Factors Affecting Trust in Market Research Relationships-Moorman-1993

Factors Affecting Trust in Market Research Relationships
Author(s): Christine Moorman, Rohit Deshpandé and Gerald Zaltman

Source: Journal of Marketing, Vol. 57, No. 1 (Jan., 1993), pp. 81-101
Published by: American Marketing Association
Stable URL: http://www.jstor.org/stable/1252059 .
Accessed: 12/04/2013 07:51
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
American Marketing Association is collaborating with JSTOR to digitize, preserve and extend access to
Journal of Marketing.
http://www.jstor.org
This content downloaded from 161.6.94.245 on Fri, 12 Apr 2013 07:51:47 AM

All use subject to JSTOR Terms and Conditions
Christine Moorman, Rohit Deshpande, & Gerald Zaltman
Factors Affecting Trust in Market

Research Relationships
Building on previous work suggesting that trust is critical in facilitating exchange relationships, the au-
thors describe a comprehensive theory of trust in market research relationships. This theory focuses on
the factors that determine users' trust in their researchers, including individual, interpersonal, organiza-
tional, interorganizational/interdepartmental, and project factors. The theory is tested in a sample of 779
users. Results indicate that the interpersonal factors are the most predictive of trust. Among these factors,
perceived researcher integrity, willingness to reduce research uncertainty, confidentiality, expertise, tact-
fulness, sincerity, congeniality, and timeliness are most strongly associated with trust. Among the re-
maining factors, the formalization of the user's organization, the culture of the researcher's department
or organization, the research organization's or department's power, and the extent to which the research
is customized also affect trust. These findings generally do not change across different types of dyadic
relationships.
THE use of information has been identified as a tion (Moorman, Zaltman, and Deshpand6 1992). Trust
source of a firm's market orientation (Kohli and is important to research relationships because, among
Jaworski 1990) and sustainable competitive advantage other things, users must frequently rely on market re-
(Day 1991; Glazer 1991; Porter and Millar 1985). One search to design and evaluate marketing strategies,
factor that distinguishes firms that merely possess in- though they are often unable to evaluate research
formation from those that use information is the level quality. Hence, being able to trust researchers to en-
of trust users have in producers of information. Trust sure quality and to interpret implications correctly for
has been found to influence the perceived quality of the firm is critical to the user's reliance on research
user-researcher interactions, the level of researcher in- in decision making.
volvement, the level of user commitment to the re- Despite the importance of trust, scholarly inquiry
lationship, and the level of market research' utiliza- on the topic has been hampered in two ways. First,
very little academic research has attempted to docu-
ment empirically the factors that affect trust in mar-
ChristineMoorman is AssistantProfessorof Marketing,
Graduate
School keting relationships (cf. Anderson and Weitz 1990;
of Business,University of Wisconsin-Madison. RohitDeshpande is
Professorof Marketing, AmosTuckSchoolof Business Dwyer and Oh 1987). In fact, no study has attempted
Administration, to
Dartmouth College.Gerald Zaltman is JosephC.WilsonProfessor of develop a theoretical framework of factors that in-
Marketing,Graduate Schoolof BusinessAdministration, Harvard Uni- fluence trust in research relationships. Second, re-
The
versity. first
author thanksthe Graduate of
School,University Wis- search has not systematically distinguished trust from
consin-Madison, the secondauthorthanksthe TuckAssociates Pro- related factors. For example, Sullivan and Peterson
gram,andall threeauthorsthankthe Marketing ScienceInstitute
for (1982) assess trust by measuring sincerity, caution,
supportingtheresearch, as wellas JonAustin,BobDwyer, AjayKohli,
andRobert Spekman fortheircomments on a previous
versionof the effort in establishing a relationship, equality, goal
manuscript,andJill Orumfor her assistancein datacollectionand congruence, consistency, and expectations of coop-
manuscriptpreparation. eration. Crosby, Evans, and Cowles (1990) assess trust
'We use the term "market" research rather than "marketing" re-
search throughout to refer to the information that is collected rather research information though we consider relationships between var-
than the functional area or department. Hence, our focus is on market ious users and marketing researchers.
Journal of Marketing
Vol. 57 (January 1993), 81-101 FactorsAffectingTrust/ 81

by measuring sincerity, competitive behaviors, hon- must rely on researchersto give "credence"to or en-
esty, and beliefs about informationsharing. As we ar- sure a level of informationquality. Other users are
gue, some of these dimensions are viewed more ap- unable to interpretresearchfindings and assess their
propriately as factors that influence trust than as implications, which means they must rely on re-
components of trust itself. searchersto performsuch interpretivefunctions. Fur-
This article develops a theory about factors af- thermore,users who rely on researchersfor scanning
fecting user trust in marketingresearchers.After de- and informationcollectionare vulnerablebecausetheir
scribingour conceptualizationof trust,we presentthis knowledge of the environmentdepends, to a large ex-
theory and formal hypotheses. We then reportthe re- tent, on their researchers'efficacy. Also increasing
sults of an empirical study designed to test the hy- the extent to which users are vulnerableto researchers
potheses. Finally, we discuss the implications and is the fact that researchersoften provide information
limitations of our study and offer suggestions for fu- that is used to evaluate users' decisions (Perkinsand
ture research. Rao 1990; Zaltmanand Moorman 1989). Finally, in
relationshipsbetween users and researchersin an ex-
ternalresearchorganization,users must share propri-
Trust etary informationabout future marketing strategies,
Trust is defined as a a willingness to rely on an ex- placing themselves at the mercy of researchers'prud-
change partner in whom one has confidence (Moor- ence in maintainingconfidentiality.
man, Zaltman, and Deshpande 1992). This definition
spans the two general approachesto trust in the lit-
erature(see also Dwyer and Lagace 1986). First, con- Theoretical Relationships
siderableresearchin marketingviews trustas a belief, A model of the antecedentsand consequencesof users'
confidence, or expectation about an exchange part- trust in marketingresearchersis illustratedin Figure
ner's trustworthinessthat results from the partner's 1. A variety of factors that affect trustare shown, in-
expertise, reliability, or intentionality(Anderson and cluding individual, interpersonal,organizational,in-
Weitz 1990; Blau 1964; Dwyer and Oh 1987; Pruitt terdepartmental/interorganizational, and project fac-
1981; Rotter 1967; Schurr and Ozanne 1985). Sec- tors. Trust,in turn, influencesa numberof relationship
ond, trust has been viewed as a behavioralintention processes (only trust is shown), which affect the ex-
or behavior that reflects a reliance on a partnerand tent to which marketresearchis used. Trust and re-
involves vulnerabilityand uncertaintyon the part of searchutilizationcan, in turn,feed back to affectusers'
the trustor(Coleman1990; Deutsch 1962; Giffin 1967; perceptionsof researchers'characteristics.We do not
Schlenker,Helm, and Tedeschi 1973;Zand 1972). This formally investigate this feedbackeffect. Instead, our
view suggests that, without vulnerability,trust is un- focus is on the factors affecting trust.
necessary because outcomes are inconsequentialfor
the trustor.It is consistent with Deutsch's (1962) def- The Role of Main-Effect Characteristics in
inition of trust as "actionsthat increase one's vulner- Trust: Individual User Characteristics
ability to another," which Coleman (1990, p. 100) Two related characteristics,user job experience and
suggests might include "voluntarilyplacing resources user firm experience, are investigatedfor their influ-
at the disposal of anotheror transferringcontrol over ence on user trustin researchers.Experiencerefers to
resources to another." This view also suggests that the user's tenure in the job or with the firm. Users
uncertaintyis critical to trust, because trust is unnec- with lower levels of experienceareexpectedto be more
essary if the trustorcan control an exchange partner's willing to trust researchersbecause of their lack of
actions or has complete knowledge about those ac- company, marketing, or research knowledge. Expe-
tions (Coleman 1990; Deutsch 1958). rienced users, in contrast, are likely to have more
We argue that both belief and behavioralintention knowledge and confidence in their own ability to use
components must be present for trust to exist. Ac- research and to manage relationships (McDaniel,
cordingly, a person who believes that a partneris Schmidt, and Hunter 1988). These qualitiesmake ex-
trustworthyand yet is unwilling to rely on thatpartner perienced users less likely to rely on researchersin
has only limited trust. Further,reliance on a partner
relationships.Hence:
withouta concomitantbelief aboutthatpartner'strust-
worthinessmay indicatepower and control more than Hi: User trust in researchers is higher when user (a) job
it does trust. experience or (b) firm experience is lower.
In the context of market research relationships,
vulnerabilityand uncertaintyarise for several reasons. Researcher Interpersonal Characteristics
Many users, for example, are unable to evaluate the Interpersonalcharacteristicsrefer to a broadset of re-
quality of research services, which means that they searcher qualities that are demonstrated in user-
82 / Journalof Marketing,
January1993

FIGURE 1
Theoretical Framework
Ant----ee--nt --roest-------------
,Antecedents - Processes e. Outcomes
Antecedents _- Processes Outcomes
--___-___-----------------------------------------------
Individual User Characteristics

. ActualJob Experience
ActualFirmExperience
PerceivedResearcherInterpersonalCharacteristics
-- ResearchAbilities&Motivations
* NonresearchAbilities&Motivations
PerceivedUserOrganizationalCharacteristics Utilization
UserTrust of Market
L_ * OrganizationalStructure in the ~
* OrganizationalCulture --R sear Research
Researcher Information
* UserLocation
Perceived Interorganizational /Interdepartmental

Characteristics
* UserOrg/ Dept Power
? ResearcherOrg/ Dept Power
? Researcher Org / Dept Culture
? Researcher Location
PerceivedProjectCharacteristics
-._ * Importance Level
* CustomizationLevel
- Currentexchange
---- Feedback to future exchanges
researcher interactions. One set of interpersonal char- is a key factor affecting trust, and that it should be
acteristics involves a researcher's motivation and abil- distinguished from technical competence in two ways
ity to engage in certain types of research activities. (see also Root and Kinnear 1991). First, whereas it is
Another set involves nonresearch motivations and generally understood that technical errors can happen
abilities that reflect a researcher's general orientation honestly, violations of trust are generally viewed as
toward a user. This orientation may be exhibited in errors of commission. Second, technical competence
greater sensitivity toward, a willingness to protect, or can be observed and assessed much more readily than
acting in earnest toward a user. can acts based in trust. Given this distinction, exper-
tise is theorized to encourage trust because a user's
Research abilities and motivations. Perceived ex- reliance on a researcher is highly dependent on his or
pertise is a researcher's perceived knowledge and her appraisal of the researcher's ability to plan and
technical competence. Investigations into the persua-
implement research activities. Research by Crosby,
sive effects of source credibility have demonstrated Evans, and Cowles (1990) supports this view by
that highly trustworthy and expert spokespersons in-
showing that perceived expertise is a significant pre-
duce more positive attitudes toward the ideas they ad- dictor of trust (see also Busch and Wilson 1976).
vocate than do spokespersons who are perceived as Perceived willingness to reduce research uncer-
less trustworthy and expert (Giffin 1967; Hovland, tainty is a researcher's perceived motivation to inter-
Janis, and Kelley 1953). Research has also found this pret ambiguous research findings. Zaltman and Moor-
effect to be more likely when the message recipient man (1988) suggest that researchers who are willing
is negatively oriented toward the issue (Sternthal, to reduce research uncertainty by reflecting creatively
Dholakia, and Leavitt 1978), a situation that mirrors on their experience provide important value-added
the skepticism many users have for research. services for users. Barabba and Zaltman (1991) fur-
Extending prior research, we argue that expertise ther suggest that the interpretation, extrapolation, and
FactorsAffectingTrust/ 83

