrng)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Explicate & Analyze

Shrink Issues
Food Safety Issues
Operational Issues

Workrnanship Issues
Handling Issues
Packaging Issues

 

 

 

5.2.1 Walmart’s Conversion to a Case-Ready Program

To date only Walmart, the world’s largest retailer has committed to outsourcing a
100 percent of their fresh meat offerings. With this unprecedented dedication and.
commitment to a case-ready strategy Walmart no longer processes and packages fresh
meats at store locations. One plausible explanation of this strategy would be Walmart
views outsourcing, not from the lens of ﬂesh meat products being core or non-core to

their business model, but whether the decision to outsource improves their overall

business model.

199

All retailers, whether they have adopted a case-ready strategy or process their
fresh meats in-stores face several demand factors associated with ﬁesh meat products
they include: consumers’ tastes and preferences24 for good tasting fresh meat products,
safe fresh meat products, the governmental/ environmental issues surrounding the
processing and packaging of fresh meat products are uncertain and a failure to comply
with current and future dictums could be daunting and costly, new processing and
packaging technologies associated with safe production of meats are expensive to
replicate for corporations that have thousands of retail locations and ﬁnally, the legal
liability25 associated with a food safety violation or tainted ﬁ'esh meat product.
Paradoxically, these demand factors were all similar to the ones the poultry industry
faced when they considered the adoption of case-ready programs and they are all still
relevant.

In its current form case-ready programs could concede that in-store processing of
fresh meats has a better control for supplying consumer’s demand for good tasting
products, because they have direct control of the inputs and hence control the output.
The same rationale goes for pack size, lean-to-fat ratios for ground beef products, and
cuts of whole muscle products.

Concerning a safe supply of fresh meat products, both in-store and case-ready
offer a safe supply. The critical factor is control; both alternatives offer controls and or
metrics in an effort to produce a 100 percent supply of safe meat for sale. Case-ready

products are processed under HAACP certiﬁed “clean rooms.” Case—ready has a

 

2" Also within tastes and preferences is quality, pack size, lean-to-fat ratio (grormd beef), desired cuts
(whole muscle meat products).
25 Legal liability has two perspectives: monetary damages and loss of brand equity.

200

competitive advantage in meat processing/packaging and the literature conﬁrms this
point.

Current governmental and environmental regulations are strict and have penalties
for non-compliance. Both alternatives process and comply under the current regulations.
These regulations as with all regulations can be changed at a moment’s notice, and non-
compliance is not an alternative. A key point about future regulations is what will be the
cost of complying with new regulations. Given the stringent amount of regulations and
the uncertain future of these regulations, the project acknowledges a case-ready program
has the advantage.

The legal liability associated with processing fresh meats has two components
monetary and brand equity. The monetary component of a legal liability is self deﬁning
and well documented. The loss of brand equity associated with processing/packaging
fresh meats must also be considered, the legal liability in most cases damages the entire
brand. In the retail grocery store business the loss of brand equity or conﬁdence in the
brand could be more expensive than monetary damages. In some instances, the entire
business model may take a signiﬁcant amount of time to regain market share and
consumer conﬁdence. In extreme cases, the entire business model could fail to recover.
The project gives case-ready programs the advantage.

Reviewing the above demand factors is a partial explanation for Walmart’s
commitment to a 100 percent case-ready program The point about consumer’s taste and
preferences for quality products is conceded to in-store processing. It should be noted the
project documented in literature and analysis within the questionnaire that the supply of

qualiﬁed meat department personnel is dwindling and the wages paid for qualiﬁed meat

201

joumeymen are on the rise, driving cost of maintaining meat cutters employed in the
retail setting to prohibitive levels. In total, the risks associated with in-store processing

could prove to be too for costly retailers in the future.
5.3 Implications and Concluding Remarks

5.3.1 Managerial Implications

The managerial implications are based upon the economic realities Limited
Assortment Store retailers would face if there is a foodbome illness outbreak caused by a
retailer’s processing and packaging units having failed to follow prescribed food safety
metrics and protocols. Many companies implicated in food safety situations have lost or
seriously damaged their businesses. To the extent it can be determined that case-ready
programs offer safer processing methods; retailers should give greater consideration to
ad0pting case-ready programs.

A serious consideration in retailers deciding on whether to outsource meat
operations relates to control of the factors of quality and speciﬁcations relating to
packaging, etc. Central processing of meat packages enable standards to be established
to a degree that decentralized processing in stores is not possible. The tradeoff is that
individual stores have less ﬂexibility in differentiating its meat offerings.

Analysis of the research propositions revealed for case-ready products to be
viable strategy processors must offer retailers the cuts, portions and requisite quality the
need. As case-ready demand increases production allocations will shift accordingly.
Because of these market place realities case-ready cuts and portion demands are slowly

being fulﬁlled, hampering adoption rates.

202

During the case study quality issues associated with the cmrent offering of case-
ready meat products available for Limited Assortment Stores to purchase did not arise.
However, interviews with the CEO of a private ﬁrm that controls 18 SAL licensed stores
did have concerns about the grades of beef available in the future. This would be
especially pertinent when the economy tightens, because more customers of Limited
Assortment stores will demand a higher grade and quality of beef. During the interview
only select grades of case-ready beef were available, in the future, processors were
planning to offer higher grades of beef products.

Case-ready programs should be considered from two perspectives: price and cost.
The study discovered that SAL senior management and several retail managers (those
currently not selling case-ready products) have the impression the cost of case-ready
products would be higher than meat products processed in-stores, with consequent higher
prices to shoppers. Asked if case-ready products as compared to in-store processed
products were more “cost effective”, only 23 percent of licensed stores agreed or strongly
agreed case-ready products were more cost effective; and 54 percent of managers of
corporate stores agreed or strongly agreed with that statement. Managers of licensed
stores, especially, (which cut and package meat in stores) were not convinced of the cost
effectiveness of case-ready programs. If greater adoption of case-ready programs is to be
realized, it will be important that a complete cost beneﬁt analysis be conducted taking
into account such factors as shrink, product availability in meat cases, labor factors

among other factors identiﬁed in this research.

203

5.3.2 Relationships with Customers

From the perspective of retailers that process meats in-stores there is another
concern of adopting a case-ready meat strategy. Currently, their butchers/meat cutters
have a direct relationship with customers which respond to customers’ questions about
meat products, meal preparation and concerns with quality / freshness. Many meat
department personnel offer cooking advice and other product suggestions. These
employees play an important role as the “face” of the meat department. One suggestion
offered by meat processors to maintain the personal relationship with shoppers in a case-
ready setting would be to train meat department personnel to manage all perishables
(meat and produce). The principal responsibilities would be inventory control, product
orderings, forecasting. However, they would be able to play the important customer
relations role. This role could include advice about complementary products throughout
the store, assisting in launching new products. These new positions are envisioned to
help market and grow the two perishable categories, categories where retailers garner
higher proﬁt margins and drive the proﬁtability of the entire store.

Case-ready programs offer retailers and processors a necessary stimulus to be
creative and proactive in improve each other’s business models. Outsourcing the core
function of meat operations may contradict the conventional wisdom: however, this

research reveals there are compelling factors that may support retailers adopting case-

ready programs.

5.3.3 Point of Sale Information Coordination
Retailers collect an extensive amount of data from the point of sale (cash register

data), which is useful to retailers and processors in developing models to project demand.

204

Point of sale data could provide meat processors with valuable data about product sales
including: velocity of sales, amounts, portions, grades, and quality. Coupled with
demographic information, processors could improve their demand models signiﬁcantly
while expanding their economies of scale in production and providing the Limited

Assortment Store retailer with more of the products their customers prefer.

5.4 Suggestions for Future Research

Further research efforts could include expanding the sample to include all retail
grocery store formats including Super Stores, Conventional grocery stores, and other
Limited Assortment Store grocery retailers. New research into case-ready products
would beneﬁt from access to sales and cost data of case-ready products, new product
offerings, pack sizes, grade, quality, velocity/timing of sales and the possibility of Direct
Store Delivery (DSD) of case-ready.

These suggestions should further the use and understanding of outsourcing core

functions through information sharing and metrics for each supply chain partner.

205

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

APPENDIX A. SUPPLY CHAIN PROCESS MAP AND SURVEY
Steps Critical Activities Along Flow Time Cold-Chain
the “Vulnerability” Operational Issues
Case-Ready Cold Chain Index
1. SAL buyers order product 0 0 Lead times (7-10 days)
2 Product shipped ﬁom —-> 10
processor
3. Tractor/Trailer arrives at D 7-8
DC
0 5 Handling issues problems
4. Unloaded onto pallets of unloading cases of
CRP M product and cold
chain issues.
O 5 Handling issues, time
5. Staged on DC loading spent on the dock,
dock especially when
temperature of dock can
be an issue.
6. Pallets moved to storage 0 1
slot
7. Product inventoried/ A 0 FIFO, inventory
slotted / stored problems
8. Product 0 0 Tum-around times for
inventoried/ordered at the receiving new order
store level
9. Corporate SAL orders 0 0 Tum-around times for
product receiving new order
10. Store transmits order 0 O Tum-around times for
receiving new order
1 1. DC picks/builds store 0 1
order ﬁ‘om inventory
12. Order is staged on dock CD 5
13. Product is loaded on to O 3 Handling issues
truck (with other food
items)
14. Order arrives at store —> 7 Temperature and
handling issues
15. Order is moved from truck 0 5
to dock
16. Order is checked in O 5
17. Product is moved to cold 0 1 Handling, especially how

 

storage

 

 

 

 

long it may take to get
into storage.

 

206

 

Critical Activities Along Cold-Chain

 

 

 

Steps the Flow Time “Vulnerability” Operational Issues
Case-Ready Cold Chain Index
18. Product stored in walk-in A l FIFO, inventory issues.
cooler
19. Product 0 1 Qualiﬁed personnel,
bloomed/weighed/priced shrink, broken seal etc.
20. Meat cases are stocked O 2 Workrnanship issues,

inventory issues FIFO
Family pack issues
Temperature issues
Temperature issues

 

21. SAL shoppers select
22. Shoppers check out
23 Shoppers drive home
24. Stored until use

25. Meal Jireparation

 

 

 

 

 

 

 

 

 

 

 

UJDCDMCD
OUrxlOO

 

creates or adds
Moves from one to another.

Occurs when is held ' ' further action.
Occurs when is until a later time.
activities.

 

The Cold Chain “Vulnerability” Index is based upon the product’s vulnerability
factors exposed while moving through the cold chain including time, temperature
and handling that may lead to product degradation. This index attributes a number
value to potential risks. The ranking is from to 1-10, 10 being the greatest risk or

potential for degradation.

