 

 

 

 

 

 

.1:
1. a 2.1

.2} .a. ..
as... r.

1131.... . .J
:2... la. .. \

xx ’7 .a:
3...:er

.r. 11v :1.) .2

5.. ..............mmﬁ.i.xw..seh “an Aubuuggﬂg
:5. v. .11}: .A( xiii..- lzL...§. » I. .52 .
.mﬁﬂpﬂ. ruﬁuzﬂﬁ 9. numb... my:

a... . . .1. . WW
3 . 3 ch
‘
..

. 93.”? ..

:1: 3,.

.tﬁﬂhgati v5

..\ .

“int"“nl’uu4.
S ..

.1 rt...

:\
9:

.3. .n‘ ..
r i. z... .
x 1153‘

S t
3.
mg

.
$3...

t::h:§.!.£n ‘3

n....ia.....x: .3!

.. I... 9.13:3...

. .3 ‘ 3.... {a

.1.

8. :55: In
.5 ..
i. is E n: x. 3.. hi

1...}!

v 9
3 . Ell-r
53...)! a.“ 3gp. «“15:

323%:
.1??? -

.3
‘3

.. 3.3. .
Etrnuamgn}
i. ....,m..;ir...

x... .

i... 1:.
.- unlit-ottluiutﬁ Fungi ADS
.' 1 .S‘A‘iﬁ 3.1.! I ‘ ~’-¢v‘ I.
u ‘7'". g ‘3 no: ‘t p l ‘1
L3,...zin. . . 5...}... § . .355. ug. its!
33%;... 3.43.... .. .2

I: .i
in

.I .t
.73.): 3..

. u i, ‘IMAJI-n'.v:-
L2. {—olul-cii A. vi.- E I. it Q}!’

I.
Q

.0.
. 5394 (.5

5
3|: :5! 4::
hhﬁvﬂxuihx . ... six-it
E .. . . y .-
thu‘.3tl:t:
. L... 5.0.»... I t
. ezlxnl$ﬁba¥a
3531.5!!! $.03.
5.6:;sz

3151:!-

.;.l 1......
9.. 13. .1335:

.ﬁ’ﬂ

:xmuuﬁ,

.- is
main:

is

.I it?!

liggu b 1 34.1
$333. 3."..- H:-
..t...v )lv

-11.. c. s. ..
(than... u. 2: . Raf... :

..\2r..t...l..:...a .
§;..$3\ .a....tll.~s$bh

tar-.17 I: h. .QXk.
lei—1...... v.1": . r

‘
.Y.
. gatfninﬁv.
I... :9;

£51K.

1...... .2...
5...}

£5.23:

3...! .1 3...... .
$13.3...“ mitll‘euﬁt ‘9...an
.25 . 1, #3..

i 45%.... ﬁnd
5 5....

p... :5.

$1.58, .93..
4. .235. .Si
ill

Eli.‘ll3..t.
.21’1'313.’ ‘
I... III...“ 3
131.1%.) in! 1

n. .a..ﬁ:h...1h.i..ﬂa.i.... h
. .1. . .. 2 .3
... :fr.i!.£a§d.3u1 :3:
r... , t

.I I'
z :15. 12‘:
V 0.5

 

THESIS

7.600

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the

dissertation entitled

THE INFLUENCE OF NETWORK CHARACTERISTICS ON

INFORMATION ACCESS, MARKETING COMPETENCE AND

PERCEPTIONS OF PERFORMANCE IN SMALL, RURAL
BUSINESSES

presented by

Barbara J. Frazier

has been accepted towards fulﬁllment
of the requirements for

PhtD- ckgnwinﬂnman_Enxinonment & Design:
Merchandise Management

T/zWM/Méﬁm

Major professor

 

Date 5/?j}OOO

MSU i: an Afﬁrmative Action/Equal Opportunity Institution 0-12771

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

11/00 WM

THE INFLUENCE OF NETWORK CHARACTERISTICS ON
INFORMATION ACCESS, MARKETING COMPETENCE
AND PERCEPTIONS OF PERFORMANCE IN SMALL RURAL BUSINESSES

By

Barbara J. Frazier

A DISSERTATION
Submitted to
Michigan State University

in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Human Environment: Design and Management
Merchandising Management Specialization

2000

ABSTRACT

THE INFLUENCE OF NETWORK CHARACTERISTICS ON
INFORMATION ACCESS, MARKETING COMPETENCE
AND PERCEPTIONS OF PERFORMANCE IN SMALL RURAL BUSINESSES

By

Barbara J. Frazier

This study focused on the inﬂuence of network relationships on the ability of retail
entrepreneurs in small communities to access and use business information. I drew upon
social network tmory to propose a model linking an entrepreneur’s network
characteristics to the level of social capital available in the network. I suggested that
social capital positively inﬂuences the quality of information that entrepreneurs can access
through their strong-tie and weak-tie networks. Information quath is then transformed
into marketing competence, which positively impacts ﬁrm performance.

Data were collected from 112 independent gift retailers in small towns in
Midwestem states, using a mailed survey instrument. Exploratory and conﬁrmatory factor
analysis were used to test validity and reliability of model constructs. A network was
conceptualized as a latent variable, which explained the density, centrality, ﬁ'iendship and
perceptual homophily features of the entrepreneur’s network ties. Social capital was a
latent variable, which explained the level of trust, commitment and reciprocal intentions
among identiﬁed network members. Information quality measured the relevancy,

timeliness and speciﬁcity of business information received from network members.

Marketing competence was characterized as the ability to assess customer needs, provide
quality customer service, and introduce innovation. Performance was measured by
perceptions of success relative to industry and competitors.

Structural equation techniques were used to test causal relationships in the model.
Results showed that network ties inﬂuenced the level of social capital in both strong-tie
and weak-tie infornmtion networks. Social capital inﬂuenced the richness of information
received from these networks. Social capital did not have a direct inﬂuence on perceptions
of ﬁrm performance.

Information received ﬁ'om weak-tie networks inﬂuenced marketing competence in
introducing innovation. No link between information ﬁom strong-tie networks and
marketing competence was found. There was a signiﬁcant relationship between both local
and innovative marketing competence and performance.

Results support social network theory contentions that both strong and weak ties
facilitate the ﬂow of information. Implications for retail entrepreneurs suggest that
networking is an important activity for gathering business information, and that strength of
network relationships can inﬂuence the quality of informtion. This research also
highlights the need for retail entrepreneurs to better use the information they receive to

build marketing skills.

Cepyright by
BARBARA J. FRAZIER

2000

DEDICATION

To my husband, Lon and children, Megan, Josh and Alex.

Your love and support carried me through.

ACKNOWLEDGMENTS

I would like to express my sincere appreciation to those who have supported my
work on this dissertation. My Committee Chair, Dr. Patricia Huddleston provided
support and guidance all the way through the project. I am pleased to be her ﬁrst doctoral
graduate. I would also like to extend my thanks to Dr. Linda, Good and Dr. Brenda
Sternquist for the inspiration, support and challenges they gave me throughout the
program at Michigan State. Gratitude also goes to Dr. Roger Calantone, who supported

my eﬂ‘orts at structural equation modeling, even while he was on sabbatical.

Iwouldalsoliketothankthefacultyandstaﬂ‘inFamily&ConsumerSciencesat
Western Michigan University, where I taught during the time I was completing my degree.
Dr. Linda Dannison, Dr. Nancy Steinhaus, and Dr. Marlene Breu were wonderful sources

of support and encouragement during this process. _

I will also treasure the camaraderie and support ﬁom the Doctoral students in
Human Environment and Design at Michigan State. Thanks to Jae-Bun Chung, Madeline

Plaster, Linda Niehm, Linda Fernandes-Plank, Leslie Stoel and Vanessa Wickﬁﬂe.

An ﬁnally, thank you to my family for all of the weekends, nights and early

momings you allowed me to have to ﬁnish my degree!

TABLE OF CONTENTS

LIST OF TABLES ................................................................................................ ix
LIST OF FIGURES ............................................................................................. xi
Chapter 1
Introduction. ......................................................................................................... 1
Signiﬁcance of Study ................................................................................. 2
Problem Deﬁnition. .................................................................................... 4
Objective of Study ..................................................................................... 8
Chapter 2
Literature Review ................................................................................................. 10
Exchange Theory ..................................................................................... 10
Network Theory ....................................................................................... 11
Entrepreneurial Exchange .......................................................................... 12
Embeddedness ........................................................................................... 13
Model Development ................................................................................. 14
Networks ...................................................................................... 14
Network Structure ............................................................. 17
Network Interaction. .......................................................... 22
Strength of Ties ................................................................. 25
Social Capital ...................................................................... 27
Information. ........................................................................ 32
Distinctive Marketing Competencies ................................... 39
Performance ....................................................................... 42
Chapter 3
Method ................................................................................................................. 45
Sample ....................................................................................................... 45
Instrument ................................................................................................. 45
Instrument Development ............................................................................ 46
Procedure .................................................................................................. 53
Sample Description. ................................................................................... 54
Chapter 4
Results .................................................................................................................. 56
Factor Development .................................................................................. 57
Exploratory Factor Analysis ...................................................................... 57
Conﬁrmatory Factor Analysis ................................................................... 59
Hypothesis Testing ................................................................................... 63
Structural Model Testing .......................................................................... 66

Summary of Results ................................................................................. 70

Chapter 5
Discussion and Conclusion .................................................................................... 72
Network Structure and Interaction ............................................................ 74
Social Capital ............................................................................................ 76
Network Relationships as a Source of Social Capital ................................. 77
Social Capital Inﬂuence on Information Quality ......................................... 77
Information Quality and Marketing Competence ........................................ 79
Marketing Competence and F irm Performance .......................................... 81
Limitations ............................................................................................... 82
Recommendations ...................................................................................... 83
Further Research. ....................................................................................... 83
Practical Implications ................................................................................. 84
Conclusion. ................................................................................................ 85
Appendix A: Survey Instrument .............................................................................. 88
Appendix B: Tables ............................................................................................... 99
Appendix C: Formulas .......................................................................................... 139
Appendix D: Figures ............................................................................................. 141
References ............................................................................................................ 147

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.

Table 8.

Table 9.

Table 10.

Table 11.

Table 12.

Table 13.

Table 14.

Table 15.

Table 16.

LIST OF TABLES

Summary of Social Capital in Literature .............................................. 99
Instrument Development ..................................................................... 102
Item Content ....................................................................................... 1 12
Characteristics of Sample .................................................................... 116
Factor Loadings and Cronbach Alpha for Scale Items ......................... 117

Parameter Estimates for First Order CFA for NETCHAR

(MI Networks) ................................................................................... 120

Parameter Estimates for First Order CFA for NETCHAR

(INV Networks) ................................................................................. 121

Parameter Estimates for Second Order CFA for NETCHAR

(MI Networks) ................................................................................... 122

Parameter Estimates for Second Order CFA for NETCHAR

(INV Networks) ................................................................................. 123

Parameter Estimates for First Order CFA for SC (MI Networks) ........ 124
Parameter Estimates for First Order CFA for SC (INV Networks) ...... 125

Parameter Estimates for Second Order CFA for SC (MI Networks)....126

Parameter Estimates for Second Order CFA for SC
(INV Networks) ................................................................................. 127

Differences in Latent Means for DENS, DENT, FRND
and PERHOM .................................................................................... 128

Parameter Estimates for First Order CFA Structural Model
(MI Networks) .................................................................................. 129

Parameter Estimates for First Order CFA Structural Model
INV Networks) ................................................................................. 131

Table 17.

Table 18.

Table 19.

Table 20.

Parameter Estimates for Structural Model (MI Networks) ................. 133
Parameter Estimates for Structural Model (INV Networks) ............... 134
Coefﬁcients of Determination for Dependent Variables

in the Causal Model ........................................................................... 135
Results of Hypothesis Testing ............................................................ 136

Figure 1.
Figure 2.
Figure 3.
Figure 4.

Figure 5.

LIST OF FIGURES

Theoretical Model .......................................................................... 141
Initial Measurement Model ............................................................. 142
Second Order Factor Model for NETCHAR. .................................. 143
Second Order Factor Model for SC ................................................ 144
Structural Model ............................................................................ 145

CHAPTER 1

INTRODUCTION

Independent retailers face enormous challenges in the modern retail environment.
Smaller retail ﬁrms have been dramatically affected by changing economic forces in the
structure of retailing in the United States. Over the last several decades, shifts in
population, changes in demographics and attitudes and lifestyles of US. consumers lmve
led to retail activity becoming increasingly concentrated in large-scale retail formats
(Dalal, Al-Khatib, Da Costa & Decker, 1994; Stone, 1995). New retail formats, such as
direct selling through catalogues, television shopping and the Internet are giving
consumers more options in how, when and where they conduct retail transactions

(Barlow, 1994; Goodman, 1995).

These changes have been devastating to many small, independent retailers, many of
whom are not able to compete with the economies of scale of large-scale retailers. The
broad assortments and low-prices offered by discount chains and category killers have
resulted in the closing of many small stores (McCune, 1994). Many retail sectors are
consolidating as bigger stores gain market share and drive independents out of business.
The density of retail establishments in the US. was 60 per 10,000 people in 1992 versus

62 in 1987, with smaller stores accounting for the decline (Du & Apfel, 1995).

5' '[i [S I

In spite of the failure rates, owning a small store remains an attractive idea for
many individuals. Retail entrepreneurship offers autonomy, ﬂexibility and satisfaction
(Buss, 1996; Cooper and Artz, 1995). Thousands of new retail businesses are started
each year in the US, offering an avenue for independent employment. Retail
entrepreneurship remains a key source of innovation and job creation in the US. economy.
Small retailers dominate ownership in hardware stores, sporting goods stores, jewelry

stores and gift stores (McCune, 1994).

Small stores can satisfy consumer demand by ﬁlling gaps in the market that are not
proﬁtable for larger retailers (Buss, 1996; Julien, 1993). As one forecaster put it, large
ﬁrms are like boulders dropped in a hole, and entrepreneurial opportunities are the spaces
created between the boulders (Williams, 1999). Retail entreprenem's can ﬁll these spaces
by providing unique products and distinctive personal service. Several authors have noted
that people miss the personal attention of “mom and pop” stores (Barlow, 1994; Casison,
1998). Smaller stores are able to compete successfully by tailoring their assortments to
complement, rather than compete with, larger discount chains and category killers.
Personal service is another way in which independent retailers can differentiate (Stone,

1995).

The threat posed by large scale retailing has had a particularly devastating toll on
independent retailers in smaller U.S. conmiunities. Urban migration due to changes in the
agricultural sector have resulted in a steady erosion of the population base and declining
consumer demand (F uguitt, Brown & Beale, 1989). Competition ﬁom discount stores,
along with easier access to nearby larger communities with regional shopping centers,
have further cut into the market share of independent home town stores (Dalal, et al,
1994). Downtown shopping districts in nonmetropolitan areas, which have historically
been populated with independent merchants, have undergone profound changes due to
economic and social forces (Lawhead, 1995). In many communities, storeﬁ'onts stand
empty, or are occupied by non-retail business establishments (Henderson & Wallace,

1992).

Strengthening the nonmetropolitan retail sector is an important part of rural
development programs (Flora & Flora, 1990; Lawhead, 1995; Markley & McNamara,
1997). Smaller, more rural communities must create a positive social and economic
environment to attract and retain residents. A healthy retail sector in a rural community
can provide off-farm employment for local residents and keep local dollars circulating in
the community (Henderson & Hines, 1990) Making a town more appealing to potential
residents may attract more good jobs (Lawhead, 1995). Recent studies of small towns
suggest that small retail establishments can serve as the glue that keeps residents in a

community by providing a place for informal public life (Irwin, Tolbert & Lyson, 1997).

Residents are more positive about their communities when they perceive convenient

3

access to meded goods and services (Brown, 1993). These factors point to the need for

more problem-solving to support independent retail entrepreneurship for rural

communities.

Research focused on causes of rural retail decline has centered primarily on rural
consumer behavior (Miller & Kean, 1997a and 1997b; Riecken & Yavas, 1988; Samli,
Riecken & Yavas, 1983). Rural residents often cite dissatisﬁmtion with local retailers as
reasons for their outshopping behavior (Dalal et al, 1994; Miller & Kean, 1997a). Studies
show that rural residents want local retailers to provide adequate assortments of well-
priced, quality merchandise, personal service and an enjoyable, convenient atm03phere. A
recent study by Miller and Kean (1997a) found that local shoppers who had positive
attitudes about retailers in their communities were generally satisﬁed with retailer offerings
in their communities. Subramanian (1993) found that as retailers adapted their product
and service offerings to suit customers, the exchange utility perceived by consumers

increased and outshopping decreased.

What comes through clearly in these studies is that relying exclusively on
hometown loyalty is no longer sufﬁcient for the survival of independent retailers in rural
communities. Local retailers in nonmetropolitan communities must possess superior
marketing competencies in order to compete with large scale retailers and specialty

retailers from other communities. The literature on rural consumer behavior suggests that

4

 

in order to retain local customers in the community and draw customers ﬁom nearby
communities, a rural retailer must develop and sustain marketing competence tlurt results

in the delivery of products and services that are valued by rural consumers.

Small town merchants must react to constant and rapid change in order to maintain
and build market share. Those small retail ﬁrms that have survived in the shadow of the
retail giants have focused on differentiation or niche strategies that complement, rather
than compete with chain stores (Barlow, 1994; Kean, Gaskill, Leistritz, Jasper, Bastow-
Shoop, Jolly & Sternquist, 1998; Nation’s Business, 1993; Stone, 1995). Execution of
effective strategie$ requires development of distinctive marketing competencies that
enable retailers to respond to environmental changes more effectively than competitors

(Conant, Smart & Solano-Mendez, 1993).

Distinctive competence refers to marketing activities that an organization performs
especially well in comparison to competitors (Day & Wensley, 1988). Competence relates
to the ability of a ﬁrm to deliver products and/or services in an eﬂicient manner relative to
its competitors. These competencies can become sustainable competitive advantages if
consumers perceive them as valuable, and competitors cannot easily duplicate them
(Bharadwaj, Varadarajan & Fahy, 1993). Distinctive marketing competence has been

shown to lead to better organizational performance in small retailers (Conant, Smart &
Salerno-Mendez, 1993). A ﬁrm creates competence by articulating its objectives and
executing the processes that are necessary to meet the goals it has set (Sinkula, 1994).

This process requires identifying and interpreting information in the environment that is

5

relevant to the development of competence (McGrath & MacMillan, 1992; Sinkula,
1994). Exploiting information asymmetries that allow an entrepreneur to identify and ﬁll
unmet customer needs can lead to sustainable competitive advantage (Lado, Boyd &
Wright, 1992). Thus, ability to access strategic information can ultimately determine the

successofaﬁrm.

The role of any retailer in marketing exchange is to create value for consumers by
delivering combinations of assortment, price, promotion, display, customer service and
location that meets the needs of the target market (Cam, Rabianski & Vernor 1995). The
ﬁrm’s strategy is the set of decisions it makes about how it will adjust to environmental
change (Miles, Snow, Meyer & Coleman, 1977). Information is critical to the
entrepreneurial, operational and administrative solutions that are necessary to compete in
the nmket. Furthermore, retailers must have the ﬂexrbility to respond more quickly than
competitors to changes in the business environment (Kean et a], 1998). An effective retail
strategy requires knowledge of current and prospective customers and industry trends,
along with access to innovative customer service, promotion and merchandising
techniques (Conant, Mokwa & Varadarajan, 1990; Mche, 1994; Pearson, 1994).

Effective strategy building requires extensive information input from many sources.

Extensive work has focused on the information seeking behavior of entrepreneurs
in small businesses. The literature on information search indicates that small business
managers seek information about rurming their businesses ﬁom multiple sources, including

S”PIJIJ'ers, other retailers, local business owners, professional advisors, employees and

 

customers (Baal, 2000; Birley, 1985; Dollinger, 1985; Gales &. Blackburn, 1990; Shafer,
1990; Specht, 1988). Many studies conclude that small business owners and managers
prefer personal sources of information as input for strategic decision making. Most of this
work focuses on the presence of relationships, but neglects the relational qualities between

the information seeker and the source.

Recent work regarding networks suggests that ties with network members can
inﬂuence performance by providing entrepreneurs with richer, more timely information
than can be attained by arms length relationships (Burt, 1992; Uzzi, 1996). The nature of
the relationship among network members appears to inﬂuence the ability to access
information in a network. Granovetter (1985) suggests that some information is more
easily accessed through what he describes as “weak ties”, where network members with
whom one has less frequent contact and fewer incidences of network interaction are richer
sources of valuable information than close ﬁiends. Weirnann (1983), on the other hand,
showed in a study of information ﬂow that “strong” ties promoted the ﬂow of information

within groups, while weak ties facilitated inter-group information ﬂow.

Some authors have conceptualized the ability to access resources ﬁom an
individual’s set of personal networks as “social capital”. Access to social capital means
that people have connections to individuals with greater amounts of economic and cultrnal
capital, and who can help with advice, further cormections and access to other resources

(Burt, 1997; Nahapiet & Ghoshal, 1998. Social capital is produced through embedded

ties characterized by frequent contact with individuals with whom one has a close,
personal relationship. Embedded relationships yield levels of trust, obligation and

rec'mrocity that can provide a competitive advantage to entrepreneurs (Burt, 1997).

Ill" [51

I conducted a study focused on exemplary small retailers in rural areas in Michigan
in 1997 (Frazier, 1999). These “superpreneurs” were identiﬁed by peer nomination as
successful retailers who possessed extraordinary vision, passion and leadership. This
study revealed that superpreneurs were skilled networkers, using multiple personal
contacts to access and ﬁlter business infornmtion. They maintained close ties with their
family members and ﬁ'iends, with customers and with business colleagues in and out of the
community. They also developed long-term relationships with people that acted as
bridges to other networks. These ties are a source of inspiration which leads to innovation

insnmllﬁrms.

Information beneﬁts may build intellectual capital that can be used to develop and
sustain marketing competence, leading to better ﬁnancial performance (Nahapiet &
Ghoshal, 1998). No empirical work Ins tested these assumptions in the context of retail

entrepreneurship. The objective of this study is to explore empirically the impact of
independent retailers’ network ties on the creation of social capital, the acquisition of

inlbrrmtion, and the development of marketing competence. The central proposition of

 

this paper is that the nature of ties in an entrepreneur’s network of suppliers, trade
associates, family, ﬁiends, and community residents inﬂuences access to the information
necessary to build competitive advantage. Higher quality information about the market and
the environment enable the development of superior marketing competence, leading to

better ﬁrm performance.

CHAPTER 2

LITERATURE REVIEW

MW

Social exchange theory looks at resource exchange in the context of
interdependent, long-term relationships embedded in the dense fabric of social
relationships. Unlike the impersonal, one-time exchange depicted in traditional marketing
theory, social exchange rests on the premise of on-going relationship with other actors
who have reasonably predictable traits that can enhance or diminish the value of the

exchange (Emerson, 1973).

Marketing exchange is a special case of social exchange, which argues that people
enter into exchange relationships for goal-oriented reasons that are not entirely based on
cost(Bagozzi, 1975). People engage in both social and economic exchange to satisfy
needs by inﬂuencing or complying with the norms and traditions of the exchange network.
On-going relationships are characterized by rules and norms that facilitate the exchange.
The expectation of future exchanges reduces the likelihood of malfeasance and
opportunism (Easton & Araujo, 1994). Exchange relationships can be characterized as
communal (caring) to instrumental (tit-for-tat) relationships (Winstead, Derlega,

Montgomery & Pilkington, 1995).

10

Networklhecnz
Traditional economic theory argues that market exchanges are independent events

conducted by self-interested actors with perfect information. Price is the mechanism that
controls the market. Social structures are not taken into account, or are accounted for
only peripherally (Coleman, 1988; Granovetter, 1973; Portes & Sensenbrenner, 1993).
Network theory, on the other hand, integrates the concept of relationships into the
exchange equation. Networks include the set of relevant exchange relationships between
actors; network analysis is concerned with the inﬂuence of structure and interaction of
those relationships on performance and outcomes (Cook & Whitemeyer, 1992; Gilly,
Graham, Wolﬁnbarger & Yale, 1998). This approach allows analysis of marketing
exchange behavior taking the effects of personal relationships into account (Uzzi, 1996).
Other research has shown that social relationships can build social capital of the exchange
partners. Network characteristics have been used to explain career mobility (Burt, 1992;
Granovetter, 1985), word-of-mouth communication (Frenzen & Nakamoto, 1993),
consumer buying behavior (Frenzen & Davis, 1990; Miller & Kean, 1997), returns to
education (Coleman, 1988; Friedmn & Krackhardt, 1997; Morgan & Sorensen, 1999),
immigration (Portes and Sensenbrenner, 1993) and successful adoption of new innovations

(Swan & Newell, 1995).

The exchange framework is particularly attractive for evaluating the performance
of rural retail entrepreneurs. First, small ﬁrm behavior is often embodied in the behavior
of the owner/manager. Decision making in small ﬁrms is often highly personalized,

reﬂecting the personality and attitudes of the owner/manager (Jennings & Beaver, 1997),

11

making the role of social structure in ﬁrm behavior a relevant topic. Further, although
rural entrepreneurs appear to share the same psychological traits as their urban
counterparts (Babb & Babb, 1992) residents in rural communities tend to have different
social strucmres than individuals from urban areas. T ies are more likely based on kinship
and neighborhood solidarities rather than on ﬁ'iendships. Networks of rural residents tend
to be denser, smaller and more homogemous (Beggs, Haines & Hurlbert, 1996; Wall,

Ferrazzi, & Schryer,l998).

WW

An abundance of entrepreneurial research has focused on the question of what
constitutes an entrepreneur. In the context of the marketplace, most concur that
entreprneurs are individuals that perform the ﬁrnction of identifying opportunities and
converting them into economic value (Baumol 1996; Carland, Hoy and Carland 1988;
Gartner 1988; Schumpeter 1942). Burt (1992) characterizes an entrepreneur as one who
has the opportunity, ability and motivation to take advantage of “structural holes”, or gaps
in information in networks. Gaps are created when certain members of a network are not
connected, providing opportunities for the entrepreneur to capitalize on opportunities.
For example, if A (the entrepreneur) knows B and C, but B and C do not know each
other, the structural hole between B and C creates opportunities for A. In terms of market
information, the structural hole between B and C provides A with more non-redundant
information that can be used to increase the rate of return. Thus, networks that are rich in
structural holes have more potential opportunities, and bridging the gaps requires social

activity. Effective social relations with network members lmve the potential to oﬂ‘er

12

higher rates of returns to well-connected players by providing access to the information

gaps in the market.