creation of meaning from research data constitute the cause users can expect them to adhere to higher stan-
primary value-adding function that research performs. dards and thereby remain more objective throughout
Furthermore, Blattberg and Hoch's (1990) findings that the research process.
individuals who use both marketing models and in- Perceived dependability is a researcher's per-
tuition make better marketing decisions than do in- ceived predictability (Rempel and Holmes 1986). Other
dividuals who use models or intuition support the im- researchers have distinguished between predictability
portance of judgment in the interpretation of research and dependability, claiming that an individual's ac-
findings. These findings and suggestions follow from tions are regarded as "predictable" whereas an indi-
the fact that research studies rarely provide all of the vidual is regarded as "dependable" (Rempel, Holmes,
information needed for important decisions and that and Zanna 1985). Used in the latter way, dependa-
every research approach has limitations, particularly bility has some basis in other interpersonal anteced-
in terms of how the data can be analyzed and inter- ents, such as sincerity and integrity (see Johnson-George
preted. Therefore, because research often creates un- and Swap 1982 for a similar distinction). However,
certainty that users must reduce in order to make de- dependability that is based in a sense of predictability
cisions, users trust researchers who display a willingness (as used in our research) is distinct from these higher
to assist in this process. This discussion suggests: order interpersonal qualities. Trust increases with de-
H2: User trustin researchersis higherwhenresearcher(a) pendability as users come to rely on the predictability
expertiseor (b) willingnessto reduceuncertaintyis and consistency of researcher actions. High variance
perceivedto be higherratherthanlower. behavior, in contrast, reduces trust.
Perceived collective orientation is a researcher's
Nonresearch abilities and motivations. Perceived perceived willingness to cooperate with users (Zaltman
sincerity is the extent to which a researcher is per- and Moorman 1988). Research has found that indi-
ceived to be honest and to be someone who makes viduals are more willing to commit to another party
promises with the intention of fulfilling them (Larzeleve if that party is believed to be cooperative as opposed
and Huston 1980). Research suggests that when a to competitive or individualistically oriented (Anderson
source's past communications are truthful, receivers and Weitz 1990; Pruitt 1981). Macneil (1980) refers
are more likely to rely on current communications from to this orientation as a flexibility that partners allow
that source (Gahagan and Tedeschi 1968; Schlenker, one another in relationships. Given this orientation,
Helm, and Tedeschi 1973). As Schlenker and his we expect trust to follow.
coauthors state, "A promiser who did not back up his Perceived tactfulness is the level of etiquette a re-
words with corresponding deeds soon would be dis- searcher displays during exchanges with users. Tact
trusted" (p. 420). Other research has suggested that is especially important in communicating research
sincerity is a subdimension of trust (see Crosby, Evans, findings that do not meet users' expectations or that
and Cowles 1990; Kaplan 1973; Rotter 1967). Con- could be embarrassing, because users strongly prefer
sistent with the first view, we believe that sincerity is research that supports the status quo or confirms prior
better described as a determinantof trust, because when beliefs (Deshpande and Zaltman 1982, 1984; Perkins
users sense that researchers are sincere or "truth tell- and Rao 1990; Root and Kinnear 1991). If users can-
ers" (Zaltman and Moorman 1988), they extend trust not count on researchers to have a sense of etiquette
because doing so lessens the vulnerability and uncer- when they discover bad news, researchers will not be
tainty associated with research relationships. trusted.
Perceived integrity is a researcher's perceived un- Perceived timeliness is a researcher's perceived
willingness to sacrifice ethical standards to achieve in- efficiency in responding to user needs. Relatedly,
dividual or organizational objectives. Past research has Zeithaml, Parasuraman, and Berry (1990) describe re-
demonstrated an empirical linkage between integrity sponsiveness as a factor that affects consumers' per-
and trust. For example, Butler and Cantrell (1984) re- ceptions of service quality. Austin (1991) also notes
port that integrity is a significant determinant of sub- the importance of timeliness to user satisfaction with
ordinate trust in superiors. In the context of research and trust in researchers and suggests that it involves
relationships, Hunt, Chonko, and Wilcox (1984, p. paying bills on time, sending requested information
312) suggest that threats to integrity standards all have or materials in a timely manner, and providing feed-
"the common theme of deliberate production of dis- back within a reasonable time period. Finally, time-
honest or less-than-completely-honest research," in- liness has been reported to be correlated positively with
volving falsifying figures, altering research results, manager satisfaction and the utilization of information
misusing statistics, or misinterpreting the results of a systems (Bailey and Pearson 1983).
research project to support a predetermined personal Perceived confidentiality is the researcher's per-
or corporate point of view. Consequently, researchers ceived willingness to keep proprietary research find-
who demonstrate integrity are likely to be trusted be- ings safe from the user's competitors. Other research
84 / Journalof Marketing,January1993

has demonstratedthe importanceof confidentialityto reaucracy on trust. One view suggests that bureau-
trust and related exchange processes (Bailey and craticallyarrangedorganizationsproducetrustthrough
Pearson 1983; Hunt, Chonko, and Wilcox 1984; structuralandproceduralcontrols,especiallywhen other
Zeithaml, Parasuraman,and Berry 1990). For ex- person-based and process-based modes are absent
ample, the perceived confidentialityof disclosures in- (Shapiro 1987; Zucker 1986). The other view is that
creases users' perceptions of counselors' credibility bureaucratic structures established to ensure effi-
(Corcoran1988), increasesthe willingness to relate in ciency (Arrow 1974) reduce the likelihood of trustin
social supportgroup settings (Posey 1988), and is a organizationalrelationships. This view is supported
criticalcomponentof social exchange and trustamong by research noting that bureaucraticstructuringin-
community members (Aguilar 1984). creases opportunism(John 1984) and that managers
Perceived congeniality is the extent to which a re- deprived of decision-making authority have corre-
searcher is perceived to be friendly, courteous, and spondingly less trustin their organizations(Hrebiniak
positively disposed toward the user. This dimension 1974).
has also been linked to satisfaction(Ives, Olson, and Given that we view trust as developing from in-
Baroudi 1983) and perceptions of service quality terpersonalrelationships, organizationalbureaucrati-
(Zeithaml, Parasuraman,and Berry 1990). Though zation is expected to reduce trust because it discour-
congeniality may not always be necessary to establish ages interpersonalrisk-taking, including displays of
good workingrelationships(Crouchand Yetton 1988), uncertaintyand vulnerability(Fox 1974). A bureau-
it seems reasonable that, given the intangible nature cratic environmentalso discouragesflexibility toward
of researchexchanges, users would be likely to make exchangepartners,thus reducingtrust(Johnand Martin
attributionsabout researchers'trustworthinesson the 1984). As Dwyer, Schurr,and Oh (1987, p. 349) note,
basis of a variety of cues. One simple cue, which re- "
... .reliance on bureaucraticstructuringwill affect
flects a minimal level of user orientation, is a re- trustworthinessadversely in that unilateraldecisions,
searcher's courtesy to the user. Being courteous, of centralized authority, and operations 'strictly by the
course, is not expected to be sufficient to maintain book' do not communicatedesired coordination,rec-
trustwithoutother researcherqualities;however, con- iprocity, or commitmentto the relationship."
geniality should contributeto trust. Perceived organizational complexity is the degree
Given that the aforementionedqualitiesreducerisk of formal structuraldifferentiationwithin an organi-
and uncertainty,researchersdisplaying attitudes and zation (Blau and Schoenherr1971). Price and Mueller
behaviors that reflect these abilities and motivations (1986) note that highly complex organizations are
should be able to increase users' beliefs in their trust- characterizedby many occupationalroles, divisions,
worthiness and increase users' willingness to rely on
departments,levels of authority,and operatingsites.
them. Hence, we formally hypothesize:
Complexity can be found in the horizontaldifferen-
H3: User trustin researchersis higherwhenresearcher(a) tiation of divisions and departments,the vertical dif-
sincerity,(b) integrity,(c) dependability,(d) collec- ferentiationof levels of authority,and the spatial dif-
tive orientation,(e) tactfulness, (f) timeliness, (g)
ferentiation of operating sites. Complexity should
confidentiality,or (h) congenialityis perceivedto be
higherratherthanlower. reduce trustin researchrelationshipsfor two reasons.
First, greater complexity may mean that researchers
User OrganizationalCharacteristics and usersare not physicallyclose to one another,which
interfereswith their ability to build trust in relation-
User's organizational structure. Two aspects of firm
structure, bureaucratizationand complexity, are re- ships. Additionally, complexity may threaten trust-
lated to trust. building because of the likelihood of greaterdissim-
Perceived organizational bureaucratization is the
ilarities in beliefs and norms as firms add more
divisions, departments,and roles. Hence, we hypoth-
degree to which a user views his or her organization esize:
as managed primarily through formalized relation-
ships and a centralized authority (John 1984). For- H4: User trustin researchersis higherwhen the user or-
malization is the degree to which rules define orga- ganization's(a) formalization,(b) centralization,or
nizationalroles, authorityrelations, communications, (c) complexity is perceived to be lower rather than
normsand sanctions, and procedures(Hall, Haas, and higher.
Johnson 1967; John and Martin 1984), including the User's perceived organizational culture. Organ-
flexibility managershave when handling a particular izationalculture, defined as "thepatternof sharedval-
task (Deshpande 1982). Centralizationis the degree ues and beliefs that help individuals understandor-
of delegation in organizationaldecision-making au- ganizational functioning and that provide norms for
thority (Aiken and Hage 1968). behavior in the organization"(Deshpandeand Webs-
There are two general views of the effect of bu- ter 1989, p. 4), is theorized to affect trust according
FactorsAffecting
Trust/ 85