207

Michigan State University
Case-Ready Meat Research ProLect Survey

Store Proﬁle Questions

1. Please check the box that describes your store.
[ ] Save-A-Lot Corporate Store [ ] Save-A-Lot Licensed Store

2. Does your store sell the following case-ready products?

Ground Beef [ ] Yes [ ] No
Whole Muscle Beef (Beef Cuts) [ ] Yes [ ] No
Pork Whole Muscle Cuts or Grinds [ ] Yes [ ] No
3. Does your store grind beef and/or pork products? [ ] Yes [ ] No

4. When is it difﬁcult to maintain adequately stocked meat cases? Check all that apply.
9 am. - 10 am.

10 am. — 1 pm.
1 pm. — 4 pm.
4 pm. —- 7 pm.

”HF-1H”
h—ll—ll—lL—JH

7 pm. — close

5. How would your customers rank these attributes of ﬂesh beef and pork products?
Rank by preference, use each preference only once.

1=First Preference 2=Second Preference 3=Third Preference

[ ] Color

[ ] Servings per package

[ ] Package price
6. Rank these ﬁ'esh meat products according to how sensitive your customers are to the
prices charged per pound.

1=Most Sensitive =Moderately Sensitive 3=Least Sensitive

[ ] Fresh Whole Muscle Beef (Beef Cuts)

[ ] Fresh Ground Beef

[ ] Fresh Pork Cuts and Grinds

7. How satisﬁed are you with the way your

store veriﬁes that fresh meat deliveries l 2 3 4 S

from the Distribution Center (DC) are N 0* at all Very satisﬁed
within acceptable temperature ranges?

208

8. How satisﬁed are you with the way your 1 - 2 3 4 5
store transfers fresh meat products from the Not at all Very satisﬁed
receiving area to the walk-in coolers?

9. How satisﬁed are you with the way your 1 2 3 4 5
store writes the delivery date on each fresh Not at all Very satisﬁed
meat box received from the DC?

Case-Ready Questions
Compare fresh meat processed in-stores with case-ready meat products. Check the box of

the product that you believe is superior for each Key Success Factor listed. Leave blank if

you believe the products are similar.

Key Success Factors Fresh Meat Case-Ready
Processed In-Stores
10. Product appearance (trim, color and [ ] [ ]
quality)
11. Shrink management (Shrink is . [ ] [ 1

deﬁned as any dollar amount thatis
less than full-expected retail price for

and all fresh meat products.)
12. Proﬁtability [ 1 l ]
13. Shelf-life [ 1 [ l
14. Food safety (processed under the safest [ ] [ ]

possible conditions including
sanitation and temperature control)

15. Outofstocks [ l l l
16. Product consistency . [ ] l l
17. SKU varieties l l l l

209

These questions are for store managers that previously sold case-ready fresh meat
products and switched to processing fresh meat in-stores.

34. When you started selling case-ready meats your sales went up or down by what
percentage? .......................................................................... %

35. Fresh meat costs associated with case-ready meats went up by what percentage:
%

ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo

36. Case-ready meat products that were offered for sale had the same price as meats
processed in the store. ........................................ Yes No

37. When you switched to case-ready meats did your customers prefer meat products
processed in the store? Yes No -

If your answer is yes, what reasons did they express to you?

 

 

 

 

 

 

 

 

 

38. What were your perceptions of overall costs associated with switching to case-ready
meats? '

 

 

 

 

 

 

 

 

 

212

Using the scale listed to the right, circle a number on each line to indicate how much you

agree or disagree with each statement about case-ready meats.

I9. Case-ready ground beef products are available
in the lean — fat ratios my customers prefer.

20. Case-ready whole muscle beef (beef cuts)
products are available in the “cuts” my customers
prefer.

21. Case-ready pork products are available in the
“cuts” my customers prefer.

22. Case-ready ground beef products are available
in the pack sizes my customers prefer.

23. Case-ready whole muscle beef (beef cuts)
products are available in the pack sizes my
customers prefer.

24. Case-ready pork products are available in the
pack sizes my customers prefer.

25. Case-ready ground beef (not chubs) with a
75/25 lean — fat ratio have a “cherry red” color.

26. Case-ready ground beef (not chubs) with a
85/15 lean — fat ratio have a “cherry red” color.

27. Case-ready ground beef (not chubs) with a 90 /
10 lean — fat ratio have a “cherry red” color.

1
Strongly
Disagree

1

Stun-sly
Disagree

Stun-sly
Disagree

2

Strongly

Strongly

Sum-air

28. Ifyou are selling case-ready beef and pork, is it a more employee-ﬁiendly program

compared to processing fresh meat in-store?

[]Yes

210

30. If you are selling case-ready beef and pork, is it a more manager-ﬁfendly program
compared to processing fresh meat in-store?

[]Yes []No

If yes, why is case-ready more manager friendly:

 

 

 

31. What case-ready products do you think your customers would purchase that you are
not currently offering or is not available?

 

 

 

 

 

32. This question is for managers that currently do not carry case-ready meat products.

What are the main reasons your store continues to process fresh meat products on-site?

 

 

 

 

 

 

 

 

33. Currently, if you do not sell case-ready meats, under what conditions would you be
persuaded to switch to a case-ready meat program?

 

 

 

 

 

211

Quesﬁons About Shrink

Shrink is deﬁned as any dollar amount that is less than theﬁtll-expected retail price for
any and all fresh meat products.

Listed below are several sources of shrink associated with ﬁ'esh meat products. For each
sources of shrink, assign the appropriate number that describes the magnitude of the
concern at your store.

39.
40.
41.
42.
43.

44.

45.
46.

Customer theft

Defective and damaged packaging
Workrnanship issues

Incorrect inventory issues

Non-bloom issues associated with
case- ready ground beef
Non-bloom issues associated with
case-ready beef cuts

Employee theﬁ

Out of date product

1
Big Concern

1

Big Concern
1

Big Concern

1

Big Concern

1
Big Concern

1
Big Concern

1

Big Concern

1
Big Concern

NNNNN

WWWUJUJ

AA-h-h-h

5

Not a Concern

5

Not a Concern

5

Not a Concern

5

Not a Concern

5

Not a Concern

5

Not a Concern

5

Not a Concern

5

Not a Concern

47. Rank what you believe are the three most important sources of ﬂesh meat shrink for
your store. ’

1=Highest Source
Customer theft
Workrnanship issues

Incorrect inventory issues

Employee theft
Out of date product

HHHHf—lf—IHH
I—Jl—li—Jh—lh‘h—Jl—dh—l

Defective and damaged packaging

213

=Second Highest

Non-bloom issues associated with case-ready ground beef
Non-bloom issues associated with case-ready beef cuts

3=Third Highest

48. Are there other sources of shrink?

 

 

 

49. How satisﬁed are you with the way

your store records (or accounts for) every 1 2 3 4 5
fresh meat package (product) not sold at N“ ”t “'1 Very min-ﬂed
full retail?

50. How satisfied are you with the way

your store stocks fresh meat cases to ensure

that First-In-First—Out (FIFO) methods 1 2 3 4 5

are followed? Not at all Very satisﬁed

51. How satisﬁed are you with the way

your store maintains adequate inventory

levels of fresh beef and pork products 1 2 3 4 5
stored in the walk-in coolers? N“ “t '1' "“1““

52. How satisﬁed are you with the way

your store manages fresh beef and pork

inventory to ensure the oldest product in l 2 3 4 5

the walk-in cooler is stacked in a way it is N“ “ ‘1' V“? “W
the next product to be picked for restocking

the fresh meat cases?

53. How satisﬁed are you with the way
your store maintains adequate fresh meat

case levels of fresh beef and pork 1 2 3 4 5
prOdUCtS? Not at all Very satisﬁed

54. How satisﬁed are you with the way
your store prepares for the increased sales

experienced during the ﬁrst days of the l 2 3 4 5
month? Not at all Very satisﬁed

55. How satisﬁed are you with the way

your store prepares for increased sales

experienced associated with special fresh 1 2 3 4 5
meat promotions Not at all Very satisﬁed

214

Food Safety and Packaging Questions

56. Currently, there are two popular case-ready packaging alternatives. The ﬁrst is
High-Oxygen and the second is Low-Oxygen. Note: High-Oxygen packaging has high
“head-space ” to accommodate more oxygen gas, Low-Oxygen has low-head space and
resembles traditional packaging associated with meats processed in-stores. Which

alternative do you believe would be more successful? Check the appropriate box.

[ ] High-Oxygen / High Head Space

Using the scale listed to the right of the food safety and packaging issues, circle a

[ ] Low-Oxygen / Low Head Space

number on each line to indicate how much you agree or disagree with each statement.

57. A food borne illness outbreak associated with
fresh meat purchased at my store would be
catastrophic to overall store sales.

58. A food borne illness outbreak associated with
fresh meat purchased at my store would be
catastrophic to meat sales only.

59. My customers would respond by purchasing
more fresh beef and pork products if they believed
those products were the safest available.

60. Case-ready beef and pork products are “safer”
than beef and pork products processed in-store.

61. Case-ready meat products are the safest meats
available.

62. Low-Oxygen / Low Head Space packaging of
case-ready ground beef is not a detriment to sales.

63. High-Oxygen / High Head Space packaging of
case-ready gound beef is not a detriment to sales.

64. Low-Oxygen/ Low Head Space packaging of
case-ready beef cuts is not a detriment to sales.

65. High-Oxygen/ High Head Space packaging of
case-ready beef cuts is not a detriment to sales.

215

Strongly
Disagree

Strongly
Disagree

Strongly

Strongly
Disagree

2

3

4

i

“i

Strongly

'5

66. Low-Oxygen / Low Head Space packaging of l 2 3 4 5

 

:aais-ready Bork cuts & gm ds is not a detriment to 5"“!‘3' 52m
67. High-Oxygen / High Head Space packaging of l 2 3 4 5
case-ready pork cuts & grinds is not a detriment to SM 5mm!”
$31 Disagree Agree
es.

68. Case—ready meat products that require 1 2 3 4 5
u 0 ,9 ' Strongly Strongly
blooming cause operatlonal problems. Wm
69. My customers know the difference between 1 2 3 4 5

- ' ‘ - ' Strongly Stu-sly
case ready packaging and in store packaging.
70. Case-ready beef and pork products are l 2 3 4 5

' ° sow-sly sun-sly

packaged. in attractive trays.

71. What packaging improvements can you suggest for the fresh meat program currently
in place at your store?

 

 

Operations Questions
Using the scale listed to the right of the operational issues, circle a number on each line
to indicate how much you agree or disagree with each statement.

72. It has become increasingly more difficult to l 2 3 4 5
ﬁnd qualified meat cutters for my store SW” 82:03?
73. A traditional meat department manager and /
or meat cutter could be trained to manage all 1 2 3 4 5
' ‘ Strongly Strongly
perishables not just meats. Ame
74. Implementing a case-ready meat program will
shift meat management from production activities I 2 3 4 5
' ' Snongy Strongly
to forecastrng and inventory management of fresh AW
meats. l
75. Case-ready beef and pork products compared 1 2 3 4 5
to rnjstore produced meat products are more cost SW‘E'Y 52:3"
effective.
76. Cost of workers compensation for employees 1 2 3 4 S
- ' ' ' - Strongly Strongly
that cut meat rs becoming prohibitive. W

216

77. It has become increasingly more difﬁcult to I 2 3 4 5

ﬁnd ualiﬁed meat c tters. 5"“th Surreal!
q ll Disagree Agree

78. When conﬂonted with an out-of-stock ﬂesh meat product, what percentages of your
customers do not buy a substitute ﬂesh meat product ﬂom your store (i.e., choose to shop
elsewhere for the fresh meat product or not to purchase at all)? %.