In the context of retail activity, retailers act as links in the marketing channel,
spanning the information gaps between producers and consumers. The best performing
retailers effectively bridge gaps in assortment, quantity, place and time (Lewison, 1994).
Offering the right product in the right place at the right time requires that a retailer have

better access to the “structural holes” in the marketplace than its competitors.

thcddedncss
Embeddedness refers to a logic of exchange where social ties inﬂuence
entrepreneurial behavior. Uzzi (1996) argues that:
“Organizations operate in an embedded logic of exchange that promotes economic
performance through inter-ﬁrm resource pooling, cooperation, and coordinated
adaptation. (p. 675)”.
Embeddedness suggests that entrepreneurs are motivated beyond purely economic goals
to pursue the enrichment of relationships through trust and reciprocity (Powell, 1990;
Smitka, 1991). Embedded relationships inﬂuence the value of a transaction and emich the
soc'ml capital of members in the network (Portes and Sensenbrenner, 1993).
Embeddedness within a group refers to the fact that exchanges within a group have an
ongoing social structure that inﬂuences action. Rather than the arms length relationships
characterized by isolated transactions where cost is everything, embedded ties involve
special, close relationships with trusted others. Social capital is the governance

mechanism that promotes voluntary transfer of information (Uzzi, 1996). The level of

13

embeddedness has been shown to be a factor in channel decisions for small ﬁrms. Uzzi
(1993; 1996) found that embedded relationships among small ﬁrms in the apparel

mamfactming industry inﬂuenced chances of survival.

Studying the outcomes of entrepreneurial network activities can provide new
insights into patterns of success and faihrre among retail ﬁrms in rural markets. In the
following section I propose a model that relates the level of embeddedness in retailer
networks to the development of social capital, information access, marketing competence

and ﬁnanc'ml perforrmnce.

Mcdslllcyclcumem
New
One’s personal network is the totality of all persons connected by a certain type of

relationship. From an ego-centered perspective, a network consists of the relevant
menmers of one’s social lamlscape at a particular point in time. Social networks may
include family, ﬁ'iends and acquaintances with whom the entrepreneur relates at a social
level. Suppliers, customers, trade associations, local business and government
organizations, and friendship and kin groups may also be part of the social networks of
small retailers (Aldrich & Zimmer, 1986; Nelson, 1989). Many network studies which are
focused on small ﬁrms deﬁne networks in the context of inter-organimtional networking
such as supplier-buyer networks (Barringer, 1997; Golden & Dollinger, 1993; Johannisson
& Monsted, 1992; Larson, 1991; Provan, 1993), competitor networks (Brown & Butler,

1 995; Human & Provan, 1997), professional advisors (Curran, Jarvis, Blackburn &

l4

Black, 1993), stakeholders (Rowley, 1997), and trade associations and wholesaler-
sponsored groups (Reijnders & Verhallen, 1996). Some studies have included more
informal networks comprised of family, friends and community residents (Bates, 1994;
Birley, 1985; Carroll & Teo, 1996; Dodd, 1997; Donckels & Lambrecht, 1997; Ostgaard

& Birley, 1996; Ramchandran & Ramnarayan, 1993).

Networks are rich in the resources that entrepreneurs need to grow and sustain
their businesses. Information acquired through network activity creates new knowledge
that can be used for decision making. Interpersonal communication in networks is
important to the diﬂhsion of new ideas. Networks promote social learning and adaptive

responses to an uncertain environment (Aldrich & Zimmer, 1986).

Entrepreneurs in small ﬁrms prefer personal rather than non-personal sources of
information (Birley, 1985; Cooper, Woo & Dunkleberg, 1989; Peters & Brush, 1996;
Shafer, 1991; Smeltzer, Fann & Nikoliasen, 1988; Young & Welsh, 1983). Brush (1992)
found that small ﬁrm managers conducted person-to-person and telephone networking
more than any other type of environmental scanning. Dollinger (1985) and Peters and
Brush (1992) found that the amount of time spent seeking information was related
positively to performance. Ostgaard and Birley (1996) associated higher and more diverse
levels of communication and networking activity with better performance. Van deVen,
Hudson and Schroeder (1984) related network activity to ﬁrm growth. In my
superpreneur study, best-in-class retailers preferred personal sources of information.

Participants in the identiﬁed mentioned personal sources more frequently than nonpersonal

15

sources when asked to name the most valuable sources of business-related inforrmtion

(Frazier, 1999).

For all of the interest in network activity in small ﬁrms, some studies suggest that
the frequency and importance of more formal networks are overstated. Bates (1994)
found that Asian immigrants who relied heavily on social networks were less proﬁtable
and more failure prone. Birley (1985) found that although small manufacturers used
networks to access resources, networking was not related to ﬁrm growth. Smaller ﬁrms
whose owner is also the manager and primary decision nnker, may simply be too busy
with day-to-day operating concerns to be able to devote very much time to formal
network activities. For these entrepreneurs, networking may be conducted in a more
informal manner. Trading information with other retailers at trade shows, building business
friendships with supplier representatives, and interacting with customers in social
situations may be more effective in developing social capital than participation in more

formal network activities.

There is evidence that the social structtue of one’s networks and position in the
structure can be a source of competitive advantage for entrepreneurs (Burt, 1992;
Granovetter, 1985 Uzzi, 1996). Burt suggests that successful entrepreneurs are those
individuals who are strategically positioned and connected in networks so that they are
able to take advantage of opportunities before others are aware of them. Networks are
places where individuals trade resources, and successful entrepreneurs are those that are

positioned to activate ties in order to gain access to business information and to attract

l6

customers (Aldrich & Zimmer, 1986).

Much of the analysis of network eﬁ‘ects has been approached ﬁom a structural
perspective. This approach focuses upon analyzing network effects by nupping the social
structure of a network from data about actual relationships in a network ﬁom all of its
members. Kilduﬂ‘ and Krackhardt (1994) suggest that the structural method can be
enhanced by focusing on “cognitive maps” of perceived relationships in a network.
Individual perceptions of network structure have been found to be eﬂ’ective in predicting
attitudes and opinions of focal network members (Marsden, 1990). Individuals use these
naps to Operate in their social environment. These cognitive maps reﬂect the perceptions
of structure in the minds of network members. Perceived ties are useful in measuring
' social inﬂuence, attitudes and opinions (Marsden, 1990). Perceived relationships were
shown to be more predictive of reputation than actual structure within an organization
(Kilduﬂ‘ & Krackhardt, 1994). I focus on the individual entrepreneur’s perceptions of the
structural and relational properties of their communication networks in this study. This
approach allows comparison of the individual networks of entrepreneurs to gain insight

into optimal structures for obtaining information that yields higher returns to information.

Recent work by network analysts suggests that certain types of networks provide
optinnl access to information (Burt, 1992; Granovetter, 1985; Greve, 1995). A person’s

network can be characterized by both its structure and by the nature of the interpersonal

l7

interaction in the network. Structure refers to the conﬁguration of an actor’s ties, or
social bonds with others (Davern, 1997; Hall & Wellman, 1985). Important dimensions of
structure with respect to information ﬂow include the density of the network, and the
centrality of a particular individual in the network (Burt, 1992; Rowley, 1997). Different
positions in the social structure means that individuals have access to differing levels of
power, prestige and wealth, leading to different opporttmities, constraints and outcomes

(Adams & Blieszner, 1994).

mm. Density is a characteristic of the whole network, and refers to the number
of ties that link network members compared to the total possible ties in the network.
Density increases as the number of ties within a network grows, and is often associated
with spatial proximity or kinship (Marsden, 1993). A network with high density would be
one where everyone in the network knows everyone else. Highly interconnected networks
facilitate ﬂow of norms and values among players, creating implicit behaviors and
expectations among members (Oliver, 1991). This meam that people in close networks are
more willing to share information with each other (Greve, 1995). Close-knit groups tend
to have little variation in norms, which leads to less ambiguity about emectations in the
group (Bienenstock, Bonacich & Oliver, 1990; Rowley, 1997). To a point, as density
increases, communication becomes more efﬁcient (Rowley, 1997; Uzzi, 1996). Density is
an indicator of actor-to-actor inﬂuence and is positively correlated with diffusion of
innovation (Rogers, 1983). New tacit knowledge ﬂows more easily through interpersonal

contacts (Lundvall, 1988).

18

Density can build social capital by facilitating the diﬂiision of norms through the
establishment of behavioral expectations (Rowley, 1997). Density allows social attitudes
to travel across the network more quickly (Bienenstock, Bonacich & Oliver, 1990).
Density is an indicator of cohesiveness of a group, and cohesiveness establishes trust
(Axelrod, 1984; Greve, 1995). Norms of trust means actors consider the probability of
long term and continued exchange in transactions with network members. Cooperation

and commitment develop as a natural basis of social relations (Axelrod, 1984).

Density has a positive effect on the speed and accuracy of certain types of
information. In diffusion of innovation studies, new ideas are transmitted among
interconnected individuals more rapidly at certain stages of the diffusion process (Rogers,
1983). Yamaguchi (1994) demonstrated that low density contributed to the ineﬂiciency of
information ﬂow through social networks. Ryan and Gross (1943) found that neighbors
were an important source of information for farmers adopting hybrid seed corn. New
drugs were more likely to be adopted by doctors who worked together than by those who
did not (Coleman, 1966). Weimann (1983) discovered that news and gossip traveled
faster and more accurately in interconmcted groups in an Israeli kibbutz. Liedka (1991)
formd that network density within a niche serves as a resource for organizatioml survival.
Private colleges belonging to the same consortia were more likely to adopt new programs
than those who were less densely connected (Baptista, 1999). Higher density of
organizations in a speciﬁc technological niche positively inﬂuenced the diffusion and

adoption of new technologies in manufacturing ﬁrms (Podolny, Stuart & Hannan, 1996).

19

Dense networks can be a constraint when redundant ties provide the entrepreneur
with similar information. Some evidence exists to support the value of sparse network
structures. Less dense networks can yield access to new information not available in one’s
primary network. Granovetter’s (1973) classic study on information access for persons
seeking employment provides support for the contention tlmt ties outside one’s primary
network offer access to valuable information. Spatially weaker links may provide more
important resources for a ﬁrm (Ostgaard & Birley, 1996). Falemo (1989) found tint
contact with persons living outside the community were more important in channeling

marketing resources to entrepreneurs.

Centrality. Centrality refers to where one is located in the ﬂow of information
relative to others in a network. Centrality is a predictor of inﬂuence or rank. An
individual centrally located in a network has status in the hierarchy, implying better access
and control of resources in a network. This position my be derived formally, through an
elected or appointed ofﬁce, or informally, built upon reputation and expertise (Ibarra,
1993). Entrepreneurs rmy ﬁnd themselves in a central position in the network hierarchy
by virtue of their socio-economic status or personality characteristics. As is well
documented in the diffusion of innovation literature, “opinion leaders” are central in the
ﬂow of information in their networks (Rogers, 1983). Through their position in a group,

they have the capacity to control or interrupt the ﬂow of communication.

The degree of centrality measures an actor’s ability to access independently all

other players in a network, the most central actors having the shortest aggregate distances

20

to all other actors (Rowley, 1997). Centrality in organizational networks is associated
with perceptions of power, adoption of innovations and access to critical information
(Burkhardt & Brass 1990; Ibarra and Andrews 1993; Krackhardt, 1990; Rogers &
Kincaid, 1981). Centrality in a network insures that information is easily accessed at low
cost. Inequality in centrality negatively impacts the ﬂow of information (Y amaguchi,
1994). Ibarra and Andrews (1993) found that advice network centrath in an advertising
agency inﬂuenced access to information, resources and legitimacy. Podolny, Stuart and
Harman ( 1996) found that organizational centrality was related to organizational growth in
technology based industries. Donckels and Lambrecht (1997) found that small enterprises
that were more central in business networks experienced higher growth. Leavitt (1951)
found that centrality in communication networks was correlated with inﬂuence on
outcomes. Centrality was shown to be an important factor in the level of administrative
innovation in an organization (Ibarra, 1992). Centrality suggests many alternative source
of inforrmtion (Rowley, 1997). Within organizations, higher centrality leads to more
boundary spmning behavior, because those in higher positions perceived more uncertainty

(Seror, 1989).

Together, high levels of density and centrality form a “tight” network structure,
whereanactorhaseasyaccesstoallofthe informationinanetwork. Thiscanbean
advantage when the information sought is of a highly tacit nature, or when information
changes too rapidly to be codiﬁed (Hansen, 1999). A tight network can be a disadvantage
when one is seeking to innovate. Looser structures where network members are weakly

connected to other networks provide access to the ideas not available in one’s immediate

2]

circle of friends and acquaintances.

Winn

Interaction refers to dimensions of solidarity and homogeneity of networks (Adams
& Blieszner, 1994). The type of interaction in a network structure can have an affect on
access to inforrmtion. The emotional intensity, intimacy, and perceived commonalities
shared between network members can affect the ﬂow of resources (Granovetter, 1985;
Marsden, 1990). Berg and Clarke (1986) note that close relationships ﬁrcilitate the

exchange ofa greater variety and higher quality ofresources than those in casual
relationships.

Emmy. Frequency measures the number of times a resource ﬂows between
two network members. The more one has contact with another, the more opportunity
there is to build a close relationship, which may facilitate the exchange of information
among network members (Foa, Converse, Tomblom & Foa, 1993). Frequency of contact
is especially important for information that is complex and changing (Alange, Jacobsson &
Jamehammer, 1998). In small business marketing, data from the immediate environment
is generally considered to be most critical, and is consequently collected on a nrore
ﬁequent basis (Brush, 1992). Van de Ven, Hudson and Schroeder (1984) found that
higher performing managers had more ﬁequent contact with employees, customers and
ﬁnanciers. Aldrich, Rosen and Woodward (1987) found that the frequency of contact

with network members positively inﬂuenced performance in entrepreneurial new ventures.

22

W. Emotional intensity measures the closeness of a relationship.
It can be equated with friendship (Marsden & Campbell, 1984). Indicators of emotional
intensity are the mutual assessment of the level of friendship in a relationship, and the
degree to which the focal individuals spend time together socially (Granovetter, 1973;
Marsden & Campbell, 1984; Schaefer & Olson, 1981). Friendship quality is based in part
on the willingness to spend free time together (Winstead, Derleger, Montgomery &
Pilkington, 1995). Closeness infers self-disclosure, help and support, shared interest,
expression, trust and acceptance (Parks & Floyd, 1996). Emotional intensity was found to

be the best indicator of unobserved tie strength (Marsden & Campbell, 1988).

Studies indicate that people often mix work and friendship (Haythornwaite &
Wellrnan, 1998). Real estate agents studied by Halpern (1996) relied on friendly
relationships in understanding and using information that they obtained in a business
context. Further, the lack of ﬁiendship among participants in the study interfered with
business transactions. Friendly relations between students in an M.B.A. program ind
positive eﬂ”ects on perceptions of team effectiveness and performance. Halpern (1996),
Specht (1987), Dollinger (1985) and others have formd that ﬁiends are an often-used

source of business information among small business owners.

Intimacy. An intimate relationship is one where an individual shares experiences in
several areas, along with an expectation that the experiences and the relationship will
persist over time (Olson, 1975) Intimacy measures the perceived level of mutual conﬁding

present in a relationship. It measures the depth of the exchange, both verbally and

23

nonverbally, between two persons. Intimacy implies commitment and acceptance, and
positively inﬂuences the level of self-disclosure (Gilbert, 1976). More intimate
relationships would be ones where such personal matters as family concerns or political
subjects are likely to be discussed (Marsden & Campbell, 1984). More intimate

relationships are likely to share common friends, similar ideas and interests (Olson, 1975).

WWW. People tend to associate with others who are like
themselves (Blau, 1961). Gilly et aL (1998) refer to perceptual homophily, or the degree
to which network members share values and experiences. Shared values and norms can
facilitate the ﬂow of information in a network and provide better access to information.
Tsai and Ghoshal (1998) demonstrated that social ties and shared vision contribute to the
creation of trust, which in turn increased the ﬂow of resources between business units in a
ﬁrm. The leveloftrustandgoalcongruence betweenindividuals inanetworkcan
determine the “thickness” of information and inﬂuence when information is received
(Borch & Arthur, 1995; Uzzi, 1996). Shared values, norms, interpersonal aﬂiliation and
respect help a ﬁrm cope with complexity and reduce uncertainty (Borch & Arthur, 1995).
Referral information is more likely to be transferred between strong, homophilous ties
(Brown & Reingen, 1987). Institutionally homogeneous networks of private colleges were
more likely to share information about curriculum than less similar groups (Kraatz, 1998).
Consumer behavior studies support the notion that people seek information ﬁom those
that are perceived to be similar to themselves. Feldrnan and Spencer (1965) found that
respondents used perceived similarity of sources rather than perceived expertise when

seeking physician referrals. Gilly et a1 ( 1998) found that perceptual homophily was a

24

stronger predictor than demographic homophily in a consumer behavior context.

Wis:

Frequency, emotional intensity, intimacy and perceptual homophily as a measure of
the level of interaction can be expressed as the strength or weakness of ties among
members (Granovetter, 1973). Strong ties exist where network members have ﬁ'equent
contact with close friends. Weak ties are relationships characterized by less ﬁ'equent
contact and less intimate, more instrumental communication (Ashrmn, Brown & Zwick,

1998).

Both strong and weak ties are vital in the ﬂow of information. Weak ties act as
bridges, permitting information to travel from one network to another (Brown & Reingen,
1987; Weimann, 1983). Weak ties are indicators of non-redundant information (Hansen,
1999). Granovetter’s (1973) seminal work on the strength of weak ties showed that
individuals searching for a job received the most valuable information from infrequent,
distant ties, rather than from strong ties. He expla'md this outcome by reasoning that
close ties are more likely to have the same information as the job searcher, and that
valuable information about new opportunities resides in weaker ties. Burt (1992) found
that top managers’ promotions within high technology organizations were enhanced by
weak, rather than strong ties. In a study of nonproﬁt organizations, those with primarily
weak ties did better in acquiring donations from external sources (Ashman, Brown &
Zwick, 1998). Weak ties with national and international networks were associated with

ﬁrm growth for entrepreneurs in medium sized manufacturing ﬁrms (Donckels &

25

Lambrecht, 1995). Swan and Newell (1995) found a correlations between use of
professional associations (weak ties) and new technology diﬁusion Hansen (1999) found
that weak ties between units of a ﬁrm are sources of new knowledge, but impede the
transfer of complex knowledge. Nelson (1991) found that churches which had more inter-

group weak ties performed better than churches without bridges to other networks.

The entrepreneurial network research tends to view dense, cliquish network
structures, where everyone in the network knows everyone else, as a disadvantage (Burt,
1992; Granovetter, 1985). However, some studies show that close relationships can
facilitate the ﬂow of sensitive, complex and rapidly changing information (Weimann, 1983;
Hansen, 1999). While strong ties may constrain access to new information, these
relationships facilitate transfer of some types of knowledge. In a study of new product
development projects in a large ﬁrm, strong ties produced better task-related outcomes
when the transfer of complex information was involved (Hansen, 1999). Baptista (1999)
found that strong ties among liberal arts colleges increased adoption of professional
programs. Weimann (1983) found that within subgroups within an Israeli kibbutz, gossip,
news and consumer information was communicated faster and more accurately through

strong ties than through weak ties.

Several studies indicate that multiple network structures may be optimal. Nelson
(1991) found that churches which lmd more weak inter-group ties, combined with strong
within group ties were higher performers. Greve (1995) formd that entrepreneurs in later

phases of entrepreneurship had more weak ties than start-up businesses, suggesting that

26

strong ties are beneﬁcial in accessing resources for start up, while weaker ties are

instrumental in acquiring resources needed for growth.

S . l C . 1
Social capital is an intangible asset that resides in the relationships present in

networks. Relationships promote social well-being and provide such rewards as emotional
support and encouragement (Coleman, 1988; Winstead et al., 1995). Close relationships
create trust and obligations and deﬁne expectations and norms among trading partners
(Gulati, 1995). Those individuals who are able to build trust, reciprocity and commitment
through their network relationships have a comparative advantage which leads to richer
and ﬁner grained exchange of information (Burt, 1997; Tsai & Ghoshal, 1998). An actor
in a network has social capital if that person can draw other resources ﬁ'om the network

because of her/his social relationships with network members (Hoﬁ'erth & Iceland, 1998).

Social capital provides the insurance necessary to facilitate transactions in the
marketplace through the presence of trust, reciprocity and commitment. Portes and
Sensenbrenner (1993, p. 1323) conceptualize social capital as “those expectations for
action within a collectivity that affect economic goals and goal-seeking behavior of its
members, even those in the economic sphere.” Expectations include the belief that others
willact irramarmerthat will facilitate actionwithinthe social structure. These
expectations are built upon a common set of values and norms drawn ﬁom an underlying
moral order. Reciprocity and trust enable non-contractual transactions and block

malfeasance and opportunistic behavior. Cormnitrnent in a network is derived from

27

cormnon awareness, collective sentiment and collective self-interest (Coleman, 19880).

In the model I developed for this study, the presence of social capital (SC) is
deﬁned by the level of trustworthiness, reciprocity and commitment perceived in the
network. The more trustworthy the focal actor, and the greater the reciprocity and

commitment perceived to be present among network members, the higher the level of SC.

Imst Being perceived as trustworthy, dependable and sincere by network
members encourages exchange among network members (Doney, Cannon & Mullen,
1998; Hawes Rao & Baker, 1995; Lane & Bachmann, 1996; McAllister, 1995; Tsai &
GhoshaL 1997). Knowledge and information are more likely to be exchanged when parties
are sure about the moral and ethical basis of another’s actions and motivations (Jones &
George, 1998). Tsai and Ghoshal (1998) found that trust between business units in a hrge
ﬁrm positively inﬂuenced information sharing between units. Trust between supervisors
and employees explained a signiﬁcant amount of variation in information sharing between

the two groups (Ramaswami, Srinivasan & Gorton, 1997).

Network interaction inﬂuences the development of trust through the characteristics
of the network member, through experience, and through affect (Jones & George, 1998).
Currall and Judge (1995) found that the longevity of prior work relationships was
associated with increased trust between managers in an organization. Persons who were
perceived to be more trustworthy were more likely to be given information by network

members.

28

Commitmm, Conﬁdence that a partner will cooperate and pursue mutually
compatible interests is derived from collective experiences and common awareness created
through group interaction (Tsai & Ghoshal, 1997). Common awareness is created when a
group is affected by common events or situations. This collective experience can lead to
internal solidarity and collective sentiment that fosters altruistic conduct and mutual
support. Characterized as shared vision by Tsai & Ghoshal (1998), this quality promotes
proper ways of acting in a social system. Coleman (1988) suggests that the expectations
created through social interaction aﬂ‘ect goal-seeking behavior of group members,
including economic ones. Commitment reﬂects the willingness of network members to

help one another by providing support, encouragement and inforrmtion.

Recjpmcity. Reciprocity contributes to SC through self-interested transactions
where network members accumulate favors and other vahrable items that can be called
upon as resources when needed (Portes & Sensenbrenner, 1993). Exchanges across
strong ties are inﬂuenced by previous encounters, which may have created outstanding
debts owed to an individual That individual can “call in’ ” the favors owed in subsequent
exchanges, which may provide better access to tacit knowledge of exchange partners
(Portes & Sensenbrenner, 1993). The desire for social acceptance may motivate actors to

provide valuable resources in return for admiration (Miller & Kean, 1997a).

Marketing transactions are inﬂuenced by feelings of reciprocity. Frenzen and
Davis (1990) found that intention to buy was inﬂuenced by reciprocal sent’mrents of home

party consumers. Miller and Kean (1997a) found that rural consumers were more

29

favorable toward local retailers when they were satisﬁed with the levels of reciprocity in
their community. In a study of manufacturer-dealer networks, the level of information
shared by manufacturers was positively correlated with the volume of other resources
ﬂowing between network members (Usdiken, 1990). Reciprocal norms in a network may
inﬂuence the willingness of network members to share valuable information with others in
the network. Relationships generate trust, reciprocal intentions, and commitment, which
are the foundation of SC. Different network structures and interaction levels produce
varying levels of social capital. Burt (1997) suggests that SC capital can be brokered into
higher returns by facilitating access to information. Higher levels of trust, reciprocity and
commitment in a network enable a person to access information when it is needed. Key
indicators of social capital include the willingness of one’s network contacts to share
information and provide help. Other mdicators include memberships in organizations and
voluntary groups (Bourdieu, 1986), the number of ﬁiendship ties, and the quality of

relationships (Wall et al., 1998).

Social capital has been operationalized in various ways, but generally refers to the
amount of help available in a person’s network of contacts. See Table 1 for a summary of
the use of social capital in previous literature. High levels of social capital have been
related to a variety of outcomes. College attendance by rural residents was positively
associated with high levels of family and community-based social capital, measured in
terms of available family resources, (Smith, Beaulieu & Seraphine, 1995), Charitable

giving was more prevalent in commrmities with higher levels of social capital, measured as

30

residents’ involvement with the community (Weissman, 1998). Coleman (1988) found that
school performance was positively inﬂuenced by Emily social capital derived ﬁom
network relationships among Emily and friends. Friedman and Krackhardt (1997) related
lower returns to college education to lower levels of social capital among Asian
immigrants. Ashrnan, Brown and Zwick (1998) deﬁned social capital as relationships that
provide resources, information and social legitimacy, and showed that higher levels of

social capital led to long-term effectiveness among nonproﬁt organizations.

Social capital may create competitive advantage for a ﬁrm through the exchange of
information among network members (Nahapiet & Ghoshal, 1997). Access to social
capital means that people have connections to individuals who can help with advice,
provide further connections and access to other resources. Interaction with various
environments promotes learning, and increases the likelihood that the entrepreneur will be
confronted with new ideas (McKee, Conant, Varadarajan & Mokwa, 1992).

Entrepreneurs with high levels of social capital are also able to beneﬁt ﬁ'om increased
referral advantages by building a reputation with others in the network (Burt, 1992;

Granovetter, 1985).

Social capital is useful for enhancing learning, economic growth, power and status
for individuals (Bourdieu, 1986; Nahapiet & Ghoshal, 1998). Chung and Gibbons (1997)
suggest that a socio-economic structure that facilitates the emergence of SC is key to the
effective creation and control of entrepreneurial behavior. Participation in networks

provides members with credentials in the form of obligations or institutionally guaranteed

31

rights. Network membership also provides access to resources via contacts and
connectiom. Both strong and weak ties can build source of social capital. The type of
network tie that best facilitates access to information depends upon the type of

information that is being accessed.

lnformaticn

Resource theory posits that people interact and associate with others because they
depend on them for resources. Information is an important resource exchanged between a
ﬁrm and its environment. Information is derived ﬁom data that ﬂows into and out of an
organization in the form of advice, opinions, instructions or enlightenment. A proportion
of information gathering activity involves contact and exchange with networks of
individuals who are linked by some common purpose or interests. Information ﬂows
between points in the structure at different rates and volumes, depending on the nature of
the relationships in the network (Borch & Arthur, 1995; Foa, Converse, Tomblom &
Foa, 1993; Leifer & Delbecq, 1977). Knowledge obtained through social interaction with
network members can lead to new combinations that drive strategic and tactical decisions

for ﬁrms (Nahapiet & Ghoshal, 1998).

Effective interaction between entrepreneurs and the environment is essential to
developing informed decisions. There is a critical need for small business owners to obtain
accurate and timely information about customer preferences and motivations and

competitor activity. Small retailers spend considerable time monitoring the environment

32

for information that will guide their strategic, managerial and technical decision making
(Dollinger 1984; Schafer 1990; Smeltzer & Fann, 1989). The intensity of information
search and number of sources have been shown to have a positive affect on growth in
small ﬁrms (Dollinger, 1984; Jarillo, 1989; Young and Welsh, 1983). Information
asymmetries that result ﬁom differences in the ability of the scanner to access information
contribute to adaptive behaviors and variation in performance outcomes (Weedman,

1992).