to the type of culture present. Deshpande, Farley, and higher level positions, in contrast, may be able to ac-
Webster (1992) assess four modal types of organiza- complish their goals without trusting relationships. We
tional cultures: clans, adhocracies, hierarchies, and hypothesize:
markets.
H6: User trust in researchers is higher when users hold
Clans emphasize cohesiveness, participation, and organizational positions with lower rather than higher
teamwork. Members of organizations with strong clan levels of authority.
cultures should be willing to trust their researchers be-
cause the cultural norm is the establishment and main-
tenance of successful working relationships (Ouchi Interorganizational/lnterdepartmental
Characteristics
1980). Moreover, clan members are generally viewed
as part of an extended family, a situation ideal for the When a relationship involves an internal user and an
creation and maintenance of trusting relationships. external researcher, the relationship is interorganiza-
Adhocracies emphasize entrepreneurship, creativity, tional, whereas if the relationship involves an internal
and adaptability. Hence, members of organizations with user and an internal researcher, the relationship is in-
strong adhocracies should be willing to trust research- terdepartmental.Three general categories comprise this
ers because the norms emphasize tolerance and flex- set of characteristics. One set involves users' percep-
ibility (Mintzberg 1979) and because such cultures are tions of power in organizational or departmental re-
likely to place a premium on employee skills (includ- lationships; the other two involve users' perceptions
ing research skills) that offer stability in dynamic set- of two research organizational or departmental char-
tings. Hierarchies emphasize order, uniformity, and acteristics.
efficiency. Lower levels of trust are expected because Perceived power in the relationship. User de-
such cultures provide sufficient stability and control
partmental/organizational power generally is due to
to make trust less necessary or salient. Markets em- the user's status as research initiator and buyer whereas
phasize competitiveness and goal achievement. Trust researcher organizational/departmental power is most
is expected to be lower in these cultures because per- likely to emanate from the researcher's specialized ex-
sonal relationships (i.e., team membership) are viewed periential, informational, or technological assets. Pre-
as instrumental and are used opportunistically vious research on power in relationships has led to
(Williamson 1975). We hypothesize: two general findings. First, power breeds trust (Frost,
H5: User trust in researchers is higher when the user or- Stimpson, and Maughan 1978; Sullivan and Peterson
ganization is perceived to be a (a) stronger ratherthan 1982), functional conflict (from the perspective of the
weaker, clan or adhocracy culture or (b) weaker rather firm having power; Anderson and Narus 1990), and
than stronger, hierarchical or market culture.
respect from exchange partners (Anderson, Lodish,
and Weitz 1987). Second, power creates conflict (from
User organizational location. Research indicates the perspective of the firm against which the power is
that a manager's position within an organization af- used; Anderson and Narus 1990) and mistrust of the
fects a variety of interpersonal and decision-making
powerful parties' intentions (Anderson and Weitz 1990).
processes. For example, organizational position has Following Anderson and Narus (1990), we predict
been reported to influence a manager's knowledge about that perceived power will affect trust differently de-
organizational phenomena (Walker 1985), self-esteem pending on which organization or department has that
(Tharenou and Harker 1982), leadership style (Adams power. Specifically, given that research organizations
1977), and perceived fairness (Stolte 1983). More or departments have achieved power because of their
prominent organizational positions also have been specialized skills, trust is likely to follow as users rely
shown to be related to trust by lower status employees on those skills in the design and evaluation of mar-
(Tjosvold 1977) and general trust in the organization, keting strategies. Moreover, when researchers have
including the degree to which it is perceived as be- power due to specialized skills, users may lack the
nign, cooperative, or consistent (Hrebiniak 1974). expertise to evaluate research adequately and hence
Conversely, other research has shown no linkage be- the use of research may involve uncertainty and vul-
tween organizational location and trust (Zalesny,
nerability-ideal conditions for trust to emerge (Bar-
Farace, and Kurcher-Hawkins 1985). ber 1983). In contrast, perceived user organization/
Given the view of trust in this research-as in-
department power may be used to direct researchers'
volving a willingness to rely on another party-a the- behaviors, which reduces users' need to trust in the
oretical position that departs from the literature is of-
relationship. We formally hypothesize:
fered. We argue that users with lower levels of
authority, in general, are less able to influence the be- H7: User trust in researchers is higher when the perception
is that (a) research organizational/departmental power
haviors of others in the firm. Hence, they develop more
is higher ratherthan lower or (b) user organizational/
trusting relationships to accomplish goals. Users in departmental power is lower rather than higher.

User's perceptions of a researcher's culture. Past Perceived customization of research project.
researchhas demonstratedthat perceptionsof an ex- Heide and Miner (1991) reportthat customizationin-
changepartner'scultureaffect trustin that partner(see creases the generationof relationship-specificassets,
Peterson and Shimada 1978; Sullivan et al. 1981). which other research has linked to increased collab-
Hence, we believe that a user's perceptions of a re- orationin (Williamson1975), continuationof (Levinthal
searcher'sculturewill influence user trustin research- and Fichman 1988), and positive evaluation of
ers. As with usercultures,we expectthe sense of family (Surprenantand Solomon 1987) relationships. Like
in clans and the tolerance and flexibility of adhocra- other customized exchanges, custom researchentails
cies to be associatedwith strongertrustin researchers. greaterlevels of risk for users than does noncustom-
Conversely, a user will have lower trustin researcher ized researchbecause users must depend on research-
cultures that are hierarchies or markets. Given the ers to make relationship-specific decisions and in-
presence of different subcultureswithin a single or- vestments throughout the project. This situation
ganization (Deshpande, Farley, and Webster 1992; produces a more likely context for the development
Deshpande and Webster 1989; Smircich 1983), this of trustthandoes a syndicatedresearchprojectin which
effect is expected regardless of whether the unit of less researcherdiscretionis necessaryand, hence, there
is less opportunityfor user uncertaintyand vulnera-
analysis is two organizationsor two departments.
bility. We hypothesize:
H8: User trust in researchers is higher when a user per-
ceives a researcher's organizational/departmental
Hlo: User trust in researchers is higher when the project
cul-
is perceived to be more ratherthan less (a) important
ture to be a stronger clan or adhocracy culture, or to
or (b) customized.
be a weaker hierarchical or market culture.
The Effects of Individualand Organizational

User's perception of a researcher's location. Moderators on H1-Ho1
Researcherslocated in higher levels of the organiza-
tion are morelikely to influencedecisionmakingwithin Differences between users and researcherscan facil-
the organizationthan are researchersat lower levels. itate or inhibit the role of trust in research relation-
This capability increases researcherperceived credi- ships. Following Moorman,Zaltman, and Deshpande
bility and power which, accordingto the theory pro- (1992), we theorizeaboutthe effects of two individual
differences and one organizationaldifference on the
posed up to this point, increases user trust in them.
hypothesized relationships. As we consider this part
H9: User trust in researchers is higher when the research- of our researchto be exploratory,the hypotheses are
er's organizational location is perceived to be higher more speculative. We do, however, offer formal hy-
rather than lower.
potheses (aside from the null) as a way of furthering
researchon these moderatingeffects.
Individual differences: members of the same and
Project Characteristics
different communities. Previous work has described
Perceived importance of research project. A research community differences between users and providers
projectmight be importantfor a varietyof reasons. It of knowledge(Caplan,Morrison,and Stambaugh1975)
may, for example, involve a strategicdecision (e.g., and how these differences affect knowledge use and
product addition or deletion), portend importantfi- other outcomes (Crosby, Evans, and Cowles 1990;
nancial or competitive considerationsfor the firm, or Deshpandeand Zaltman1984, 1987). We examinetwo
hold importantcareer implications for an individual. general types of communitydifferences:(1) when the
Higher project importancelevels are expected to in- user and provider are both researchersversus when
creaseusertrustin researchersbecausesuch levels raise one is a manager, and (2) when the user and the pro-
users' vulnerabilityin the relationship,which compels vider are both in marketingversus when one is a non-
them to work hard at developing effective research marketer(e.g., an engineer).
relationships and causes increased reliance on re- In general, our theorizingaboutthe effects of sim-
searchers. Unimportantprojects involve less vulner- ilarities and differences is based in the principle of
ability, which means that trustis correspondinglyless salience in social psychological research. Salience is
salient in these relationships.2 typically used to refer to stimuli that are prominentin
relationto a particularcontext because of their innate
qualities, the situational environment, or a person's
2We acknowledge that users are likely to choose researchers they
trust for important projects. However, if we control for prior trust in
priorknowledge or frame of reference. Priorresearch
the current exchange, users are also likely to trust researchers more suggests that individuals tend to pay more attention
when they are engaged in important, as opposed to unimportant, proj- to, to attributemore causality to, and to be more ca-
ects.
pable of retrieving salient stimuli (Taylor et al 1979;
Factors Trust
Affecting / 87

Tversky and Kahneman 1974). Applied to the issue Method
of how dyadic conditionsaffect the importanceof var-
ious characteristicsto trust, we suggest that charac- Measure Development
teristics that are shared between dyad members are On the basis of the construct definitions, we either
less salient to the partnersand, hence, have a weaker developed measures or borrowedthem from past re-
influence on how the relationshipoperates. In con- search (see Appendix) and then performedtwo pre-
trast, characteristicsthat are not shared are more sa- tests. The first pretest assisted in item refinement. A
lient to the partnersand, hence, have a strongerim- list of defined measuresand items for those measures
pact on the relationship. was mailed to 10 academiciansand 10 practitioners.
Therefore, users working with researchers from Following Churchill (1979), we requested recipients
different communities (i.e., manager-researcherand to assign each item to the measure they thought ap-
marketer-nonmarketer dyads)are expectedto rely more propriate, as well as to note when they thought the
heavily on interpersonalcharacteristicsin the decision item could be representedby more than one measure.
to trustbecausethis unsharedinformationwill be more The second pretest, which determinedthat trust
salient in the exchange process. These interpersonal could be discriminatedadequatelyfrom the interper-
characteristicsare not expectedto be salientwhen dyads sonal characteristics, involved a sample of 50 ran-
contain members of the same community (i.e., re- domly selected membersof the sample frame. Recip-
searcher-researcher and marketer-marketer dyads), as ients were asked to evaluate their most recent
both parties are likely to display these characteristics relationship with either an internal or external re-
in their exchanges. The effects of the remainingchar- searcher. Responses (54%) were analyzed for dis-
acteristics (e.g, organizationalor project factors) on criminantvalidity with relatedconstructsby means of
trust are not sensitive to these communitydifferences an exploratoryfactor analysis (principalcomponents
because they do not vary systematicallyacross similar analysis with a varimaxrotation)and for reliabilityby
and dissimilar dyads. Hence, we hypothesize: examining item-to-total correlations. Results indi-
cated that trust could be differentiatedfrom related
H,1: The relationships between trust and interpersonal constructsand that it was reliable.
characteristics are stronger for relationships involv-
ing dissimilar parties than for relationships involving Procedure
similar parties. A sample of 1680 research users was generated by
phoningeach firm and division on the AdvertisingAge
Organizational differences: intraorganizational and 1990 list of the top 200 advertisers. The users con-
interorganizational relationships. Given that two de- sisted of (1) marketing managers (e.g., vice presi-
partmentsare more likely to be similarto one another dents of marketing, marketingdirectors, or product
if they are within a single firm than if they are in two and brand managerswho evaluated either internalor
different organizations, it follows that organizational externalresearchers),(2) marketingresearcherswithin
characteristics should be more salient, and hence a firm who evaluated external researchers, and (3)
strongerpredictorsof trust, in interorganizationalre- nonmarketingmanagers(e.g., engineers, productde-
lationships than in intraorganizationalrelationships. velopment managers, and manufacturingmanagers)
We expect no differences, however, in the effects of who evaluated internalresearchers.
Each sample memberwas mailed a cover letter, a
individual, interpersonal, interorganizational/inter-
departmental, and project characteristics on trust questionnaire, and a brief author profile. The cover
because such characteristicsare not expected to vary letterexplained the purposeof the researchand prom-
ised a summaryof the results to all who returnedtheir
systematically in interorganizationaland intraorgani- business cards with completed questionnaires. The
zational dyads.3 Hence, we hypothesize:
questionnaires,though identical in structure,directed
H12: The relationships between trust and the organiza- respondents to evaluate different types of relation-
tional characteristics are stronger for interorganiza- ships. One new dollar bill was affixed to the cover
tional relationships than for intraorganizationalrela- letter as an advance token of appreciation. Finally,
tionships. respondentswere asked for the name and address of
one nonmarketingmanager who used marketingre-
search. These recommendationsyielded 188 eligible
3Though clearly the interorganizational/interdepartmental charac- respondents,increasingthe sample size to 1868. Three
teristics will vary depending on whether the relationship is interor- weeks after the first mailing, randomlyselected non-
ganizational or intraorganizational, in this case the variability is a
function of the nature of the variable and its conceptualization rather respondentswere phoned, remindedof the survey, and
than a condition that reflects differences that can be examined em- encouragedto complete and returnthe questionnaire.
pirically. Two weeks after the phoning, a second mailing was
January1993