79. If your customers do not purchase ﬂesh beef and pork at your store, where do they
most likely make those purchases?

 

 

80. How would you suggest information about your store’s needs concerning case-ready
meat products be communicated to the DC and the case-ready meat processor?

 

 

 

 

 

 

81. If you switched to a case-ready meat program, what beneﬁcial use could you make
with the space previously used to cut meat?

 

 

Thank you very much for your time and expertise. Please place the questionnaire back
into the envelope, seal the closure and return to your District Manager.

Dr. John Allen, Professor Emeritus Food Marketing
George Young, Graduate Research Assistant

217

BIBLIOGRAPHY

Adams, C. E., et a1. (1992) HACCP System in Regulatory Inspection Programs: Case
Studies of the USDA, USDC and DOD, ed M. D. Pierson, and J. D.A. Corlett.
New York, Chapman & Hall.

Adams, L. "Mapping Yields Manufacturing Insights." Quality 39, no. 5(2000): 62-66.

Aldrich, 1H. and RD. Nelson. Linear Probability, Logit and Probit Models. Beverly
Hills: Sage University Press, 1984.

Al-Mashari, M., and M. Zairi. "BPR Implementation Process: An Analysis of Key
Success and Failure Factors." Business Process Management Joumal 5, no.
1(1999).

Alexander, M., and D. Young. “Strategic Outsourcing.” Long Range Planning. 29, no.
1(1996): 116-119.

Alexander, M. and D. Young. "Outsourcing: Where' s the Value?" Long Range Planning
29, no. 5(1996): 728- 730.

Allen, J.W. “Centrally Packaged Fresh Red Meat: Success or Failure.” National
Provisioner 201, no. 8(1989): 1-8.

Anj ard, R. "Process Mapping: A Valuable Tool for Construction Management and Other
Professionals." Facilities 16, no. 3/4(1998): 79-94.

Anonymous. “Freezing Meats and Seafood.” Clemson University Extension Bulletin:
HGIC 3064 hm://hgic.clemson.edu/factsheets/HGIC3064.htm. Accessed January
27, 2007

Anonymous. “Case-Ready: The Innovation that Changed the Industry.” National
Provisioner 220 no. 6(2006): 173-176.

Anonymous. “Case-Ready Clarity.” National Provisioner 219 no. 10(2005): 146.

Anonymous. "Manufacturing: The Ins and Outs of Outsourcing." The Economist 320, no.
7722(1991): 54-55.

Anonymous. "MAP-ing Produce, Meat in the Supermarket." Packaging Digest 31, no.
1(1994): 88-90.

218

Anonymous. “MAP Quest.” National Provisioner 220 no. 7(2006): 56-60.

Anonymous. "Modiﬁed Atmosphere Packaging." Paper, Film & Foil Converter 70, no.
7(1996): 20-21.

Anonymous. "Want in on the MAP Action?" Plastics Technology 44, no. 11(1998): 48-
51.

Anonymous. "Excel Converting Hog Plant to Case-Ready Facility." National Provisioner
215, no. 6(2001): 26.

Anonymous (2001) ERS Estimates F oodbome Disease Costs at $6.9 Billion per Year.

Anonymous. "Meat-Safety Maneuvers: Retail Food Stores are Protecting Their Meat
Departments Against Environmental and Product Contamination by Following
Solid Protocols Related to Cleaning, Sanitizing, and Proper Handling." Meat
Retailer 56, no. 3(2002): 1-3.

Anonymous. "Enthusiasts Tout Central Processing Beneﬁts." DSN Retailer Today 41, no.
14(2002): 7-8.

Anonymous. "Proﬁting From Case-Ready." National Provisioner 217, no. 11(2003):
18E.

Anonymous. "Natural Protection: Patented Ingredient Targets Harmful Bacteria."
National Provisioner 217, no. 3(2003): 24—44.

Amemiya, T. “Qualitative Response Models: A Survey.” Journal of Economic Literature.
19, no. 4(1981): 1483-1536.

Arnold, S. J. "A Test for Clusters." Jll/Hi, Journal of Marketing Research 16, no. 4(1979):
545-55 1 .

Arya, A., J. Fellingham and B. Mittendorf. “Make-or-Buy in the Presence of Uncertainty
and Private Information.” Journal of Accounting Education 23, no. 3 (2005): 189-
203.

Ashford, JR. and RR. Sowden. “Multi-Variate Probit Analysis.” Biometrics 26 no.
3(1970):535-546.

Bagozzi, R., and J. Edwards. "A General Approach for Representing Constructs in
Organizational Research." Organizational Research Methods 1(1999): 45-87.

Bagozzi, R, and L. Phillips. "Representing and Testing Organizational Theories: A
Holistic Construal." Administrative Science Quarterly 27, no. 3(1982): 459-489.

219

Bardic, A. "Outsourcing Expertise." National Provisioner 217, no. 1(2003): 46-48.
Bardic, A. "Strategic Partnering." National Provisioner 217, no. 1(2003): 40-44.

Bardic, A. “State of the Industry 2005: Easier, Better and Faster!” National Provisioner
219, no. 8(2005): 24-48.

Bardic, A. "Packaging by Design." National Provisioner 217, no. 1(2003): 70-76.

Barney, J. B. "How a Firm's Capabilities Affect Boundary Decisions." MT Sloan
Management Review 40, no. 3(1999): 137-145.

Barney, J. B. “Firm Resources and Sustained Competitive Advantage.” Journal of
Management. 17, no. 1 (1991): 99-120.

Barry, C. "Case Closed. Case-Ready Meat is Here to Stay: As Wal*Mart Leads the Way
for Retailers, the Future for Case-Ready Meat is NownEven if its Growth is Slow
but Steady." Food & Drug Packaging 66, no. 7(2002): 28-34.

Barthelemy, J. "The Hidden Costs of IT Outsourcing." MIT Sloan Management Review
42, no. 3(2001): 60-69.

Baum, C. "How Long Before MAP Reaches it Full Potential?" Packaging 39, no.
10(1994): 26-3 1.

Bauman, H. E. (1992) Introduction to HACCP, ed. M. D. Pierson, and J. D.A. Corlett.
New York, Chapman & Hall, pp. 1-6.

Belcher, J .N. “Industrial Packaging Developments for the Global Meat Market.” Meat
Science 74(1), 2006. 143-148.

Bell, B. P., et al. "A Multi-state Outbreak of Escherichia coli 0157:H7—Associated
Bloody Diarrhea and Hemolytic Uremic Syndrome ﬂom Hamburgers: The
Washington Experience." Journal of the American Medical Association 272, no.
1(1994): 1349-1353.

Bendor-Samuel, P. "Leverage Best Practices." Executive Excellence 18, no. 2(2001): 17-
22.

Benn, I. a. J. P. Strategic Outsourcing: Exploiting the Skills of Third Parties. London:
Hodder & Stoughton, 2002.

Beme, S. "Map-ping the Future Capabilities." Prepared Foods 163, no. 3(1994): 101-
106.

220

Bertrand, K. “Cargill Meat Solutions (CMS) Beefs Up Case-Ready Packaging
Operations.” National Provisioner 220, no. 6(2006): 106-107.

Bettis, R. A., S. P. Bradley, and G. Hamel. "Outsourcing and Industrial Decline."
Academy of Management Executive 6, no. 1(1992): 7-22.

Biazzo, S. "Approaches to Business Process Analysis: A Review." Business Process
Management Journal 6, no. 2(2000): 99-110.

Biazzo, S. "Process Mapping Techniques and Organizational Analysis." Business Process
Management Journal 8, no. 1(2002): 42- 52.

Bienstock, C. C., and J. T. Mentzer. "An Experimental Investigation of the Outsourcing
Decision for Motor Carrier Transportation." Transportation Journal 39, no.
1(1999): 42-59.

Birch, D. "Should We Make-or-Buy?" Supply Management 6, no. 25(2001): 38-39.

Bliss, C]. “The Calculation of the Dosage-Mortality Curve.” Annas oﬂ Applied Biology.
22 no. 7 (1935); 307—330 (134-167).

Bond, T. C. "System Analysis and Business Process Mapping: A Symbiosis." Business
Process Management Journal 5, no. 2(1999): 164-176.

Bonoma, T. V. "Case Research in Marketing: Opporttmities, Problems and Process."
JMR, Journal of Marketing Research 22, no. 2(1985): 199-208.

Bowersox, D. J., D. J. Closs, and M. B. Cooper. Supply Chain Logistics Management.
Boston: McGraw Hill, 2002.

Bradley, P. "How quality is Changing Make-or-Buy Decisions." Purchasing 116, no.
1(1994): 89-91.

Brethauer, D. M. The Power of Strategic Costing: A Proactive Competitive
Approach for Setting Future Strategic Plans. New York: American Management
Association (AMACOM), 2000.

Breyer, R. F. "Some Preliminary Problems of Sample Design for a Survey of Retail
Trade F low." Journal of Marketing 10, no. 4(1946): 343-353.

Brody, A. L. "Active Packaging Becomes More Active" Food Technology 59, no.
12(2005): 82-86.

Brody, A. L. (1993) The Market, ed. R. T. Parry. Glasgow, Blackie Academic &
Professional, pp. 20-40.

221

Brody, A. L. "The Case for-or Against-Case-Ready Fresh Red Meat in the United
States." Food Technology 54, no. 8(2000): 153-156.

Brody, A. L. "What's Active about Intelligent Packaging?" Food Technology 55, no.
6(2001): 75-77.

Brody, A. L. "Case-Ready Fresh Red Meat: Is it Here or Not?" Food Technology 56, no.
1(2002): 77-78.

Brody, A. L. "Case-Ready Meat Systems Battle for Dominance." Brand Packaging 6, no.
1(2002): 11-12.

Brody, A. L. "Active Packaging Becomes More Active" Food Technology 59, no.
12(2005): 82-84.

Brody, A. L. "Food Packaging Climbs to the Summit" Food T echnologv 59, no. 7(2005):
73-75.

Brody, A. L. "Packaging Innovation as a Marketing Tool" Food Technology 60, no.
6(2006): 111-116.

Brody, A.L. “Sustainability: A New Buzz in Food Packaging.” Food Technology 60 no.
10(2006): 76-78.

Brown, M. (2000) HACCP in the Meat Industry. Cambridge, Woodhead Publishing
Limited, pp. 329.