The value of information depends upon its accuracy, relevancy, reliability,
speciﬁcity and timeliness (O’Reilly, 1982). Information sources vary in their perceived
ability to provide higher quality information. In situations where the environment is
uncertain and ambiguous, face-to-face information is often considered to be richer because
of its ability to provide irmnediate feedback and multiple cues to interpret corrrplex

subjective messages (Daﬁ & Mginton, 1979).

Information that is more complex is often accessed through personal ties.
Complexity refers to the level of codiﬁcation and the dependent nature of the information.
Information characterized by low codiﬁcation (not expressed in writing) is similar to tacit
knowledge, and is transferred more easily through strong ties (Hansen, 1999; Weimann,
1983). Dependency refers to the degree to which information is interdependent with
another set of information (Winter, 1987; Zander & Kogut, 1995) . Stronger ties help to

interpret dependent information within its relevant context.

33

Information that inspires innovation is often located outside the immediate
environment of the searcher (Alange, Jacobsson & Jarnehanrrner, 1998; Freeman, 1991;
Rogers, 1983). Innovative Indian entrepreneurs networked extensively with external
sources (Ramachandran & Rarnnarayan, 1993). Swan and Newell (1995) reported a link
between boundary spanning activity and technological innovation in professional
associations. Weak ties are more valuable in accessing information that contains new

ideas.

1 l l |° I ll .
Marketing information is considered by small business managers to be the most
important type of information used in business planning (Smeltzer, Fann & Nickolaisen,
1988). For retailers, key marketing information is located both in the local environment
(data about customers, competitors and local economic and regulatory conditions, for
example). Information about new products, processes and technical innovations is found

in the remote environment (Brush, 1992).

Successful implementation of a retail strategy requires access to both local and
remote market information sources. The structural hole argument postulates that because
of optimal network ties, the successful retailer is in a position to bridge the information
gaps between local and remote markets and thereby create competitive advantage. In

other words, social capital yields higher quality information.

34

Wu, Local marketplace information includes

information about consumer demand, competitor moves and local economic and
regulatory conditions (Brush, 1992). Because this information tends to be complex and
ever-changing, strong ties which contain common cultural norms and values are important
in information transfer. When information is localized, those who are closest to the source
can obtain information more cheaply than others (Hansen, 1999). Strong ties,
characterized by frequent contact and close relationships, foster intra-group ﬂow of
information. Weimann (1993) suggests that strong ties facilitate the ﬂow of information
within a close-knit group, such as among Emily, ﬁ'iends and neighbors. High ﬁequency of
interaction, along with the intimacy and emotional intensity found in close relationships,
nukes information ﬂow quickly between network members. Cultivating tacit knowledge

requires an environment of trust, respect and commitment (Durrance, 1998).

The superpreneur study (Frazier, 1999; Frazier & Niehm, 1999) revealed that
high-performing retail “superpreneurs” are adept at extracting information about the local
market from their local network of friends, Emily, business associates, government
oﬁicials and other community residents. Their local networks were built on ﬁiendship,
kinship and geographic proximity, and provided quick access to thicker, richer and less
costly information about the local marketplace than could be obtained through armslength
methods. These “Market Intelligence” networks were useful for tapping into information
that was ever-changing and uncodiﬁed. Superpreneurs were able to assess consumer
demand, evaluate competitive threats, and tailor advertising, customer service and

merchandising strategies based on the information they received from their networks.

35

WW Remote marketplace information includes data
about broader socio-cultural, political and economic trends, as well as technological trends

(Brush, 1992). Hartrmn, Tower and Sebora (1994) found that innovative introductions
were a product of interacting with the external environment. Pioneering entrepreneurs
who emphasized new products, markets and technology were found to be active scanners
of remote information sources (Ramachandran & Ramnarayan, 1993). Sources external to
manufacturing ﬁrms accounted for up to two-thirds of inputs to innovative development
(Conway, 1995). Innovative Norwegian ﬁrms sought information primarily from trusted,
similar ﬁrms (Johamrisson & Dolva, 1995). Ties that reach outside a dense core group into
distant and less frequent contacts, or “Innovative” networks are more important for
radical change (Alange et a1, 1998). Granovetter (1973; 1985) suggests that weak ties
(acquaintances) are crucial in accessing information that is not readily available in the near
environment. The argument here is that close ﬁiends and kin would likely have the same
information as the information seeker, so new information is most likely to reside outside

the network of close ties (Burt, 1992; Granovetter, 1973; Uzzi, 1996).

The superpreneur study identiﬁed unique networks of individuals who provided
information which inspired innovative marketing ideas, and identiﬁed new sources of
merchandise and merchandising techniques. These innovative networks consisted of
retailers, supplier representatives, and other business professionals who acted as bridges to
networks with new information. Retailers in this study emphasized that long-term

relationships resembling “weak ties” gave them to access ideas and information in the

36

remote market (Frazier, I999).

 

Schumpeter (1942) suggests that the essence of entrepreneurship is the ability of
certain individuals to recognize gaps in products, services and distribution before others,
and to respond by creating new combinations which meet the needs in the marketplace.
This activity requires access to information in a time frame that results in competitive
advantage. Kaish and Gilad (1991) found that entrepreneurs exposed themselves to more
infornuuion, looked in kss obvious places, and used different types of information cues
than traditional managers. Christensen, Madsen and Peterson (1986) postulate that
opportunity identiﬁcation is contingent on profound market knowledge. Environmental
scanning research links boundary-spanning information search to market-based learning
and the development of marketing competence (Beal, 2000; Brush, 1992; McKee, Conant,

Varadarajan & Mokwa, 1992; Mohan-Neill, 1995; Peters & Brush, 1996).

Compared to large-scale retail ﬁrms, a small retailer may be at a disadvantage
when it comes to having access to key information sources about new trends, new
products, or new merchandising processes. This disparity may be overcome by creating
network relationships that provide access to key information. Likewise, insuring that
information is received in a timely manner is important in meeting customer demand.
Entrepreneurs also use control of information as a source of advantage. Because there is

always more information available than can be attended to, being able to tap into

37

infomurtion that is relevant to a speciﬁc situation may be a beneﬁt of network ties. For
example, retailers in some rural communities may not have immediate use for information
about a Eshion trend that is popular in urban areas. Close ties with suppliers who have a
sense of what the retailer’s customer prefers can help to limit information to the most

relevant, saving time and other costs.

Wigs} For independent retailers operating in rural markets, it
seems that both strong and weak ties would be valuable in accessing high quality
information from networks. Strong ties within the community would facilitate the ﬂow of
information about market preferences, competitor moves, and local economic and
regulatory conditions. Weak ties, on the other band, would provide access to information
that could be used to spark innovation in merchandising and marketing practices. Previous
network studies suggest that Market Intelligence (MI) networks will contain tighter,

stronger ties than Innovation (INV) networks.

Based on the preceding literature, the ﬁrst set of hypotheses are:

[1,: Market Intelligence Network (MI) structures will be more highly
connected (higher density/higher centrality) than Innovation Network
(INV) structures.

11,: MI Networks will contain stronger ties (higher levels of Emotional
Intensity, Intimacy, Perceptual Homophily, and Frequency of
Interaction) than INV Networks.

H,: For MI Networks, denser, more central and stronger ties will lead to
higher levels of Social Capital (SC).

H4: For MI Networks, higher levels of Social Capital (SC) will lead to
higher Information Quality (IQUAL).

38

11,: For INV Networks, denser, more central and stronger ties will lead to
higher leveb of Social Capital (SC).

H‘: For INV Networks, higher levels of Social Capital will lead to higher
levels of Information Quality (IQUAL).

D . . . l l l . C .
Independent retailers must develop marketing abilities that are highly visible and
valued by consumers to remain competitive in the rural marketplace. These competencies
can be used to build a sustainable competitive advantage by oﬁ‘ering superior product
assortments, better service, and/ or shopping experiences (Conant, Smart & Solo-Mendez,
1993). Regardless of size, superior retailers possess something special and hard to imitate,
which enables them to outperform their competitors by delivering value to their
customers. Every value activity uses and creates information, making information quality a

critical part of marketing competence (Porter & Millar, 1985).

Bharadwaj, Varadarajan & Fahy (1993) note that competitive advantage is
developed and sustained through the mobilization of unique resources and distinctive
skills. Superior skills are those capabilities that set a ﬁrm apart ﬁ'om its competitors.
Conant et aL (1993) operationalized distinctive nmrketing competencies for smaller
retailers. Knowledge of customers, competitors and industry trends, skill in segmenting
markets and the ability to select, price and advertise product litres were identiﬁed as
source-of-skill advantages. Other functional activities considered to be relevant were
awareness of store strengths and weaknesses, developing store image, eﬂ‘ectiveness in

conducting public relations, civic involvement, employee development and control and

39

evaluation of retail programs. Their study found that the most successﬁrl small retailers

were those with clearly deﬁned strategies and the greatest number of distinctive marketing

competencies.

Becausehisuﬂikelythatnmﬂerﬁmrscancompetewﬁhhrgescalechamsusinga
cost leadership strategy, most successful small retailers use a differentiation or niche
strategy. These approaches require a ﬁrm to be unique in some aspect that is important to
consumers (Conant et al., 1993; Porter, 1985). Analysis of the environment is critical in
the development of a differentiation strategy (Neil, 1986). Retailers must be connected to
channels that provide information about product, promotion, service delivery, consumer
demand and competitor activity. They must determine consumer demand and match it to
information they have about available products and trends. Small retailers may gain
competitive advantage by being able to access information and synthesize it in a more

responsive manner than their competitors.

The ability to maintain competitive advantage through distinctive marketing
competencies requires continual adaptation to changing consumer demand. A successful
retail strategy emerges from the process of studying the environment to discover new
opportunities for responding to consumer needs and wants. Retailers interpret consumer
demand by possessing a thorough understanding of the economic, social, demographic,
technological and political trends that impact consumer demand. In order to respond to

clunges, the retailer must constantly scan the environment for innovative products and

40

services that ﬁt the expectations of the market. Resources and skills are prone to
deteriorate over time and must be upgraded (Bharadwaj et al, 1993). Firms can adapt

more quickly than competitors by being better informed about the environment than

competitors.

The nature of linkages to network members affects the learning of marketing skills.
Interaction with various environments promotes learning by increasing the likelihood that
the organization will be confronted with new ideas (McKee, Conant, Varadarajan &
Mokwa, 1992). Using the argument provided by network theory, embedded network ties
can inﬂuence the quality of information. The social capital created by embedded ties can
lead to distinctive marketing competence by providing the information necessary to
develop and maintain these skills. Networking activity has been shown to inﬂuence
marketing competence by providing market access, cost savings, shorter lead times,
technology and process innovation, market feedback and ﬁnancial resources (Larson,

1991).

An entrepreneur’s ability to create and use social capital may lead to easier access
to the information required to develop and maintain marketing competencies. Richer
information about the nature of consumer demand may be accessed by those individuals
who have strong social ties to potential customers and others in the immediate marketing
environment. Advantages relating to innovation may be enjoyed by retailers who have

developed close friendships with suppliers. Retailers may also be able to learn about

41

industry and economic trends through social ties with individuals and ﬁrms that have

access to information.

Levitas, Hitt and Dacin (1997) suggest that knowledge gained from collaboration
with other ﬁrms can lead to innovation. Normann (1985) postulates that environmental
scanning promotes learning by increasing the likelihood that a ﬁrm will be confronted with
new ideas. Research has linked the ability to innovate with increased environmental
scanning (Conway, 1995; Hartman, Tower and Sebora; 1994; McKee, Conant,
Varadarajan & Mokwa, 1992; Rarnachandran & Ramnarayan, 1993). Larson (1991)
found that free and rapid movement of information between exchange partners increased a
ﬁrm’s tacit knowledge, expanded innovative capacity, improved product quality, reduced
costs, and enhanced market competitiveness. This suggests that better information quality

can lead to higher levels of marketing competence.

[1,: For Market Intelligence (MI) Networks, higher Information Quality
will lead to higher Marketing Competence related to Local Markets

(MC/MI).
H,: For Innovative Networks, higher Information Quality will lead to
higher Marketing Competence related to Remote Markets (MC/INV).

Wines

Strategy research is based on the notion that strategy inﬂuences performance
(Lubatkin & Shrieves, 1986). The relationship between information seeking and
performance has been revealed in the literature. Dollinger (1984) showed that intensity of

search was related positively to performance for small retailers. Peters and Brush (1996)

42

found that scanning the environment for information related to competitors and market
share was related to ﬁnancial growth in small ﬁrms. Scanning intensity was also related to
employee growth in new manufacturing ﬁrms (Box, White & Barr, 1993). The use of

professional advisors was related to ﬁnancial success by Lussier (1996).

Marketing competence is associated with better performance (Bharadway,
Varadarajan & Fahy, 1993; Conant, Smart & Mowka, 1993; Snow & Hrebiniak, 1980).
This model suggests that retailers who obtain higher quality information develop better
marketing competence, and thus perform better than less informed retailers. Therefore, I
suggest that higher levels of marketing competence (both Local Market Competence and

Remote Market Competence) will lead to higher growth and proﬁtability.

11,: For MI networks, higher Local Market Marketing Competence
(MC/MI) will lead to higher perceptions of performance (PERF).

Hm: For INV Networks, higher Innovative Marketing Competence
(MC/INV) will lead to higher perceptions of performance PERF.
Network social capital has also been linked to access to other types of resources
which lead to better ﬁnancial performance in small ﬁrms. Network support has been
linked to survival, ﬁrm growth, and overall success of ﬁrms (Duchesneau & Gartner,
1988; Gina & Sexton, 1989; Ostgaard & Birley, 1996). Besides providing valuable
information, networks also provide access to resources such as ﬁnancial capital, emotional

support and change capability. This leads to the ﬁnal two hypotheses:

43

H": For MI Networks, higher SC will lead to higher perceptions of
performance (PERF).

H”: For INV Networks, higher SC leads to higher PERF.

Chapter 3

METHOD
Sample

The population considered in this study consisted of owners and/or primary

managers of small retail stores in smaller communities in the Midwestern U.S. Only
businesses located in smaller communities with populations less than 25,000 were part of
the sample. I controlled for community size because of research that indicates that people
in smaller towns tend to have different networks than their urban counterparts (Babb &
Babb, 1992). As the commercial database did not have a way to identify communities as
ruraLpopulationsizewasusedasameasure. Thesamplewasdrawnrandomlyfroma
commercial database of owners and managers of independently owned retail gift shops
(SIC code 5947) in Midwestern states. A single merchandise category (gift shops) was
sampled in order to control for variations in information search patterns and ﬁnancial
performance by merchandise category. Because the focus of this study was small ﬁrms,
the sample for this study was drawn from the set of retail ﬁrms where the owner was the
primary decision maker. Only ﬁrms with less than twenty employees were included in the

sample. Chain stores and franchise operations were excluded ﬁom the study.

Instnnncnt

A mailed, self-administered, questionnaire was used to measure the constructs in
the model. I developed the preliminary instrument both from existing scales and scales
developed ﬁom a review of the network literature in several domains: social exchange,

network analysis, and business strategy.

45

Insmmmmxelcnnent

The Network Structure, Network Interaction, and Social Capital scales were
previously undeveloped. I ﬁrst generated items for each construct ﬁom a review of the
relevant literature (See Table 2 for details on instrument development.) Preliminary
scales were developed, then presented to three experts familiar with the social network
research. They were given a deﬁnition ofeach construct, along with a random list ofscale
items,andaskedto placetheitemswiththeconstructthatbest ﬁteach item. Iftwo of
threeexpertsagreedonanitem’splacementwithaconstruct, theitemwasretainedas
originally developed. Where less than two experts agrwd on an item’s placement with a
construct, items were either revised, replaced or eliminated. This resulted in a 3-item scale
for Density, a 6-item scale for Centrality, 4-item scale for Emotional Intensity, a 3-item
Intimacy scale, a 4-item Perceptual Homophily Scale, a 3-item Frequency scale, a 4-item
Trust scale, 3-item Commitment scale, and a 3-item Reciprocity scale. (See Table 3 for

item content.)

The Marketing Competence and Information Quality scales were previously
developed, but were adapted for use in the study context. Marketing Competence relates
to the superior skills and capabilities that set a ﬁrm apart from its competitors. Conant,
Smart and Solo-Mendez (1993) operationalized distinctive marketing competencies for
smaller retailers. Knowledge of customers, competitors and industry trends, skill in
segmenting markets and the ability to select, price and advertise product lines were
identiﬁed as source-of-skill advantages. Other functional activities considered to be

relevant were awareness of store strengths and weaknesses, developing store image,

46

eﬁctiveness in conducting public relations, and civic involvement. Content validity of the
scale was addressed by the researchers, however factor analysis and reliability was not

reported in their study.

The scale developed by Conant et al. (1993) was adapted for use in this study.
The original scale consisted of 25 items. I was interested in measuring marketing
competence related to knowledge of customer meds and preferences, competitors, local
rmrket conditions, and adopting new merchandise, marketing ideas and business
techniques. Based on the superpreneur study, I ﬁrst eliminated items that did not appear
to relate directly to competence gained from local or remote rmrket networking. These
items related to employee training, store location, allocation of ﬁnancial resources, sales

forecasting, and control and evaluation of programs.

Content validity for Marketing Competence was assessed ﬁrst through an expert
panel consisting of three researchers familiar with small business and retailing. Experts
were given a deﬁnition of local market and remote market competence, and were asked to
sort competencies from the original scale into one or both of those categories, or indicate
that the item did not belong to either category. As a result of this stage, seven items were
retained which measured local Marketing Competence. Four items remained after revision

for measming innovative Marketing Competence. (See Table 3.)

The Information Quality scale was used with modiﬁcation to the study context.

Experts were given the study deﬁnition of information quality, and asked to rate each

47

question on the following scale: (1) clearly representative of information quality, (2)
somewhat representative of information quality, and (3) not representative of information
quality. All items were rated either (1) or (2) by the expert panel. Wording for those
items rated (2) was revised based on expert comments. This stage led to a 5-item scale
for information quality, measuring the accuracy, relevance, speciﬁcity, reliability and

timeliness of information received ﬁ'orn the network.

Using the scales that resulted ﬁorn the expert panel stage, I pre-tested the
instrument to assess content and construct validity and internal reliability. 1 identiﬁed
potential participants for the pretest ﬁom telephone directory listings of gift stores, and
contacted them by phone to solicit their participation. Those agreeing to participate were
sent a questionnaire. Twelve questionnaires were returned complete after one week. I
interviewed participants alter administering the questionnaire to identify problems with
conrprehension and determine time needed to complete the questionnaire. I also assessed
reliabilities for each scale, which ranged ﬁom .69 to .90, except for the scale measuring
ﬁequency of contact. The alpha for this scale was .47. Because of concerns about
questionnaire length, and lack of evidence in the literature that frequency should be
measured as a latent variable, I decided to measure ﬁequency using a single item in the

ﬁnal questionnaire.

Mm
This study focused on networks that provided business-related information to the

respondent. Respondents were asked to identify two information networks. First, they

48

were asked to identify individuals to whom they talk about local market information
regarding customers, competitors and local market conditions. Respondents were asked
to list the ﬁrst names or initials of all the people to whom they talk about these topics on
the instrument. Ten spaces were provided, with instructions to make additional spaces if
mcessary. These people comprised the local, or market intelligence (MI) network. The
same procedure was used to identify networks for remote marketplace information
(information about new merchandise, new marketing ideas and new business techniques).

This group is identiﬁed in the study as the INV network.

Previous network research used the recall method to identify network members
(Burt, 1987; 1997;1barra,1993;Marsden, 1990; Tsai & Ghoshal, 1997). Individual
perceptions of network structure are found to be eﬂ°ective in predicting attitudes and
opinions. Individuals use these maps to operate in their social environment. These
cognitive maps reﬂect the perceptions of structure in the minds of network members.
Perceived relationships were shown to be more predictive of reputation than actual
structure within an organization (Kilduﬂ‘ & Krackhardt, 1994; Weick & Bougon, 1986). I
focus on the individual entrepreneur’s perceptions in the assessment of network structure

and interaction in this study.

W Structure is a latent variable which is
represented by the conﬁguration of ties among the individuals identiﬁed as network
members. Structure was assessed through measurement of density and centrality.

Density (DEN S) refers to the number of ties that link network members compared to the

49

total possible ties in the network (Marsden, 1993). Based on work by Burt (1987),
Granovetter (1973), Greve (1995), and Marsden (1990). I measured density by asking
respondents three questions that identify the degree to which the people named as network
members interact with each other. Questions were sealed 1 to 5 (not true at all to very

true).

Centrality (CENT) refers to the degree to which one is central or peripheral in the
ﬂow of information relative to others in a network. Based on work by Baldwin, Bedell
and Johnson (1998) and Rowley (1997), I assessed centrality by asking six questions to
determine the degree to which respondents were in a position to call or talk to the
network members they named directly. Responses ranged from 1 (strongly disagree) to 5

(strongly agree).

W Interaction is a latent variable which is
represented by the closeness of a set of network relationships. The ﬁ'equency of
interaction as well as the emotional intensity, intimacy, and perceived commonalities
shared between network members are the observed variables that deﬁne levels of

interaction in the network. Frequency was measured as a single item.

Emotional intensity (EMOT) measures the closeness of a relationship. It can be
equated with ﬁ-iendship (Marsden & Campbell, 1984). Statements regarding the
closeness of the relationship were assessed using a ﬁve point scale (1=not true at all to

5=very true).

50

Intimacy (INT) measures the perceived level of mutual conﬁding present in a
relationship. Three questions were developed based on discussions by Marsden and
Campbell (1984), Schaefer and Olson (1981) and Parks and Floyd (1996), and Franzen
and Nakamoto (1993), that measured the degree to which respondents felt they would
discuss private topics such as family matters and politics. These questions assessed the
likelihood that the respondent would conﬁde in mmed network members. A ﬁve point
scale (l=very unlikely to 5=very likely) was used, with higher scores indicating more

intimate relations with named network members.

Perceptual homophily (PERHOM) measures the degree to which respondents
believe that network members are similar to themselves in shared outlook on life. Four

statements were rated on a ﬁve point scale (l=strongly disagree to 5= strongly agree).

W This latent variable consists of three dimensions: the
respondent’s self-perceptions of their own trustworthiness among network members,
respondent’s assessment of the level of reciprocal intentions among identiﬁed network
members, and respondent’s assessment of the level of commitment among identiﬁed

network members.

Trust (TRST) is the expectation by one person that another will act in an ethically
justiﬁable manner (Smeltzer, 1996). This construct was measured by respondent
perceptions of their reputation with respect to dependability, sincerity and trustworthiness

among named network members. Four questions measured agreement with statements

51

about the perceived trust placed in the respondent by the named network members.

Reciprocity (RECIP) deals with the respondent’s assessment of the level of
support, accumulation of favors owed and the fairness perceived to be present in
relationships (Franzen & Davis, 1990; Miller & Kean, 1997). Three questions measuring
this variable were used to assess the perceived level of reciprocity between named

network members.

Commitment (COM) is the third variable comprising social capital, and measures
the level of conﬁdence that a partner will cooperate and pursue mutually compatible
interests. It includes the degree to which respondents believe tint network members share
the same goals and visions, and their assessment of the vigor with which the network
supports the respondent, and the amount of mutual help that is given in the network. The
level of commitment present in the network was assessed via three questions which rated
the respondents perception of network commitment. TRST, COM and RECIP were

measured on a I (strongly disagree) to 5 (strongly agree) scale.

W

I used a scale developed by 0’ Reilly (1982) to measure the quality of information
received ﬁom both immediate and remote marketplace sources. Information quality
measures the accuracy, relevancy, reliability, speciﬁcity and timeliness of information. The
scale was originally developed to assess quality and accessibility of information ﬂour a

variety of formal and informal sources, including personal sources. The ﬁnal scale

52

included ﬁve questions on quality, measured using a seven point scale (l=not relevant at
all, to 7 = very relevant; l=not reliable at all, to 7=very reliable, etc). Cronbach alpha

was .89 in the original study. The wording was adapted for use in the current study.

WWW

Marketing Competence was measured in the context of the local market
(MC/MI), which addressed skills in responding to and communicating with customers.
Innovative marketing competence (MC/INV) dealt with assessing the ability to be ﬁrst to
identify new trends and try new business techniques. The scales masured responses using
a 7 point scale indicating how competent the respondent felt they were compared to the

top three competitors (1= not as strong, 7=much stronger).

E. . l E E [BEE E
I used subjective measures of growth, proﬁtability and overall performance
compared to industry and competitors to measure performance. Subjective assessments of
performance are generally consistent with secondary performance measures (V enkatraman

& Ramanujarn, 1986). Respondents were asked to indicate a), their assessment of the
ﬁrm’s overall performance, b), their assessment of the ﬁrm’s performance compared to
industry, and 0), compared to competitor performance, on a 5 point scale of 1 equaling

“poor” to 5 equaling “excellen ”.

53

Procedure

Iobtainedalistofnarnesandaddressesofonethousandsmallretailgiﬂ store
owners ﬁom a commercial database. Aﬁer eliminating duplicate listings and pretest
participants, questionnaires were mailed to 987 participants, along with a letter explaining
the study. An addressed, stamped reply envelope was included with the questionnaire
and letter. F ollow-up reminder/thank you postcards were mailed two weeks after the ﬁrst
mailing. As a result of the reminder postcard, nineteen participants requested that another
questiomraire be sent. Thirty-eight questionnaires were returned as undeliverable. Please

see the Appendix for the ﬁnal questionnaire, cover letter, and follow-up postcard

One hundred twelve completed questionnaires were returned, for a response rate
of 12.1%. Several reasons may exist for the low response rate. Response rates are
typically problematic when sampling small businesses (Conant & White, 1999). Authors
have cited diﬂiculties in contacting the appropriate respondent, lack of time, survey
“burnout”, and concerns about conﬁdentiality (Wmter, Fitzgerald, Heck, Haynes and
Danes, 1998). The response rate achieved in this study is in the same range as those
achieved by other studies where small retailers are the participants (e. g. Conant & White,
1999 - 13.1%; Ganesan & Weitz, 1996 - 13.8%; Robinson, Logan & Salem, 1986,

10.1%).

Questionnaire length and the nature of the questions may have contributed to the
low response rate. In the pretest, respondents noted that the questionnaire took about 30

minutes to complete, but that they were often interrupted and completed the questionnaire

54

over the span of several hours or even days. Several also stated that the questions
required a considerable amount of thought to answer. I received several replies ﬁom
retailers who indicated that they were going out of business, or that their businesses were

too small to be relevant to the study.

S l D . .
The sample consisted of 104 owners and 7 managers (one respondent did not
indicate status) of small gift stores in small towns in Michigan, Ohio, Indiana, Illinois, and
Wisconsin. Seventy-one percent were female, and twenty-six percent were male.
Respondents tended to be older, with over eighty percent over 40 years old. Three-

fourths were college educated (see Table 4.)

Nearly half(45.1%) of the respondents had owned their current business for over
ten years, and over two-thirds had more than ten years experience in retailing. Firms were
quite small, with forty-two percent reporting that they had no full-time employees. (See

Table 4 for sample descriptives.)