sent to all nonrespondents.A similar procedurewas correspondingto the number of hypothesized mea-
used for the recommendednonmarketers,with the ex- sures and trust (e.g., H3, n = 9). In each case, the
ception that the phone calls were not necessary be- predicted number of constructs had a superiorfit as
cause the response rate was above average after the indicated by standardgoodness-of-fit indices and by
first mailing. the low cross-loadings between constructs.5 These
Table 1 summarizesthe samplesand responserates findings provide evidence that the predicted number
for each of the four dyads. As noted there, some re- of measuresbest representsthe underlyingdata struc-
cipients returnedthe questionnairebecause it was in- ture, and that the factors theorizedto affect trust are
appropriatefor their organizationor experience,which distinctfrom the measureof trustitself. Table 2 notes
reduced the sample to 1719. Of these eligible indi- that fitted models retainedall measuresat acceptable
viduals, 779 (45.3%) responded. The response rates reliability levels.
varied across dyads from 35.8% for the internalmar-
keting manager-externalresearcherdyad to 58.0% for General Theory-Testing Approach
the internalresearcher-externalresearcherdyad. Such We tested the hypotheses by using a single regression
variance across types of respondentshas been noted model in which all main and interactioneffects were
previously(Deshpandeand Zaltman1984). Earlierand enteredsimultaneouslyas predictorsof the dependent
later responses were comparedon key variables and
variable, trust. Prior to that test, we created several
no differences were found. sets of variablesby using dummy and effects coding.
Measure Purification First, researcherand user organizationallocationswere
The Appendix contains the measures and items. Fol- dummycoded so that each level was a variablecoded
a 1 and all other levels were coded 0. Second, vari-
lowing Gerbingand Anderson(1988), we purifiedour ables representingthe main effects of the individual
proposed measures by assessing their reliability and and organizational moderatorswere constructed by
unidimensionality.First, we examinedthe item-to-total means of unweighted effects coding (see Cohen and
correlationsfor the items in each of the proposedscales
Cohen 1983).6 Following from a coding of the re-
and deleted items with low correlationsif they tapped
no additionaldomain of interest. Then, we performed
a confirmatoryfactor analysis on items from subsets 5In each case X2 was significantly smaller than competing confir-
matory factor models (A X1 > 3.84).
of theoreticallyrelated measures to assess the extent 6Despite the unbalanced nature of the dyad characteristics in our
to which they reflected a single dimension. The sub- sample (e.g., 276 intraorganizational dyads and 503 interorganiza-
sets correspondedto the characteristicscontainedin a tional dyads), unweighted effects coding was adopted over weighted
effects coding (which would adjust the parameter estimates by group
given hypothesis.4After deletion of items having high size) for two reasons. First, the first three dyads (see Table 1) were
correlations with items for other measures, models evenly sampled at the outset of the study and deviations in response
reflected actual differences in these samples. For example, many IMM-
ranging from one to n factors were estimated, with n EMR and IMM-IMR relationships did not exist in the firms we sam-
pled; therefore the questionnaires were passed on to an IMR who was
4This procedure was performed in subsets because of the problems instructed to complete it by focusing on an EMR relationship. Hence,
associated with purifying large numbers of constructs by confirmatory the eligible sample for dyad 3 increased from the original sample level.
factor analysis and the importance of establishing unidimensionality In the case of dyad 4, we requested that the respondents in the first
only for highly theoretically related measures. three dyads (695 respondents) provide the name of a nonmarketing
TABLE1
Dyadic Relationships and Response Rates
Relationship Original Eligible Number of Response
Type Sample Sample Respondents Rates (%)
All dyads 1868 1719 779 45.3
Dyad 1, IMM-IMR:
internal marketing manager-
internal marketing researcher 560 489 192 39.2
Dyad 2, IMM-EMR:
internal marketing manager-
external marketing researcher 560 480 172 35.8
Dyad 3, IMR-EMR:
internal marketing researcher-
external marketing researcher 560 570a 331 58.0
Dyad 4, INM-IMR:
internal nonmarketing manager-
internal marketing researcher 188 180 84 46.6
aThe eligible sample increase over the original sample is due to the reassignment of
dyad 1 and dyad 2 members to dyad 3.
FactorsAffecting
Trust/ 89

TABLE2
Properties of Purified Measures
Measure Items Range Mean S.D. Alpha
USER TRUST IN RESEARCHER 5 1-7 5.32 1.11 .84
ACTUALJOB EXPERIENCE 1 1-35 8.54 6.87
ACTUALFIRMEXPERIENCE 1 1-30 3.34 3.19
Perceived Researcher Interpersonal Characteristics
EXPERTISE 3 1-7 5.74 1.16 .92
WILLINGNESS TO REDUCEUNCERTAINTY 3 1-7 4.82 1.03 .57
SINCERITY 3 1-7 6.04 .97 .92
INTEGRITY 3 1-7 6.16 .97 .77
DEPENDABILITY 2 1-7 5.16 1.04 .80a
COLLECTIVE
ORIENTATION 3 1-7 5.11 1.06 .70
TACTFULNESS 3 1-7 5.49 1.00 .78
TIMELINESS 3 1-7 5.61 1.12 .79
CONFIDENTIALITY 3 1-7 6.21 1.08 .89
CONGENIALITY 3 1-7 5.88 .96 .77
Perceived User Organizational Characteristics
FORMALIZATION 15 1-4 2.06 .39 .79
CENTRALIZATION 8 1-4 3.05 .40 .73
COMPLEXITY 1 1-4 1.93 .93
CLANCULTURE 4 1-100 21.08 12.74 .87
ADHOCRACY CULTURE 4 1-100 21.79 12.38 .87
HIERARCHICAL CULTURE 3 1-100 25.67 14.80 .81
MARKETCULTURE 4 1-100 30.61 14.25 .87
USER LOCATIONb 1 1-3
Perceived Interorganizational/InterdepartmentalCharacteristics
RESEARCHER ORG/DEPTPOWER 3 1-7 4.00 1.37 .79
USER ORG/DEPTPOWER 3 1-7 3.53 1.64 .87
RESEARCHER ORG/DEPTCLAN 4 1-100 24.56 12.74 .91
RESEARCHER ORG/DEPTADHOCRACY 4 1-100 24.16 12.38 .91
RESEARCHER ORG/DEPTHIERARCHY 4 1-100 22.91 14.80 .91
RESEARCHER ORG/DEPTMARKET 4 1-100 28.19 14.25 .91
RESEARCHER LOCATIONb 1 1-4 - -
Perceived Project Characteristics
IMPORTANCE 2 1-7 6.02 1.02 .74a
CUSTOMIZATION 1 1-7 6.24 1.35
'Pearson correlations.
bVariablesare nominal. Hence, the means and standard deviations are not reported.
spondents' job titles and instructions to focus on tionships (coded -1) and marketer-marketerrelation-
a particular type of provider, RESDYAD consists ships (coded 1). Following from instructionsasking
of researcher-manager relationships (coded -1) respondents to focus on either an external or internal
and researcher-researcherrelationships (coded 1); researcher, ORGDYAD consists of interorganiza-
MKTDYAD consists of nonmarketer-marketer rela- tional relationships (coded -1) and intraorganiza-
tional relationships(coded 1). Finally, given the in-
manager who uses research information. Hence, the resulting 188 tercorrelatednature of the independent variables, a
sample members would appear to be proportional to the actual number
of INM-IMR relationshipsin these firms. Second, it is unclear whether procedurerecommendedby MasonandPerreault(1991)
differences in response rates reflect actual differences in the number was performedto rule out multicollinearityas a source
of these relationships (suggesting that our sample is ecologically ac- of inflatedstandarderrorsfor the parameterestimates.
curate) or represent biases in the willingness to respond to the survey.
A brief followup of nonrespondents in dyads 1 and 2 indicated that
many of the nonrespondents were no longer with the firm or did not
use research in their firms. Therefore, our response rates, though dif- Results
ferent from one another, appear to reflect the types of research rela-
tionships in the large firms in our sample. (We did experiment with
Table 3 contains the coefficients resulting from the
a weighted effects coding scheme to assess the impact on our param- regression model. As suggested there, a user's trust
eter estimates. Few changes occurred in the estimates.) in his or her researcheris significantly affected by 15
90 /Journalof Marketing,
January1993

TABLE 3
Regression Analyses8
Standardized
Independent Variablesb Beta Coefficient t-value
H1: Effect of Individual User Characteristics
ACTUALJOB EXPERIENCE .00 .19
ACTUALFIRMEXPERIENCE -.01 -.45
H2: Effect of Perceived Researcher Research Abilities and Motivations
EXPERTISE .19 6.18*
WILLINGNESS TO REDUCEUNCERTAINTY .32 9.53*
H3: Effect of Perceived Researcher Personal Abilities and Motivations
SINCERITY .15 3.41*
INTEGRITY .62 18.16*
DEPENDABILITY .01 .47
COLLECTIVE ORIENTATION .04 1.32
TACTFULNESS .17 4.73*
TIMELINESS .08 2.31**
CONFIDENTIALITY .26 8.10*
CONGENIALITY -.10 -2.45*
H4: Effect of Perceived User Organizational Structure
FORMALIZATION -.14 -1.85**
CENTRALIZATION .02 .00
COMPLEXITY .02 .66
H5: Effect of Perceived User Organizational Culture
CLANCULTURE .02 .85
ADHOCRACY CULTURE .00 .03
HIERARCHICAL CULTURE -.01 -.26
MARKETCULTURE -.002 -.09
H6: Effect of Perceived User Organizational Location
USER LOCATION(researcher) -.03 -1.04
USER LOCATION(marketing manager) -.004 -.13
USER LOCATION(vice president) .03 1.31
H7: Effect of Perceived User and Researcher Organizational/Departmental Power
RESEARCHER ORG/DEPTPOWER .08 3.45*
USER ORG/DEPTPOWER .01 .39
H8: Effect of Perceived Researcher Organizational/Departmental Culture
RESEARCHER ORG/DEPTCLAN .03 1.17
RESEARCHER ORG/DEPTADHOCRACY .01 .19
RESEARCHER ORG/DEPTHIERARCHY -.05 -1.81**
RESEARCHER ORG/DEPTMARKET -.03 -.75
Hg: Effect of Perceived Researcher Organizational Location
RESEARCHER ORG/DEPTLOCATION(report to brand/product managers) -.29 -2.36**
RESEARCHER ORG/DEPTLOCATION(report to top or general managers) -.01 -.57
RESEARCHER ORG/DEPTLOCATION(report to marketing/sales managers) .02 .82
RESEARCHER ORG/DEPTLOCATION(report to engineers/product -.02 -.72
development managers)
Hl0: Effect of Perceived Project Characteristics
IMPORTANCE -.04 -1.04
CUSTOMIZATION .04 2.01**
HI1:Effect of Individual Moderators on Hi-Ho0c
RESEARCHER TIMELINESSx RESDYAD .07 2.61**
H12:Effect of Organizational Moderators on H1-H10c
USER ORGANIZATION FORMALIZATION x ORGDYAD .06 3.83**
Total R2 = .51
"Tobe conservative,all tests of significanceare two-tailed.
RESDYAD= 1 for researcher-researcherdyads, -1 for researcher-manager dyads
MKTDYAD= 1 for marketer-marketerdyads, -1 for marketer-nonmarketerdyads
ORGDYAD= 1 for intraorganizationaldyads, -1 for interorganizationaldyads
bAll independent variables are perceived by the user, unless otherwise noted.
"Onlythe significant interactions are displayed.
*p < .001.
**p < .05.
***p < .10.