Burgelman, R A. "A Process Model of Internal Corporate Venturing in the Diversiﬁed
Major Firm " Administrative Science Quarterly 28, no. 2(1983): 223-244.

Burgess, R "Avoiding Supply Chain Management Failure: Lessons ﬂom Business
Process Re-engineering." The International Journal of Logistics Management 9,
no. 1(1998): 15-23.

Burke, R "Risks and Reputations: The Economics of Transaction Costs." Corporate
Communications 3, no. 1(1998): 5-10.

Burn, D. "Restaurants and Supermarkets Adopt Shelf Fresh/Stable to Control Labor
Costs." Food in Canada 54, no. 2(1994): 15-20.

Buzby, J. C., P. D. Frenzen, and B. Rasco (2001) Product Liability and Microbial

Foodbome Illness: ERS/USDA Report Number 799, Economic Research Service
/ USDA.

222

C énez, L. E., K. E. Platts, and D. R Probert. "Developing a Framework for Make-or-Buy
Decisions." International Journal of Operations & Production Management 20,
no. 11(2000): 1313-1330.

Carroll, GR. and DJ. Teece. Firms, Markets and Hierarchies: The Transaction Cost
Economics Perspective. Oxford, Oxford University Press, 1999.

Casale, F. J. (1998) The 2000 Outsourcing Index.

Caswell, J. A, and N. H. Hooker. "HACCP as an International Trade Standard."
American Journal of Agricultural Economics 78, no. 8(1996): 775-779.

Cavinato, J. "How to Make the Outsourcing Decision." Distribution 90, no. 5(1991): 84-
86.

Chatterjee, S., L. Jarnieson, and F. Wiseman. "Identifying Most Inﬂuential Observations
in Factor Analysis." Marketing Science 10, no. 2(1991): 145-160.

Chavosta, J., R.Rl Rucker and M.J. Watts. “Transaction Costs and Cattle Marketing: The
Information Content of Seller-Provided Presale Data at Bull Auctions.” American
Journal of Agricultural Economics 83 no. 2 (2001): 286-301.

Chelty, S. "The Case Study Method for Research in Small and Medium Sized Firms."
International Small Business Journal 15, no. 1(1996): 73-85.

Churchill, G. A. J. Marketing Research: Methodological Foundations. Fort Worth: The
Dryden Press: Harcourt Brace College Publishers, 1999.

Coase, R. H. "The Nature of the Firm." Economica 4, no. 16(1937): 386-405.
Cohen, J. "Case-Ready Mixed Reviews." Progressive Grocer 75, no. 5(1996): 213-214.

Collins, P. "Using Process Mapping as a Quality Tool." Management Services 41, no.
3(1997): 24-26.

Converse, F .M. and S. Presser. Survey Questions: Handcrafting the Standardized
Questionnaire. Beverly Hills: Sage University Press, 1986.

Cook, T. D., and D. T. Campbell. Quasi-Experimentation: Design & Analysis Issues for.
Field Settings. Dallas: Houghton Mifﬂin Company, 1979.

Cooper, R, and R. S. Kaplan. "Measure Costs Right: Make the Right Decision." The
CPA Journal 60, no. 2(1990): 38-45.

Corlett, D. A. J ., and M. D. Pierson (1992) HACCP: Deﬁnitions and Principles, ed. D. A.
J. Corlett, and M. D. Pierson. New York, Chapman & Hall, pp. 6-7.

223

Cromerford, J.W., G.l. Greaser, H.L. Moore and J.K. Harper “Beef Cow-Calf
Production.” Penn State Cooperative Extension Bulletin: Agricultural
Alternatives. 2001.

Comer, P. D. "An Integrative Model for Teaching Quantitative Research Design."
Journal of Management Education 26, no. 6(2002): 671-692.

Cotoia, M., and S. Johnson. "Applying the Axiomatic Approach to Business Process
Redesign." Business Process Management Journal 7, no. 4(2001): 304-322.

Cox, A. Business Success. Boston: Earlsgate Press, 1997.

Cox, E. P. I. "The Design and Understanding of Survey Questions." JAR, Journal of
Marketing Research 19, no. 2( 1982): 279-280.

Crespi, J.M. and S. Marette. “How Should Food Safety Certiﬁcation Be Financed?”
American Journal Agricultural Economics 83 no. 4 (2001): 852-861.

Creswell, J. W. Reserach Design: Qualitative & Quantitative Approaches. Thousand
Oaks: Sage Publications, 1994.

Crisp, R. D. "A Case Study in Copy Research." JMR, Journal of Marketing Research 17,
no. 1(1953): 347-356.

Crutchﬁeld, S. R., et al. "Economic Assessment of Food Safety Regulations: The New
Approach to Meat and Poultry Inspection." Agricultural Economics Report 755,
no. July(2001): 1-29.

Culliton, J. W. Make-or-Buy: A Consideration of the Pr0blems Fundamental to a
Decision Whether to Manufacture or Buy Materials, Accessory Equipment,
Fabricating Parts, and Supplies. Boston: Graduate School of Business
Administration: Harvard University, 1942.

Damelio, R. The Basics of Process Mapping. New York: Productivity Inc., 1996.

Davenport, R. "Faster. Better. Cheaper. Technology, Consumer/Retailer Demand are
Driving teh Case-Ready Meat Category to New Heights." Refrigerated & Frozen
Foods 13, no. 4(2002): 38-42.

Davenport, T. H., and J. E. Short. "The New Industrial Engineering: Information

Technology and Business Process Redesign." Sloan Management Review 31, no.
4(1990): 11-27.

224

de Weerd-Nederhof, P. C. "Qualitative Case Study Research: The Case of a PhD.
Research Project on Organizing and Managing New Product Development
Systems." Management Decision 39, no. 7(2001): 513-538.

Demetrakakes, P. "Making a Case for Itself." National Provisioner 215, no. 7(2001): 64-
67 .

Denton, D. K. "Process Mapping Trims Cycle Time." Human Resource Management
Magazine 40, no. 2(1995): 56-61.

Dickson, J. P. JMR, Journal of Marketing Research 34, no. 2(1997): 296-298.

DiRomualdo, A., and V. Gurbaxani. "Strategic Intent for IT Outsourcing." MIT Sloan
Management Review 39, no. 4(1998): 67-80.

Dolan, T. "Best Practices In Process Improvement." Quality 36, no. 8(2003): 23-28.
Dolan, T. G. "Case-Ready Cadence." National Provisioner 216, no. 10(2002): 66-71.

Dowdell, S. "Kroger is Set to Activate Case-Ready Meat Project." Supermarket News 48,
no. 12(1998): 33-34.

Downey, G. and D. Beauchene. “Discrimination Between Fresh and Frozen-Then-
Thawed Beef m. longissimus dorsi by Combined Visible-Near Infrared
Reﬂectance Spectroscopy: A Feasibility Study.” Meat Science 45, no. 3(1997):
35 3-363. '

Drake, D.J. “Understanding and Improving Beef Cattle Carcass Quality.” University of
California Division of Agriculture and Natural Resources Publication. ANR
Publication38130, 2004.

Drtina, R E. "The Outsourcing Decision." Management Accounting 75, no. 9(1994): 56-
62.

Drucker, P. F. Managing for the Future: The 1990's and Beyond. New York: Dutton,
1992.

Duff, M. "Case-Ready Meat Suffers Setback at Kmart." DSN Retailer Today 41, no.
14(2002): 7.

Dunn, S. W. "The Case Study Approach in Cross-Cultural Research" JMR, Journal of
Marketing Research 3, no. 1(1966): 26-31.

Dyer, W. G. J., and A. L. Wilkins. "Better Stories, Not Better Constructs, to Generate
Better Theory: A Rejoinder to Eisenhardt." (1991).

225

Earl, M. J. "The Risks of Outsourcing IT." MIT Sloan Management Review 37, no.
3(1996): 26-32.

Ehie, I. C. "Determinants of Success in Manufacturing Outsourcing Decisions: A Survey
Study." Production and Inventory Management Journal 42, no. 1(2001): 31-39.

Eilert, 8.]. “New Packaging Technologies for the 21st Century.” Meat Science 71 no.
1(2006): 122-127.

Eisenhardt, K M. "Building Theories from Case Study Research." Academy of
Management: The Academy of Management Review 14, no. 4(1989): 532-550.

Ellis, G. "Make-Or-Buy Decisions: A Simpler Approach." Management Accounting 70,
no. 6(1992): 22-23.

Ellis, G. "Solving Make-Or-Buy Problems With Linear Programming." Management
Accounting 71, no. 10(1993): 52-53.

Ellram, L. "The Use of the Case Study Method in Logistics Research." Journal of
Business Logistics 17, no. 2(1996): 93-138.

Ellram, L. M., and A. Carr. "Strategic Purchasing: A History and Review of the
Literature." International Journal of Purchasing and Materials Management 30,
no. 2(1994): 10-18.

Falkrnan, M.A. “Top 10 Trends in Packaging.” Packaging Digest 42, no. 11(2005): 12.

Farouk, M.M., K.J. Wieliczko, and I. Merts. “Ultra-Fast Freezing and Low Storage
Temperatures are not Necessary to Maintain the Functional Properties of.
Manufacturing Beef.” Meat Science 66, no. 1(2004): 171-179.

Feame, A. "The Evolution of Partnerships in the Supply Chain: Insights ﬁ'om the British
Beef Industry." Supply Chain Management 3, no. 4(1998): 214-231.

Ferrante, M. A. "MAP: Putting the Pieces Together." Food Engineering 70, no. 1(1998):
79-82.

lFill, C., and E. Visser. "The Outsourcing Dilemma: A Composite Approach to the Make-
or-Buy Decision." Management Decision 38, no. 1(2000): 43-50.

lFisher, R., and W. Ury. Getting to Yes: Negotiating Agreement Without Giving In.
Boston: Houghton Mifﬂin, 1981.

Fleming, S. and C. Stapleton. “Should Buying be Outsourced?” Supply Management 12,
no. 4 (2007): 28-30.

226

Fleury, A. and MT Fleury. “Competitive Strategies and Core Competencies:
Perspectives for the Internationalization of Industry in Brazil.” Integrated Manufacturing
Systems. 14, no. 1 (2003): 16-25.

Forcinio, H. "Mapping out a Safety Net." Prepared Foods 168, no. 2(1999): 52-56.

Ford, D., B. Cotton, D. Farmer, H. Gross and I. Wilkinson. “Make or Bury Decisions and
Their Implications.” Industrial Marketing Management 22, no. 3(1993): 207-214.

Foster, J. "Is There a Role for Transaction Cost Economics if we View Firms as Complex
Adaptive Systems?" Contemporary Economic Policy 18, no. 4(2000): 369-3 85.

Fowler, F. J. Improving Survey Questions: Design and Evaluation. Thousand Oaks: Sage
Publications, 1995.

Fox, J. A., and D. A. Hennessy. "Cost Effective Hazard Control in Food Handling."
American Journal of Agricultural Economics 81, no. 2(1999): 359-372.