55

Chapter 4

RESULTS

This chapter summarizes the process used to test hypotheses and the results of
those tests. The study generated responses to questions about the respondent’s
relationships with two groups of people. First, respondents were asked to identify people
ﬁom whom they received information about local market conditions, then evaluate aspects
of their relationships with the people they identiﬁed. These data pertain to the MI (Market
Intelligence) group. The second set of data related to people who were identiﬁed as
sources of information about new trends, ideas and merchandise. These data pertain to
the INV (Innovative) group. The goalinthe ﬁnalanalysis wasto identifyamodelthat ﬁt
both sets of data. I will ﬁrst explain the exploratory and conﬁrmatory factor analyses used

to develop ﬁnal model, then discuss testing of the hypotheses.

I used SPSS 7.0 to conduct exploratory factor analyses and assess reliability. I
used EQS 5.1 to test hypotheses related to the measurement and structural models.
Maximum likelihood procedure was used to estimate model parameters, as MLE estinutes
have shown to be quite robust to violation of normality (McCallum, 1995, p.38). I
assessed model ﬁt using several methods. First, I looked for a small, nonsigniﬁcant 12
statistic, which measures the absolute magnitude of the discrepancy between the sample
and the ﬁtted covariance matrices. For each model analyzed, I also reviewed the
standardized residual matrix, looking for large residuals as evidence of poor model ﬁt.

Lagrange Multiplier (LM) tests and Wald tests, provided by the EQS program, were used

56

to identify misspeciﬁcations in the model. The LM tests model restrictions to identify
parameters that would contribute to a signiﬁcant drop in x2 if they were to be ﬁeely
estimated. Wald tests identify parameters that could be set to zero, without loss of model
ﬁt (Byme, 1994). Model modiﬁcations were not made, however, unless a substantive

argument could be given to do so.

Because x2 may not perform well under conditions of small sample size and
nonnormal distribution, both of which characterize these data, I also assessed ﬁt with
incremental (NNFI) and comparative (CFI) indexes provided by EQS, which measure the
degree of congruence between the model and the data. These indexes adjust for
nonnormality of the data. A value of .90 or greater was considered acceptable ﬁt of the

data to the model.

WW1:

Prior to testing the hypotheses, I conducted exploratory factor analyses on items
for the hypothesized constructs. Because of the small sample size (112), the goal in this
stage of the analysis was to identify scale items that would lead to the most parsimonious
measurement model possrble. I was looking for scale items that loaded satisfactorily and
uniquely onto a priori deﬁned factors in both the MI and INV scales. For each set of
responses, I retained items in the analyses that (a), did not crossload onto other factors,

and (b), loaded greater than .50 on the hypothesized factor, for both sets of data.

57

 

measuring ﬁrm performance, local and innovative marketing competence and information
quality were factor analyzed using varimax rotation. Items which loaded greater than .50

on the hypothesized factors were retained for timber analysis (see Table 5).

WW I retained items mwsming Density.
Centrality and Perceptual Homophily that met the criteria speciﬁed above. This resulted in
a three-item Density scale, an four-item Centrality scale, and a three-item Perceptual
Homophily Scale. Factor analysis revealed that item measuring EMOT and INT loaded
on one factor, rather than two, in both groups (See Table 5). I re-conceptualized the
construct as Friendship (FRND), consisting of item v40, v41, v42, and v43. This is
consistent with the conceptualization of friendship offered by Olson (1975), and Parks and
Floyd (1996), who deﬁne friendship in terms of willingness to self-disclose. I dropped the
variable which measured ﬁ'equency of interaction as a single indicator, as it loaded on
several other factors in both sets of data. The social network literature suggests that

ﬁequency of contact is a measure of network interaction, but these results suggest that it

is not a unique concept.

W RECIP and COM loaded on a single factor in the MI network
responses, while TRST and RECIP loaded together in the INV responses (see Table 5).
Since there was not a common pattern between the network responses, and these factors
appeamdtobediﬁhﬂﬁomomanothermtheﬁterMmeJretamedmeseﬂneescalesas

separate factors.

58

Results ﬁom this step suggest, however, that in this sample, trust, reciprocity and
commitment are not unique constructs. It may be that social capital in strong-tie networks
is not the same as social capital in weak tie networks. Trust and reciprocity appear to be a
single factor in weak-tie networks, while commitment and reciprocity behave as a single
factor in strong-tie networks.

Results of factor analyses, and Cronbach alphas for each scale are displayed in
Table 5. Cronbach alphas ranged from .71 to .95, exceeding the .70 threshold
recommended by Nrmnally (1978). The measmernent model is shown in Figure 2. Item

content is given in Table 3.

W

Using the items identiﬁed in exploratory factor analysis, I tested validity of the
NETSTRUCT, NETINTER and SC constructs for both MI and INV data using
conﬁrmatory factor analysis. This was completed in two steps. In the ﬁrst step, I tested the
hypothesized relationships of ﬁrst order factors, then tested their loadings on to the
hypothesized second-order factors (NETSTRUCT, NETINTER and SC; see Figure 2). In
the second step, I created composite scores that transformed the ﬁrst order factors into
observed variables by averaging the scores for each scale. As a result, DENS, CENT
became observed variables for the latent variable NETSTRUCT. FRND and PERHOM
became observed variables for the latent variable NETCHAR. TRST, COM and RECIP
were averaged to create observed variables for SC. This step was taken because the small

sample size in this study required that I simplify the model to decrease the parameter-to-

59

subject ratio.

Wu

W A ﬁrst-order conﬁrmatory factor
analysis for DENS, CENT, FRND, and PERHOM revealed two variables which
crossloaded onto more than one factor in both the NH and INV models (v40 loaded on
both CENT and FRND; v46 loaded on PERHOM and FRND). Since the goal of this step
was to create composite scores for the ﬁrst-order factors, these variables were dropped

from the model to create a model where all variables loaded cleanly onto a single factor.

Results of the ﬁrst order CFA for M] network data after eliminating these variables
were (x1=78.527, (If 50, F106, p=.006; NNFI=.924, CFI=.942). For INV data the results
were (f=80.928, df 50, n=101, p = .002; NNFI==.935, CFI=.953). A review of the
measurement equations found that all he parameters were signiﬁcant at the .05 level for
both MI and INV models (see Tables 6 and 7). LM tests indicated that v30, which asked
whether the people in the network knew each other by name and was hypothesized to load
on DENS, also loaded signiﬁcantly on FRND for both groups. Although substantively,
the argument could be made to add this parameter, subsequent tests indicated that adding
it would not improve model ﬁt substantially, and would confound the ﬁnal model;

therefore, no modiﬁcation was made.

WWW Three-item DENS (v30, v31, v32) and four-item
CENT scales (v33, v36, v37, v3 8) were hypothesized to load on second-order factor

60

NETSTRUCT, and three-item FRND (v41, v42, v43) and two-item PERHOM scales
(v47, v49) were hypothesized to load on NETINTER. To simplify the model, I re-
conceptualized NETSTRUCT and NETINTER as one second-order factor, Network
Characteristics (NETCHAR). This adjustment is consistent with some scholars’
conceptualization of network dimensions. Granovetter (1985) and Marsden and Campbell
(1984) presented network characteristics without delineating between structural and
interaction components. The single latent factor ﬁnding suggests that the structure
features of a network (density and centrality) are not distinct ﬁom the kinds of
relationships between network members. Network studies oﬁen focus on either structure
or interaction. These results suggest that structural and relational characteristics cannot

be viewed separately when deﬁning network characteristics.

The revised model (Figure 3) produced acceptable ﬁt for both MI (x’=61.548, df
50, n=106; p < .126; NNFI=.969, CFI=.977). For the INV data, results were: xz=SS.591,
df 50, n=101; p < .240; NNFI=.987, CFI=.991). Parameter estimates are shown in Tables
8 and 9. These results suggest that network density, centrality, ﬁ'iendship and shared

values are explained by a common, second order factor, NETCHAR.

Estimates of the reliability and variance extracted measures for each construct
were computed to assess whether the speciﬁed indicators were suﬂicient in their
representation of the constructs (Hair, Anderson, Tatham and Black, 1995). Formulas
used to calculate these estimates may be found in the Appendix. An examination of these

factors reveal that reliability is above the recommended .70 level for DENS, CENT, and

61

FRND in both data sets, and near the acceptable level for PERHOM. The variance
extracted was near or above the recommended level of .50 (Hair et al, 1995) for all
variables in both groups except for PERHOM in the MI model. The low variance
extracted for PERHOM indicates that shared values, beliefs and outlook on life do not
explain a substantial portion of the variance in NETCHAR (See Table 11.) The low
composite reliability for PERHOM is an indication tint more indicators may be needed for
this factor. These results suggest that further develOpment of the scale measuring this

construct is warranted.

S . l C . l
W I then proceeded to test the relationships of TRST,
COM and RECIP, and their loadings on a second-order factor conceptualized as social
capital (SC). Exploratory factor analysis had indicated that RECIP and TRST loaded on
a single factor for INV networks, and COM and RECIP loaded on a single factor for MI
networks (See Table 5). Because these three constructs appear to be distinct in the
literature, and for practical reasons, three indicators of SC were required in later analyses,
I retained three constructs as originally proposed. I then proceeded to the ﬁrst-order
CFA, using a four-item TRST scale (v20, v21, \QZ, v23), a three item COM scale (v24,
v25, v26), and a three-item RECIP scale (v27, ra8, v29). Multivariate LM tests in EQS
indicated that v22, measuring TRST and v27 measuring RECIP loaded on multiple

factors in both data sets, and were subsequently dropped from the model.

62

Conﬁrmatory factor analyses ﬁtting MI and INV data to the trimmed model
produced acceptable ﬁt for MI (xi-4 32.710, dﬂ7, n=106; p=.01227; NNFI=.921,
CFI=.952). When ﬁt to the model, INV data provided a less than optimal ﬁt (11 =43.059,
df17, n=103, p <.001; NNFI=.873, CFI=.923). Parameter estimates were all signiﬁcant at
the .05 level, and factors were signiﬁcantly correlated, as expected.‘ See Tables 10 and

11 for parameter estirmtes.

SEW I conducted a second-order conﬁrmatory factor
analysis which hypothesized TRST, COM and RECIP as ﬁrst order factors explained by

the second order factor Social Capital (SC). See Figure 4. This analysis revealed results
for the MI network data as (177—44806, (if 18, n=106; p < .012; NNFI=.873, CFI=.918),
and for INV data (x’=43.526, df 18, n=103; p < .001; NNFI=.923, CFI=.953). Parameter
estimates for the second-order analysis are shown in Tables 12 and 13.2 The patterns of
loadings for each group were signiﬁcant and positive. TRST loaded less strongly in the
MI (1318 (.591) than in the INV data (.816), While COM loaded more strongly for MI

networks (.988) than for INV networks (.738).

Composite reliabilities found in Tables 12 and 13 for the constructs are near or

 

In the MI model, COM and RECIP had a correlation greater than 1.0. Bollen (1995)
suggests that when correlations are out of, but near admissible range, it may be due to the
factors actually being highly correlated in the population, which is expected in this case.
Other sources of this result may be small sample size, presence of outliers or model
misspeciﬁcation.

2

To account for the high correlation between F6 and F7, D6 and D7 were constrained to

be equal.

63

above the .70 threshold, indicating that the factors are reliable estimators. Variance

extracted is strong for RECIP, and moderate for TRST and COM.

H 1.1.

.- u. . if. an": mt. a n. .u.-__r.‘ '. ,u. - - l .- .u 10.0 .‘r- ,\ err. . .-

To test H1 and H2, which stated that network structure and interaction would be
different between local networks and remote networks, I used a multi-group, structured
means approach to determine whether the means of the latent variables DENS , CENT,
FRND and PERHOM, were signiﬁcantly different for the MI and INV groups. I was
interested in knowing whether these constructs were similar in MI and INV networks,
thus shedding light on the differences between clmracteristics in each type of network. In
other words, what makes a network that is used to gather local infornmtion different from
one which is used for information about new ideas, trends and business techniques? Since
NETSTRUCT and NETINTER became the single second-order factor NETCHAR in the

ﬁctor development process, H1 and H2 are tested in a single analysis.

Wis. In EQS, answering this question involves creating a
constant variable, which has a variance ﬁxed to zero. This restructures the dependent
factors so that their residuals manifest the variance and covariance information for that
variable. The two groups are then compared, with the factor intercepts in one group ﬁxed
to zero; this group then operates as a reference group against which latent means for the

other group are compared. Loadings are constrained to be equal across groups; the LM

(,4

test then tests statistically for the validity of the constraints (Byrne, 1994). To determine
whether the latent construct means are signiﬁcantly different across groups, the factor
intercepts representing latent mean values in the non-reference group are examined for
statistical signiﬁcance. Signiﬁcance in this study would indicate that latent mean

structures for MI and INV network characteristics are different across groups.

I used the baseline model representing the ﬁnal ﬁrst-order construct model for
DENS, CENT, FRND and PERHOM to test the hypothesis that the means of these latent
variables would be signiﬁcantly different for MI and INV networks. INV was designated
as the reference group; therefore the factor intercepts were ﬁxed to zero in this group.
The intercepts of measured variables were set to be equal across groups. LM tests in the
initial analysis indicated that releasing the constraint holding v36 equal for both groups
(“If I needed advice about running my business, I could call them on the telephone”),
would substantially improve model ﬁt. This suggests that the respondent may not feel that
he/she could call people in remote information networks as easily as those in local
inforrmtion networks. Because it seemed reasonable that this measure might not be the
same for both groups, I released the constraint. After releasing this constraint, good
model ﬁt was achieved (x2=148.3, df 107, n=101; p = .005; NNFI=.957, CFI=.965). All
estimates relating to the factor loadings and variable intercepts were signiﬁcant for both

swaps-

Turning to the hypotheses that the means of the latent constructs DENS, CENT,

FRND, and PERHOM would differ across MI and INV networks, I examined the factor

65

 

intercepts that represent the latent mean values. Results indicate that signiﬁcant
differences in the latent means for CENT (mean difference= .231; z=2.747, p < .05), and
FRND (mean difference = .394; z=2.475, p < .05) were signiﬁcantly higher in MI than in
INV networks. No signiﬁcant difference was found between groups for DENS (z=1.487,

p=.l4) or PERHOM (z=1.685, p=< ..09). See Table 14 for results.

These results suggest that differences exist between MI and INV networks with
respect to the degree to which an individual is centrally located in the network, and in the
strength of the ﬁiendships between the individual and identiﬁed network members,
partially supporting H, and H2. Respondents in the study were more centrally located,
and had stronger ﬁiendships with their local networks than with remote networks. No
differences were found in the density of the two networks, or in the degree to which
respondents perceived themselves to share values and beliefs with network members. This
suggests that both strong-tie and weak-tie information networks are conﬁgurations of
personal relationships where everybody knows everybody, and whom the information
seeker perceives as similar to himself. This does not support Granovetter’s (1973)

evidence that stronger ties are found among people who are similar to one another.

W
C . [C . S E I C |
To test the causal hypotheses 3 through 12, I created composite scores for the
factors relating to NETCHAR and SC. Scores ﬁ'om the varra' bles retamed' in the second

order CFA’s for DENS, CENT, FRND, PERHOM, TRST, COM and RECIP were used

66

to create average scores by averaging respondents’ responses for the items in the ﬁnal
measurement models. These scores then became the values of observed variables loading

on NETCHAR and SC, as previously hypothesized. (See Figure 5.)

 

Using composite scores for DENS, CENT, FRND, PERHOM, TRST, COM and
RECIP, I then conducted CFA’s for both sets of data for all variables in the structural
model. Results for MI data were (x2=l79.324, df 96, n=104; p < .001; NNFI=.880,
CFI=.904). Wald tests indicated that covariances between several of the factors could be
dropped without loss ofmodelﬁt intheMImodel. INVdataproducedabetterﬁtting
model (f=131.48l, df96, n=100; p = .009; NNFI=.945, CFI=.956). Tables 15 and 16
give parameter estimates for the conﬁrmatory factor analysis. LM tests for the MI data
identiﬁed additional signiﬁcant paths between PERHOM and SC and between COM and

IQUAL.

W

WWW Forthisstageoftheanalysis. scales
measuring DENS, CENT, FRND, PERHOM, TRST, COM and RECIP in each data set
were averaged to obtain a single score for each construct. The data for Marketing
Intelligence (MI) and Innovative (INV) networks were then ﬁt to the ﬁnal structural
model (see Figure 5). When I ﬁt the MI network data to the structural model, I found less
than acceptable ﬁt (x’=208.425, df 99, n=104; p < .001; NNFI=.847, CFI=.874). As

indicated by the conﬁrmatory factor analysis, LM tests for the model indicated that the

67

lack of ﬁt could be attributed to nonsigniﬁcant paths between performance (PERF) and
social capital (SC), and between Marketing Competence in local networks (MC/Ml) and
Information Quality (IQUAL). Results fail to support the contention of some literature
that social capital accrued in information networks has a direct inﬂuence on ﬁrm

performance.

Multivariate LM tests indicated that, in addition to the hypothesized loadings, a
relationship existed between commitment (COM) and IQUAL, and that perceptual
homophily (PERHOM) loaded on SC. Supported by literature that suggests that
commitment may be closely related to information sharing in dense networks, (Ashman,
Brown & Zwick, 1998; Coleman, 1988), I modiﬁed the structural model by adding a path
ﬁ'om COM to IQUAL. This provided a small improvement in 1’ (190.292), and modest
improvement in ﬁt indexes (NNFI=.870; CFI=.894). Although shared values have been
proposed to contribute to the generation of trust, commitment and feelings of reciprocity

(Coleman, 1988), a test of the PERHOM -> SC relationship did not contribute

signiﬁcantly to model ﬁt.

WWW For the Innovative network (INV) data,
statistics were jar-132.196, df 99, n=100; p < .001; NNFI=.950, CFI=.959. An
examination of parameter estimates showed that all variances were signiﬁcant, except for
one error variance (v13) relating to Marketing Competence (MC/INV). Hypothesized
variable loadings were all signiﬁcant. Multivariate LM tests suggested additions of a path

ﬁom Commitment to Information Quality (COM ->IQUAL), but a test of this

68

modiﬁcation did not contribute substantially to model ﬁt, and was not incorporated into

the model.

 

LocallnfonnaﬁanemerLRelaticnshins. For Marketing Intelligence (MI)
networks, results of the structural equation analysis revealed a positive and signiﬁcant path

between Network Characteristics (NETCHAR) and Social Capital (SC), supporting H3 ,
and between SC and Information Quality (IQUAL), supporting IL . The path between
IQUAL and Marketing Competence (MC/MI) and between SC and Performance (PERF)
were not signiﬁcant, therefore H7 and H,0 were not supported. A positive and signiﬁcant
relationship between MC/MI and PERF did exist in this model, providing support for H9 .
See Table 17 for parameter estimates. These results suggest that, for local information
(MI) networks, network characteristics positively inﬂuence the level of social capital
present in a network. Social capital in local information networks contributes to the
quality of information received ﬁom the network but does not inﬂuence the development
of MC, failing to support the argument that intellectual capital is a direct result of
information acquisition. Marketing competence does, however, exert a positive inﬂuence

on the performance of the ﬁrm.

For the INV data, hypothesized

 

regression paths from NETCHAR to SC (H, ), SC to IQUAL (H,5 ), IQUAL to MC/INV
(H, ), and MC/INV to PERF (H, ,) were all positive and signiﬁcant, supporting these

hypotheses. (See Table 18). The hypothesized relationship between SC and PERF was

69

not signiﬁcant, therefore H,2 was not supported. In innovative information networks, this
model suggests that network characteristics positively inﬂuence the level of social capital,
which leads to higher quality information. Information quality has a positive impact on the

perceived ability to irmovate, which leads to better ﬁnancial performance for the ﬁrm.

._ u - a An examination of the standardized

 

residual variances provides ﬁuther insight into the relationships hypothesized in this
model. By obtaining the coeﬁcient of determination from the standardized residual
variance provided by EQS output, an estimate of the amount of variance in the dependent
variable that can be explained by the independent variable can be calculated (Bentler,
1993). In both networks, a substantial portion of the variance in the latent variable
representing Social Capital (SC) was explained by the latent variable representing network
characteristics (NETCHAR). The estimate of SC accounted for a relatively srmll portion
of the variance in the latent variable measuring Information Quality (IQUAL) for both
networks, suggesting that other variables not speciﬁed in this model are explaining a larger
share of the variation in information quality. Although causally signiﬁcant, the estimate of
IQUAL explained less than ten percent of the variance in Marketing Competence in
Innovative networks (MC/INV). The estimate of MC/INV explained about 25% of the
variation in INV performance (PERF), and just under 20% of performance in MI

networks. (See Table 19.)

70

SumnnnLnfResults
Overall, results of the analysis suggest that the hypothesized structural model

represents a good ﬁt when applied to Innovative information (INV) networks. It supports
the relationship between an individual’s social ties, and the level of social capital present in
those relationships. In innovative information networks, social capital provides access to
richer information ﬁ'om network members, which is used to build innovative marketing
competency. Overall ﬁrm performance was related to perceptions of competence.
Performance, however, was not inﬂuenced by the social capital available from the

network.

For local information networks, the model also explained the inﬂuence of networks
on social capital, and social capital on information quality. There was no signiﬁcant
relationship between the quality of information available in local networks and perceptions
of competency relating to meeting customer needs, although marketing competence in this
area did inﬂuence overall ﬁrm performance signiﬁcantly. As with the innovative network,
the model did not support a relationship between social capital and overall ﬁrm
performance in local networks. A summary of the results of hypothesis testing are given

in Table 20.

71

CHAPTER 5

DISCUSSION AND CONCLUSION

The purpose of this study was to elaborate on the inﬂuence of personal
relationships on the ability of retail entrepreneurs in small communities to acquire and use
business information. Speciﬁcally, I focused on the inﬂuence of the relational qualities
between the entrepreneurial information seeker and the people who make up their business
information networks. I drew upon the perspective suggested by social network theory to
argue that characteristics of an entrepreneur’s network relationships can produce social
capital that yields access to information available from its members. Following the
argument suggested by Granovetter (1973) and Burt (1997), I theorized that the relational
characteristics of networks which provide local business information to the retailer would
be different than those of networks where the retailer accessed new information. I then
proposed that information accessed through these networks can build speciﬁc types of
marketing competencies related to assessing consumer demand, delivering high quality
customer service, and the ability to innovate. Finally, I suggested that nmrketing

competence in these areas would inﬂuence the overall ﬁnancial performance of the ﬁrm

I conceptualized business information networks as being made up of people ﬁ'om
whom the entrepreneur obtains information pertinent to operating their business. I
focused on two types of information networks: networks which provide information

about the local environment, and networks which access new infornmtion that can be used

72

to create innovation. These networks have both structural and relational characteristics
that reﬂect the interconnectedness, centrality, ﬁiendship ties, and perceptions of shared
values and beliefs of its members. Following the arguments set forth by Coleman (1988),
Bourdieu (1986), and Putnam (1993), the strength of these network ties predicts the
amount of social capital that is available to an information seeker. Social capital, which is
deﬁned by the amount of trust, commitment and reciprocal intentions present in network
relationships, enables the seeker to access reliable, speciﬁc and relevant information about
business conditions and events. The quaﬁty of information is seen as a predictor of the
information seeker’s perceptions of competence in assessing consumer demand, providing
quality service, and creating innovative marketing programs. Perceptions of marketing
competence in these and other areas has been shown to be a predictor of the overall

success of retail ﬁrms (Conant et al, 1993).

I proposed a latent factor model linking network characteristics, social capital,
information quality, marketing competence and ﬁrm performance. I used exploratory and
conﬁrmatory factors analysis and structural equation modeling techniques to test the

validity of the constructs and the relationships between factors.

The sample in this study was drawn from the population of individuals operating
small gift stores in small towns in Midwestern states. Data were collected using a self-
administered, written survey. The sample consisted of 112 owners and managers of small
ﬁrms. Most were college-educated, and had been in business over ten years. Nearly three

fourths of respondents were female. Firms were quite small, with forty-two percent

73

employing no full-time employees.

W

This study took a unique approach to the measurement of networks by adapting
network mapping teclmiques to survey research. This approach allowed me to measure
the network characteristics of multiple actors engaged in similar pursuits, and assess the
inﬂuence of different network characteristics on various outcomes. This study was also
unique in applying social network concepts to information search in that it developed
multiple indicators to describe the constructs in question so that theory relating the

constructs could be tested simultaneously.

Conﬁrmatory factor analysis indicated that density, centrality, ﬁ'iendship and
perceptual homophily represented a common latent factor representing an actor’s
business information network characteristics in both remote and local information
networks. This supports the basic contention of social network theory which
conceptualizes personal networks as having both structural and rehtioml characteristics
(Burt, 1992; Granovetter, 1985; Greve, 1995). I was not able to distinguish structural

network characteristics from relational characteristics, as originally proposed.

A substantial portion of the social network argument rests on the idea that peOple
belong to multiple networks, some of which contain strong ties, where network
relationships are “tight”, whereas others contain weak ties, characterized by “looser”

network structures, and less intimate relationships (Granovetter, 1973). Brown and

74

Reingen (1987), Burt (1992), Granovetter (1973), Hansen (1999), Nelson (1991),
Weimann (1983) and others suggest that different networks are used to access different
types of information. I was able to provide partial support for this idea. I showed that
local networks, which I deﬁned as a set of relationships between the information seeker
(respondent) and those who provided information about local market conditions, differed
ﬁom remote networks, which included pe0ple who provided information about new
business ideas. These differences pertained to the centrality of the information seeker in
the network, and the strength of ﬁ-iendships between network members. Local
information was obtained ﬁ-om network conﬁgurations in which the information seeker
was more centrally located, and where friendship ties were stronger, when compared to
networks where new business ideas were found. This suggests that strong ties are used
to obtain information about customer demand, competitor activity and local economic
conditions, while weaker ties were employed to access information about new
merchandise, trends, marketing ideas and business techniques. This supports Weimarm
(1983), who found that consumer information, local news and gossip ﬂowed more
eﬂiciently through strong ties. These ﬁndings also support Swan and Newell (1995), who
found a correlation between the use of weak ties and new technology diffusion, and
Hansen (1999) who found that weak ties between units ofa ﬁrm are sources ofnew
knowledge. Local and remote networks in this study did not differ signiﬁcantly based on
their density, or in the degree to which the focal member perceived the network members

to share common beliefs and values.

75

These results also support the qualitative study ﬁndings that preceded this research
(Frazier, 1999; Frazier and Niehm, 1999). In the study of highly successful
“superpreneurs”, we found that small retailers occupied central positions in strong tie
networks which provided proprietary data and tacit knowledge about local market
conditions. Strong tie networks included ﬁiends and relatives, customers, other local
retailers, employees and community residents. Retailers were less centrally located in
weak-tie networks which served as a sources of knowledge which generated new ideas
and innovation. People in these networks included people outside of the immediate
community, who shared business interests with the superpreneur and acted as bridges to

other networks.