main and interactioneffects. We review these results of the interactions involving marketing orientation
in the order of the hypotheses. differences (MKTDYAD) are significant, indicatinga
The Effects of Main-Effect Characteristics on lack of supportfor H,1. Furthermore,only one of the
Trust interpersonalcharacteristics(timeliness, P = .07, p
< .05) interactswith differences in researchorienta-
As neither of the individual user characteristicsis a
tion (RESDYAD) to predict trust, also indicating a
significantpredictorof user trustin the researcher,Hla,b
is not supported.However, most of the researcherin- general lack of support for H,1. This significant in-
teraction indicates that the timeliness-trustrelation-
terpersonalcharacteristicsare significant. Both per-
ceived researchabilities and motivationssignificantly ship is strongerin researcher-researcher relationships
than in researcher-manager relationships.These find-
predict trust: the researcher'sperceived expertise (3
= .19, p < .001) and perceived willingness to reduce ings also indicate that the relationshipsbetween trust
researchuncertainty(3 = .32, p < .001). These find- and the individual, organizational, interorganiza-
ings supportH2a,b-Many of the nonresearchabilities tional/interdepartmental,and project characteristics
and motivationsare also significantpredictorsof trust, are unaffected by community similarities and differ-
includingthe researcher'sperceivedsincerity(P = .15, ences.
p < .001), integrity (P = .62, p < .001), tactfulness For the interactionsinvolving organizationaldif-
(P = .17, p < .001), timeliness (P = .08, p < .05), ferences (ORGDYAD), results indicate a lack of sup-
and confidentiality(P = .26, p < .001). These results port for H12,which predicts that organizationalchar-
support H3a, H3b, H3e, H3f, and H3g. The researcher's acteristics will be stronger predictors of trust in
perceived congeniality, in contrast, has a significant interorganizationalrelationships.In fact, thereis only
negative effect on trust(P = -.10, p < .001), failing one significant interactioninvolving ORGDYAD and
to supportH3h.Finally, as the researcher'sperceived these characteristics, the effect of user's organiza-
dependabilityand collective orientationare unrelated tional formalizationx ORGDYAD on trust(P = .06,
to trust, H3cand H3dare not supported. p < .05). Examining this interactionby plotting the
Fewer of the organizationalvariableshave an ef- regressionlines underthe two conditionsof the mod-
fect on user trust. User organizationformalizationhas erating variable reveals an interesting result. When
the expected negative effect on trust ((3 = 14, p < formalizationis low, its effect on trust is strongerin
.10), supportingH4a.However, user organizationcen- interorganizationaldyads than in intraorganizational
tralization,complexity, and culturedo not predictuser dyads. However, as formalizationincreases, its effect
trust, failing to support H4b,cand H5a,b. Finally, user on trust is strongerin intraorganizationaldyads. This
location also has no effect on trust, which does not result should be viewed as a contingencyon the main-
support H6. effect relationshipbetweenformalizationandtrustnoted
Forthe interorganizational/interdepartmental char- previously.Finally, ORGDYADdoes not interactwith
acteristics, we find that researchorganization/depart- the individual, interpersonal,interorganizational/in-
ment power predicts user trust (3 = .08, p < .001), terdepartmental,or project characteristics.
which supports H7a. User organization/department
power is unrelatedto user trust, however, so H7bis
not supported.A user's perceptionof researcher'sor- Discussion
ganization/departmentculture is also generally un- The traditionalview of trustadoptedin marketinghas
related to trust, except when the perceived culture is been based on a purely psychological approach.Our
a hierarchy(p = -.05, p < .10), in which case the research complements and extends that view to in-
perception has a negative effect on user trust. These clude sociological theories. Hence, our definition in-
findings partially supportH8b but fail to supportH8a. cludes both a confidence in an exchange partner(the
Finally, results for researcherswho report to brand
and productmanagers(as opposed to higher levels of psychologicalcomponent)and a willingness to rely on
an exchange partner (the sociological component).
the organization)are related negatively to user trust
Confidence and reliance, in turn, indicate the critical
(P = -.29, p < .05), providingsupportfor H9. None roles of uncertaintyand vulnerabilityto trust in rela-
of the other researcherlocations are significantly re-
lated to trust.Finally, projectimportancehas no effect tionships. In particular,we argue that if a trustorhas
on trust and Hloais not supported.However, project complete knowledge about an exchange partner'sac-
customization is a weak predictorof trust (p = .04, tions, is able to control the exchange partner,or has
not transferredcritical resourcesto an exchange part-
p < .05), supporting Hiob.
ner, trust is not necessary in the relationship.
The Effects of Individual and Organizational Given this theoretical foundation, our research
Moderators on H1-H1o further establishes that trust is distinct from related
In general, the effects of the moderatorson the hy- antecedents, including a variety of interpersonal,or-
pothesized relationshipsare weak. For example, none ganizational, and interorganizationalfactors that may
January1993

affect the level of trust in relationships. Distinguish- be useful, especially given the current trend of re-
ing between trustand otherconceptsaccomplishestwo structuringinternalmarket research departmentsand
objectives. First, it assists in defining the nomological subcontracting market research to commercial re-
network of trust and related factors. Second, it pro- search firms. Relatedly, investigating the degree to
vides a preliminaryunderstandingof the conditions which individualand organizationalfactorsjointly in-
underwhich trustis facilitatedor underminedand gives fluence integrity should suggest ways to manage re-
directionto futureresearchabout the factors that vary searchers and their environmentsto foster integrity.
concomitantlywith trust. In the remainderof this sec- Finally, understandingthe types of beliefs held by high
tion we discuss these factors and provide insight into integrityresearchersabout themselves, theirroles, and
their theoreticaland managerialimplications. theirrelationshipswith othersmay provideinsightinto
Our results indicatethattrustmay be more a func- whether low integrity is a product of opportunistic,
tion of interpersonalfactorsthanof individualfactors. instrumentalthought processes or some other set of
These results are consistent with a general research beliefs.
trendfocusing on trustas a productof the relationship Following integrity, the researcher's perceived
between two parties as opposed to a personalitytrait willingness to reduce researchuncertaintyis the next
exhibited by either party. Despite this trend, we be- most importantpredictorof trust. As described pre-
lieve future research would benefit from identifying viously, reducinguncertaintymay involve using fewer
and testing other individual characteristicsthat may dataanalysis skills and more datainterpretationskills.
affect trust levels. For example, an individual's in- Such skills require researchersto utilize their broad
terpersonalorientation-the extent to which a person understandingof the marketplace and research in-
is interestedin and reactiveto otherpeople (Swap and sights to construct explanations about research find-
Rubin 1983)-might make a person more likely to ings. In this process, the researchershares his or her
trust others. tacit knowledge or frame of reference with the user.
Among the perceivedinterpersonalcharacteristics, This frame of reference is likely to be sophisticated
a researcher'sperceived integrity is the most impor- because the researcherwho participatesin the uncer-
tant predictorof trust, indicatingthat users expect re- tainty absorptionprocess is also likely to be more ex-
searchersto adhereto high standardsand to maintain perienced and hence to have an especially useful in-
objectivity throughout the research process. Re- ventory of tacit knowledge (Sternberg and Wagner
searcherintegrity is likely to be reflected in an array 1986). Attemptsto identify researcherqualities facil-
of behaviorsover the researchcycle. These behaviors itating this skill would be useful to firms attempting
can include at the outset the researcher'sclear state- to recruitsuch individuals.Finally, because of the im-
ment about what cannot be accomplishedgiven time, portance of integrity, our results indicate that re-
budget, or other constraints.They might even include searchers must carefully distinguish between uncer-
the researcher'srefusal to undertakea project when tainty absorption and strict data analysis when
he or she feels various constraintspreclude the col- discussing findings with users.
lection of sound data (Barabbaand Zaltman1991). At Researcherconfidentialityis also very important
the end of the process, a researcher'srefusal to par- to trust in research relationships, which is not sur-
ticipate in a skewed interpretationof the data also prising given that informationcan be used to secure
contributes to the development and maintenance of competitive advantage. Future research should ex-
trust. Additionally, users know that research results amine the cues that increase and decreaseperceptions
may be subject to a variety of interpretations,espe- of confidentiality. For example, is confidentialityex-
cially by people in different functional areas. Re- pressed solely in an individualresearcher'scharacter-
searcherintegrityprovidesa stabilizingeffect thatusers istics? To what degree do organizationalor depart-
come to rely on in the reconciliationof competinguser mental norms and procedurescontributeto or detract
judgments. from such perceptions?Do users trust confidentiality
The importanceof integrity to trust suggests sev- contracts signed by external research firms? It may
eral importantresearchquestions. First, given that in- also be interestingto investigate how the absence of
tegrity is likely to be importantbecause users are un- confidentialityaffects the exchanges between parties,
able to evaluateresearchquality, it may be interesting with attentionto dysfunctionalbehaviors(e.g., users'
to assess the degree to which the importanceof in- withholdingof importantbackgroundinformation)that
tegrity is diminishedas a user's ability to evaluate re- may undermine the utilization of research in firms
searchquality increases. Furthermore,futureresearch (Moorman, Zaltman, and Deshpande 1992).
might assess the importanceof individual and orga- This researchsupportsthe generally acceptedidea
nizational factors to integrity. Discovering, for ex- thatexpertiseis an importantfoundationfor trust.Our
ample, the types of organizational cultures, struc- research also highlights, however, that trust is af-
tures, and reward systems that foster integrity would fected by a variety of other factors that are not attri-
FactorsAffecting
Trust/ 93