F ulscher, J ., and S. G. Powell. "Anatomy of a Process Mapping Workshop." Business
Process Management Journal 5, no. 3(1999): 208-226.

Gaebler, D. M. "Modiﬁed Atmosphere Packaging of Ground Pork." Dissertation,
University of Nebraska, 1999.

Galosich, A. "Putting Case-Ready on the Map." National Provisioner 212, no. 10(1998):
52-53. '

Galosich, A. "IBP Captures Marketplace Value." National Provisioner 213, no. 2(1999):
22-26.

Galosich, A. "Building a Case for Pre-Packaged Meats." National Provisioner 213, no.
10(1999): 75-83.

Galosich, A. "Colorado Boxed Beef Company: Processor of the Year Trailblazers."
National Provisioner 214, no. 6(2000): 24-26.

Galosich, A. "Case-Ready of Not." National Provisioner 214, no. 5(2000): 90-96.

Gambino, A. J. The Make-or-Buy Decision. New York: National Association of
Accountants and The Society of Management Accountants of Canada, 1980.

Gardiner, S. C., and J. J. H. Blackstone. "The Theory of Constraints and the Make-or-

Buy Decision." International Journal of Purchasing and Materials Management
27, no. 3(1991): 38-43.

227

Gay, C. L., and J. Essinger. Inside Outsourcing: The Insiders Guide to Managing
Strategic Sourcing. London: Brealy Publishing, 2000.

Gellynck, X., W. Verbeke and B. Vermeire. “Pathways to Increase Consumer Trust in
Meat as a Safe and Wholesome Food” Meat Science 74, no. 1(2006): 161-171.

Ghauri, P., K Gronhaug, and I. Kristianslund. Research Methods in Business Studies.
New York: Prentice Hall, 1995.

Gibb, W. G. J., and A. L. Wilkins. "Better Stories, Not Better Constructs, To Generate
Better Theory: A Rejoinder to Eisenhardt." Academy of Management: The
Academy of Management Review 16, no. 3(1991): 613-619.

Giﬁi, C., A. V. Roth, and G. M. Seal. Competing in World Class Manufacturing:
America's 21” Century Challenge. Homewood: Business One Irwin, 1993.

Gill, C.O. “Extending the Storage Life of Raw Chilled Meats.” Meat Science 43, no.
5(1996): 99-109.

Glaser, B. G., and A. L. Strauss. The Discovery of Grounded Theory: Strategies for
Qualitative Research. Chicago: Aldine Publishing Company, 1967.

Glass, J. A. "The End of the Outsourcing Era." Information Systems Management 13, no.
2(1996): 89-91.

Globerman, S. and AR. Vining. “The Outsourcing Decision: A Strategic Framewor
Economics Working Paper EconWPA #0404007, International Trade. University
of Washington, 2004.

Golan, E., et al. (2004) Food Safety Innovation in the United States: Economic Theory
and Empirical Evidence from the Meat Industry, vol. 2004, Economic Research
Service / USDA.

Golan, E., et al. "Tracing The Costs and Beneﬁts of Improvements in Food Safety: The
Case of the Hazard Analysis and Critical Control Point Program for Meat and
Poultry." USDA (Economic Research Service), October, 2000.

Goodrich, RM., F.R. Schneider and RH. Schmidt. “HACCP and Overview.” FSHNOS-
12, Florida Cooperative Extension Service, Institute of Food and Agricultural
Sciences, University of Florida. Published August 2005. h_t_tp://edis.ifas.uﬂ.edu.

Gottfredson, M., R Puryear and S. Phillips. “Strategic Sourcing: From Periphery to the
Core.” Harvard Business Review

Gourishankar, T. (2003) A Simple Process Map, vol. 2003.

228

Greaver, M. F. Strategic Outsourcing: A Structured Approach to Outsourcing Decisions
and Initiatives. New York: American Management Association (AMACOM),
1998.

Greco, J. "Outsourcing: The New Partnership." The Journal of Business Strategy 18, no.
4(1997): 48-54.

Greenﬁeld, M. "Process Mapping's Next Step." Quality 35, no. 9(2002): 50-55.

Greer, C. R, S. A. Youngblood, and D. A. Gray. "Human Resource Management
Outsourcing: The Make-or-Buy Decision" The Academy of Management
Executive 13, no. 3(1999): 85-96.

Grieco, P. L., and M. Pilachowski. Activity Based Costing: The Key to World Class
Performance. Palm Beach Gardens: PT Publications, Inc., 1992.

Grifﬁths, D. (2001) The Theory and Practice of Outsourcing, vol. 2004.

Gross, H. "Make-or-Buy Decisions in Growing Firms." The Accounting Review 41, no.
4(1966): 745-753.

Gross, H. Make-or-Buy. Englewood Cliﬁ‘s: Prentice-Hall,1nc., 1966.

Gnmert, K.G. “Future Trends and Consumer Lifestyles with Regard to Meat
Consumption.” Meat Science

Gupta, M., and D. Zhender. "Outsourcing and Its Impact on Operations Strategy."
Production and Inventory Management Journal 35, no. 3(1994): 70-76.

Gujarati, D.N. Basic Econometrics. 3rd Edition New York: McGraw-Hill, 1995.

Hair, J. F., RE. Anderson, R.L. Tatham and WC. Black. Multivariate Data Analysis. 5‘"
Edition Upper Saddle River: Prentice Hall, 1998.

Hair, J.F., RP. Bush and DJ. Ortinau. Marketing Research: Within a Changing
Information Environment. 3”d Edition Boston: McGraw-Hill Irwin, 2006.

Hanacek, A. “Keeping the Food Supply Safe.” National Provisioner 220, no. 6(2006):
64-76.

Harrington, L. H. "Unlock Proﬁts." Transportation & Distribution 37, no. 6(1996): A-N.
Ham's, J. M., et al. "The US. Food Marketing System: Competition, Coordination, and

Technological Innovations Into the 21“ Century." USDA Economic Research
Service Agricultural Economic Report No. 811, no. June 2002(2002): 1-99.

229

Hegji, C. E. "A Neoclassical Model of the Make-Or-Buy Decision." The Mid-Atlantic
Journal of Business 37, no. 1(2001): 65-73.

Heikkila, J ., and C. Cordon. "Outsourcing: A Core or Non-Core Strategic Management
Decision?" Strategic Change 11, no. 4(2002): 183-193.

Heller, H. and M. Major. 2005. “2005 Meat Operations Review: Prime Time.”
Progressive Grocer 84(4):62-66.

Henderson, G. "Convenience Beef Products Show Rapid Growth." Food Systems
INSIDER 1, no. 2(2001): 6.

Higgins, K. T. "The Case-Ready Meat Challenge." Food Engineering 72, no. 9(2000):
27-29.

Higgins, K.T. “Core Competency: Process ControL Maintenance, Packaging, Aseptic
Processing.” Food Engineering 76, no. 8 (2004): 40-44.

Hirschheim, R, and M. Lacity. "The Myths and Realities of Information Technology
Insourcing." Communications of the Association for Computing Machinery 43,
no. 2(2000): 99-107.

Hobbs, J. E. "A Transaction Cost Approach to Supply Chain Management." Supply
Chain Management 1, no. 2(1996): 15-30. ,

Hobbs, J .E. “A TransactiOn Cost Analysis of Quality, Traceability and Animal Welfare
Issues in UK Beef Retailing.” British Food Journal 98 no. 6 (1996): 16-26.

Hodges, M. "Not Your Father's Outsourcing Deal." Optimize June (2003): 38-52.

Hodgins, M. 2002. “The Market for Case-Ready Beef.” Saskatchewan Agriculture, Food
and Rural Revitalization Publication May, 2002: 1-80.

Hoetker, G. “The Use of Logit and Probit Models in Strategic Management Research:
Critical Issues.” Strategic Management Journal 28 no. 4(2007): 331-343.

Howie, M. "Hormel, Cargill Ink Joint Venture." Feedstujjis 74, no. 38(2002): 6.

Huffman, R. “Ordinance Banning Low Oxygen and Carbon Monoxide Packaging: Bad
News for Consumers.” Remarks given at the American Meat Institute Foundation
Meeting March 23, 2006 in Chicago, IL

Hunt, M.C., RA. Mancini, K.A. Hachmeister, D.H. Kropf, M. Merriman, G. Del Duca
and G. Milliken. “Carbon Monoxide in Modiﬁed Atmosphere Packaging Affects
Color, Shelf Life, and Microorganisms of Beef Steaks and Ground Beef.” Journal
of Food Science 69 no. 1(2004): 45-52.

230

Hunt, V. D. Process Mapping. New York: John Wiley & Sons, 1996.

Hurley, A. E., et al. "Exploratory and Conﬁrrnatory Factor Analysis: Guidelines, Issues,
and Alternatives." Journal of Organizational Behavior 18, no. 6(1997): 667-683.

Hussey, D., and P. J enster. "Outsourcing: The Supplier Viewpoint." Strategic Change 12,
no. 1(2003): 7-20.

Jacka, J. M., and P. J. Keller. Business Process Mapping: Improving Customer
Satisfaction. New York: John Wiley & Sons, 2002.

Jayasingh, P.; D.P. Comforth; C.E. Carpenter and D. Whittier. “Evaluation of Carbon
Monoxide Treatment in Modiﬁed Atmosphere Packaging or Vacuum Packaging
to Increase Color Stability of Fresh Beef.” Meat Science 59(3), 2006. 317-324.

Jennings, D. "Strategic Sourcing: Beneﬁts, Problems and a Contextual Model."
Management Decision 40, no. 1/2(2002): 26-34.

Jensen, J. L., and R. Rodgers. "Cumulating the Intellectual Gold of Case Study
Research." Public Administration Review 61, no. 2(2001): 235-246.

Jenster, P., and H. S. Pedersen. "Outsourcing: Facts and. Fiction." Strategic Change 9, no.
3(2000): 147-154.

Johnson, M. Outsourcing: In Brief Oxford: Butterworth - Heinemann, 1997.
Jones, B. "Is Central Prep the Answer?" Progressive Grocer 74, no. 5(1995): 197-201.

Juran, D. C., and H. Dershin. "Lessons in Supply Chain Assessment and Improvement."
Measuring Business Excellence 4, no. 1(2000): 18-27.

Kakabadse, A. and N. Kakabadse. “Outsourcing: A Paradigm Shift.” The Journal of
Management Development. 19, no. 8 (2000): 670-728.

Kakabadse, A. and N. Kakabadse. "Outsourcing Best Practice: Transformational and
Transactional Considerations." Knowledge and Process Management 10, no.
1(2003): 60-71.

Kakabadse, A. and N. Kakabadse. “Outsourcing: Current and Future Trends.”
Thunderbird International Business Review. 47, no. 2(2005): 183-204.