Socialﬁanital

Drawing from various perspectives in the literature (Bourdieu, 1986; Burt, 1997;
Coleman, 1990; Loury, 1961; Portes and Sensenbrenner, 1993), social capital was
conceptualized as a latent construct, represented by the observed variables of trust,
commitment and reciprocal intentions among network members. Results of conﬁrmatory
factor analyses suggest that social capital in the context of business information networks
may be less complex than I have portrayed it in this study. Exploratory factor analysis and
high correlations in conﬁrmatory factor analyses between these constructs indicate a need
for further investigation of the underlying structure of social capital. In a review of the
origins and development of the concept, Wall, Ferrazzi and Schryer (1998) describe social
capital as an “elastic” concept, which varies depending on the perspective and scale of

analysis used to operationalize the term. After completing proposed tests, I tested various

76

conceptualizations of the SC construct, and did not ﬁnd that better ﬁt could be obtained

by conceptualizing SC as a two factor or single factor model.

 

I was able to provide support for the idea that networks produce social capital.
The signiﬁcant paths between the latent variables which represent network characteristics
and social capital in both networks suggest that strong networks contain more social
capital than weak networks. This ﬁnding provides empirical support for the basic
argument set forth by Coleman (1985) and others, which says that social capital formation
is facilitated by close social networks. These results support Gulati’s (1995) ﬁnding that
close relationships between ﬁrms built trust, deﬁned expectations, and created obligations
in alliance networks. It also mirrors Tsai and Ghoshal’s (1998) ﬁndings that social
interaction had a positive effect on trustworthiness in the resource networks that existed

between business units in a large ﬁrm. My conclusions should be tempered by the fact that

the latent construct SC needs to be reﬁned.

 

I found moderate support for the contention that social capital plays a role in
accessing information resources in both market intelligence and innovative business
infornmtion networks. As indicated by the regression paths between SC and IQUAL for
MI networks (.61), and the same path in INV networks (.46), the level of social capital
present in business information networks inﬂuences the richness of information obtained

from those networks. These results add support to the notion that information is shared

77

more ﬁeely in relationships that are characterized by high levels of social capital (Burt,
1997; Frenzen & Davis, 1990; Portes & Sensenbrenner, 1993). Speciﬁcally, information
ﬂows between network members at different rates and volrunes, depending on levels of
trust, commitment and reciprocity. This supports Tsai and Ghoshal’s (1998) results,
which showed that trustworthiness was positively associated with resource exchange

among business units.

I was not able to show a direct link between social capital and performance. There
appear to be no additional beneﬁts to performance created by the relationships in these
networks, as measured by this model. Although social capital has been connected to
performance through its ability to build reputation and access ﬁnancial capital, the

networks identiﬁed in this study did not supply additional beneﬁts.

Past literature on entrepreneurial information search identiﬁes a preference for
personal sources of information among small business owners (Arbuthnot, Slarna and
Sisler, 1993; Brush, 1992; Srneltzer, Fann and Nikolaisen, 1988; Specht, 1987). These
authors have found that the information gained ﬁ'om personal sources is more accessible,
relevant and reliable than information from non-personal sources. The results in this study
elaborate and extend the ﬁndings of previous studies in that they focus on the role of
embedded social relationships on the ability to access worthwhile business information.
My results suggest that entrepreneurs are able to tap into valuable business information by
creating social capital in their business relationships. Small retailers who are adept at

cultivating social relationships may have an advantage when it comes to getting the

78

information they need to make business decisions.

 

I pr0posed that the richer the information found in each of the information
networks, the more competent the entrepreneur. The argument is that a key way to
achieve competitive advantage in the retail environment is through the development of
distinctive marketing competencies. I suggested that access to certain information was
essential to building strong competencies in such activities as gauging customer demand,
providing quality customer service and offering new and distinctive merchandise. I drew
ﬁ'om research on environmental scanning, which links seaming behavior to learning and
development of marketing strategies. Scanning the task and general environment allows a
ﬁrm to increase intellectual capital regarding environmental opportunities and threats that
impact its survival (Beal, 2000). Nahapiet and Ghoshal (1998) hypothesize that new
intellectual capital is created through access to parties which enable the combination and
exchange of existing information. As an externally oriented competency, market-based

learning results in the ftmdamental bases of competitive advantage (Sinkula, 1994).

Results indicated limited support for this argument in the innovative infornntion
networks, as indicated by the regression path between IQUAL and MC (.305), and no
signiﬁcant relationship between these constructs in the local networks. This suggests that
other variables are explaining the variation in marketing competence in this sample.
Although the presence of social capital inﬂuenced the quality of information received from

one’s network, it does not necessarily mean that the information is used to build skills in

79

marketing. Competence may be due more to experience, education or cognitive ability
(Alder, 1992; Jo, Hyungrac Lee, 1996; Sinkula, 1994; Stuart & Abetti, 1990). A
univariate test of data from this sample showed that signiﬁcant differences in perceived
marketing competence existed between education levels. No differences were indicated
when respondents were grouped by years of experience, however. Cognitive ability was

not a variable captured in this study.

Sinkula (1994) also notes that in order for market information processing to
translate into organizational learning, the proper supply of tmequivocal, timely information
must be present. In other words, the information that is obtained from the network may
not be suﬂicient to develop competence, or it may not be used in a timely manner.

Sufﬁciency and timeliness were not measured in the ﬁnal model.

The lack of relationship between IQUAL and MC in MI networks may also
suggest that there are few “structural holes” in local networks. Even though the
information received from the local networks is rich, it my be redundant, and therefore
not useful in creating marketing skills. This supports Burt’s (1992) contention that higher
returns are available to well-comiected players only when they provide access to the

information gaps in the marketplace.

The weak link between IQUAL and MC may indicate a need to develop skills in
environmental scanning. Providing srmll retailers with guidance on how to use the

information available to them via networks within a framework such as SWOT analysis,

80

may focus their information gathering activities in a more productive manner.

 

The link between both local marketing competence and innovative marketing
competence and perceptions of ﬁrm performance was signiﬁcant. The path between
MC/MI -> PERF (.40) indicated a moderate inﬂuence of local marketing competence on
the respondent’s perception of the performance of the ﬁrm. The relationship was slightly
stronger for the MC/INV->PERF path (.49). These ﬁndings support Conant et a1 (1993),
who found that retailers with higher souree-of-advantage skills in a variety of marketing
competencies performed better. The relatively weak relationships indicate that, as would
be expected, other variables not included in this model explain a larger proportion of the
variation in perceptions of performance. Variation in respondents’ perceptions of their
marketing skills relating to customer demand assessment and customer service explained
just under twanty percent of the variance in performance; innovation skills explained about
twenty ﬁve percent. Day and Wensley (1988) contend that superior skills and resources
are not automatically converted into performance payoffs. This conversion is mediated by
strategic choices, ﬁrm objectives, entry timing and the quality of tactics and
implementation. Results of this study, however, suggest that distinctive skills in staying in
touch with customers and a focus on continuous innovation contribute to ﬁrm success in
small retail ﬁrms (Mintzberg, 1978). How information accessed through networks is
transformed into marketing competence that leads to better performance rennins

unexplained.

81

I . . .
Results should be examined in light of the limitations imposed by study
characteristics.
ThesmallsamplesizethatcharaeterizedthisstudyhmitsthcabﬂityofMEproceduresto
detect differences among the data. As a result of the small number of cases, I was limited

in the degree of complexity that I could introduce into the model in this study.

This study focused only on business information networks for speciﬁc types of
information, thus, it would not be appropriate to extend these ﬁndings beyond those
network deﬁnitions. As I focused on a single sector of the retail industry with respect to
product lines (gift retailers) and geographic location (small towns), these results cannot be
generalized to other types of small retailers, or to retailers operating in larger
communities. Small business owners in smaller communities may have very diﬁ‘erent
network structures than their counterparts in urban areas. Implications may also be
limited only to retail ﬁrms, as patterns of information search may be unique to the retail

enviromnent.

Measurement model results show that further reﬁnement of measures that I
developed for this study for network characteristics and social capital are needed. I also
adapted measures for information quality, marketing competence and performance to this
research setting, and these should be replicated and reﬁned. The method I used to deﬁne
networks in a survey research setting was also previously untested and needs further
investigation.

82

W

Using techniques that allow comparison of social relationship patterns among
entrepreneurs can generate interesting questions, but more research is needed to reﬁne
measures that will allow investigators to use this approach In addition to reﬁning the
measures and methods in this study relating to information networks, identifying other
types of resource networks and their beneﬁts would be a valuable line of inquiry. For
example, networksthat oﬂ'erpersonal support mayalso beinstrumentalinﬁrmsuccessby
providing the emotional support necessary to sustain entrepreneurial activity, especially in

the start-up stages.

Although comparison of MI and INV models was only hypothesized for network
characteristics, the results of CFA’s measuring social capital raise interesting questiom
about whether dimensions of this construct are universal to different types of networks.
Themodelsuggestedinthe second orderCFAexplaining socialcapital did notﬁtthelocal
network data well, suggesting that an alternative model would better explain the patterns

of social resources present in strong-tie networks.

The link between access to information and being able to use it to become more
competent in important marketing skills also needs further investigation. Experience,
education, cognitive ability and motivation may contribute to the transformation of market
information into skill. The inﬂuence of other types of marketing skills on performance,

such as competence in tactical areas such as pricing and advertising would be interesting

83

avenues of research

E . l I l' .
The very existence of small retailers is threatened by profound changes in the retail

sector. The insights ﬁom this study suggest that entrepreneurs can beneﬁt ﬁ'om using
their information networks. Small retailers can capitalize on their unique network
positions in local networks to gather input for strategic and tactical decisions. Interacting
with these networks can provide information on the wants and needs of customers, and

provide timely data on economic and competitor activity.

Local business development training programs can focus on the beneﬁts of
community involvement. In the superpreneur study, (Frazier, 1999; Frazier & Niehm,
1999) we found that exceptional retailers were very involved in community activities. One
retailer emphasized the importance of being involved in her community. She was active in
a church group and the school parent organization, as well as in business-related groups.
Although she explained that her primary motivation in belonging to these groups was not
business-related, she admitted that she gained a great deal of insight about the local

economy through these interactions.

Maintaining connections with pe0ple who can provide new ideas is also important
tothesurvivalofsmallretailers. Eveninruralareas, consumershaveeasyaccessto new
products via travel, catalogues and the Internet. Independent retailers must be

competitive with the trends offered by their large-scale competition. Superpreneurs used

84

their “weak tie” connections with knowledgeable people outside of their close networks
as catalysts for innovation. National retail trade organizations can foster these
relationships by providing venues for small retailers to interact with other professionals
and share ideas. When asked what type of business support he desired, one superpreneur
said that he would like nothing better than to go somewhere and “just talk with other

retailers” in his industry “for about three days”.

Most entrepreneurial training programs focus on ﬁnancial, legal and marketing
aspects of business ownership. Little attention is paid to developing networking skills.
Business training in retail entrepreneurship should also emphasize the importance of
building relationships, and suggest ways of using the information received ﬁom these

connections to improve marketing skills.

Ccnclusicn

In response to calls for alternative explanations of entrepreneurial success (Aldrich
and Zimmer, 1986; Tsjvold & Weicker, 1993), I used social network theory as a
Work for asking questions about the inﬂuence of social relationships on small ﬁrm
performance. The very existence of small retailers are threatened by the profound changes
in the retail sector. The “strong tie/weak tie” argument parallels the axiom “It’s not what
you know, it’s who you know”. My model has suggested that “who you know detennines
what you know”, and provides a platform for further inquiry into the inﬂuence of

networking on entrepreneurial success.

85

APPENDICES

86

APPENDIX A

Survey Instrument

87

October 14, 1999
Dear Independent Retailer,

Would you like to know wlnt nukes some retailers more succmﬁrl than
others? We are working on a research project at Michigan State University to try to
answer this question You are one of a small number of business owners and managers
that are being asked to provide information and opinions about the way they operate their
businesses.

We know that small retailers like you use unique strategies to be successful. This
study is concerned with the inﬂuence of personal relationships on business practices. Your
opinions and attitudes are instrumental in increasing understanding of this important topic.
So that we can obtain accurate and consistent information, the enclosed questionnaire
should be completed by the We: for this business.

You may be assured of complete conﬁdentiality. All of your answers will be
reported together so that you cannot be identiﬁed in any way. The responses you provide
are completely anonymous and can never be linked to you. The questionnaire has an
identiﬁcation number for mailing purposes only. This is so we can check your name off
the list when the questionnaire is returned Your name will never be placed on the
questionnaire. You may provide all or part of the information.

The questionnaire will take about 30 minutes to complete. When you have
completed the questionnaire, please follow instructions on the back page for returning by
mail. So that we may be sure to include your responses in the study, please return the
questionnaire no later than November 1, 1999.

Thank you in advance for agreeing to participate in this study. The results will be
made available to researchers and business professionals interested in the success of
independent retailers. If you would like a copy of the results, please check the box at the
end of the questionnaire. You may contact the researchers listed on the questionnaire if
you have any questions about this study.

Sincerely,

Barbara Frazier
Doctoral Student

Patricia Huddleston
Associate Professor

88

 

Small Business Survey
Michigan State University

This questionnaire should be completed by the W (the owner or
manager who makes the major decisions for the store).

     
     
 
   

BYOURﬁmmuchsﬂongerﬂnmaholt N“. mm M"-

thesaue,ornotasstrongasyonrtop3 mus g... m..."

competitors when it com to: II «Indec- than a:

err-petition competition

1. Assessment of current customers’ 1 2 3 4 S 6 7
needs and wants.

2. Assessment of prospective l 2 3 4 5 6 7
customers’ needs and wants.

4. Quality of customer service. I 2 3 4 5 6 7

5. Ability to offer competitive prices. 1 2 3 4 5 6 7

6. Creating a pleasant shopping 1 2 3 4 5 6 7
atmosphere

7. Effectiveness of store advertising. 1 2 3 4 5 6 7

8. Effectiveness of store layout and 1 2 3 4 5 6 7
merchandise presentation.

9. Ability to differentiate l 2 3 4 5 6 7
merchandise and service oﬂ'erings
from that of competitors.

10. Being ﬁrst to introduce new 1 2 3 4 5 6 7
merchandise and merchandise
lines.

ll. Introducing new ideas in my I 2 3 4 5 6 7
business.

12. Trying new marketing techniques. I 2 3 4 5 6 7

89

in this section, we are going to ask you about people who give you information that helps you make business
decisions. The ﬁrst part of this survey concerns w_ho you get information ﬁ'om about the following areas:

 

.5

Local competition
local market conditions

 

. ‘ ‘ '
.“,4\L«.v-,* t.’ . . . .. , --.., v,,v,~u “.1“. 'aq-I" -. r \

‘1

Your customers needs and preferences

 

 

Please think for a moment about the mp]: you talk to when you need information or advice about the above
areas. These might include (but are not limited to) business or professional people, family, ﬁ'iends, neighbors,
comrmmity residents, government ofﬁcials anyone who gives you useful information and advice about the

above areas.

 

 

 

 

Feel ﬂee to add additional spaces.

 

i. ." i i . . i . i
l. 4. 7. 10.

5. 8. ll.

3. 6. 9. 12.

Inthespaceprovidedbelow,plenaewﬁtetheFlRSfNAMESORmmotantheWthtyonca-thhkof
whoyoutnntoforadvleeandinformatlouabouttheaboveareas. Wearegoingtoaskywsomequestionsabomthis
groupofpeopleinthe next section. Make additional lines if necessary. The list isforrecallpm'posesonly. W

 

 

 

 

 

Very

3 True

1. These people know each other by name. 5

2. These people talk to each other about business. 5

3. These people see each other regularly in business 5

situations.

4. My relationships with these people are very close. 5

S. I do things socially with these people. 5

6. [fl had the chance. I would spend a free afternoon 5
with any of these people.

7. I consider most ofthese people my friends. 5

8. l otien share business information with these people. 5

90

Not true
Neutral at all
3 2 l
3 2 l
3 2 l
3 2 l
3 2 l
3 2 l
3 2 l
3 2 I

 

 

 

muagebeatthepeoplethetyoallstedlatheboeu "

 

 

 

 

 

 

 

 

 

 

 

page 2, please indicate how Holy or unlikely it h thet: . Very Somewhat Somewhat Very
' .- Likely Likely Neutral Unlikely Unlikely
9. You would share personal matters with them. 5 4 3 2 1
l0. You might discuss family matters with them. 5 4 3 2 l
l I. You might ask them for advice about a private matter. 5 4 3 2 l
"i. 5’0»? {.1
Ia per-al. eonparhgyo-uﬂtothe people yoa lust! la l
rtlslulxﬁowsinners-wieldyiilaesyyvaaesetothesepeopk Very Somewhat Somewhat Very
viii raped”: , 1 , z. ,2 Similar Similar Neutral Dissimilar Dissimilar
12. Your outlook on life. S 4 3 2 1
[3. Your likes and dislikes. 5 4 3 2 I
I4. Your business philosophy. 5 4 3 2 l
15. Your values and beliefs. 5 4 3 2 l
Disagree Strongly
Neutral Somewhat Disagree
17. I am considered to be dependable by these people. 5 4 3 2 1
18. These people would say that I am sincere. 5 4 3 2 I
19. These people would trust me with personal 5 4 3 2 1
information about themselves.
20. I am satisﬁed with the level of business support I get 5 4 3 2 1
from them.
2l. 'I'heywouldsaythatlamatrustworthyperson. 5 4 3 2 l
22. We do each other favors from time to time. 5 4 3 2 l
23. In general, they are fair in their business dealings 5 4 3 2 l
with me.
24. These people share the same ambitions and visions 5 4 3 2 1
about business that I do.
25. They are enthusiastic about helping me in my 5 4 3 2 1
business.
26. I talk directly with these people about business issues. 5 4 3 2 l
27. If any of them had information that would help me in 5 4 3 2 1
my business. they could tell me directly.
28. Among these people. I often pass along business 5 4 3 2 l

information from one person to another.

91

Strongly Agree Neutral Disagree Strongly

Agree Somewhat Somewhat Disagree

29. lf I needed advice about running my business, I could 5 4 3 2 I
call any of these people on the telephone.

30. They support me in my business. 5 4 3 2 l

3 I. If these people had business information that would be 5 4 3 2 l
helpful, they would tell me right away.

32. I am one ofthe ﬁrst to hear about new things from 5 4 3 2 I
this group of people.

33. I frequently talk to these people about business topics. 5 4 3 2 I

34. I would do a favor for any ofthese people ifthey 5 4 3 2 I
asked.

35. These people would be willing to do me a favor if I 5 4 3 2 l
asked.

 

 

,rm. {r'i’r’

The next set of questions are asking about the information you receive from the people ion identiﬁed

.r-_

2,93 the first page. Please circle the number that best represents your opinions and feelings about the if}

. l
. , w j ‘g
.. n§1ewakjww "13‘7“ grvrs-x vi 61.:3 WIF‘IA” ﬁzznnrwfnnv 7;“va 90v“: 'v-rs'.’ _...,,,’..._.,.',._ ie‘ri‘; Fir-13‘»

 

 

l. Wheeadnghformedoaﬁomthepeopkyoaumedabovehmvaecmuwoaﬂyunyhnﬂyb?

 

NOT 1 2 3 4 5 6 7 VERY
ACCURATE ACCURATE
AT ALL

 

 

2. So-edmutheiaformeduwegetmeygetrightmtheheanofthepmble-weenbdag. Otherd-estheiaformtioa
meyaotbeveryspeciflctooareeeds. legenenLhowrelevantistheieformatioeﬁo-thepeopleyoana-edabove?

 

 

NOT 1 2 3 4 5 6 7 VERY
RELEVANT RELEVANT
AT ALL

 

 

3. Attimeswemastgetherelotofinformetioawhichisa’tveryrelevaatinordertogetcaoaghtomelteagooddecision.
Othertinesweaeedoalyasmallamountofinformetloabecaasetheinformatlonisveryspeeiﬁcandallowsnsto
meloadecisioa. Howspeciﬁcistheinformetioeyoagetfromthepeopleyouaamedebove?

 

 

NOT 1 2 3 4 5 6 7 VERY
SPECIFIC SPECIFIC
AT ALL

 

 

4. Some information may be exactly whet we require. How often is this the case for information obtained from the people
you named above?

 

NOT l 2 3 4 5 6 7 VERY
OFTEN OFTEN

 

 

5. To be useful, information must often be available when we need it. not at some later time. How timely would you
estimate information to be from the people you named above?

 

NOT l 2 3 4 5 6 7 VERY
VERY TIMELY
TIMELY

 

92

 

 

The second section of this survey is asking about who gives you information and advice about the following areas:

 

. , .: .,,.w --,.37 o. "a 2:73 .'-~--.*~-
.New merchandise ~.’ -I;-'*
‘ _ New marketing ideas :’ :

" a . ' ,» . . . . ._ ‘ e ,. _ ,
, I N o- ‘ - a ' " ., . , .w"

 

 

 

Now think about the people you talk to when you need information or advice about the areas listed above.
These also might include (but are not limited to) family, ﬁiends, neighbors, community residents, other business
or professional people, government oﬁcials and others- anyone who gives you useful informtion and advice
about the above areas.

 

lnthespaceprovidedbelow,pleasewritethel'lRSTNAMESORlNTﬂAISofalltheiadivldaelsthet
yoacaethiakofwhoyoutnrntoforedviceandiaformationabouttheaboveereae. Wewillbeasking
thesamequestionsabwtthis groupaswedid forthe last group. Again, the Iistisforrecallpurposesonly,
andwe ‘II | . . l' |

 

l. 4. 7. IO.

2. 5. 8. l l.

3. 6 9. 12.
Feel free to add additional spaces.

 

 

 

 

 

 

Not true
Neutral at all
I. These people know each other by name. 5 4 3 2 l
2. These people talk to each other about business. 5 4 3 2 l
3. These people see each other regularly in business 5 4 3 2 I
situations.
4. My relationships with these people are very close. 5 4 3 2 l
S. I do things socially with these people. 5 4 3 2 I
6. If I had the chance. I would spend a free afternoon 5 4 3 .. I
with any of these people.
7. I consider these people my friends. 5 4 3 2 I
8. I oﬂen share business information with these people. 5 4 3 2 I

93

 

 

 

 

“MWMMw-ﬂhﬂykhm Very Somewhat Somewhat Very

- ' ' . . ' ' , ' Likely Likely Neutral Unlikely Unlikely
9. You would share personal matters with them. 5 4 3 2 I
10. You might discuss family matters with them. 5 4 3 2 I
ll. You might ask them for advice about a private matter. 5 4 3 2 I
12. You will receive business information 5 4 3 2 l

ﬁom them in the next week.

hmleuparhgynarselfhthepeepleyoaae-ediei
'MSECONDheohowd-ilerweddyoasayyoaereb ,2 Very Somewhat Somewhat Very

 

  

 

 

 

 

 

 

,these'peoplewithr-pecttot - , ,, ,_ 3‘ Similar Similar Neutral Dissimilar Dissimilar
13. Your outlook on life. 5 4 3 2 I
I4. Your likes and dislikes. 5 4 3 2 I
15. Your business philosophy. 5 4 3 2 I
16. Your values and beliefs. 5 4 3 2 l
Strongly Agree Disagree Strongly
Agree Somewhat Neutral Somewhat Disagree
1?. I am considered to be dependable by these people. 5 4 3 2 1
l8. These people would say that I am sincere. 5 4 3 2 I
19. These people would trust me with personal 5 4 3 2 I
information about themselves.
20. I am satisﬁed with the level of business support I get 5 4 3 2 I
from them.
2l. Theywouldsaythatlamatrustworthyperson. S 4 3 2 l
22. We do each other favors from time to time. 5 4 3 2 l
23. In general, they are fair in their business dealings 5 4 3 2 l
with me.
24. These people share the same ambitions and visions 5 4 3 2 I
about business that I do.
25. They are enthusiastic about helping me in my 5 4 3 2 1
business.
26. I talk directly with these people about business issues. 5 4 3 2 l
27. Ifany ofthem had information that would help me in 5 4 3 2 I

my business. they could tell me directly.

IN)

28. Among these people, I oﬂen pass along business 5 4 3
information from one person to another.

94

Strongly Agree Disagree Strongly
Agree Somewhat Neutral Somewhat Disagree

 

29. If I needed advice about running my business, I could 5 4 3 2 I
call any of these people on the telephone.

30. I frequently talk to these people about business topics. 5 4 3 2 l

3 I. If these people had business information that would be 5 4 3 2 1
helpful, they would tell me right away.

32. I am one of the ﬁrst to hear about new things li'om 5 4 3 2 I
this group of people.

33. I ﬁequently talk to these people about business topics. 5 4 3 2 l

34. I would do a favor for any of these people if they 5 4 3 2 I
asked.

35. These people would be willing to do me a favor ifl 5 4 3 2 I
asked.

 

   
 

n... at...» “ mam. w mm... mm * " ~' ammonium. m
SECTIONZ. Please circle the nurnber that bed represensyour opinions and feelings about the following

15"“

  

l. Whandnghformeduﬁ'omthepeophyonnmedebovehoweeemtewuldyonaynmuﬂyh?

NOT l 2 3 4 5 6 7 VERY
ACCURATE _ ACCURATE
AT ALL

 

2. Attimeswernnstgatherelotofinformetionwhichisn’tveryrelevantieordertogctenonghtomaloagooddecisioa.
Otherti-esweneedoalyesmellemonntofinformationbecansetheinformetionisvcryspeciﬂcandallows-to
maloedec'uion. Howspeeineietheinformationyongetfrolnthepeopleyonnemedabove?

 

 

NOT I 2 3 4 5 6 7 VERY
SPECIFIC SPECIFIC
AT ALL

 

3. Sonetimestheinformetionwereceivemaygetright totheheartofthe problemweere facing. Otherthnesthe
informationmeynotbeveryspeciﬂctoonrneeds. legenerehhowrelevantlstheinformetionfro-thepeopleyon

 

 

named above?
NOT I 2 3 4 5 6 7 VERY
RELEVANT RELEVANT
AT ALL

 

4. Some information may be exactly what we require. How often is this the case for information obtained from the
people you named above?

NOT l 2 3 4 S 6 7 VERY
OFTEN OFTEN

 

 

 

5. To be useful. information must often be available when we need it, not at some later time. How timely would you
estimate information to be from the people you named above?

 

 

NOT I 2 3 4 5 6 7 VERY
VERY TIMELY
TIMELY

95

 

 

 

 

 

How would you describe the overall performance of your store(s) last year?
Poor Average Excellent
I 2 3 4 5
How would you describe your performance relative to your major competitors?
Poor Average Excellent
I 2 3 4 5
Howwonldyou describe your performance relative tootberstorea likeyours in the industry?
Poor Average Excellent
1 2 3 4 5
In 1998, did your store...? (Circle one)
Make a proﬁt Break even Lose money

 

What is your title? _Owner Manager _Other (Please specify)

How many people do you employ full time (besides yourself)? __
How many people do you employ part time ?_

What is your age? _
What is your gender? (circle one) Male Female

How many years have you owned or managed this business? __ years _months
How many years of experience in retailing do you have? _ years _months

How long have you owned a business in this community? years months

Please indicate the highest level of education completed
_Some high school _Some College
_High school _College

___Post-graduate

Thank you very much for your time. Please place the questionnaire in the enclosed. postage-paid return envelope and mail no
later than November I. 1999. Questions may be directed to Barbara Frazier at (616) 387-37I9. or Dr. Patricia Huddleston at
(517) 353-9907.