butionstied exclusively to technicalcompetence. Spe- For the interorganizational/interdepartmental fac-
cifically, given that research is essentially a shared tors, several interestingfindings emerge. In the case
form of knowledge (Hunt 1990), its truthfulnessis af- of researcherorganizationalculture,bureaucracieshad
fected in large part by a researcher's demonstrated a negative effect on trust, as expected. This finding
sincerity. Hence, the "truth"of researchmay be de- is consistent with the negative effect of user organi-
terminedin part by a "trust"test in addition to other zationformalization,which also reducedtrust,and with
means of validating knowledge claims (Zaltman, previousresearchdescribingthe effect of bureaucratic
LeMasters, and Heffring 1982). In fact, this "trust" control on trust in channels of distribution(Dwyer,
test is a componentof the authoritytruthtest noted in Schurr, and Oh 1987; John and Martin 1984). These
the sociology of knowledge literature(Holzner and findings indicate that trust can be negatively influ-
Marx 1979). Futureresearchcould extend some of the enced by bureaucraticcontrols within one's own or-
preliminaryapplied sociology work that identifies dif- ganizationand by such controls in an exchange part-
ferent kinds of "truthtests" (Deshpande 1986; Weiss ner's organization. As a result, large firms may be
and Bucuvalas 1980) by examining the conditions un- especially vulnerable to low levels of trust, because
der which various tests (including the trust test) are such organizationsare typically highly layered. Re-
employed individually or jointly to influence knowl- cent moves by firms to delayer and flatten organiza-
edge acceptance. tional structuresseem likely, therefore,to producethe
Two other perceivedresearchercharacteristicsim- added benefit of an increasedatmosphereof trustbe-
portantto trust are a researcher'stimeliness and con- tween researchersand users. This possibility suggests
geniality. Of the two, the negative relationship be- that an importantavenue for futureresearchis inves-
tween congenialityandtrustwarrantsthe most attention. tigating the longitudinalimpactsof organizationalde-
Past research has typically found a positive relation- layeringon marketingperformance.In particular,given
ship between sources' courtesy, friendliness, or lik- the substantialefforts being made in this directionby
ability and their persuasivenessor the extent to which some firms, it is critical to examine whethera lasting
they are trusted. Perhaps because market research is improvement in trust, research utilization, or other
viewed as a "science" and researchersare considered outcomes occurs.
"scientists"ratherthan "businesspeople,"congenial- Anotherinterorganizational/interdepartmental fac-
ity may run counter to the schema that many users tor, researcher power, increases users' trust in re-
have of researchers. An overly congenial researcher searchers. This finding is inconsistent with that of
therefore may cause users to make negative attribu- Anderson and Weitz (1990), who reportthat an im-
tions, includingquestioningthe researcher'sskills and balanceof power in relationshipsdecreasesmutualtrust
knowledge. Future research should investigate the between parties. One possible explanationis thata re-
conditions under which congeniality fosters trust, in- searcher'spower translatesinto credibilityand exper-
cluding the presence of other research and nonre- tise, not the ability to threatenor control an exchange
search abilities and motivations. partneras has been suggested in other research.More
Though much of our theory about the interper- broadly,the issue of how trustoperatesin conjunction
sonal characteristicsis supported,the same cannot be with other relationship factors, such as power and
said for our theory aboutthe user's organizationstruc- control, warrantsgreater research attention. For ex-
ture, culture, and location. Only formalizationis sig- ample, Shapiro (1987) refers to the need for controls
nificantly and negatively related to trust. Other re- so that the "trusted"do not behave opportunistically;
searchhas establishedthatorganizationalcharacteristics however, she also claims thatsuch controlsreducetrust
affect the decision to adopt and implementnew ideas in relationships. Future research should utilize more
(cf. Rogers 1983;Zaltman,Duncan,and Holbek 1973). controlled approachesto separatethe effects of trust,
Moreover,these factorshave been found to affect how power, and bureaucraticcontrol on how relationships
organizationsacquireand process information(Desh- operate.
pandeand Kohli 1989). It is surprising,therefore,that Given that researcherpower affects user trust, it
these effects do not extend uniformlyto the formation is importantto consider the factors that influence re-
of users' trust in researchers.Perhapsfuture research searcher power. First, our results indicate that re-
should consider how other processes mediate or mod- searchersmay be empoweredby the ownershipof ex-
erate the effects of organizationalvariableson trustin clusive experiential, informational, or technological
relationships. For example, additional research may assets. Second, the acquisitionof individualresearch-
demonstratethat organizationalcharacteristicsinflu- ers (with specializationsor importantskills not readily
ence trustonly when their impactis mediatedby more found among other researchers)may also createa per-
micro-level processes, as when organizationalbureau- ception of power. Researchorganization/department
cratization influences the level of exchange bureau- power, then, may actually be an aggregatedsense of
cratization, which in turn decreases trust. the researchabilities discussed previously. Third, re-
January1993

searcher power may be a function of organizational capabilities, or procedural rules for executing the
location, as our results indicate that researchersare project?Examiningthese and other compensatorybe-
trusted less when they report to product and brand haviorsin futureresearchwould provideadditionalin-
managersthan when they reportto higher level mar- sight.
keting or general managers. Designing research re- Considering our results in the aggregate, we see
lationshipswith these factors in mind should increase that the effects of various characteristicson trust do
user trust in researchers. not generallychange across differenttypes of research
With the above-notedexceptions, most of the or- relationships.This stability suggests that several core
ganizational and interorganizational/interdepartmen- characteristicsapparentlyare valued in these relation-
tal factors do not directly affect trust, suggesting that ships and that context does not affect the perceived
trust is generally less sensitive to these more macro salience of these characteristicsto trust, as was ex-
factors. Futureresearch would do well, therefore, to pected. This stability also suggests that our theory of
focus on the interpersonalfactors that appear to ac- trustmay be applicableto other information-basedre-
count for the greatest variance in trust. Another ap- lationships and perhaps to other relationships more
proach would be to examine the potential moderating generally. One difference, alluded to previously, be-
(as opposed to main) effects of structure,culture, lo- tween information relationships and other relation-
cation, and power on the relationships between the ships that may cause some variancein the character-
interpersonalcharacteristicsand trust. Such investi- istics affecting trustis whetherthe exchange involves
gations would extend the extant literatureby provid- search, experience, or credence goods. Given the in-
ing insight into whether certain organizationalchar- tangibleand sophisticatednatureof many information
acteristicscan add to or detractfrom the effectiveness relationships,more experience and credence qualities
of various interpersonalcharacteristics.Our finding are likely to be present, increasing the need for the
that the interpersonalcharacteristicsgenerally do not user to rely on the provider of informationand the
affect trustdifferentlyin intra-and interorganizational interpersonalqualities he or she displays. However,
settings is an initial step toward this end. if the exchangeinvolves moresearchgoods (i.e., goods
Results are mixed on project importanceand cus- whose quality can be assessed by direct observation),
tomization factors. Customization of the research the importanceof trustand of the interpersonalfactors
projectaffectstrustin researchrelationshipsandis likely to trust is likely to diminish.
to generate relationship-specificassets that past re- Finally, though our researchprovides insight into
searchhas linked with exchange cooperationand con- the relativeimportanceof variouscharacteristics to trust,
tinuity(Levinthaland Fichman 1988). Unlike custom- it leaves unansweredthe questionof how these factors
ization, the strategicimportanceof the researchproject are processed during early and later encounters be-
has no effect on users' trustin the researchers.Future tween exchange partnersand whetherthe factors that
research should seek to understandhow users offset develop trust at the outset of a relationshipare the
the risk and vulnerabilityassociated with having re- same as those that maintainit in later stages. Future
searchershandle strategicallyimportantprojectswith- researchusing a longitudinaldesign could addressthis
out trusting relationships. For example, do users es- and related questions about how trust actually devel-
tablish various controls such as contracts, monitoring ops.
Appendix
Measures and Itemsa Source
User Trust in Researcher Moorman, Zaltman, and
a. If I or someone from my firm could not be reached by our researcher, I would Deshpande (1992)
be willing to let my researcher make important research decisions without my
involvement.
b. If I or someone from my department was unable to monitor my researcher's
activities, I would be willing to trust my researcher to get the job done right.
c. I trust my researcher to do things I can't do myself.
d. I trust my researcher to do things my department can't do itself.
e. I generally do not trust my researcher.b
Actual Job Experience New item
How many years have you been in this position?
Actual Firm Experience New item
How many years have you worked for this firm?

Appendix (Continued)
Measures and Items' Source

Perceived Researcher Interpersonal Characteristics
Expertise New scale
a. Expert-nonexpert (7 a great deal, 1 none)
b. Trained-untrained(7 a great deal, 1 none)
c. Experienced-inexperienced (7 great deal, 1 none)
Uncertainty Reduction New scale
a. My researcher reflects on his/her experience to fill in the gaps left by
research.
b. My researcher is unable to provide convincing interpretations when the data
appear inconclusive.b
c. My researcher is creative in interpreting uncertain research findings.
Sincerity New scale
a. My researcher is sincere in his/her promises.
b. My researcher is honest with me.
c. My researcher will act sincerely in future dealings.
Integrity New scale
a. My researcher does not have a great deal of integrity.b
b. My researcher brings high personal standards to his/her work.
c. Integrity-no integrity (7 a great deal, 1 none)
Dependability New items
a. I have a good sense of what my researcher will say and do in most situations.
b. My researcher is a very predictable person.
Collective Orientation New scale
a. My researcher is a team player.
b. My researcher is flexible in meeting my needs even if it sometimes means
sacrificing methodological perfection.
c. My researcher makes wise judgments concerning when to express his/her
opinions and when to allow me to do things my way.
Tact New scale
a. Tactful-untactful(7 great deal, 1 none)
b. My researcher would be tactful in presenting research findings that may be
embarrassing for someone.
c. My researcher would be diplomatic when presenting unexpected research
findings.
Timeliness New scale
a. My researcher usually accommodates my last minute requests.
b. My researcher is punctual in meeting deadlines.
c. My researcher returns phone calls promptly.
Confidentiality New scale
a. The information we share with my researcher will not be shared with
competitors.
b. Our research findings are safe with my researcher.
c. My researcher can be expected to keep confidential what he/she learns about
our organization beyond the specific research project.
Congeniality New scale
a. Disagreeable-agreeableb (7 a great deal, 1 none)
b. Friendly-unfriendly(7 a great deal, 1 none)
c. Good disposition-bad disposition (7 a great deal, 1 none)
Perceived User Organizational Characteristics

Centralization(1 never, 2 seldom, 3 often, 4 always, 9 not applicable) Deshpande and Zaltman
a. How frequently do you usually participate in decisions on the adoption of new (1982)
products?b
b. How frequently do you usually participate in decisions on the modification of
existing products?b
c. How frequently do you usually participate in decisions to delete existing
products?b
d. There is little action taken in my job until a superior approves the decision.
e. If I wished to make my own decisions, I would be quickly discouraged.
f. Even small matters on this job have to be referred to someone higher up for
final answers.

Measures and Itemsa Source
g. I have to ask my boss before I do almost anything.
h. Any decision I make has to have my boss' approval.
Formalization (1 definitely true, 2 more true than false, 3 more false than true, 4 Deshpande and Zaltman
definitely false) (1982)
a. Generally, I feel that I am my own boss in most matters relating to my job.
b. I can make my own decisions in my job without checking with anybody else.
c. How things are done around here is pretty much up to me.
d. I am allowed to do almost as I please.
e. I make up my own rules on this job.
f. I am constantly being checked on for rule violations.b
g. I feel that I am constantly being watched to see that I obey all the rules.b
h. There is no specific rule manual detailing what I do.
i. There is a complete written job description for my positionb
j. Whenever we have a problem, we are supposed to go to the same person for
an answer.b
k. In this organization everyone has a specific job to do.b
I. The organization keeps a written record of everyone's performance.b
m. We follow strict operational procedures at all times.b
n. Whatever situations arise, we have procedures to follow in dealing with
them.b
o. Going through the proper channels in getting a job done is constantly
stressed.b
Hall, Haas, and Johnson
Complexity (1 definitely true, 2 more true than false, 3 more false than true, 4 Adapted from Price and
definitely false) Mueller (1986)
This organization has a large number of separate departments.b
Culture: Clan (1A, 2A, 3A, 4A), Adhocracy (1B, 2B, 3B, 4B), Hierarchy (1C, 2C, 3C, Adapted from Quinn 1988 by
4C), Market (1D, 2D, 3D, 4D) Deshpand6, Farley, and
Webster (1992).
1. Kind of Organization (Please distribute 100 points)
(A) The organization is a very personal place. It is like an extended
family. People seem to share a lot of themselves.
(B) The organization is a very dynamic and entrepreneurial place.
People are willing to stick their necks out and take risks.
(C) The organization is a very formalized and structured place.
Bureaucratic procedures generally govern what people do.
(D) The organization is very production oriented. The major concern is
100 with getting the job done. People aren't very personally involved.
2. Leadership (Please distribute 100 points)
(A) The head of the organization is generally considered to be a
mentor, sage, or a father or mother figure.
(B) The head of the organization is generally considered to be an
entrepreneur, an innovator, or a risk taker.
(C) The head of the organization is generally considered to be a
coordinator, an organizer, or an administrator.
(D) The head of the organization is generally considered to be
100 a producer, a technician, or a hard-driver.
3. What Holds the Organization Together (Please distribute 100 points)
(A) The glue that holds the organization together is loyalty and
tradition. Commitment to this firm runs high.
(B) The glue that holds the organization together is a commitment to
innovation and development. There is an emphasis on being first.
(C) The glue that holds the organization together is formal rules and
policies. Maintaining a smooth-running institution is important here.
(D) The glue that holds the organization together is the emphasis on
100 tasks and goal accomplishment. A production orientation is
commonly shared.
4. What Is Important (Please distribute 100 points)
(A) The organization emphasizes human resources.
High cohesion and morale in the firm are important.