Kannan, V. R, and K C. Tan. "Supplier Selection and Assessment: Their Impact on

Business Performance." Journal of Supply Chain Management 38, no. 4(2002):
11-21.

231

Katz, D., and R Kahn. The Social Psychology of Organizations. New York: Wiley, 1978.

Keith, H. "The Changing Processing Sector." National Provisioner 216, no. 6(2002): 64-
69.

Keith, H. "Case-Ready Savings." National Provisioner 216, no. 6(2002): 46-69.

Keith, H. "Case-Ready Savings. Special Report: Case-Ready Solutions." National
Provisioner 216, no. 6(2002): 46-52.

Keller, P. J., and J. M. Jacka. "Process Mapping." Internal Auditor 56, no. 5(1999): 60-
64.

Ketchen, D. J. J., and C. L. Shook. "The Application of Cluster Analysis in Strategic
Management Research: An Analysis and Critique." Strategic Management
Journal 17, no. 6(1996): 441-458.

Kilman, S. (2002) Branding Beef: A Roast is a Roast? Not in the New Game of
Marketing Meat. New York, pp. 1-3.

Klass, B. S., J. G. McClendon, and T. W. Gainey. "Human Resource Outsourcing and Its
Impact: The Role of Transaction Cost Economics." Personnel Psychology 52, no.
1(1999): 113-136.

Klastorin, T. D. "Assessing Cluster Analysis." JMR, Journal of Marketing Research 20,
no. 1(1983): 92-98.

Klein, M. (2004) Making a Strong Case for Case-Ready Meat Packaging,
www.cargillcom/news/498pgﬁ.

Klepacki, L. "Ready for Case-Ready? The Prepackaged Fresh Meat Movement is
Gaining Support, But Questions Remain." Supermarket News 44, no. 31(1994):
31-37.

Kmenta, J. Elements of Econometrics. New York: Macmillan Publishing, 1986.

Kropko, Jonathan. "Choosing Between Multinomial Logit and Multinomial Probit
Models for Analysis of Unordered Choice Data" Paper presented at the annual
meeting of the MPSA Annual National Conference, Palmer House Hotel, Hilton,
Chicago, Illinois, April 03, 2008.
http://www.allacademic.com/meta/p265696_index.htrnl. ‘

Kruger, R. (2002) Meat Case Study, vol. 2003.

232

 

Kufeld, W. F., R D. Tobias, and M. Garratt "Efﬁcient Experimental Design with
Marketing Research Applications." JAM, Journal of Marketing Research 31, no.
4(1994): 545-557.

Kvenberg, J ., P. Stolfa, D. Stringfellow and ES. Garrett. “HACCP Development and
Regulatory Assessment in the United States of America.” Food Control 11, no.
5(2000): 387-401.

Lacity, M. C., and R. Hirschheim. "The Information Systems Outsourcing Bandwagon."
MIT Sloan Management Review 35, no. 1(1993): 73-86.

Laios, L., and S. Moschuris. "An Empirical Investigation of Outsourcing Decisions."
Journal of Supply Chain Management 35, no. 1(1999): 33-41.

Lalman, D.L. “Nutrient Requirements of Beef Cattle.” Oklahoma Cooperative Extension
Service Report: E-9 74. March, 2004.

Lambert, D. M., and T. L. Pohlen. "Supply Chain Metrics." The International Journal of
Logistics Management 12, no. 1(2001): 1-19.

Langrnan, B. "What's Up With Case-Ready Meat?" National Provisioner 216, no.
6(2002): 58-62.

Langrnan, B. "What's New in Packaging." National Provisioner 261, no. 1(2002): 46-51.

Lankford, W. M., and F. Parsa. "Outsourcing: A Primer." Management Decision 37, no.
4(1999): 310-318.

Laseter, T. M. Balanced Sourcing: Cooperation and Competition in Supplier
Relationships. San Francisco: Hossey-Boss Publishers, 1998.

Lai, W.L. and P.Y. Chan. “The Formation of Owner’s Corporations in Hong Kong’s
Private Housing Estates.” Property Management. 22 no. 1 (2004): 55-68.

Lau, R S. M., and C. N. Hurley. "Outsourcing Through Strategic Alliances." Production
and Inventory Management Journal 38, no. 2(1997): 6-10.

Libecap, G. D. "The Rise of the Chicago Meatpacking and the Origin of Meat Inspections
and Antitrust." Economic Inquiry 30, no. 2(1992): 242-262.

Lientz, B. P., and K. P. Rea. Process Visualization: An Executive Guide to Business
Process Design. New York: John Wiley & Sons, 2002.

Linder, J. C., M. 1. Cole, and A. L. Jacobson. "Business Transformation Through
Outsourcing." Strategy & Leadership 30, no. 4(2002): 23-28.

233

Lingle, R. "A Case-Ready Meat Tour: A Visit to a Wal*Mart Supercenter."
Packworld.com 7, no. 3(2003): 70-76.

Little, D., et al. "Responsive Manufacturing Demands Alignment of Production Control
Methods to Business Drivers." Integrated Manufacturing 12, no. 3(2001): 170-
178.

Logan, M. S. "Using Agency Theory to Design Successful Outsourcing Relationships."
The International Journal of Logistics Management 1 1, no. 2(2000): 21-32.

Loh, L., and N. Venkatraman. "Determinants of Informational Technology Outsourcing:
A Cross-Sectional Analysis." Journal of Management Information 9, no. 1(1992):
7-25.

Lonsdale, C. "Locked-In to Supplier Dominance: On the Dangers of Asset Speciﬁcity for
the Outsourcing Decision." Journal of Supply Chain Management 37, no.
2(2001): 22-27.

Lonsdale, C., and A. Cox. "Outsourcing: The Risks and Rewards." Supply Management
2, no. 14(1997): 32-34.

Louis, P. J. "Food Packaging in the Next Millennium." Packaging Technology and
Science 12(1999).

Lynch, C. F. "Price vs. Value: The Outsourcing Conundrum." Logistics Management &
Distribution Report 41, no. 2(2002): 35.

Maddala, G.S. Introduction to Econometrics. New York: Macmillan Publishing
Company, 1992.

Major, M. “Main Course: The Annual Meat Conference Put Consumer Needs and
Attitudes at the Center of the Plate.” Progressive Grocer 85, no. 6(2006): 88-90.

Maltz, A., and L. Ellram. "Outsourcing Supply Management." Journal of Supply Chain
Management 35, no. 2(1999): 4-17.

Maltz, A., and L. M. Ellram. "Total Cost of Relationship: An Analytical Framework for
the Logistics Outsourcing Decision." Journal of Business Logistics 18, no.
1(1997): 45-67.

Markowitz, H. M. "Analysis and Intuitive Decision Making Models in Business: A
Forward." The Mid-Atlantic Journal of Business 30, no. 1(1994): 5-7.

Martin, K "Packaging Innovations Bring Farm Freshness to the Food Market." Food
Engineering International 24, no. 1(1999): 41 -47.

234

Mason, F. "Mapping a Better Process." Manufacturing Engineering 1 18, no. 4(1997): 58-
68.

Mazzocco, M. A. "HACCP as a Business Management Tool." American Journal of
Agricultural Economics 78, no. 8(1996): 770-774.

Mclvor, R. T. "A Practical Framework for Understanding the Outsourcing Process."
Supply Chain Management 5, no. 1(2000): 22-36.

Mclvor, R. T., P. K Humphreys, and W. E. McAleer. "A Strategic Model for the
Formulation of an Effective Make-or-Buy Decision." Management 35, no.
2(1997): 169-180.

McKenna, S. D. "The Darker Side of Entrepreneurship." Leadership & Organization
Development 17, no. 6(1996): 41-52.

McNally, R. C., and A. Grifﬁn. "Firm and Individual Choice Drivers in Make-or-Buy
Decisions: A Diminishing Role for Transaction Cost Economics?" Journal of
Supply Chain Management 40, no. 1(2004): 4—17.

McTaggart, J., D. Chanil and S. Dowdell. 2006. “Report of the Grocery Industry:
Sru'vival of the Fittest.” Progressive Grocer 85(6): 22-57.

Mead, P. S., et al. (1999) Food Related Illness and Death in the United States.

Meyers, J.L.; G.C. Gamst and A]. Guarino. Applied Multivariate Research: Design and
Interpretation. Thousand Oaks: Sage Publications, 2006.

Miles, M. B. "Qualitative Data as an Attractive Nuisance: The Problem of Analysis."
Administrative Science Quarterly 24, no. 4(1979): 590-601.

Miles, M. B., and A. M. Huberman. Qualitative Data Analysis: A Sourcebook of New
Methods. Beverly Hills: Sage Publications, 1984.

Miller, L. “Bad Math Still Grinding Into Proﬁts: Meat Study.” Supermarket News 49(20),
2001.

Miller, L. "Annual Meat Conference 2002 Case-Ready Review CouldReveal Beneﬁts:
Levin." Supermarket News 79, no. 5(2002): 79-80.

Miller, M. "What's In a Name?" Food Svstems INSIDER 1, no. 2(2001): 1-4.

Milliken, A. L. "Key Ingredients of Successful Performance Metrics in the Supply
Chain." The Journal of Business Forecasting Summer (2001): 23-24.

235

Minoli, D. Analyzing Outsourcing: Reengineering Information and Communication
Systems. 303 ed. San Francisco: McGraw Hill, Inc., 1995.

Mintert, J., J.L. Lusk, T.C. Schroeder, J.A. Fox and M. Koohmaraie. “Valuing Beef
Tenderness.” Kansas State University Agricultural Experiment Station and
Cooperative Extension Service Report: MF-2464. May, 2000.

Mitchell, R. Wal *Mart Sets the Pace. Meat and Deli Retailer, November, 2005.

Mock, E. J ., and D. F. Miller. "The Make-or-Buy Decision Under Uncertainty."
Integrated Manufacturing Systems 7, no. 2(1970): 44-51.

Monson, F., C. Sanudo and 1. Sierra. “Inﬂuence of Breed and Ageing Time on the
Sensory Meat Quality and Consumer Acceptability in Intensively Reared Beef.”
Meat Science 71, no. 3(2005): 471-479.

Moody, P. E. "Strategic Purchasing Remains an Oxymoron." MIT Sloan Management
Review 42, no. 2(2001): 18.

Morgan, J .B. “Extending Shelf-Life of Beef Cuts Utilizing Low Level Carbon Monoxide
in Modiﬁed Atmosphere Packaging Systems: Project Summary.” www.beef.org

Prepared on behalf of the Cattlemen’s Beef Board by the National Cattlemen’s
Beef Association Center for Research & Knowledge Management. July, 2003.

Morris, S. “Production System Feed Efﬁciency.” Massey University Institute of
Veterinary and Animal Sciences Presentation. ” March, 2001.

Moschandreas, M. "The Role of Opportunism in Transaction Cost Economics." Journal
of Economic Issues 31, no. 3(1997): 39—57.