If you would like a copy of the results. please send a postcard to: Barbara Frazier

Print "COPY OF RESULTS REQUESTED" on the card. Michigan State University
204 Human Ecology Building
East Lansing, MI 48824

96

Follow-up Postcard

   

Two weeks ago, a questionnaire seeking your opinions about i
business relationships was mailed to you. Your name was drawn l
in a random sample of retail store owners in the Midwest. i

If you have already completed and returned it to us please accept
an sincere thanks. If not, please do so today. Because it has been ,
sent to only a limited number of small retailers, we need your input. It is extremely important to us i
that your opinions be included in the study i
if the results are to accurately represent the opinions of independent
retailers.

If by some chance you did not receive the questionnaire, or it
got misplaced, please call me at (616) 387-3719 and I will get
another one in the mail to you today.

Sincerely,
Barbara Frazier

Project Director
‘ Michigan State UniversityI ___l

 

97

APPENDIX B

Tables

98

 

 

Author(s)

Deﬁnition

Context

 

Jacobs,
(1 965)

Networks of cm ss-cutting personal
relationships developed over time provide the
basis for trust, cooperation and collective
action

Neighborhood network
structures

 

Loury,
(1977)

The set of resources inherent in family
relations and in community social
organizations

Child development

 

Bourdieu
(1 986)

SC is one of a number of separate though
related, forms of capital. The creation and
efﬁcacy ofSC depends on membership in a
social group whose members establish group
boundaries through the exchange of symbols

or things.

Italian manuﬁicturing
ﬁrms

 

(1993)

SC is the networks, relations and obligations
existing in social situations. It is a product of
interaction. SC includes trust, norms and
networks which enhance cooperative action.
Sc includes the assurance that altruistic
actions will be rewarded in the future
(generalized reciprocity). SC is a resource of
a network; it plays a role in outcomes of
other qualities. Indicators of SC are:
memberships in organizations/voluntary
groups, number of ﬁiendship ties, offers of
help, quality of relationships.

Immigration

 

 

Coleman,
(1988)

 

SC is a resource that can be used to achieve
goals. SC exists in structures with
reciprocity, expectations, norms, values and
trust. Individuals with high levels of SC have
more obligations outstanding. Norms foster
collectivity where members forego self-
interest and act in the interest of the group.
SC is useful in providing information that
facilitates action.

 

Theoretical
development; used SC
to explain differences in
individuals’ chances to
improve their human
capital by staying in
school

 

99

 

IablaJJQntinucd

 

 

 

 

 

 

Bates, SC is present in the form of a captive market, Asian immigrant
(1994) and derives from culturally based tastes that entrepreneurs

can only be served by co—ethnic businesses
Smith, Within the context of the community, social Families in the context
Beaulieu & capital exists in the norms, social networks of college attendance of
Seraphine and interaction between members. It is youth in a rural
(1995) represented by genuine concern or interest community.

that adult members have about another

person’s child. Signs of its presence include:

enforcement of norms, monitoring activities

of other peeple’s children, oﬂ‘ering programs

for youth. SC presence is determined by the

structure and process of social relations in

the family and in the community.
Burt SCisaqualitycreatedbetweenpeople. SC Managersinan
(1997) predicts that returns to intelligence, electronics ﬁrm

education and seniority depend in some part

on a person’s location in the social structures

of a market or hierarchy. The network that

ﬁlters information also directs, concentrates

and legitimates information
Chung & The concept of social capital refers to the Inna-organizational
Gibbons, value that certain aspects of social structure entrepreneurship
(1997) have for actors as resources that can be used

to achieve their ends.
Ashman, SC is found in the form of social relationships Nonproﬁt organizations
Brown & within and between diverse social groups. It and fund—raising eﬂ‘orts
Zwick is a resource developed by maintaining
(1998) relationships with people and organizations.

 

Sc provides social legitimacy and social
cooperation among and between
organizations.

 

 

100

 

W

 

 

 

 

 

 

Flora SC enhances the beneﬁts of investments in Rural community
(1998) physical and human capital. SC thrives when economic development
individuals interact in a social system in
multiple roles over a period of time.
Hoﬁ‘erth & Contact, exchange among network members Urban and rural
Iceland indicate the existence of SC. Provision of populations
(1998) help or assistance can reﬂect either
reciprocation or investment in new social
ties.
Nahapiet & SC is a multidimensional construct. Theoretical
Ghoshal Networks of relationships constitute a development of social
(1998) valuable resource for the conduct of social capital and
affairs, providing members with collectively organizational
owned capital, which entitles them to credit. development
It is embedded in network relationships.
Summers A global construct that is intended to include Rural economic
& Brown several dimensions development
(1998)
Wall, Sc is subject to a variety of interpretations Review of social capital
Ferrazzi & reﬂecting different trends and perspectives. literature in education,
Schryer Concept is found in sociology, economics economics and
(1998) and education. Coleman’s ﬁ'amework is sociology.

 

predominant in education. Sociologists use
Bourdieu. Access to SC means that people
have connections to individuals who possess
greater amounts of economic and cultural
capital, might help with advice, further
connections, loans and so forth.

 

 

lOl

 

 

SEVERE

 

wand—deco

9350253825“

 

2312* 32:58:2—

 

g I 238

map—ewoom e». e aim—n 239.3% £33.. Eu area 8 35»
Bee? ire cardamom 8 m3 @599: sage—.5 E :5 Emma—grow.
€38: E3553 @2258 8 883. 83. 9.3.8308 8 RE
mammog— eg 9.”. :68 RE: e Raccoon age—u. e». 9.8 8
an 885.

 

we: :23

233:” AFB W08 Omen—s. mcoE meadow. 39:5

233% mom—Ga 3. 355m mam area—mum 39. 9o Fm wen Becca.
ﬁre m8 9o 38% «39 £68 we: Emoemmom Benn; 562.85.. 8
vases 22:51” 38 gm egg on ma 9:... poem—o Ea.

 

HE .v Graver Coed

0836.5— Eﬁm «$95 men—8 mu:
”.mgﬁnoonﬁagomgmoaoa "game—583555»
Ram—o 0838:“ 93. $63 Bron 8 335 earn. (an. SE2.
:58 «72.8 88:88 Ema—.938? p895. 383% BE
83.3

 

 

we: 2 mod

 

moaen Ema e». ea $85.38 8533.
2295:» 8:808 52:an 33.5: E6 «0388—. 9.93on q-
MN Edam wagon.

 

 

102

 

3388 :88

”nausea page? «enema—me: 8972.». $8030 aware—£0
@5988 A595? 8830 88:88 98.559: «WE 5&9.
5:869. ewe—88 En: 8.38: 38o «03383 Ashram m2. 338
cm poem—o £58 5488 me: E wave—.83 Bung... ><9ﬁmn
2589. 3.338 £8 a £3. 88”? axe—gin 8989 we $9
858: 835.

 

2912..— man—3:3

 

E

 

we: :23

$395 290 Ewen 8 go 5.39. “38:98: 3363
£96 968% 088 8 can Beg. use 239. £98 88.—
9353 8 can 9889.

 

038839. 233

madam no page?“ Ed :88 £396 383 5889 88»
agqgegﬂaaoqwﬁgwmannogmoggew
5.5 @929: 98 8880. Be £58 :8 9d :gamam: 8 889
sage—.5 9. ~38.

 

 

gunman: 2008

 

gangaeﬁegommom signage—W manna—3:5
2589. camouﬂage :8. E9. anew one». 3E9. @953.

 

 

103

E

 

9.98 38$

U38 8923.8 9d .58 £58 «53.3% 585 3333a 82
gamma-$88988; E88 own—o moon—Raceﬁmev E
:5 833w go :3 .892 one: 889..

HEM 99¢. em Zeniames g5 8898 @965. em 9.»
E sum—E5. ©9883. 69.99% £8 8.5:an 3. £883 :5
889:5: went :5 38: C3 <9u. in... A8 83259
Saga. 9. A8 .88 be” 582 8a 889 peace

 

0935

 

wanes? man: an .888: 383

3b.? 9.89.8

C8 838:0. manor £399 828m :5 89.8 8 £39 a:
game». mm 908 8 m: 2:3 883 8 a gear 999 9.. $95.
on E95 evince“ we men—2&5— ire W 83835. 998 w E803.
358m 8 a: 089. 3289* 80838.

 

 

82—8. Coed

 

085.3. For—mom 388:8”... one egg...
08889.. 80858 game” 889.. 8 e: :38? g.

$339598. gnu—8m Ea 33355. 8 833— 88338:
magnum 8263 go? Snag.

 

 

104

 

203.0...» 8839.0.-

 

E

 

39.9.9. 8 095.00.. Cow...

9.5.0 2098.... 3.5.99

51959... €09 magma...“ Egg 3.00089 £53....
0.. 0 0.009.000 .009. 89.9.30 3. 89.0.50 8 ....0£ 8.8.. 8.5
9.9.. 0... 078...... G 0a a 00.... 08.9 38¢. 8 308 :5: 9.00 0
£85

 

€9.39. 293.

.50 9.95... 8959.9. 3. @383 0.. 00:89. 56039.00
989.9. 8 no. 9... 9.3.8.. 03.0

 

03:088.. 238

: 90 9.9.9339. 38.09.800.30 $3808... .89....
005895.80 0......“ 9.5:... 03.30. 80 0.52.0.5. .988...» 90
.398... p38089§§<§m§30§01§§303
A... .3... >99. .00 «00.88 .82 039. 83. 89 9.6.05.5...
02.890 8 889.3 :0 980mm. 859. u 9 .0000 3.00 a £8...
000090.55. "808 89. 0000 0 <00... a... .000 99. 3.00 0 £00...
and? u 9.00 a v9... 019$.

 

 

mg Coon.

 

vaEwoom 028.88% 9.... 0.89.80 9.9.09.8. mm 9......9 0..
".88 ... m Q38. £00.. 8... ”.0 89 80.5.... 9.... 8-90....9 As... 9
36.8. £8... .82 0:9. 00 «0.. m0 8 .50... 309 050.00 90.... 0..
8.3 9.0.5.. 9.... 80.. 380......

 

 

105

Egg—EB.

 

Hush @ Grow—B— 283

r}

E3 omgwm E can enmgmou
3% 0.838% sacs Scar can no v5: «33 Smomﬁn F
8am»— 03oz

 

n.2,

 

35 a 38a 963

E25 3.3. c». @025 3&an man—Ema E
é um oggm 30.88%

 

:a w 08mg 285

E3 cm .53 28 3mg:
38313 0.8389 3550: :38 BREE: 05mm 80E REEB
«<3. <2: ~53:

 

3028: .v Zara—88 283

:35..me 3538 m: £36m.8o:9 0088:3838 96382:
gag nag” alas Ema—v. soc—a v6: co 8 maﬁa a man
panacea £9 A888 v9.88?

 

E

 

 

3388 .w 02:25: 233

 

>318: at $8" $233 3E8.

Eng 8 83899: own—cg“ Bonus—.8 aw «585m 885
cm £593 88: o». ax 838 Ag. 38%. 3.58. 59: 96:6.
<3? 25 Ema—.3 £8 E8589 5 m 3.308%.
ZESEégggwaauongiwogumn—aﬁ
:88. Ba «:6 90% Summon E.

 

 

106

E;

 

3% a Boa :83

~35: SEES Ramona
95555 v.83 comm-on gm c». 05888. 5 was 55585
5538. 2.03% 8855 own—0888 £3 Egan.

3:38 S @8528 awn—FEE.

 

magma an 98: C3:

Mains ow 50883 on 5538.

6558 5558. mm 3.896% 3.5853 555: 35m on m
g cm 65w... >b 5538 8553er mm 25 5 £59 B.
55398— mgaom 55558 oggoam 5 83:: 88m. 53m <15
3 “36355: 55 En ago—538 Ba 3.55858 9a.: 385
39. 55 A058. 33.

 

 

Edmund 5 238568 383

 

C5523. «2538 5 ﬂoaém.5c:5 8555583: «€585.

giét.§mﬂ8§§uﬁ5oﬁ§m¢8
35 9505 56 m8 £55 8 £58 389:»— Snmaasoam. mos
58¢. 26 v5: 8 mama 33035 553.35: £5. A2258
38034..

 

 

107

ENE

 

E5

 

038539 203v

Owomanﬁgoégoognroaonmﬁﬁomﬁﬁoaoaé

 

 

Q5. GEES. 935:ng an «in
388

 

dﬂ ESQ 33833? 088355 Ego—on 3. gr— Ba
39.8: 333. 2:85 Sun 38» ESE: Sago: i:
§S§§mmo§§wm8wﬂ£ro§m§§§i
38333?
Egéﬁggﬁagnﬁwaagﬁom
8% 353:5» 936% Ba @3835... man 998
853.58 m3 36g 5a 883% $8. Won— g cm
2:3er 8: p98. €03 395.8 38938. g: 5 959.9:
£3»... 38325— roBovK 388mg 5o .38an Ba Boa
8:383 REE—EH? $9 39.88 8:63 8 9936.830
3892.5.

>353 Eon a mg was: mow—o 8 g $2.038»: 38°35.

 

 

108

Ebb—HE.

 

g

 

E; a 95%: :83

E3 own—5m E can Swan—g:

385m 35” CV é 53.85 we: a? on £355 83
ggégaagoomvdﬁgﬁgﬁnﬁogog
918%. A8 2? mono—i. gov—a W08 £39 5.5 3: ~83
Egg—surge «on»:

 

Dog. 95o: an EEK: 283

.5588 882E. 35 m: m 358 :38 ca Emma 3338 o».
:5 85:35. Runways. magma; ma 8.363 o». E
v9.85 333 F m gnaw £3»

 

wane. 23: H35 w SHEER: on 383.8? «E. 856238.
833. g 586.5» 53:? c.6898. @338
29.82: man 883%

$28. 55 a" wows. 28$ >353 cmgugﬁg.

 

 

 

H5035; 350858 52.58 «Boo—.5. AGREE?
REE—5 Ba 39¢: 5.33%.

 

 

109

 

w..

 

3633 @ 09mm :83

moan 95v. 5&3». Ba gang

389.8 03035303” :Emoﬁa £599. Go 3308 9:8 v6: 9
$31. A -N u mango? 9.3m 5o 8 +~ nu moan—w 35 can m
@55

 

E2» 5828:

3828a magma: £5. §€8n§ cm 28.— 8g

.59: Eamon 86 we: $9 :5 «53: v5: m3 ~38 90
83%.

sacs 8338 m8 «5: 25. En mum—down. egg—Ema 23. Bag
om v66. 855.5%:

amoi gamma 93 3: £9 :5 35503819.“ 3 ES
8333..

 

 

an: # 9.5m?— :88

 

73ng cm :33 E 26 among?“—

3828 manna immoau

:05. .5: gm 88a 25358 SE. 099. :38 u» :51":
.6820 5 OS. Em» Ed amp—Emma? 38% ug 8:896 «2%
ma Samoa own—o £6? cam—55:3... Ch“ £8an 9% 8
madam? £88 88539. Sad—mac: .3

 

 

110

 

mane—.3939: 0.3—5.

 

0.35. O? 22$

maiowoom 5 m maﬁagn £33“ Bog—Ba 95:? BE 8888‘
3.386 8568 c». 53353:.
n "be

Zoogggomcﬁowggémozainaﬁgo.

 

wanna—.538

 

0°35». Kori» .w <§W=
283

Gonna. m8»: @ wogo-Zn=A—nn
:83

moan @206on 83m 9.5.». cm 330$ 8 389.3 83.8%
883g 3.ng

Ragga 3. 0253. may: .w maggoag won :8 .39 m3»:
833— Rgaa. 2o EQS 8.5.me Q. REE? $3 8338 5
$99. «8%.

0033853 EmwnoRan 37:85 «Hanan Odom Eamon—baa.
@8388: mum—v82 man 8883 5 co:— in?

 

 

<o=w§g mu game. 3mm

 

mcgoomg gin om comma; m3 manna—5. 8:933»
«i»: 88:35 Ramona—So 3085.8 .

 

 

lll

W
F1: Performance (PERF) 1=Poor; 5=Bxce11ent

 

V1 How would you describe the overall performance of your store(s) last year?

 

V2 How would you describe your performance relative to your major competitors?

 

V3 How would you describe your performance relative to other stores like yours in
the industry?

 

 

F2: Local Marketing Competence (MC/MI) 1=Not as strong as competition; 7= Much
stronger than competition

 

V4 Assessment of current customers’ needs and wants.

 

V5 Assessment of prospective customers’ needs and wants.

 

V6 Quality of customer service.

 

V7 Ability to offer competitive prices.

 

V8 Creating a pleasant shopping atmosphere

 

V9 Effectiveness of store advertising.

 

V10 Effectiveness of store layout and merchandise presentation.

 

 

F3: Innovative Marketing Competence (MC/INV) 1=Not as strong as competition; 7=
Much stronger than competition.

 

V11 Ability to differentiate merchandise and service oﬁ‘erings from that of
competitors.

 

V12 Being ﬁrst to introduce new merchandise and merchandise lines.

 

V13 Introducing new ideas in my business.

 

 

 

V14 Trying new marketing techniques.

 

 

112

Wm

F4: Information Quality (IQUAL)

 

 

V15

When using information ﬁom the people you named above, how accurate
would you say it usually is? l= Not accurate at all; 7 = Very accurate

 

V16

Sometimes the information we get may get right to the heart of the problem
we are facing. Other times the information may not be very speciﬁc to our

needs. In general, how relevant is the information ﬁom the people you named
above? 1=Not relevant at all; 7=Very relevant

 

V17

At times we must gather a lot of information which isn’t very relevant in order
to get enough to make a good decision. Other times we need only a small
amount of information because the information is very speciﬁc and allows us
to make a decision. How speciﬁc is the information you get from the people
you namd above? 1=Not speciﬁc at all; 7=Very speciﬁc

 

V18

Some information may be exactly what we require. How often is this the case
for information obtained ﬁom the people you named above? 1=Not oﬁen;
7=Very often

 

V19

 

To be useful, information must oﬁen be available when we need it, not at
some later time. How timely would you estimate information to be ﬁom the
people you named above? 1=Not very timely; 7=Very timely

 

F5: Trust (TRST) 1= Strongly disagree; 5= Strongly agree

 

 

 

 

 

V20 I am considered to be dependable by these people.

V21 These people would say that I am sincere.

V22 These people would trust me with personal information about themselves.
V23 They would say that I am a trustworthy person.

 

F6: Commitment (COM)1= Strongly disagree; 5= Strongly agree

 

 

 

 

 

V24 I am satisﬁed with the level of business support I get ﬁ'om them.
V25 These people share the same ambitions and visions about business that I do.
V26 They are enthusiastic about helping me in my business.

 

113

 

Wm

F7: Reciprocity (RECIP) l= Strongly disagree; 5= Strongly agree

 

 

 

 

V27 I would do a favor for any of these people if they asked.
V28 These people would be willing to do me a favor if I asked.
V29 We do each other favors from time to time.

 

 

F9: Density (DENS) l= Strongly disagree; 5= Strongly agree

 

 

 

 

V30 These people know each other by name.
V31 These people talk to each other about business.
V32 These people see each other regularly in business situations.

 

F10: Centrality (CENT) 1= Not true at all; 5= Very true

 

 

 

 

 

 

 

V33 I talk directly with these people about business issues.

V34 If any of them had information that would help me in my business, they could
tell me directly.

V35 Among these people, I often pass along business information from one person
to another.

V36 If I needed advice about running my business, I could call any of these people
on the telephone.

V37 If these people bad business information that would be helpful, they would tell
me right away.

V38 I am one of the ﬁrst to hear about new things from this group of people.

 

F 12: Emotional Intensity (EMOT) l= Not true at all; 5= Very true

 

 

 

 

 

V39 My relationships with these people are very close.

V40 I do things socially with these people.

V41 IfI hadthechance, I would spendaﬁ'eeaﬁernoonwithanyofthese people.
V42 1 consider most of these people my ﬁ'iends.

 

 

114

 

Wary;

F 13: Intimacy (INT) 1=Very unlikely; 5= Very likely

 

 

 

 

V43 You would share personal matters with them
V44 You might discuss family matters with them.
V45 You might ask them for advice about a private matter.

 

F14: Perceptual Homophily (PERHOM) l= Very dissimilar; 5=Very similar

 

 

 

 

 

 

V46 Your outlook on life.

V47 Your likes and dislikes.
V48 Your business philosophy.
V49 Your values and beliefs.

 

115

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample Characteristic Frequency Percentage"
Owner 104 92.9
Manager 7 6.3

Gender
Female 80 71.4
Male 29 25.9

Age
Less than 40 years 18 16.4
41-50 years 36 32.7
Over 50 years 56 49.1

Education
High school 25 22.7
College 76 69.1
Post-graduate 9 8.2

Years Owned Retail Business
5 or less 30 27.0

6-10 30 27.0
Over 10 50 45.1
Years experience in Retailing
5 or less 14 12.6
6-10 18 16.2
Over 10 79 70.3
Number of Full-time
Employees
None 47 42.0
1 30 27.0
2 or more 44 31.2
Number of Part-time
Employees
None 16 14.3
1 to 5 60 54.1
6 or more 35 30.7
*Less than 100% due to missing data

 

116

 

a O
;

‘Vaﬁaﬂh

.
J E.0_II_’ Q...
_—

Pedbnmmxw

um, I ‘1”: _o

hﬁnkeﬁng
(Connxnence

 

Ihﬁbnnaﬁon

 

DAI

1h“!

hAI

INF!

 

v1

.824

.856

 

v2

.825

.850

 

v3

.767

.826

 

alpha

.839

.839

 

v4

.793

 

v5

.819

 

.804

 

alpha

.81

 

v11

.655

 

v12

.812

 

v13

.883

 

v14

 

alpha

 

v15

 

v16

.877

.938

 

v17

.915

.919

 

v18

.865

 

v19

 

 

alpha

 

 

 

 

 

 

.89

 

117

 

 

O
- 0 .' J. "1.5...

O
z. o 0.31;. :13

.10.“?! 51003

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variable Trust Commitment
MI A INV MI INV

v20 .662 .687
v21 .829 .781
v22 .566 .720
v23 .828 .686
alpha .71 .75
v24 .798 .582
v25 .751 .824
V26 .854 .844
alpha .79 .75

Reci- Reci-

procity procity
v27 .757" .540‘
v28 .655" .727‘
v29 .731" .555’
alpha .87 .76
‘ v27, v28, v29 loaded with COM in MI networks
b v27, v28, v29 loaded with TRST in INV
networks

 

 

118

 

Variable

I
a. 9 1’ . .

Density

I
O
-

C
.' 2.. l.’

Centrality

’ O
5 .\

{O I I;:.

I {.log.

Emotional
Intensity &
Intimacy

Perceptual
Homophily

 

MI

INV

MIINV

 

v30

.749

.674

 

V3]

.878

.895

 

v32

.827

.867

 

alpha

.89

 

v33

.699

.807

 

v34

 

v35

 

v36

.626

.757

 

v37

.763

.635

 

v38

.673

.630

 

alpha

.90

 

v39

 

v40

.651

.632

 

V4]

.643

.769

 

v42

.687

.729

 

v43

.857

.868

 

v44

 

v45

 

alpha

.94

.87

 

v46

 

v47

.722

.841

 

v48

.613

.732

 

v49

.762

.713

 

 

alpha

 

 

 

 

 

 

 

.76

 

.87

 

 

119

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.. - 'mu 1 mm o 81 0)" a o kl calm! act...
Standardized
Parameter Standardized Residual

Path Label Estimate z-value Estimate Variance
v30, DENS 1.00 .755 .656
V3], DENS 1.17 7.905” .850 .526
v32, DENS 1.125 7.638“ .792 .611
v33, DENS 1.00 .645 .765
v36, CENT 1.360 6.185* .755 .656
v37, CENT 1.074 6.447"I .805 .589
v38, CENT 1.065 5.669* .671 .742
v41, FRND 1.00 .782 .623
v42, FRND 1.001 8727* .924 .383
V43, FRND .480 5.848" .572 .820
v47, PERHOM 1.00 .658 .753
v49, PERHOM 1.270 4.393" .704 .710
DENS, CENT .313 3.496“ .519
DENS, FRND .425 3.225“ .422
DENS, PERHOM .270 2.929" .461
CENT, FRND .362 4.045“ .666
CENT, PERHOM .178 3.196“ .561
FRND, PERHOM .335 3.629“ .634
*p<.05

 

 

x2 (50, n=106) = 78.257, p=.006; NNFI= .924, CFA=.942

 

120

 

 

 

ulna. ‘ ‘ u

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

an 'itSH-f-l‘vl m 01-: s u \l as: an 1.1.. .