Measures and Items' Source
(B) The organization emphasizes growth and acquiring new resources.
Readiness to meet new challenges is important.
(C) The organization emphasizes permanence and stability.
Efficient, smooth operations are important.
(D) The organization emphasizes competitive actions and achievement.
100 Measurable goals are important.
User Location
What is your current position?
Perceived Interorganizational/lnterdepartmental Characteristics
Researcher Organization/Department Power New scale
a. My research firm (department) has special expertise not found in other
research firms (department).
b. Other research firms (department) could not provide me with the insights that
my present research firm (department) can.
c. My research firm's (department) expertise is common in the industry.b
User Organization/Department Power New scale
a. My research firm relies on our company's business for their continued growth.
b. My firm accounts for a large proportion of my research firm's revenues.
c. My research firm is dependent on our firm.
Researcher Organization/Department Culture: Adapted from Quinn (1988)
Clan (A's), Adhocracy (B's), Hierarchical(C's), Market (D's) by Deshpande, Farley, and
Same measures as the User's Organizational Culture except that the Webster (1992)
respondent is directed to evaluate the researcher's organization/department
culture
Researcher Location Adapted from Kinnear and
To whom do your researchers primarily report? (1 top management, 2 general Root (1988)
management, 3 marketing or sales management, 4 product or brand
management, 5 engineering and product development)
Perceived Project Characteristics

Importance of the Research Project New items
a. Important to the firm-not important to the firm (7 a great deal, 1 not at all)
b. Important to me-not important to me (7 a great deal, 1 not at all)
Customization of the Research Project New item
Syndicated-Customb (7 a great deal, 1 not at all)
'All items used the scale 7 = stronglyagree, 4 = neitheragree nor disagree,1 = stronglydisagree,unless otherwisenoted.
bReverse coded.
REFERENCES
Adams, Edward F. (1977), "An Investigation of the Influence and Barton Weitz (1990), "Determinants of Con-
of Job Level and Functional Specialty on Job Attitudes and tinuity in Conventional Industrial Channel Dyads," Mar-
Perceptions," Journal of Applied Psychology, 62 (June), keting Science, 8 (Fall), 310-23.
335-43. Anderson, James C. and James A. Narus (1984), "A Model
Aguilar, John L. (1984), "Trustand Exchange: Expressive and of the Distributor's Perspective of Distributor-Manufacturer
InstrumentalDimensions of Reciprocity in a Peasant Com- Working Relationships," Journal of Marketing, 48 (Fall),
munity," Ethos, 12 (Spring), 3-29. 62-74.
Aiken, Michael and Jerald Hage (1968), "Organizational and (1990), "A Model of Distributor
Interdependence and IntraorganizationalStructure," Amer- Firm and ManufacturerFirm Working Partnerships,"Jour-
ican Sociological Review, 33 (December), 912-29. nal of Marketing, 54 (January), 42-58.
Anderson, Erin, Leonard Lodish, and Barton Weitz (1987), Arrow, Kenneth (1974), The Limits of Organization. New York:
"Resource Allocation Behavior in Conventional Chan- W. W. Norton & Co., Inc.
nels," Journal of Marketing Research, 24 (February), 254- Austin, Jon (1991), "An Exploratory Examination of the De-
62. velopment of Marketing Research Service Relationships:

An Assessment of Exchange Evaluation Dimensions," in Structural Determinants of Information-Processing Break-
1991 AMA Educators' Proceedings, Mary C. Gilly and F. down in Organizations," Knowledge: Creation, Diffusion,
Robert Dwyer, eds. Chicago: American Marketing Asso- and Utilization, 11 (December), 155-69.
ciation, 133-41. and Frederick E. Webster, Jr. (1989), "Organiza-
Bailey, J. and S. Pearson (1983), "Development of a Tool for tional Culture and Marketing: Defining the Research
Measuring and Analyzing Computer User Satisfaction," Agenda," Journal of Marketing, 53 (January), 3-15.
Management Science, 29 (May), 519-29. and Gerald Zaltman (1982), "Factors Affecting the
Barabba,Vincent and GeraldZaltman(1991), Hearing the Voice Use of Market Research Information: A Path Analysis,"
of the Market. Cambridge, MA: Harvard Business School Journal of Marketing Research, 19 (February), 14-31.
Press. and (1984), "A Comparison of Factors
Barber, Bernard (1983), The Logic and Limits of Trust. New Affecting Researcher and Manager Perceptions of Market
Brunswick, NJ: Rutgers University Press. Research Use," Journal of Marketing Research, 21 (Feb-
Blattberg, Robert C. and Stephen J. Hoch (1990), "Database ruary), 32-8.
Model and Managerial Intuition: 50% Model + 50% Man- and (1987), "A Comparison of Factors
ager," Management Science, 36 (August), 887-99. Affecting Use of Marketing Information in Consumer and
Blau, Peter (1964), Exchange and Power in Social Life. New Industrial Firms,," Journal of Marketing Research, 24
York: John Wiley & Sons, Inc. (February), 114-8.
and R. A. Schoenherr (1971), The Structure of Or- Deutsch, Morten (1958), "Trust and Suspicion," Journal of
ganizations. New York: Basic Books, Inc. Conflict Resolution, 2 (September), 265-79.
Busch, Paul and David T. Wilson (1976), "An Experimental (1962), "Cooperation and Trust: Some Theoretical
Analysis of a Salesman's Expert and Referent Bases of So- Notes," in Nebraska Symposium on Motivation, Marshall
cial Power in the Buyer-Seller Dyad," Journal of Market- R. Jones, ed. Lincoln, NE: University of Nebraska Press,
ing Research, 13 (February), 3-11. 275-320.
Butler, John K. and Stephen R. Cantrell (1984), "A Behav- Dwyer, F. Robert and Rosemary R. Lagace (1986), "On the
ioral Decision Theory Approach to Modeling Dyadic Trust Nature and Role of Buyer-Seller Trust," in 1986 AMA Ed-
in Superiors and Subordinates," Psychological Reports, 55 ucators' Proceedings, T. Shimp et al., eds. Chicago:
(August), 19-28. American Marketing Association, 40-5.
Caplan, Nathan S., Andrea Morrison, and Russell J. Stam- and Sejo Oh (1987), "Output Sector Munificence
baugh (1975), The Use of Social Science Knowledge in Public Effects on the Internal Political Economy of Marketing
Policy Decisions at the National Level. Ann Arbor, MI: Channels," Journal of Marketing Research, 24 (November)
Institute for Social Research. 347-58.
Churchill, Gilbert A., Jr. (1979), "A Paradigm for Developing , Paul Schurr, and Sejo Oh (1987), "Developing
Better Measures of Marketing Constructs," Journal of Mar- Buyer-Seller Relationships," Journal of Marketing, 51
keting Research, 16 (February), 64-73. (April), 11-27.
Cohen, Jacob and Patricia Cohen (1983), Applied Multiple Fox, Alan (1974), Beyond Contract: Work, Power, and Trust
Regression for the Behavioral Sciences, 2nd ed. Hillsdale, Relations. London: Faber & Faber Ltd.
NJ: Lawrence Erlbaum Associates. Frost, Taggart, David V. Stimpson, and Micol R. C. Maughan
Coleman, James S. (1990), Foundations of Social Theory. (1978), "Some Correlates of Trust," Journal of Psychol-
Cambridge, MA: The Belknap Press. ogy, 99 (May), 103-8.
Corcoran, Kevin J. (1988), "The Relationship of Interpersonal Gahagan, J. P. and J. T. Tedeschi (1968), "Strategy and the
Trust to Self-Disclosure When Confidentiality Is Assured," Credibility of Promises in the Prisoner's Dilemma Game,"
Journal of Psychology, 122 (March), 193-5. Journal of Conflict Resolution, 12 (June), 224-34.
Crosby, Lawrence A., Kenneth R. Evans, and Deborah Cowles Gerbing, David W. and James C. Anderson (1988), "An Up-
(1990), "Relationship Quality in Services Selling: An In- dated Paradigm for Scale Development Incorporating Un-
terpersonal Influence Perspective," Journal of Marketing, idimensionality and Its Assessment," Journal of Marketing
54 (July), 68-81. Research, 25 (May), 186-92.
Crouch, Andrew and Philip Yetton (1988), "Manager-Subor- Giffin, Kim (1967), "The Contribution of Studies of Source
dinate Dyads: Relationships Among Task and Social Con- Credibility to a Theory of InterpersonalTrust in the Com-
tact, Manager Friendliness and Subordinate Performance in munication Process," Psychological Bulletin, 68 (August),
Management Groups," Organizational Behavior and Hu- 104-20.
man Decision Processes, 41 (February), 65-82. Glazer, Rashi (1991), "Marketing in an Information-Intensive
Day, George S. (1991), "Learning About Markets," Market- Environment: Strategic Implications of Knowledge as an
ing Science Institute Report #91-117 (June). Cambridge, Asset," Journal of Marketing, 55 (October), 1-19.
MA: Marketing Science Institute. Granovetter, Mark (1985), "Economic Action and Social
Deshpande, Rohit (1982), "The Organizational Context of Structure: The Problem of Embeddedness," American
Market Research Use," Journal of Marketing, 46 (Fall), Journal of Sociology, 91 (November), 481-510.
91-101. Hall, R., E. G. Haas, and N. F. Johnson (1967), "Organi-
(1986), "TruthTests and Utility Tests: Private Sec- zational Size, Complexity, and Formalization," American
tor Decision Makers' Frames of Reference for Social Sci-
Sociological Review, 32 (December), 903-11.
ence Research," Social Science Quarterly, 67 (March), 39- Heide, Jan and Anne Miner (1992), "The Shadow of the Fu-
52. ture: The Effects of the Anticipated Interaction and Fre-
, John U. Farley, and Frederick E. Webster, Jr. quency of Contact on Buyer-Seller Cooperation," Academy
(1992), "CorporateCulture, Customer Orientation, and In- of Management Journal, 35, 265-91.
novativeness in Japanese Firms: A Quadrad Analysis," Holzner, Burkart and John H. Marx (1979), Knowledge Ap-
Working Paper #92-100. Cambridge, MA: Marketing Sci- plication: The Knowledge System in Society. Boston: Allyn
ence Institute. and Bacon, Inc.
and Ajay Kohli (1989), "Knowledge Disavowal: Hovland, C. I., I. L. Janis, and H. H. Kelley (1953), Com-