Moschuris, 8.]. “Organizational Participants in the Make-or-Buy Process.” Industrial
Marketing Management. (Article in Press), 2007.

Mullin, R. "Managing the Outsourced Enterprise." The Journal of Business Strategy 17,
no. 4(1996): 28-36.

Nelson, P.B. “Information and Consumer Behavior.” The Journal of Political Economy.
78 no. 2 (1970): 311-329.

Norris, P.E., E.M. Brown, and SS. Batie. An Analysis of Situation, Structure, Conduct
and Performance in Air Emission and Watershed Effluent Markets. Staff Paper
2002-06. Department of Agricultural Economics Michigan State University,
2002.

236

Myers, B. L., N. L. Enrick, and A. J. Melcher. "Strategy in Research Design and
Hypothesis Testing." Academy of Marketing Science Journal 2, no. 1(1974): 249-
261.

O'Brien, L. "Purchasing: The Final Frontier." Supply Management 31, no. 3(2002): 22-
25.

Ollinger, M., and D. Moore. "Food Safety Investment in the Meat and Poultry
Industries." Economic Reserach Service / USDA Food Safety Innovation AER-
83l(2004): 13-20.

Ollinger, M., D. Moore, and R Chandran. "Meat and Poultry Plants' Food Safety
Investments: Srn'vey Findings." Economic Reserach Service / USDA Food Safety
Innovation May 2004, no. 1911(2004): 1-4.

Pagnoncelli, D. "Managed Outsourcing: A Strategy for a Competitive Company in the
1990's." Management Decision 31, no. 7(1993): 15-22.

Paper, D.; J., J. A. Rodger, and P. C. Pendharkar. "A BPR Case Study at Honeywell."
Business Process Management Journal 7, no. 2(2001): 85-99.

Park, H. Y., C. S. Reddy, and S. Sarkar. "Make or Buy Strategy of Firms in the US."
Multinational Business Review 8, no. 2(2000): 89-97.

Parry, R. r. (1993 ) Introduction to MAP, ed R. r. Parry, First Edition. Glasgow, Blackie
Academic & Professional.

Patton, E., and S. H. Applebaum. "The Case for Case Studies in Management Research."
Management Research 26, no. 5(2003): 60-71.

Patton, W. S. "Use A Map, Not A Hammer." Quality 36, no. 6(2002): 3436.

Pedhazur, E., and L. Schmelkin. Measurement, Design and Analysis: An Integrated
Approach. Hillsdale: Lawrence Erlbaum, 1991.

Pennington, T. H. (2000) Introduction, ed A. Brown. Cambridge, CRC Press, pp. 3-9.
Perren, L., and M. Ram "Case Study Method in Small Business and Entrepreneurial
Research: Mapping Boundaries and Perspectives." International Small Business

Journal 22, no. 1(2004): 83-99.

Perry, C. "Processes of a Case Study Methodology for Postgraduate Research in
Marketing." European Journal of Marketing 32, no. 9/10(1998): 785-806.

Peters, J., and K. Howard. "Looking for Good Research in Management: A Publisher's
Case Study." Management Decision 39, no. 7(2001): 594-598.

237

Peters, T. J. The T om Peters Seminar: Crazy Times Call for Crazy Organizations. New
York: Vintage Books, 1994.

Petersen, J. H., et al. "Evaluation of Retail Fresh Meat Packagings Covered with Stretch
Films of PVC and non-PVC Alternatives." Packaging Technology and Science
17, no. 2(2004): 53-66.

Petrak, L. "The Case for Change." National Provisioner 215, no. 9(2001): 29-34.

Petrak, L. “Red Hot Controversy: The Debate Over High Oxygen Modiﬁed Atmosphere
Packaging is in Full Bloom.” National Provisioner. 1, no. 1(2006): 46-47.

Pettigrew, A. M. "Longitudinal Field Research on Change: Theory and Practice."
Organizational Science 1, no. 3(1990): 267-292.

Phipps, P. A., S. J. Butani, and Y. I. Chun. "Research on Establishment-Survey
Questionnaire Design." Journal of Business & Economic Statistics 13, no.
3(1995): 337-346.

Pierson, M. D., and J. D.A. Corlett (1992) HACCP Principles and Applications. New
York, Chapman & Hall, pp. 212.

Piggott, NE. and TL. Marsh. “Does Food Safety Information Impact U.S. Meat
Demand?” American Journal of Agricultural Economist 86, no. 1(2004): 154-174.

Pitelis, C. N. "Transaction Cost and the Historical Evolution of the Capitalist Firm."
Journal of Economic Issues 32, no. 4(1998): 999-1017.

Poppo, L., and T. Zenger. "Testing Alternative Theories of the Firm: Transaction Cost,
Knowledge Based, and Measurement Explanations for Make-or-Buy Decisions in
Information Services." Strategic Management Journal 19, no. 9(1998): 853-877.

Poore, M.H. “Managing Beef Cattle on Forage.” North Carolina State University
Department of Animal Science Beef Cattle Report. 1998

Porter, C. "The Case for Case-Ready." Meat International 12, no. 4(2002): 34—36.
Porter, M.E. Competitive Strategy. New York: The Free Press, 1980.

Porter, ME. “The Contributions of Industrial Organization to Strategic Management.”
Academy of Management Review. 6, no. 1 (1981): 609-620.

Porter, M. E. Competitive Advantage: Creating and Sustaining Superior Performance.
London: The Free Press, 1985.

238

Prahalad, C. K., and G. Hamel. "The Core Competence of theiCorporation." Harvard
Business Review 68, no. 3(1990): 79-91.

Prahalad, C. K., and G. Hamel. “Strategy as a Field of Study: Why Search for a New
Paradigm.” Strategic Management Journal. 15, Summer Issue (1994): 5-16.

Probert, D. R "The Practical Development of a Make-or-Buy Strategy: The Issue of
Process Positioning." Integrated Manufacturing 7, no. 2(1996): 44-52.

Punj, G., and D. W. Stewart. "Cluster Analysis in Marketing Research: Review and
Suggestions for Application." JMR, Journal of Marketing Research 20, no.
5(1983): 134-148.

Quinn, J. B. "Strategic Outsourcing: Leveraging Knowledge Capabilities." MIT Sloan
Management Review 40, no. 4(1999): 9-21.

Quinn, J. B. "Outsourcing Innovation: The New Engine of Growth." MIT Sloan
Management Review 41 , no. 4(2000): 13-28.

Quinn, J. B., and F. G. Hilmer. "Strategic Outsourcing." MIT Sloan Management Review
35, no. 4(1994): 43-55.

Raunick, D. A., and A. G. Fisher. "A Probabilistic Make-or-Buy Model." Journal of
Purchasing 8, no. 1(1972): 63-74.

Reichardt, E., and L. J. Nichols. "SCOR Your ISO Certiﬁcation." Quality 42, no.
2(2003): 44.47.

Remenyi, D., et al. "The Creation of Knowledge Through Case Study Research." Irish
Journal of Management 23, no. 2(2002): 1-17.

Rindﬂeisch, A. and J .B. Heide. “Transaction Cost Analysis: Past, Present, and Future
Applications.” Journal of Marketing 61 no. 4 (1997): 30-54.

Roberts, T. "Economics of F oodbome Disease Brieﬁng Room." USDA Economic
Research Service The Economics of Food, Farming, Natural Resources, and Rural
America, no. Food Safety: Features (2001): l.

Rothery, B., and I. Robertson. The Truth About Outsourcing. Hampshire: Gower
Publishing Limited, 1995.

Rowley, J. "Using Case Studies in Research." Management Research News 25, no.
1(2002): 16-27.

239

Saghir, M. "Packaging Information Needed for Evaluation in the Supply Chain: The Case
of the Swedish Grocery Retail Industry." Packaging Technology and Science 15,
no. 1(2002): 37-46.

Saghir, M., and G. Jonson. "Packaging Handling Evaluation Methods in the Grocery
Retail Industry." Packaging Technology and Science 14, no. 1(2001): 21-29.

Salay, E., J. A. Caswell, and T. Roberts (2003) Survey Instrument for Case Studies of
Food Safety Innovation, University of Massachusetts at Amherst.

Salvage, B. “Case-Ready Penetration.” National Provisioner. 219(6) 2005.

Salvage, B. "Case-Ready and Natural." National Provisioner 217, no. 1(2003): 24-28.

Salvage, B. "Case-Ready...or not?" National Provisioner 217, no. 6(2003): 88-97.

Salvage, B. “Tracking European Packaging Trends.” National Provisioner 219 no.
10(2005): 130-132.

Sami, A.S., C. Augustini and F .J. Schwarz. “Effects of Feeding Intensity and Time on
Feed on Performance, Carcass Characteristics and Meat Quality of Simmental
Bulls.” Meat Science

Sarantakos, S. Social Research. South Yarra, Australia: Macmillan Education, 1998.

Savory, P., and J. Olson. "Guidelines for Using Process Mapping to Aid Improvement
Efforts." Hospital Materiel Management Quarterly 22, no. 3(2001): 10-16.

Schmid, AA. Property, Power, and Public Choice-An Inquiry into Law and Economics.
Praeger, New York, NY. 1987.

Schmid, A.A. Institutional and Behavioral Economics. ” East Lansing, Michigan
Walbridge Press. 2001. -

Schoemaker, P. J. H. "How to Link Strategic Vision to Core Capabilities." MIT Sloan
Management Review 34, no. 1(1992): 67-81.

Schroeder, T. C. C. E. Ward, J. R Mintert and D. S. Peel. “Value-Based Pricing of Fed
Cattle. Challenges and Research Agenda.” Review of Agricultural Economics 20
no. 1(1998):125-134.

Schwarz, N. "Self-Reports: How the Questions Shape the Answers." American
Psychologist 54, no. 2(1999): 93-105.

Sekaran, U. Research Methods for Business. New York: John Wiley, 1992.

Selltiz, C., L. S. Wrightsman, and S. W. Cook. Research Methods in Social Relations.
New York: Holt, Reinhart and Winston, 1976.

240

Sexton, D. E. J. "A Cluster Analytic Approach to Market Response Function." JMR,
Journal of Marketing Research 11, no. 1(1974): 109-114.

Shah, J ., and N. Singh. "Benchmarking Internal Supply Chain Performance:
Development of a Framework." The Journal of Supply Chain Management 37, no.
1(2001): 37-47.

Sheth, J. N. "Using Factor Analysis to Estimate Parameters." Management Science 15,
no. 6(1969): 350-352.

Simchi-Levi, D., P. Kaminsky, and E. Simchi-Levi. Designing and Managing the Supply
Chain. New York: McGraw Hill, 2004.

Simon, A., A. Sohal, and A. Brown. "Generative and Case Study Research in Quality
Management Part 1: Theoretical Considerations." The International Journal of
Quality & Reliability Management 13, no. 1(1996): 32-41.

Simon, H. Administrative Behavior. New York: Macmillan, 1961.