Path Label Parameter Standardized Standardized
Estimate z—value Estimate Residual

Variance

v30, DENS 1.00 .807 .591

v31, DENS 1.076 9.961“ .902 .431

v32, DENS .993 9.495" .850 .526

v33, CENT 1.00 .628 .778

v36, CENT 1.660 6.112“ .770 .638

v37, CENT 1.476 6.365“ .821 .571

v38, CENT 1.601 6.115* .771 .637

V4], FRND 1.00 .869 .494

v42, FRND .975 12.539‘ .938 .345

V43, FRND .923 9.316" .765 .645

v47, PERHOM 1.00 .742 .670

v49, PERHOM 1.140 5.996"I .837 .547

DENS, CENT .287 3.574" .529

DENS, FRND .688 3.926“ .510

DENS, PERHOM .354 3.351“ .480

CENT, FRND .371 4.269" .708

CENT, PERHOM .177 3588* .618

FRND, PERHOM .450 4.086“ .632

‘p < .05

 

 

x2 (50, n=101) = 80.928, p=.002; NNFI'f' .935, CFI=.953

 

121

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

sa-z 'erIC-l‘d ‘1": ~. ‘..Iu';c‘ a. . u _._;:__Iult-l.o.~
Construct Path Label Parameter Standardize Standardized
Estimate z-value d Estimate Residual
Variance
DENS v30, F9 1.00 .763 .646
V3], F9 1.159 7.953‘ .847 .531
v32,F9 1.106 7.663" .787 .617
CENT v33, F10 1.00 .636 .772
v36, F10 1.372 6.089‘I .751 .660
v37, F10 1.099 6.389" .816 .578
v38, F10 1.081 5.612" .671 .741
FRND v41,F15 1.00 .782 .623
v42, F15 1.002 8.684" .924 .383
v43, F15 .481 5.856‘ .573 .819
PERHOM v47,F 13 1.00 .647 .763
v49,F 16 1.316 4.317“ .717 .697
Construct Equations
DENS, NETCHAR .625 4.799" .585 .811
CENT, NETCHAR .462 5.543" .821 .572
FRND, NETCHAR .767 6.334" .805 .593
PERHOM, NETCHAR .405 4.569“ .743 .669
*p< . 05
x2 (50, n=106) = 61.548, p=.126; NNFI= .969, CFI=.977
Variable Composite Reliability Variance Extracted
DENS .762 .516
CENT .750 .431
FRND .740 .495
PERHOM .560 .390

 

 

 

122

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

._.,' 'zzuca 1w" . 2.5333; .3 . sl_.___2__{__ll\| 1.1.0.:
Construct Path Label Parameter Standardized Standardize
Estimate z-value Estimate d Residual
Variance
DENS v30, F9 1.00 A .602 .533
V3], F9 1.492 7.677"I .938 .347
v32,F9 1.310 7.646" .842 .540
CENT v33, F10 1.00 .628 .778
v36, F 10 1.663 6.121 “ .772 .635
v37, F 10 1.473 6.356“ .820 .573
v38, F10 1.600 6.114" .771 .637
FRND v41,F15 1.00 .870 .492
v42, F15 .966 12.646* .931 .366
v43, F15 .938 9.567“ .777 .629
PERHOM v47,F13 1.00 .743 .670
v49,F 16 1.138 6.015" .736 .548
Construct Equations
DENS, NETCHAR .479 4.307" .540 .841
CENT, NETCHAR .388 5.661" .845 .535
FRND, NETCHAR .947 7.61 1"I .829 .559
PERHOM, NETCHAR .472 5.484“ .757 .653
* p < .05
{(50, n=101) =55.91, p=.24; NNFI== .987, CFI=.991
Variable Composite Variance Extracted
Reliability
DENS .799 .579
CENT .773 .462
FRND .817 .60
PERHOM .671 .506

 

 

 

 

123

 

“low _ . . 1
-1‘ '

20‘ l _szuz.‘ a .‘lnw " 0 ml .1”: t. o a Cu! union

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Path Label Parameter Standardized Standardized }

Estimate z-value Estimate Residual
Variance

v 20, TRST 1.00 .653 I .757

v21, TRST .895 5.838" .787 .617

v23, TRST .613 5.584‘ .706 .708

v24, COM 1.00 .815 .579

v25, COM .988 6.771“ .655 .756

v27, COM 1.073 8.421 ‘ .794 .608

v28, RECIP 1.00 .736 .677

v29, RECIP .851 6.449’ .615 .788

TRST, COM .113 3.316“ .490

TRST, RECIP .162 4.158“ .790

COM, RECIP .335 5.483“ 1.023

*p < .05

{(17, n=106) = 32.71, p=.01227; NNFI= .921, CFI=.952

 

 

124

 

..- ::u._"-l in.“ 0 «1.196; a 0. [15‘ Iran 0

Path Label Parameter Standardized Standardized

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Estimate z-value Estimate Residual
Variance

v 20, TRST 1.00 .735 .678
v21, TRST .944 6.204"I .677 .736
v23, TRST 1.163 7.149“ .808 .590
v24, COM 1.00 .715 .699
v25, COM 1.321 5.916"I .682 .731
v27, COM 1.221 6.185“ .722 .692
v28, RECIP 1.00 .844 .536
v29, RECIP 1.040 7.332“ .752 .659
TRST, COM .123 4.278"I .764
TRST, RECIP .225 4.717* .792
COM, RECIP .266 4.649" .803
I"p < .05

 

 

x2 (17, n=103) = 43.059, p < .001; NNFI= .873, CFI=.923

 

125

 

 

 

HOUR" O

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

:u ':¢:m,‘ 3; a "-1.0 u 0; ya a o lo :7, £01.11 t - 1.0 1.
Construct Path Label Parameter Standardized Standardized
Estimate z-value Estimate Residual
Variance

TRST v20, F5 1.00 .659 .752
V2], F5 .895 5.702"I .794 .608
v23, F5 .597 5.481“ .694 .720

COM v24, F6 1.00 .806 .591
v25, F6 .994 6.674" .650 .760
v27, F6 1.075 8.253" .785 .620

RECIP v28, F7 1.00 .792 .610
V29, F7 .811 6.269‘I .632 .775

Construct Equations

TRST, SC .227 4.461“ .591 .807

COM, SC .591 9.062" .988 .158

RECIP, SC .574 8937* .987 .162

‘p < .05

{(17, n=106) = 44.806, p=.012; NNFI= .873, CFI=.918

Variable Composite Reliability Variance Extracted

TRST .690 .429

COM .719 .461

RECIP .665 .557

 

 

 

126

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

=.- c a u. in; ' o . ”-1.0 m .m- ; o -o '. .._ .' :_. an
M
Construct Path Label Parameter Standardized Standardized
Estimate z-value Estimate Residual
Variance
TRST v20, F5 1.00 .739 .674
v21, F5 .931 6.230“ .672 .741
v23, F5 1.156 7.290"I .897 .590
COM v24, F6 1.00 .467 .711
v25, F6 3.299 2.471 * .749 .633
v27, F6 3.021 2.459‘ .785 .620
RECIP v28, F7 1.00 .839 .544
v29, F7 1.045 7.426“ .750 .661
Construct Equations
TRST, SC .304 6.452“ .816 .578
COM, SC .141 2.364" .738 .675
RECIP, SC .750 9.213* .986 .169
* p < .05

 

 

{(18, n=103) = 43.526, p < .001; NNFI= .923, CFI=.953

 

 

 

 

 

 

Variable Composite Reliability Variance Extracted

TRST .726 .473
COM .671 .415
RECIP .676 .512

 

 

 

127

 

 

Ill ll D'fﬂ .1 H E DENS :EIU:
W

 

 

 

 

 

 

 

 

 

 

 

 

Mean

Difference

between MI
Latent and INV Standard
Factor Networks Error z— value
DENS .257 .173 1.487
CENT .23 1* .084 2.747
FRND .394‘ .159 2.475
PERHOM .183 .109 1.685
I"p < .05
x2 (107, n=101) = 148.3, p=.005; NNFI= .957, CFI=.965

 

 

 

128

 

o
-.
H-

M .106;

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. r u: a a, [.115 ucr lul who .
Path Label Parameter Standardized Standardized
Estimate z-value Estimate Residual
Variance
DENS, NETCHAR 1.000 .447 .894
CENT, NETCHAR 1.598 4.645 " .942 .335
FRND, NETCHAR 1.587 4.413‘ .742 .671
PERHOM, .662 3.879“ .555 .832
NETCHAR
TRST, SC 1.000 .568 .823
COM, SC 2.794 5.969“ .843 .537
RECIP, sc 2.589 6.017“ .858 .518
v16, IQUAL 1.000 .803 .595
v17, IQUAL 1.179 9.583’ .880 .475
v18, IQUAL 1.222 8.845“ .812 .584
v4, MC/MI 1.000 .949 .316
v5, MC/MI .981 14.500" .893 .449
v6, MC/MI .396 4.641“ .437 .899
v1, PERF 1.000 .711 .703
v2, PERF 1.408 6.965"I .856 .517
v3,PERF 1.189 6.788" .772 .636
SC, NETCHAR .094 3.257‘ .787
NETCHAR, IQUAL .192 2.709“ .395
NETCHAR, MC .308
NETCHAR, PERF .050 1.440 .173
SC, IQUAL .130 4.050‘ .728
SC, MC .017 1.088 .144
SC, PERF .013 #996 .139
IQUAL, MC .113

 

 

 

 

 

129

 

 

 

my .01 um Waring-1‘ ‘1112‘15.’ an 01-61 s.u.~ll_r:..uu‘

 

 

 

mm , I
IQUAL,PERF I .067I 1.322 I .185 I
MC, PERF I .235 I 3233* I .414 I
*p<.05

 

 

x’ (96, F104) = 179.324, p < .001; NNFI= .880, CFI=.904

 

130

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . ’51:. .1 am.» . mm.- . ,1. 11.: o..- l1\_l_ucv1...
Path Label Parameter Standardized Standardized'
Estimate z-value Estimate Residual
Variance
DENS, NETCHAR 1.00 .606 .795
CENT, NETCHAR 1.357 6.843" .940 .341
FRND, NETCHAR 1.404 6425* .825 .565
PERHOM, .628 5.137” .605 .796
NETCHAR
TRST, SC 1.000 .736 .677
COM, SC 1.314 6.825“ .717 .697
RECIP, SC 2.389 8.079‘ .859 .512
v16, IQUAL 1.000 .933 .361
v17, IQUAL 1.032 13.167“ .909 .416
v18, IQUAL .894 9.726" .759 .651
v12, MC/INV 1.000 .608 .794
v13, MC/INV 1.279 5.479" .944 .331
v14, MC/INV 1.056 5.290“ .638 .770
v1, PERF 1.000 .71 .703
v2, PERF 1.471 7.214" .894 .448
v3,PERF 1.174 6.857“ .760 .650
SC, NETCHAR .217 4.272“ .850
NETCHAR, IQUAL .285 2.879‘ .361
NETCHAR, MC/INV -.094 -1.194 -.138
NETCHAR, PERF -.022 -.457 -.052
SC, IQUAL .157 3.637“ .481
SC, MC/INV -.010 -2.98* -.034
SC, PERF .022 1.067 .128
MC/INV, IQUAL .178 1.741 .203

 

 

 

 

 

131

 

 

 

 

23‘ o H 11.7! Unit; '7 _‘111112 ‘ to 1:1 .1 «’1'

 

 

 

 

. . __a . u -12- AL‘!‘ Ill.
mm ,

PERF/IQUAL I .113 1.796 I .209 I

PERF, MC I .224 3.068“ I .476 I

1"p < .05 7

 

12 (96, n=100) = 131.48, p=.009; NNFI= .945, CFI=.956

 

 

132

 

 

 

 

 

 

 

 

 

 

 

 

“"7

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Path Label Parameter Standardized Standardized
Estimate z-value Estimate Residual
Variance
DENS, NETCHAR 1.00 .432 .902
CENT, NETCHAR 1.718 4.470" .978 .207
FRND, NETCHAR 1.623 4.353" .733 .680
PERHOM, NETCHAR 1.835 3.620" .618 .758
TRST, SC 1.00 .621 .784
COM, SC 1.818 5.164“ .599 .505
RECIP, SC 2.370 6.578“ .858 .514
v16, IQUAL 1.00 .803 .596
v17, IQUAL 1.157 9.410" .863 .505
v18, IQUAL 1.249 9.005“ ;826 .564
v4, MC/Ml 1.00 .943 .334
v5, MC/MI .998 9.051“ .897 .443
v6, MC/MI .400 4.481* .437 .900
v1, PERF 1.00 .705 .709
v2, PERF 1.413 6.924" .852 .524
v3, PERF 1.215 6.805“ .781 .624
Construct Equations
NETCHAR, SC .317 3.749“ .756 .654
SC, IQUAL 2.288 4.603* .613 .790
SC, PERF .272 1.072 .116 .904
IQUAL, MC/MI .183 1.347 .146 .989
MC/MI, PERF .201 3.482" .402 .903
" p < .05

 

 

x’ (99, n=104) = 208.425, p < .001; NNFI= .847, CFI=.874

 

 

133

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. v "v ‘1 w ‘ ' ___'____.t- 1 'i_.\1£_’_"_\.‘__\_’__‘3_~.;
Standardized
Parameter Standardized Residual
Path Label Estimate z-value Estimate Variance
DENS, NETCHAR 1.00 .606 .795
CENT, NETCHAR 1.357 6.831" .941 .339
FRND, NETCHAR 1.401 6.413“ .823 .568
PERHOM, NETCHAR .631 5.154“ .608 .794
TRST, SC 1.00 .740 .672
COM, SC 1.300 6.804“ .714 .701
‘ RECIP, SC 2.381 8.093“ .861 .508
v16, IQUAL 1.00 .933 .359
v17, IQUAL 1.030 13.159“ .909 .417
v18, IQUAL .893 9.718“ .759 .651
v12, MC/INV 1.00 .607 .795
v13, MC/INV 1.28 5.520" .944 .330
v14, MC/INV 1.056 5284* .638 .770
v1, PERF 1.00 I .706 .708
v2, PERF 1.498 7.126’ .904 .427
v3, PERF 1.169 6.787“ .752 .659
Construct Equations
NETCHAR, SC .351 5.403“ .846 .532
SC, IQUAL 1.434 4.184* .466 .885
SC, PERF .180 1.043 .110 .960
IQUAL, MC/INV .265 2.282“ .305
MC/INV, PERF .299 3.655“ .487 .871
*p < .001 2

 

 

36 (99, n=100) = 132.196, p <.001; NNFI= .950, CFI=.959

 

 

134

 

 

. . 0 C O O I
801‘ 1“" ‘l 0 .“OHHJI 0 "H11"! :IH‘ 11 I‘ 5..-; U0"

 

 

 

 

 

 

 

 

 

 

 

 

Standardized Residual Variance Explained in Dependent

Variance Variable“

MI INV Nﬂ INV

Network Network Network Network
Path Model Model Model Model
NETCHAR, SC .654 .532 .527 .717
SC, IQUAL .790 .885 .376 .217
SC, PERF .904 .871 not sig not sig
IQUAL, MC 4 .989 .960 not sig .078
MC, PERF .903 .871 .184 .241
‘1 - (standardized residual variance’)

 

 

 

 

135

 

 

 

 

 

 

 

 

 

 

 

Evian—8m... ”8:5
Ewan 58308 202.8% 95 «wane—.8 ea: .8 :53 Emma. 8388: 9573 man; 96:28:
96.53% 8:ng Em: HESS—no: Zagenw 925 53:3.
K— anonwm SE 8:85 $359. mom 9E5 556 on magma?— FSEWQ. 55.8%. 55% 9.32.8:
wean—:8:— moBovrm—S Ed 3.35:8. em ~3§~63 :5: 92 23518.
v.2. 7.: 2322*.» mag. :63 8:3 3: 38mg :8 £5 .25 8 Egan 55% em mew—venom
mean— 0eEE 60V.
v.2. E 202618. Emuﬁ. 33w 0». mean: 0:38— Am0v a: g 8 Enron Ewe—Sumo: mauve—.3:
Dew—3. 90995.
men :12 23.3.5. mag. :53 8:5: 3: 3:39 :8 «5.: .8: 8 Egon .956 cm $32.8:
meow: 0:38— 69.
v.9. 92 2322.5. Emrﬂ. rim—m om meow: 0:38— ; 5:: 8 Egon _oSw—m ow mauve—.8:
59353: 95—5. 00925.
me 3:18" 58—538 95 29235. Ema 93353:: 0..me in. We: 8 3%.. 22 832.8:
Ewangm 02:38:08 858: 8 Eco:— Kmawwa EQZU.
men 3:52:38 233:5. Enron Ewe—433:: 95:0. 25: We: 8 35:3 Zﬁwoﬂﬁm 96:38:
0:338:96 8.3:: 8 ”0:38 Ema 20545.
m2. 2: 89235. 3mg Hoe:— zﬁwﬂim 00:69.38 9309.5 2:. We: 8 Emrﬁ. mauvonoa

 

 

 

38358 cm takeaway.

 

 

 

136

 

 

 

 

 

3

men 92 28351:... gran 888:8 Eng 0:88:88 20:25 £5 :8: 8

 

 

 

 

mega—8:
Enron 88:88 em 3888 mama...
3 m9. 7.: 238:8. 3E5. m0 in: .8: 8 Enron 88388 3883580 Qmwmv. 22 26:98:
_N v.9. 2< 208:5. 58:. m0 £5 :8: 8 584 38:88 owwmau. 2o: mauveno:

 

 

137

APPENDIX C

Formulas

138

 

 

Formulas for Composite Reliability and Variance Extracted

Composite Reliability

(Sum of Staming!$51.14;;anm5)2
(Sum of Standardized Loadings)2 + Sum of Indicator Measurement Error

 

 

Variance Extracted

Sum of SquamLSJandaLdiEiLQadings

Sum of Squared Standardized Loadings + Sum of Indicator Measurement Error

 

 

 

139

 

APPENDIX D

Figures

140

\

. 0:209:00

382 3328,: ._ 2:me

      

wage—32 Alli

£95

aoﬂgoﬁm

 

  

Edmonton
I. \9
x8382

5%:an

  

A _ no
\ e
ago—Sm
Alli“: 38382
J a

141

E

 

gown—Ema 8% 885820 8282? Sum

3:809:00 maﬁa—BE >79. 8833800 maze—32 :5

E

     

382 50808302 BEE .~ 2:5

EE

 

2:88am
x8362

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

143

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

58:98? 8.2 33280 3335, 33030 .«o mooqatg sum 95
Ra WED .53 WEE ﬁrm .3 an. n «3?: sQQ 38%22
5.qu “San—Be; 32.1. $3.3 u Gang 5me "€252 5
826220320 828382 ”—082 88mm .820 983m .m 25me 53/
$5
ace.
«.3.
Q5
:5
£3
Em .

momﬁcouogﬁo x5382 . hm>

NR.
ban—~50 IV 35

5.
..IV mm>

«hm. \1 c _
mam. Na.

~m>

IVE.
& malv :5

56:00 53.
. \1 or»

:w men—V

Atom—mama 8m 32:8 333.?» 33030 no 805% 8km
manic ..§.u~r,22 award .83? a2": .3 Q £32.22
3.qu “mare—:2 5...".— .eaailéﬁa .5 «x "33592 _E

3&8 38m £32 8.5 520 88% .v 2&3

 

 

emu.
«3.
E ,3.
338 38m i
«ma.
me. \1
2%. ”mm.
Sm.
mum.
Ba. \1

5
380.5802

3
Eon—£8800

EN .

\\\V
5.
j

c.—

As.

%
l\

fa}

o.—

l 2.13%

 

 

QN>

 

 

 

 

wm>

 

 

 

 

cm>

 

 

 

 

mN>

 

 

 

 

vm>

 

 

 

 

mm>

 

 

 

 

_N>

 

 

 

 

om>

 

 

144

 

28%? 82...
$338? 8m Badman—o 3235, Sam
88288 3332 >2: mSaﬁaaco 3332 a):

 

.082 _aéoﬁm :5 .m 2:5
E E

E as

   

c.—

 

 

qa .utu 53 .uﬁzz
53¢ @ENMTSSug «a «a
£33qu
gauze mEVE?

 

 

     

 

  
    

 

 

 

 

 

 

 

 

 

 

 

%M . .
S we as 8.7.. .mavgnlviu: .3 «x
5. a: a 8».
x3 K 4% \4 E 3.33:: _2
. . ..
3:30 3 3335:
5.39583 _ _ 080m
3».
.5.
mm». nimgochm
V. . . .
. 2.». 85:8 .
_ H _ >\o_ .33. am «So _ €030”an . . 5:55:00
3%. 73m 0_ . m wmb. / ‘
A\\\. 3
an» S

 

anooamoum

:.
N k an.

 

 

 

 

 

 

 

 

Hangman—o0

 

 

 

£85

 

 

 

145

REFERENCES

146

REFERENCES

Adams, Rebecca & Blieszner, Rosemary (1994). An Integrative conceptual

Framework for Friendship Research MW
11(1224).163-184.

Alange, S., Jacobsson, S & Jarnehammar, A. (1998). Some Aspects of an
Analytical Framework for Studying the Diffusion of Organizational Innovation.

 

Alder, H. (1992). A Model for Personal Success. WM
23-25.

Aldrich, H., Rosen, B. and Woodward, W. (1986). A Social Role Perspective of
Entrepreneurship: Preliminary Findings ﬁo an Empirical Study, (unpublished paper,
Department of Sociology, University of North Carolina at Chapel Hill, April 1986)

Aldrich, H., & Zimmer, C. (1986). Entrepreneurship Through Social Networks 1n

Whip D L Sexton & K Smilor (edS) Ballinger:
Cambridge MA, 3- 23.

Anderson, J. C. & Gerbig, D. W. (1988). Some Methods for Respecifying
Measurement Models to Obtain Unidimensional Construct Measurement, loumalgf

WWW). 453-460.

Arbuthnot, Jeanette, Slama, Mark, and Sisler, Grovelyn (1993). Selection Criteria
and Information Sources in. the Purchase Decisions of Apparel Buyers of Small Retailing
1 1 ‘ . 1993 (April), 12-23.

 

Ashman, D. Brown, L. D. & Zwick, E. (1998). The Strength of Strong and Weak
Ties: Building Social Capital for the Formation of Governance of Civil Society Resource
Organizations, . . ; : . _ . 153- 171.

 

Axelrod, RM. (1984). MW New York: Basic Books.

Babb, E. & Babb, S. (1992). Psychological Traits of Rural Entrepreneurs The
loumalpﬂSmeponpmsilLﬂ). 353-362.

Bagozzi, R. (1975). u_-._ .
Wig-39-

Baldwin, T., Bedell, M. & Johnson, J. (1997). The Social Fabric of a Team-based
M.B.A. Program: Network Effects on Student Satisfaction and Performance,_Apad_e_m3LQf

Managemenﬁmunalﬂiﬂ. 1368-1397-

 

147

Baptista, Rui(1999). The Diﬂ‘usion Process of Innovations: A Selective Review.
:, . 107-129.

 

Barlow S. (1994). Conmeting with the Giants W
78-81.

Barrera, M. (1996). Distinctions Between Social Support Concepts, Measures and
Models MW 413-445

Barringer, B. (1997). The Effects of Relational Channel Exchange on the Small
Firm: AConceptual Framework o -11.. z . -. 65- 79.

 

Bass, L. & Stein, C. H. (1997). Comparing the Structure and Stability of Network
Ties Using the Social Support Questionnaire and the Social Network List ,loumalpf

Wmmu 123-132

Bates, T. (1994). Social Resources Generated by Group Support Networks May
Not Be Beneﬁcial to Asian Immigrant-Owned Small Businesses W
671 -689.

Baumol, W. (1996). Entrepreneurship: Productive, Unproductive, and
Destructive W11. 3- 22

Beal, R(2000). Competing Effectively: Environmental Scanning, Competitive
Strategy, and Organizational Performance in Small Firms, MW

WM). 27-47

Bearden, S. & Teel, J. (1982). Sampling Size Effects on Chi Square Statistics
Used 1n Evaluating Causal Models, _ 1, . ,, - ._
430.

 

Beggs, J., Haines, V., Hurlbert, J. (1996). Revisiting the Rural-Urban Contrast:
Personal Networks in Nonmetropolitan and Metropolitan Settings MW
(2), 306-325.

Bentler, P. (1993). n ' . _.
CA: BMDP Statistical Software.

 

Berg, J., Piner, K., & Frank, S. (1986). in Resource Theory 1W
61129119319115 Foa, Converse, Torblom & Foa (eds) San Diego : Academic Press, 1993.

Bharadwaj, S. & Menon, A. (1993). Determinants of Success 1n Service

Industries: A PIMS- based Empirical Investigation W
19-40.

148

Bharadwaj, S., Varadarajan, P. R., & Fahy, J. (1993). Sustainable Competitive
Advantage in Service Industries: A Conceptual Model and Research Propositions Journal
oﬁMarketiiiaﬂchiobeﬂ. 83- 99

Bienenstock, E.J., Bonacich, P., & Oliver, M. (1990). The Effect of Network
Density and Homogeneity on Attitude Polarization W,Jﬂl22QL153-172.

Birley, S. (1985). The Role of Networks in the Entrepreneurial Process Journal
W111). 107-113

Blau, Z. S. (1961). Structural Constraints on Friendship in Old Age. American
WW 429-440

Bollen, K- (1989). W New York: John
Wiley & Sons.

 

Borch, O. J, & Arthur, M. B. (1995). Strategic Networks Among Small Firms:

Implications for Strategy Research Methodology NW),
419-441.

Bourdieu, P (1986) The Forms of Capital in Handbookoﬂheomandkcssarch
WWW John Richardson, ed. New York: Greenwood Press.

Box, T., White, M. & Barr, S. (1993). A Contingency Model of New
Manufacturing Firm Performance BMW 31-44

Brown, R. (1993). Rural Community Satisfaction and Attachment in Mass
Consumer Society Wham, 387-403.

Brown, B. & Butler, J. (1995). Competitors as Allies: A Study of Entrepreneurial
NetworksintheUHS Wine Industry .. _ __ . . ._ .. - _ . 1.
66.

 

Brown, J. & Reingen, P. (1987). Social Ties and Word-of-Mouth Referral
Behavior loumaipiﬂonaimeLResearshJAﬂ). 350-362-

Brush, C. (1992). Marketplace Information Scanning Activities of New
Manufacturing Ventures 1 __11_.__ _ - ._ .1 _ . '

 

Burkhardt, M. & Brass, D. (1990). Changing Patterns or Patterns of Change? The
Effects of a Change in Technology on Social Network Structure and Power

AdministratiiLe.Ssieiis;e_Qiiartszihz_li1(1).104-127

149

Burt, R S. (1987) A Note on Strangers, Friends and Happiness W
(4), 311-331.

Burt, R. S. (1992). W. Cambridge, MA: Harvard University Press.

Burt, Ronald S. (1997). The Contingent Value of Social Capital Adminislnalixe
WW 339-365.

Busenitz, L. (1996). Research on Entrepreneurial Alertness. JoumaloﬂSmail
WW 3544-

Buss, D. D. (1996). The Little Guys Strike Back NaIiQn’s anm' 955 35 (1111!! l)
18.

Butler, J. (1991 ). Toward Understanding and Measuring Conditions of Trust:
Evolution of a Conditions of Trust Inventory W3). 643-663.

Byme, B. (1994). _ . _ -:
Thousand Oaks, CA. Sage Publications.

 

Carland, J. W., Hoy, W. & Carland, J. (1988). Differentiating Entrepreneurs from

Small Business Owners: A Conceptualization WW2).
354-359.

Cam, N, Rabianski, J. & Vernor, J. (1996). Structural Trends Impacting Retail
Business WW 1042

Carroll, G. and Teo, A. (1996). On the Social Networks of Managers Academuzf
WW2). 421-433-

Casison, J. (1998). Bring Back Mom-and-Pop? WM 8.
Christensen, P., Madsen, O. & Petersen, R. (1986). Conceptualizing

Entrepreneurial Opportunity Identiﬁcation 1n I1 - .
D. L. Sexton & R. Smilor, eds, Ballinger: Cambridge MA, 61- 75.

 

Chung, L. H. & Gibbons, P. (1997). Corporate Entrepreneurship. The Roles of
Ideology and Social Capital 1 1 _1 . . . 1 , 10-30.

 

Churchill, G. (1979). A Paradigm for Developing Better Measures of Marketing
Constructs W). 64-73

Coleman, J. S. (1966). u.‘ '
Merrill.

. New York: Bobbs-

 

150

 

Coleman, J. S. (1988). Social Capital in the Creation of Human Capital Amrican
W194. 395- $120.

Conant, J ., Mokwa, M., & Varadarajan, P. (1990). Strategic Types, Distinctive
Marketing Competencies and organizational Performance: A Multiple Measures-based
Study WWWEL 365-383.

Conant, J., Smart, D., and Solano- Mendez, R. (1993). Generic Retailing Types,

Distinctive Marketing Competencies, and Competitive Advantage. W62
(3). 254-279.

Conant, J. & White, J. (1999). Marketing Program Planning, Process Beneﬁts, and
Store Performance: An Initial Study Among Small Retail Firms IIQ;.ima.l_9_fRe_tail.ing.15.
(4),. 525- 541.

Conway, S. (1995). Informal Boundary- -spanning cormnunication in the Innovation
Process: An Empirical Study _ ‘ - . 327-
342.

 

Cook, K., and Whitmeyer, J. (1992). Two Approaches to Social Structure:
Exchange Theory and Network Amiysis W 109-127.