munication and Persuasion. New Haven, CT: Yale Uni- Implications for Future Research," Journal of Marketing,
versity Press. 49 (Fall), 41-51.
Hrebiniak, Lawrence G. (1974), "Effects of Job Level and Perkins, W. Steven and Ram C. Rao (1990), "The Role of
Participation on Employee Attitudes and Perceptions of In- Experience in Information Use and Decision Making by
fluence," Academy of Management Journal, 17 (Decem- Marketing Managers," Journal of Marketing Research, 27
ber), 649-62. (February), 1-16.
Hunt, Shelby D. (1990), "Truthin Marketing Theory and Re- Peterson, Richard B. and Justin Y. Shimada (1978), "Sources
search," Journal of Marketing, 54 (July), 1-15. of Management Problems in Japanese-AmericanJoint Ven-
, Lawrence B. Chonko, and James B. Wilcox (1984), tures," Academy of ManagementReview, 3 (October), 796-
"Ethical Problems of Marketing Researchers," Journal of 804.
Marketing Research, 21 (August), 309-24. Porter, Michael E. and Victor E. Millar (1985), "How Infor-
Ives, B., M. Olson, and J. Baroudi (1983), "The Measure- mation Gives You Competitive Advantage," Harvard Busi-
ment of User Information Satisfaction," Communications ness Review, 85 (July-August), 149-60.
of the ACM, 26 (10), 785-93. Posey, E. Carol (1988), "Confidentiality in an AIDS Support
John, George (1984), "An Empirical Investigation of Some Group," Journal of Counseling and Development, 66 (Jan-
Antecedents of Opportunism in a Marketing Channel," uary), 226-7.
Journal of Marketing Research, 21 (August), 278-89. Price, James L. and Charles W. Mueller (1986), Handbook of
and John Martin (1984), "Effects of Organizational Organizational Measurement. Marchfield, MA: Pitman
Structure of Marketing Planning on Credibility and Utili- Publishing.
zation of Plan Output," Journal of Marketing Research, 21 Pruitt, Dean G. (1981), Negotiation Behavior. New York: Ac-
(May), 170-83. ademic Press, Inc.
Johnson-George, C. and W. C. Swap (1982), "Measurement Quinn, Robert E. (1988), Beyond Rational Management. San
of Specific Interpersonal Trust: Construction and Valida- Francisco: Jossey-Bass Inc., Publishers.
tion of a Scale to Assess Trust in a Specific Other," Journal Rempel, John K. and John G. Holmes (1986), "How Do I
of Personality and Social Psychology, 43 (December), 1306- Trust Thee," Psychology Today, 20 (February), 28-34.
17. , , and Mark P. Zanna (1985), "Trust in
Kaplan, Robert M. (1973), "Components of Trust: Note on Close Relationships," Journal of Personality and Social
Use of Rotter's Scale," Psychological Reports, 33 (Au- Psychology, 49 (July), 95-112.
gust), 13-14. Rogers, EverettM. (1983), Diffusion of Innovations. New York:
Kinnear, Thomas C. and Ann R. Root, eds. (1988), Survey of The Free Press.
Marketing Research. Chicago: American Marketing As- Root, Ann and Thomas C. Kinnear (1991), "The Effects of
sociation. Content Agreement, Decision Making Stage and Supplier
Kohli, Ajay K. and Bernard J. Jaworski (1990), "MarketOri- Credibility on the Evaluation and Use of Marketing Re-
entation: The Construct, Research Propositions, and Man- search," paper presented at the Marketing Science Institute,
agerial Implications," Journal of Marketing, 54 (April), 1- Cambridge, MA (June 4).
18. Rotter, Julian (1967), "A New Scale for the Measurement of
Larzeleve, R. E., and T. L. Huston (1980), "The Dyadic Trust Interpersonal Trust," Journal of Personality, 35 (Decem-
Scale: Toward Understanding Interpersonal Trust in Close ber), 651-65.
Relationships," Journal of Marriage and the Family, 42 Schlenker, BarryR., Bob Helm, and James T. Tedeschi (1973),
(August), 595-604. "The Effects of Personality and Situational Variables on
Levinthal, Daniel and Mark Fichman (1988), "Dynamics of Behavioral Trust," Journal of Personality and Social Psy-
Interorganizational Attachments: Auditor-Client Relation- chology, 25 (3), 419-27.
ships," Administrative Science Quarterly, 33 (September), Schurr, Paul H. and Julie L. Ozanne (1985), "Influence on
345-69. Exchange Processes: Buyers' Preconceptions of a Seller's
Macneil, Ian R. (1980), The New Social Contract: An Inquiry Trustworthiness and Bargaining Toughness," Journal of
Into Modern Contractual Relations. New Haven, CT: Yale Consumer Research, 11 (March), 939-53.
University Press. Shapiro, Susan P. (1987), "The Social Control of Impersonal
Mason, Charlotte H. and William D. Perreault, Jr. (1991), Trust," American Journal of Sociology, 93 (3), 623-58.
"Collinearity, Power, and Interpretation of Multiple Smircich, Linda (1983), "Concepts of Culture and Organiza-
Regression Analysis," Journal of Marketing Research, 28 tional Analysis," Administrative Science Quarterly, 28
(August), 268-80. (September), 339-58.
McDaniel, M. A., F. L. Schmidt, and J. E. Hunter (1988), Stemberg, Robert J. and Richard K. Wagner, eds. (1986),
"Job Experience Correlates of Performance," Journal of Practical Intelligence: Nature and Origins of Competence
Applied Psychology, 73 (May), 327-30. in the Everyday World. New York: Cambridge University
Mintzberg, Henry (1979), The Structuring of Organizations: Press.
A Synthesis of the Research. Englewood Cliffs, NJ: Pren- Sterthal, Brian, Ruby Roy Dholakia, and Clark Leavitt (1978),
tice-Hall, Inc. "The Persuasive Effect of Source Credibility: Tests of Cog-
Moorman, Christine, Gerald Zaltman, and Rohit Deshpand6 nitive Response," Journal of ConsumerResearch, 4 (March),
(1992), "Relationships Between Providers and Users of 252-60.
Market Research: The Dynamics of Trust Within and Be- Stolte, John F. (1983), "Positional Power, Personality, and
tween Organizations," Journal of Marketing Research, 29
Perceptions of Fairness in Exchange," Journal of Social
(August), 314-29. Psychology, 120 (August), 287-8.
Ouchi, William G. (1980), "Markets, Bureaucracies, and Sullivan, Jeremiah and Richard B. Peterson (1982), "Factors
Clans," AdministrativeScience Quarterly, 25 (March), 129- Associated With Trust in Japanese-American Joint Ven-
41. tures," Management International Review, 22 (2), 30-40.
Parasuraman,A., Valarie A. Ziethaml, and Leonard L. Berry , , Naoki Kameda, and Justin Shimada
(1985), "A Conceptual Model of Service Quality and Its (1981), "The Relationship Between Conflict Resolution

January1993

Approaches and Trust-A Cross Cultural Study," Academy ence Research and Decision Making. New York: Columbia
of Management Journal, 24 (December), 803-15. University Press.
Surprenant, Carol F. and Michael R. Solomon (1987), "Pre- Williamson, Oliver (1975), Markets and Hierarchies: Analysis
dictability and Personalization in the Service Encounter," and Antitrust Implications. New York: The Free Press.
Journal of Marketing, 51 (April), 86-96. Zalesny, Mary D., Richard V. Farace, and Ronnie Kurcher-
Swap, Walter C. and Jeffrey Z. Rubin (1983), "Measurement Hawkins (1985), "Determinants of Employee Work Per-
of Interpersonal Orientation," Journal of Personality and ceptions and Attitudes: Perceived Work Environment and
Social Psychology, 44 (1), 208-19. Organizational Level," Environment and Behavior, 17
(September), 567-92.
Taylor, Shelley E., Jennifer Crocker, Susan T. Fiske, Merle
Zaltman, Gerald, Robert Duncan, and Johny Holbek (1973),
Sprinzen, and Joachim D. Winkler (1979), "The General- Innovations and Organizations. New York: John Wiley &
izability of Salience Effects," Journal of Personality and Sons, Inc.
Social Psychology, 57 (September), 357-68. , Karen LeMasters, and Michael Heffring (1982),
Tharenou, Phyllis and Phillip Harker (1982), "Organizational Theory Construction in Marketing: Some Thoughts on
Correlates of Employee Self-Esteem," Journal of Applied Thinking. New York: John Wiley & Sons, Inc.
Psychology, 67 (December), 797-805. and Christine Moorman (1988), "The Role of Per-
Tjosvold, Dean (1977), "The Effects of the Constituent's Af- sonal Trust in the Use of Research," Journal of Advertising
firmation and the Opposing Negotiator's Self-Presentation Research, 28 (October/November), 16-24.
in Bargaining Between Unequal Status Groups," Organi- and (1989), "The Management and Use
zational Behavior and Human Performance, 18 (February), of Advertising Research," Journal of Advertising Re-
146-57. search, 28 (December/January), 11-18.
Tversky, Amos and Daniel Kahneman (1974), "JudgmentUn- Zand, Dale E. (1972), "Trust and Managerial Problem Solv-
der Uncertainty: Heuristics and Biases," Science, 185 (Sep- ing," AdministrativeScience Quarterly, 17 (June), 229-39.
Zeithaml, Valarie, A. Parasuraman,and Leonard Barry (1990),
tember), 1124-31.
Delivering Quality Service. New York: The Free Press.
Walker, Gordon (1985), "Network Position and Cognition in Zucker, Lynne G. (1986), "Production of Trust: Institutional
a Computer Software Firm," Administrative Science Quar- Sources of Economic Structure, 1840-1920," Research in
terly, 30 (March), 103-30. Organizational Behavior, 8, 53-111.
Weiss, Carol H. and Michael J. Bucuvalas (1980), Social Sci- Reprint No. JM571105
I _
TITLE PAGE REPRINT
NUMBER
Measuring,and ManagingCustomer-Based
"Conceptualizing, Brand 1 JM571100
Equity"(Keller)
"CorporateCulture,CustomerOrientation,and Innovativeness
in 23 JM571101
JapaneseFirms:A QuadradAnalysis"(Deshpande,Farley,&Webster)
"Taxonomy of BuyingDecisionApproaches" (Bunn) 38 JM571102
"ControlCombinationsin Marketing:ConceptualFramework and 57 JM571103
EmpiricalEvidence" &
(Jaworski,Stathakopoulos Krishnan)
"TheImpactof OrganizationalCitizenshipBehavioron Evaluations
of 70 JM571104
SalespersonPerformance" (MacKenzie,Podsakoff&Fetter)
"FactorsAffectingTrustin MarketResearchRelationships" 81 JM571105
(Moorman, Deshpand6&Zaltman)
LegalDevelopmentsin Marketing 102 JM571106
MarketingLiterature
Review 111 JM571107
BookReviews 127 JM571108
To order REPRINTS,
contactthe AmericanMarketing
AssociationPublications
Group, 250 South Wacker Drive, Chicago, IL60606. Prices for reprintsmay be
AMBaC4N
found on page 37 of this issue. AaQ
ASMCt4nON
AsOssN
FactorsAffecting
Trust/ 101


Factors Affecting Trust in Market Research Relationships-Moorman-1993

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Factors Affecting Trust in Market Research Relationships-Moorman-1993

Uploaded by

Copyright:

Available Formats

Factors Affecting Trust in Market Research Relationships

Author(s): Christine Moorman, Rohit Deshpandé and Gerald Zaltman

This content downloaded from 161.6.94.245 on Fri, 12 Apr 2013 07:51:47 AM