Sislian, E., and A. Satir. "Strategic Sourcing: A Framework and a Case Study." Journal
of Supply Chain Management 36, no. 3(2000): 4-11.

Slater, G., and D. A. Spencer. "The Uncertain Foundations of Transaction Costs
Economics." Journal of Economic Issues 34, no. 1(2000): 61-87.

Sloan, E. A. "Prime Time for Meats and Poultry." Food Technology 60, no. 3(2006): 19.

Sloan, E. A. "What, When, and Where Americans Eat: 2003." Food Technology 57, no.
8(2003): 48-66.

Smith, B. S. "A Case-Ready Modiﬁed Atmosphere Packaging System for Beef."
Dissertation, Louisiana State University, 1997.

Smulder, F.J.M., J. Hiesberger, P. Hobauer, B. Dogl and E. Dransﬁeld. “Modiﬁed
Atmosphere Storage Under Subatrnospheric Pressure and Beef Quality: Color,
Drip, Cooking Loss, Sarcomere Length, and Tenderness.” Journal of Animal
Science. 84, no. 9(2006): 2456-2462.

Sohal, A., A. Simon, and E. Lu. "Generative and Case Study Research in Quality

Management Part H: Practical Examples." The International Journal of Quality &
Reliability Management 13, no. 2(1996): 75-89.

241

Sorheim, O., T. Aune and T. Nesbakken. “Technological, Hygenic and Toxicological
Aspects of Carbon Monoxide Used in Modiﬁed Atmosphere Packaging of Meat.”
Trends in Food Science & Technology 8 no. 9(1997): 307-312.

Sorheim, 0., H. Nissen and T. Nesakken. “The Storage Life of Beef and Pork Packaged
in an Atmosphere with Low Carbon Monoxide and High Carbon Dioxide.” Meat
Science 52, no. 8(1999): 157-164.

Spector, LC. and M. Mazzeo. “Probit Analysis and Economic Education.” Journal of
Economic Education. 11 no.2(1980): 37-44.

Springer, J. (2001) Case-Ready Already. www.ideabeatcom.

Stalk, G., P. Evans, and L. Shulman. "Competing on Capabilities." Harvard Business
Review 70, no. 2(1992): 57-69.

Starbird, S.A. “Moral Hazard, Inspection Policy, and Food Safety.” American Journal of
Agricultural Economics. 87 no. 1(2005): 15-27.

Stewart, D. W. "The Application and Misapplication of Factor Analysis in Marketing
Research." JMR, Journal of Marketing Research 18, no. 1(1981): 51-62.

Stewart, G. "Supply Chain Performance Benchmarking Study Reveals Keys to Supply
Chain Excellence." Logistics Information Management 8, no. 2(1995): 38-44.

Summers, J. 0. "Guidelines for Conducting Research Publishing in Marketing: From
Conceptualization Through the Review Process." Academy of Marketing Science
Journal 29, no. 4(2001): 405-415.

Suter, D. and SR. Henneberry. “An Examination of the Structure, Conduct and
Performance of the US. Food Processing Industry.” Journal of Food Products
Marketing. 3 no. 2(1996): 65-85.

Symons, R. T., and R. A. Jacobs. "Multi-Level Process Mapping: A Tool for Cross-
Functional Quality Analysis." Production and Inventory Management Journal 38,
no. 4(1997): 71-76.

Tamagnini, V., and A. Tregear. "An Assessment of Niche Marketing Opportunities in the
Delicatessen Meat Sector." British Food Journal 100, no. 5(1998): 228-235.

Tayles, M. and C. Dnuy. “Moving From Make/Buy of Strategic Sourcing: The
Outsourcing Decision Process.” Long Range Planning 34, no. 5(2001):605-622.

Tewari, G., D. S. Jayas, and R A. Holley. "Centralized Packaging of Retail Meat Cuts: A
Review." Journal of Food Protection 62, no. 4(1999): 418-425.

242

Tosh, M. "Meatier Issues: Consumer Demand and Category Management Reshape
Operations." Progressive Grocer 78, no. 1(1999): 59-60.

Torkkeli, M. and M. Tuominen. “The contribution of Technology Selection to Core
Competencies.” International Journal of Production Economics. 77, no. 3 (2002):
271-284.

Thompkins, J.A. “Don’t Outsource the Relationship.” Industrial Engineer 37 no. 11
(2005): 28-33.

Turcsik, R. "Meating Change." Progressive Grocer 80, no. 8(2001): 39-45.

Turcsik, R. "Meating Change." Progressive Grocer 80, no. 8(2001): 39-45.

Unnevehr, L. J ., and H. H. Jensen. "HACCP as a Regulatory Innovation to Improve Food
Safety in the Meat Industry." American Journal of Agricultural Economics 78, no.
3(1996): 764-769.

(USFDA) US. Food and Drug Administration. 2002 GRAS notice nr GRN 000083.
Washington DC: USFDA.

Van Hoek, R. 1. "Measuring the Unmeasurable: Measuring and Improving Performance
in the Supply Chain." Supply Chain Management 3, no. 4(1998): 187-192.

Veall, MR. and KP. Zimmermann. “Pseudo-R2 Measures for Some Common Limited
Dependent Variable Models.” Journal of Economic Surveys. 10 no. 3(1996):24l-
259.

Venkatesan, R "Strategic Sourcing: To Make or Not to Make." Harvard Business Review
70, no. 6(1992): 98-107.

Vernimmen, T., W. Verbeke, and G. v. Huylenbroeck. "Transaction Cost Analysis of
Outsourcing Farm Administration by Belgian Farmers." European Review of
Agricultural Economics 27, no. 3(2000): 325-345.

Vosburgh, R. “Retailers Consider Modiﬁed Atmosphere Packaging for Meat.”
Supermarket News 54 no. 9(2006):l-2.

Voldseth, S. (2002) Beef Industry at the Foreﬁ'ont of Beef Safety Initiatives.
Wagar, K. (2005) “Zero Hour.” Progressive Grocer 84(17): 106-107.

Walker, G., and D. Weber. "A Transaction Cost Approach to Make-or-Buy Decisions."
Administrative Science Quarterly 29, no. 3(1984): 373-391.

243

Walker, G., and D. Weber. "Supplier Competition, Uncertainty, and Make-or-Buy
Decisions." Academy of Management Journal 30, no. 3(1987): 589-596.

Welch, J. A., and P. R. Nayak. "Strategic Sourcing: A Progressive Approach to the
Make-or-Buy Decision." The Academy of Management Executive 6, no. 1(1992):
23-31.

Wellman, D. "Meat, Produce Lead Reﬁigerated Opportunities." Frozen Food Age 50, no.
10(2002): 1-3.

Westgren, R, and K Zering. "Case Study Research Methods for Firm and Market
Research." Agribusiness 14, no. 5(1998): 415-423.

White, L. “Keeping Meat Safe.” National Provisioner 220, no. 9(2006) 78-86.

White, R, and B. James. The Outsourcing Manual. Hampshire: Gower Publishing
Limited, 1996.

Wicklund, R.A., D.D. Paulson, E.M. Tucker, A.J. Stetzer, F. DeSantos, M. Rojas, B.J.
MacFarlane and MS. Brewer. “Effect of Carbon Monoxide and High Oxygen
Modified Atmosphere Packaging and Phosphate Enhanced, Case-Ready Pork
Chops.” Meat Science 74(4), 2006. 704-709.

Wilkinson, B.H.P., J.A.M. Janz, P.C.H. Morel, RW. Purchas and W.H. Hendriks. “The
Effect of Modiﬁed Atmosphere Packaging with Carbon Monoxide on the Storage
Quality of Master-Packaged Fresh Pork.” Meat Science 73, no.4 (2006): 605:610.

Williamson, 0. E. Markets and Hierarchies. New York: Free Press, 1975.

Williamson, 0. E. The Economic Institution of Capitalism. New York: The Free Press,
1985.

Wilson, R. D. "Research Design: Qualitative and Quantitative Approaches." JMR,
Journal of Marketing Research 33, no. 2(1996): 252-255.

Winicur, B. "Analysis of Decision Making." The National Public Accountant 38, no.
2(1993): 6-8.

Winkleman, M., D. Dole, L. Pinkard, J. Molloy et al. “The Outsourcing Source Book.”
Journal of Business Strategy. 14 no. 3(1993): 52-56.

Wolfe, D. F. "A New Questionnaire Design." Journal of Marketing 21(1956): 186-190.

Wolpert, V. "MAP Shows Potential for Growth." Paper, Film & Foil Converter 70, no.
7(1996): 20-21.

244

Woodward, B. “Beef Cattle Management Update.” University of Minnesota Extension
Service. Issue 42, September, 1996.

Wooster, J. "MAP Guide to Freshness." Canadian Packaging 51, no. 4(1998): 39-40.
Wooster, J. "MAP Guide to F reshness." Canadian Packaging 51, no. 4(1998): 39-40.

Wulf, D. and L. Muench. “Retail Display Life of Case-Ready Beef.” National
Provisioner 220, no. 3(2006):]0-11.

Yin, R. K. Case Study Research: Design and Methods. 3rd ed. Thousand Oaks: Sage,
1994.

Yoon, K. P., and G. Naadimuthu. "A Make-or-Buy Decision Analysis Involving Precise
Data." International Journal of Operations & Production Management 14, no.
2(1994): 62-69.

Young, B. "Production Partners." National Provisioner 215, no. 6(2001): 18-25.

Young, B. “Engineering Food Safety.” National Provisioner 220, no. 9(2006): 44-62.

Young, B. "Value-Added Growth." National Provisioner 215, no. 10(2001): 9-14.

Young, B. "Excel is in the Case." National Provisioner 215, no. 9(2001): 20-25.

Young, B. "Progressive Packaging." National Provisioner 216, no. 8(2002): 78-82.

Young, B. "Friendly Competitors: Brutality Does Not Necessarily Deﬁne the
Marketplace Battleﬁeld Based on the Recently Announced Partnership Between
Two Meat Industry Leaders—Excel and Hormel." National Provisioner 216, no.
9(2002): 30.

Young, B. "Partnering to Compete." National Provisioner 216, no. 9(2002): 10-15.

Yue, J ., and H. Cooper. "A Quantitative Review of Research Design Effects on Response
Rates to Questionnaires." JMR, Journal of Marketing Research 20, no. 1(1983):
36-44.

Zanderighi, L. "How to Design Perforated Polymeric Films for Modiﬁed Atmosphere
Packaging (MAP)." Packaging Technology and Science 14, no. 6(2001): 253-266.

Zhu, 2., K. Hsu, and J. Lillie. "Outsourcing - A Strategic Move: The Process and the
Ingredients for Success." Management Decision 39, no. 5/6(2001): 373-378.

Zsidisin, G. A. "Managerial Perceptions of Supply Risk." Journal of Supply Chain
Management 39, no. 1(2003): 14-25.

245

   

M'lll'llillljﬂlllﬁliiﬂliiiﬂljijil“