Cooper, A. & Amtz, K. (1995). Determinants of Satisfaction for Entrepreneurs
W19. 439-457

Cooper, A. F,olta, T. & Woo, C. (1995). Entrepreneurial Information Search
MW 107-120

Cooper, A., Woo, C. & Dunkelberg, W. (1989). Entrepreneurship and the Initial
Size ofFirms W2). 317- 332

Currall, S. & Judge, T. (1995). Measuring Trust Between Organizational
Boundary Role Persons . : _ _ ._ _. _ . '
151- 170.

 

Curran,J., Jarvis, R., Blackburn,R., and Black, S. (1993) NetworksandSmall
Firms: Constructs, Methodological Strategies .1 -111,.,. .z . .
13- 26.

 

Daﬁ, R& Lengel, R. (1986). Organizational Information Requirements, Media
Richness and Structural Design WW 554-571.

Daﬁ, R. L. & Mginton, J. (1979). language and Organization, Agademf
ManmmenLReximAﬂ). 179-191

151

Dalal, M., Al-Khatib, J. Dacosta, M., and Decker, R. (1994). Why Do Small
Towns Lose Retail Business?: An Empirical Investigation mm

mm 241-252

Das, T. K., & Teng, B. S. (1998). Between Trust and Control: Developing

Conﬁdence in Partner Cooperation in Alliances MW
491 -.512

Davern, M. (1997). Social Networks and Economic Sociology: A Proposed
Research Agenda for a More Complete Social Science WW

Econoiuicsandiocioiogx 5613). 287-302.

Day, G. & Wensley, R. (1988). Assessing Advantage: A Framework for
Diagnosing Competitive Superiority WW 1-20.

DeNoble, A. & Moliver, D. (1983). The Small Business Dilemma: Can
Cooperation Help? 1 - . = 51 -58.

 

Dodd, S. (1997). Social Network Membership and Activity Rates. Some
Comparative Data _ _ 1 z z -

 

Dollinger, M. (1984). Environmental Boundary Spanning and Information

Processing Effects on Organizational Performance W21.
(2) 351-368.

  

National Culture on the Development of Trust . -
601 -620.

Du, F. & Apfel, I. (1995). The Future of Retailing, WW1
(2). 26-39.

Duchesneau, D, & Gartner, W. (1990). A Proﬁle of New Ventur Success and
Failure In an Emerging Industry W115). 297-312

Durrance, B. (1998). Some Explicit Thoughts on Tacit Learning Irainingjnd
W2). 24-29-

Easton, G. and Araujo, L. (1994). Market Exchange, Social Structures and Time
Empeanloumalprarkeimmu 72-84

152

Eckenstahler, C. (1995). Reﬁlling Small Town Retail Space £9.9an
DexelppmenLBeﬁemMZ). 92- 93

Emerson, R. (1973). Social Exchange Theory WW
335-362.

Falemo, B. (1989,) "The. Firrns' External Persons: Entrepreneurs or Network
._ . 167-177.

 

Feldman, S. & Spencer, M. (1965). The Effect of Personal Inﬂuence in the

Selection of Personal Services, WW, Peter Bennett,
ed. Chicago: American Marketing Association.

Flora, J. (1998). Social Capital and Communities of Place WM)
481-506.

Flora, J. and Flora, C. (1990). Developing Entrepreneurial Communities
W 197- 207.

Flora, J., Sharp, J. Flora, C., & Newlon, B. (1997). Entrepreneurial Social
Inﬁastructure and Locally Initiated Economic Development in the Nonmetropolitan

United SMtCSWWﬂL 623-645

Foa, U., Converse, J., Tomblom, K., and Foa, E. (1993). W. San
Diego, CA: Academic Press.

Frazier, B. (1999).Embedded Network Relationships as a Source of Competitive
Advantage for Rural Retailers. Working paper, Michigan State University, Department
of Hurmn Environment & Design: Merchandising Management, East Lansing.

Frazier, B. & Niehm, L. (1999). Embedded Community Relationships as a
Competitive Strategy for Rural Retailers. Paper presented at 1999 Snowbird Conference
on Rural Retailing, Snowbird, Utah.

Freeman, C. (1991). Networks of Innovators. A Synthesis of Research Issues
ResearchBQhQLMSJ. 499-514

Frenzen, J ., and Davis, H. (1990). Purchasing Behavior in Embedded Markets
W 1-12-

Frenzen, J. & Nakamoto, K.(1993). Structure, cooperation, and the Flow of
Market Information 1 , 111:. .. ' 360-374.

 

Friedman, R. & Krackhardt, D. (1997). Social Capital and Career Mobility
WWW 316-334

153

Fuguitt, G. V., Brown, D. L., & Beale, C. L (1989). W
America. New York: Russell Sage Foundation.

Gales, L. & Blackburn, R. (1990). An Analysis of the Impact of Supplier
Strategies and Relationships on Small Retailer Actions, Perceptions, and Performance

W 7-21

Ganesan, S. & Weitz, B. (1996). The Impact of Stafﬁng Policies on Retail Buyer
Job Attitudes and Behaviors WM). 31 -56.

Gardner, D. (1986). Marketing/Entrepreneurship Interface: A Conceptualization

in W D L Sexton & K Smilor eds” Ballinger:
Cambridge MA, 35- 54.

Gartner, W. (1988). Who IS an Entrepreneur? is the Wrong Question.
Mummumhmmm “-32

Gatignon, H. & Robertson, T. (1985). A Propositional Inventory for New
Diffusion Research, WA). 349-367-

Giﬂ'ord, S. (1998). Limited Entrepreneurial Attention and Economic Development
W119. 17- 30

Gilbert, SJ. (1976). Self-Disclosure, Intimacy and Communication in Family,
W1 221-231-

Gilly, M., Graham, J., Wolﬁnbarger, M., and Yale, L. (1998). ADyadic Study of
Interpersonal Information Search 1 ..11:__ -
83- 100.

 

Ginn, C. & Sexton, D. (1990). A Comparison of the Personality Type Dimensions
of the 1987 INC. 500 Company Founders/CEO’s with Those of Slower-Growth Firms,
W115). 313- 326

Golden, P. & Dollinger, M. (1993). Cooperative Alliances and Competitive
Strategies 1n Small Manufacturing Firms _ , .1
43- 56.

 

Goodman, L. R. (1995). Changes 1n Retailing and the Smaller Community 1n
North Dﬂommmmmtmﬁpﬁm 91 --92

Granovetter, M. (1973). The Strength of Weak Ties Americanlonmal
W 1360-1381.

154

Granovetter, M. (1985). Economic Action and Social Structure: The Problem of
Embeddedness AmeﬁoanlmmaloﬁSooiolouilm 431 --510

Greve, A. (1995). Networks and Entrepreneurship-An Analysis of Social
Relations, Occupational Background, and Use of Contacts During the Establishment

Process SeandhaﬂnangementlmmalJJﬁ). 1 -24

Gulati, R. (1995). Does Familiarity Breed Trust? The Implications of Repeated
Ties for Contractual Choices in Alliances . - -... . _ .. I,

 

Hair, J., Anderson, R., Tatham, R. & Black, W. (1995). W
Analysis, Upper Saddle River, N.J.: Simon & Shuster.

Hall, A.& Wellman, B. (1985). Social Networks and Social Support Socjal
W S. Cohen and S.L. Syme, eds., Orlando, Fla: Academic Press.

Halpern J. (1996). The Effects of Friendship on Decisions: Field Studies of Real
Estate Transactions W12). 1519-1547.

Hanlon, D. & Scott, M. (1986). Strategy Formulation 1n the Entrepreneurial Small

 

Hansen, M. (1999). The Search-Transfer Problem: The Role of Weak Ties 1n

Sharing Knowledge across Organization Subunits WW
(1), 82-111.

Hartman, E. Alan, Tower, C. B. & Sebora, T. C. (1994). Information Sources and
Their Relationship to Organizational Innovation in Small Businesses W

W211). 36-47

Haslam, N. (1995). Factor Structure of Social Relationships: An Examination of
Relational Models and Resource Exchange Theories WW
Relationships 12121217 227.

Hawes, J., Rao, C. P. & Baker, T. (1994). Retail Salesperson Attributes and the

Role of Dependability in the Selection of Durable Goods WW
SalesManagement11ﬁﬂL6l- 7-1

Haythomwaite, C., & Wellman, B. (1998). Work, Friendship and Media Use for

Information Exchange in a Networked Organization, WWW
InformahonSmmAﬂlZ). 1101- 1114

Henderson, D. and Wallace, G. (1992). Retail Business Adjustments in Rural
Hierarchies WM 30-93.

155

Hoﬁ‘erth, S. & Iceland, J. (1998). Social Capital in Rural and Urban Communities
RMSoeiolosiLoue). 574-598

Hoy, F., and Vaught, B. C. (1980). The Rural Entrepreneur: A Study in
Frustration I1- .. . 1.1.

 

Human, S. E. and Provan, K. G. (1997). An Emergent Theory of Structure and

Outcomes in Small-Firm Strategic Manufacturing Networks Wm
M2). 368-403-

Ibarra, H. (1992). Homophily and Differential Returns: Sex Differences in

Network Structure and Access in an Advertising Firm Wk
32 (3 1, 422-447.

Ibarra, H. (1993). Network Centrality, Power and Innovation Involvement.
Determinants of Technical and Administrative Roles 1. 1 1 =
(3), 47 1- 501 .

 

Ibarra, H. & Andrews, S. (1993). Power, Social Inﬂuence, and Sense Making.
Effects of Network Centrality and Proximity on Employee Perceptions Administraﬁye

Smemeﬂuartedxﬂﬂ). 277 303

Irwin, M, Tolbert, C. and Lyson, Thomas (1997). How to Build Strong Home
Towns WW2). 42-47

Jacobs, J. (1965). I1: ,
Random House.

 

Jarillo,J.C.(1988). OnStrategic Networks -1
31-41.

 

Jennings, P. & Beaver, G. (1997). The Performance and Competitive Advantage
omeall Firms: A Management Perspective ,1 1111., =
63- 75.

 

Jo, H. & Lee, J. (1996). The Relationship between an Entrepreneur’s Background
and Performance in aNew Venture, W15). 161-171.

Johaimissen, J. and Dolva, J. (1995). Innovative Companies’ External Information
SearchinRussia 1 - 1 _ -=_ _ _ - . ,, 367- 376.

 

Johannisson, B. & Monsted, M. (1992). Contextualizmg Entrepreneurial
Networking 1 - -=_ . , 1 _ , . 109-136

 

156

Johnson, J. Lynn and Kuehn, R. (1987). The Small Business Owner/Manager’ s
Search for External Information 1 -11...- z .
60.

 

Jones, G. & George, J. (1998). The Experience and Evolution of Trust:

Implications for Cooperation and Teamwork W
531 -.546

Julien, P. (1993). Small Business as a Research Subject. Some Reﬂections on
Knowledge of Small Business and Its Affect on Economic Theory Smallﬁusiness
Eronomies, 5 (2), 157-166.

Kaish, S. ,& Gilad, B. (1991). Characteristics of Opportunities Search of
Entrepreneurs Versus Executives: Sources, Interests, General Alertness W
W16. 45-61

Kean, R.,Gaski11, L. Leistritz, L. Jasper, C. Bastow-Shoop, H., Jolly, L., &
Stemquist. B (1998). Effects of Community Characteristics, Business Environment, and
Competitive Strategies on Rural Retail Business Performance W55

Management._lo12). 45- 57.

Kean, R., Niemeyer, S. & Miller, N. (1996). Competitive Strategies in the Craft
Product Retailing Industry . . . 1 13-22.

 

Kilduﬂ‘, M. & Krackhardt, D. (1994). Bringing the Individual Back In: A
structural Analysis of the Internal Market for Reputation in Organizations Academgf

ManagemenLloumal 51(1). 87-108.

Kirzner. I. (1973). Competittonnpiﬁntreprenehrship, (311108180411: The
University of Chicago Press.

Kraatz, M. (1998). Learning by Association? Interorganizational Networks and
Adaptation to Environmental Change WWW, 621-643.

Krackhardt, D. (1995). Entrepreneurial Opportunities in an Entrepreneurial Firm:
AStructural Approach ' - 1 . . ’ z.

 

Lado, A., Boyd, N. & Wright, P. (1992). A Competency-Based Model of
Sustainable Competitive Advantage: Toward a Conceptual Integration W

Managemenuiﬂ). 77-91-

Lane, C. & Bachman, R. (1996). The Social Constitution of Trust. Supplier
Relations in Britain and Germany, WM 365- 395.

157

Larson, A. (1991). Partner Networks: Leveraging External Ties to Improve
Entrepreneurial Performance W 173-188.

Lawhead, T. (1995). A Comprehensive Strategy for Rural Downtowns Emngmic
W112). 75- 81.

Leavitt, H. (1951). Some Effects of Certain Communications Patterns on Group
Performance W 38- 50

Leifer, R. and Delbecq, A. (1976). Organizational/Environmental Interchange; A
Model of Boundary Spanning Activity AcademufManagetuentRendeiUt. 40-50.

Lewin, J. & Johnson, W. (1997). Relationship Marketing Theory in Practice: A
Case Study WW 23- 31

Lewison, Dale M. (1994). Retaih_ng. New York: Macmillan.

Levitas, E. Hitt, M. & Dacin, M. T. (1997). Competitive Intelligence and Tacit

Knowledge Development in Strategic Alliances WW).
20-27.

Liedka, R. (1991). Who Do You Know in the Group?: Location of Organizations
in Interpersonal Networks SW2). 455-474.

Loury, G. (1977). A Dynamic Theory of Racial Income Diﬂ‘erences, in Women,
1 :_ ' 1 1 Toronto. Lexington Books.

 

Lubatkin, M. & Shrieves, R. (1986). Towards Reconciliation of Market

Performance Measures to Strategic Management Research, AcademyoﬁManagemem
W3). 497-512

Lundvall, B. (1998). Why Study National Systems and National Styles of
Innovations? - .. _ 1 407-421.

 

Lussier, R. (1996). A Startup Business Success versus Failure prediction Model
for the Retail Industry Mid;Atlantic_lottrm1o£B_usiness_212). 79-92

Marsden, P. (1990). Network Data and Measurement mm
mm 435-463

Marsden, P.(1993). The Reliability of Network Density and Composition
Measures SociaiNeMorksJiﬂ). 399-421-

Marsden, P. & CampbelL K. (1984). Measuring Tie Strength W
(2), 482-501.

158

Markley, D. & McNamara, K. (1995). Sustaining Rural Economic Opportunity

 

McAllister, D. (1995). Aﬁ‘ect- and Cognition-based Trust as Foundations for
Interpersonal Cooperation 111 Organizations, . - = . z .
59.

 

McCallurn, R. C. (1995). Model Speciﬁcation: Procedures, Strategies and Related

Issues in Rick H Hoyle (ed.,) StruohuaLEquationModeLngLConceptslssuesand
Applications. Thousand Oaks, CA: Sage.

McCune, J. (1994). In the Shadow of Wal-Mart WM 10-
12.

McGrath, R., MacMillan, I. (1992). More Like Each Other than Anyone Else? A

Cross-Cultural Study of Entrepreneurial Perceptions. MW
(5), 419-429.

McKee, Daryl, Conant, Jeﬂi‘ey, Varadarajan, P., & Mokwa, Michael (1992).
Success-Producer and Failure-Preventer Marketing Skills: A Social Learning Theory

Inteipretation WM). ”-26

Midgley, D., Morrison, P. & Roberts, J. (1992). The Effect of Network Structure
in Industrial Diﬂiision Processes Researehﬁeliexll, 53 3-552.

Miles, R., Snow, C., Meyer, A. & Coleman, J. (1977). Organizational Strategy,
StructureandProcess . . . - . .1 - . 546-562.

 

Miller, N. & Kean, R.(1997a). Reciprocal Exchange in Rural Communities:
Consurners’ Inducement to Inshop WW 637-661.

Miller, N. & Kean, R. (1997b). Factors Contributing to Inshopping Behavior in
Rural Trade Areas: Implications for Local Retailers Wines

ManagementlﬂZ). 80-94

Mintzberg, H., Quinn, J. & Voyer, J. (1995). Was. Englewood
Cliffs, NJ: Prentice-Hall.

Mintzberg, Henry (1998). - z. -1 "
Willem. New York. Free Press.

Mohan-Neill, S. (1995). The Inﬂuence of Firm 3 Age and Size on Its
Environmental Scanning Activities 1 .111- . . 1 11; 1 (October), 10-
21.

 

 

159

Morgan, S, & Sorensen, A. (1999). Parental Networks, Social Closure, and
Mathematics Learning: A Test of Coleman’s Social Capital Explanation of School Effects

WWW66L631

Nahapiet, J. & Ghoshal, S. (1998). Social Capital, Intellectual Capital and the
Organizational Advantage AcademufManagemenLReximM). 242-266.

NatienlaBus'mess (1993). Survival Tips for Small Retailers, 8] (6). 26.

Neil, T. ”(1986). Distinctive Competence: A Marketing Strategy for Survival

 

Nelson, G. (1989). Factors of Friendship: Relevance of Signiﬁcant Others to
Female Business Owners Enuepreaemhialheomandlfmotimlﬂﬂ). 7-18

Nelson, R. (1989). The Strength if Strong Ties: Social Networks and Intergroup
Conﬂict inOrganizations .1. - _

  

Niederkoﬂer, M. (1991). The Evolution of Strategic Alliances: Opportunities for
Managerial Inﬂuence louma1o£Busmesslenturing§1mm 237- 257

Normann, R. (1985). Developing Capabilities for Organizational Learning in
.1 ° 1 1111. .1: “A 1.1-11 1....°1'\ . .im 1 11,1111 1.12111111'11-11 111

W. Johannes M. Pennings and Associates. San Francisco: Jossey-Bass.

Nunnally, J. C.(1970). Introduction to Psychological Measurement. New York:
McGraw-Hill.

O’Brien, D. & Hassinger, E. (1992). Community Attachment Among Leaders in
Five Rural Communities RmaLSeeiejegyiuﬂ). 521 -.534

Oliver, C. (1991). Strategic Responses to Institutional Processes Aeademef
ManagemenLRexienLlo. 145-179.

Olson, D. H. (1975). Intimacy and the Aging Family. W College
of Home Economics, University of Minnesota.

0’ Reilly, C. (1982). Variations in Decision Makers’ Use of Information Sources:
The Impact of Quality and Accessibility of Information . - _. , -,

_5_(4), 756-771.

 

160

Ostgaard, T. & Birley, S. (1996). New Venture Growth and Personal Networks,
WED. 37- 50

Parks, M. & Floyd, K. (1996). Meanings for Closeness and Intimacy in Friendship
WWW). 35-107

Pearson, B. (1994). Competing in the 90’s. 319115, 26 (5), 79-80.

Peters, M. & Brush, C. (1996). Market Information Scanning Activities and
Growth in New Ventures. A Comparison of Service and Manufacturing Businesses

W81 -89

Pineda, R., Lerner, L. ,,Mi11er C. & Phillips, S. (1998). An Investigation of
Factors Affecting the Information-Search Activities of Small Business Managers Journal

 

Podolny, J ., Stuart, T. & Harman, M. (1996). Networks, Knowledge and Niches:
Competition in the Worldwide Semi-Conductor Industry, 1984-1991. W

W 659-689-

Porter, M. (1985). .- '° 1.
W New York: The Free Press.

 

Porter, M. & Millar, V. (1985). How Information Gives You Competitive
Advantage WW 149-160

Portes, A., and Sensenbrenner, J. (1993). Embeddedness and Immigration: Notes

on the Social Determinants of Economic Action WWW
1320-1350.

Powell, G. (1990). One More Time. Do Female and Male Managers Diﬂ’er?
W31. 68-75-

Provan, K. (1993). Embeddedness, Interdependence and Opportunism in
Organizational Supplier-Buyer Networks, W23). 841-856.

Putnam,R.(1993). u:._...1 ‘ =..
Princeton, N. J.: Princeton University Press.

 

Ramachandran, K., & Rarnnarayan, S. (1993). Entrepreneurial Orientation and
Networking. Some Indian Ev1dence MW}, 513-524.

Ramaswami, S. Srinivasan, Srini S & Gorton, S. (1997). InforImtion Asymmetry
Between Salesperson and Supervisor: Postulates from Agency and Social Exchange
Theories, . _. 1': - .. , - ,- 29-50.

 

Riecken, G. & Yavas, U. (1988). A Taxonomy of Outbuyers: A New Perspective
Intematianalloumalﬂetaihngﬂl). 5-15.

Reijnders, W. & Verhallen, R. (1996). Strategic Alliances Among Small Retailing
Firms: Empirical Evidence for the Netherlands .- _ . . __ .
lﬂﬂﬂlanuarﬂ. 3545

Robinson, J., Logan, E., & Salem, M. (1986). Strategic versus Operational
PlamiIIg In Small Retail Firms WWW 7-161

Robinson, J. & Sexton, P. (1994). The Effect of Education and Experience on
Self-Employment Success JournalntBusinesslemurinL

Rogers, E. (1983). Wm New York: The Free Press.

 

Rowley, T. (1997). Moving Beyond Dyadic Ties: A Network Theory of
Stakeholder InﬂuencesAeadeMManaganReﬂmJZﬂ). 887-910-

Rogers E M & Kineaid1D L (1981) W
W New York: Free Press; London: Collier Macmillan.

Ryan, B. & Gross, N. (1943). The Diﬁ’usion of Hybrid seed Com in Two Iowa
Communities, W 15-24.

Sarnli, A. C. & Riecken, G. & Yavas, U. (1983). Intermarket Shopping Behavior
and the Small Community: Problems, Prospects of a Widespread Phenomenon lomlgf

AcademmiMarkethciemLLLUZLl- 14.
Schaefer, M. & Olson, D. (1981). Assessing Intimacy: The Pair Inventory .[omnal

 

Schafer, S. (1991). Level of Entrepreneurship and Scanning Source Usage in Very
Small Businesses WW1 19- 31

Schumpeter, J.A., (1942). - '

Harper &
Brothers.

New York:

 

Seror, A. (1989). A Study of Ind1v1dual Boundary- -spanning Communication
Patterns in a Research and Development Setting .- - - . . 1

(1239). 279-290.

Shah, P. (1998). Who are Employees’ Social Referents? Using a Network
Perspective to Determine Referent Others 1 - . 1
268.

 

249-

 

162

Sinkula, J. (1994). Market Inforrmtion Processing and Organizational Learning
lmunalsiMarkeiinLiﬂlanuam 35145.

Smart, D. & Conant, J. (1994). Entrepreneurial Orientation, Distinctive Marketing

Competencies and Organizational Performance W
(3), 28-41.

Smeltzer, L. (1996). The Meaning and Origin of Trust in Buyer-Supplier
Relationships WWW, 40-48

Smeltzer, L. & Fann, G. (1989). Comparison of Managerial Communication
Patterns in Small Entrepreneurial organizations and Large, Mature Firms mm

W 14291.4 198-2151

Smeltzer, Larry, Fann, G. andNikolaisen, N. (1988). Environmental Scanning
PracticesinSmallBusiness 1 1.11.- - _ _. - _ 11 . _ ._

 

Smith, M. Beaulieu, L. & Seraphine, A. (1995). Social Capital, Place of
Residence, and College Attendance W 363-380.

Smitka, M. (1991). W
W, New York. University Press.

Snow, C. & Hrebiniak (1980). Strategy, Distinctive Competence, and
OrganizationalPerformance1- 1f . . , -. ,_ .1 317-336.

 

Specht, P. (1987). Information Sources Used for Strategic Planning Decisions in
Small memmmmmm 21-34-

Stone, K. (1995). - '1. - '
W New York: John Wiley & Sons, Inc.

 

Stuart, R. & Abetti, P. (1 990). Impact of Entrepreneurial and Management
EXperienee on Early Perfonnance WWW 151-162

 

W. Unpublished dissertation, University of Nebraska.

“Summers, G. & Brown, D. (1998). A Sociological Perspective on Rural Studies
. . : 640-643.

 

Swan, J. & Newell, S. (1995). The Role of Professional Associations in
Technology Diffusion WW 847-874.

163

Tsai, W. & Ghoshal, S. (1998). Social Capital and Value Creation: The Role of
Intraﬁrm Networks AcademioLManagemenLloumaLAuiQ). 464-476-

Usdiken, B. (1990). Reciprocity, Asymmetry, and Information sharing in
Manufacturer-Dealer Networks ScandinaxianloumaLoLManagemenLﬂﬂ). 309-3211

Uzzi, B. (1993). The Dynamics of Interorganizational Networks: Embeddedness
and Economic Action. PlLD. Dissertation, Sociology Department, State University of
New York, Stony Brook, NY.

Uzzi, B. (1996). The Sources and Consequences of Embeddedness for the
Economic Performance of Organizations: The Network Effect Amerjganﬁociolngieal

WM). 674-693

Van de Ven, A., Hudson, R. & Schroeder, D. (1984). Designing New Business
Startups: Entrepreneurial, Organizational, and Ecological Considerations. W
ManageanJlLtl). 87-107-

Venkatrarnan, N. & Ramanujam, V. (1986). Measurement of Business
Performance in Strategy Research: A Comparison of Approaches, W

W. 11191211112111. 801-8141

Wagner, M. (1996). The Role of Business Retention in Downtown Revitalization
WM 41 -42.

Wall, E., Ferrazzi, G. & Schryer, F. (1998). Getting the Goods on Social Capital
Wmoazz.

Weedman, J. (1992). Informal and Formal Channels in ”Boundary- -Spanning
_- -.1 .1._- . _ 257-

 

Weiek. K & Bougon, M. (1976). The Equity Context QrganizationalBehaidor
andHumanPerfonnaimJéJl). 32-65.

Weimann, G. (1983). The Strength of Weak Conversational Ties in the Flow of
Information and Inﬂuence W, 5 (1283 ), 245-267.

Weissrmn, R (1998). Who’ll Pay for the Christmas Goose? W
DemogranhiesZQIlZJ. 46-47

Williams, G. (1999). 2001: An Entrepreneurial Odyssey Entrennenem 106-109.

164

Winstead, B. ,Derlega, V. Montgomery, M. & Pilkington, C. (1995). The Quality

of Friendships at Work and Job Satisfaction MW
12120), 199-215.

Winter, S. (1987). Knowledge and Competence as Strategic Assets, in David J.
Teece (ed11) Winnie:- Cambridge, MA. Ballinger, 159-184.

Winter, M. Fitzgerald, M. Heck, Haynes, G. & Danes, S. (1998). Revisiting the
Study of Family Businesses: Methodological Challenges, Dilemmas, and Alternative

Approaches EamﬂxBusinessReximJLﬂ). 239- 2521

Yamaguchi, K (1994). Autonomy in Exchange Networks and the Structural
Determinants of Autonomy: a Rational Choice Theory. Association Paper. International
Sociological Association.

Young, E.C. & Welsh, H. P. (1983.). Information Source Selection Patterns as

Determined by Small Business Problems MW 42-
49.

Zander & Ko gut (1995). Knowledge and the Speed of the Transfer and Imitation
of Organizational Capabilities. An Empirical Test WW1), 76-92.

165

 

 

In1711117ij:Mlllﬂiﬂllﬂﬂlﬂll