Content uploaded by Gwen Jones
Author content
All content in this area was uploaded by Gwen Jones on Aug 15, 2014
Content may be subject to copyright.
Address correspondence to Gwen E. Jones, Department of Management, 285 Madison
Avenue, Fairleigh Dickinson University, Madison, NJ 07940; gjones@fdu.edu (e-mail).
311
The Journal of Social Psychology, 2004, 144(3), 311–321
Trust in the Workplace: Factors Affecting
Trust Formation Between Team Members
MICHELE D. SPECTOR
GWEN E. JONES
Fairleigh Dickinson University
Madison, NJ
ABSTRACT. The authors used survey data from 127 professional-level employees work-
ing in 8 industries to assess the effects of respondent’s trusting stance and (a) the trustee’s
organization membership (internal or external), (b) the hierarchical relationship (super-
visor or peer), and (c) the gender of the trustee, on initial trust level for a new project
team member. The authors found that trusting stance was positively related to initial trust
level. The authors also found an interaction effect between respondent gender and trustee
gender on initial trust. Specifically, male initial trust level was higher for a new male
team member and lower for a new female team member. The present study provided
additional understanding of the formation of initial trust levels and its importance for
team functioning.
Key words: gender trust, team trust, workplace trust
KARL (2000) noted that trust among organizational members is at an all-time
low, and Morris (1995) found that 56% of nonmanagement employees in 57 ser-
vice and manufacturing organizations viewed a lack of trust as a problem in their
respective organizations. A comprehensive review of the trust literature (Whiten-
er, Brodt, Korsgaard, & Werner, 1998) showed that personal trust is linked to
cooperation, performance, and quality of communication in organizations. Work-
ing well together requires some degree of trust, and the type of work encounters
that occur in today’s organizations require trust to be formed rather quickly
(McKnight, Cummings, & Chervany, 1998).
Investigators can view trust as an attitude held by one individual, the truster,
toward another individual, the trustee. Trust in individuals is an expectation or
belief that actions from another party will be motivated by good intentions.
Moreover, individuals take a risk in this belief because the other party may not
act out of benevolence (Whitener et al., 1998). A critical time for the develop-
ment of trust for another person in an organizational context is the beginning of
the relationship. Models of trust levels (Creed & Miles, 1996) and initial trust
(McKnight et al., 1998) among employees show that trust is based on an indi-
vidual’s propensity to trust (trusting stance) and on variables associated with the
situation. The present study empirically tested initial trust formation
antecedents, including a trust formation variable and several situational variables
that are relatively common among the increasingly diverse composition of work-
force teams.
Trusting Stance
Trusting stance is the degree to which an individual consistently deals with
people as if they are well-meaning and reliable across situations and persons. It
is a conscious choice to trust people until they prove untrustworthy (McKnight
et al., 1998). The indications of empirical research have been mixed on whether
trusting stance leads to the development of interpersonal trust, and a link has not
yet been established. However, McKnight et al. asserted that in new relationships
among organizational members, an individual’s trusting stance will positively
impact the degree of initial trust for another individual.
Hypothesis 1: Individuals with a high trusting stance are more likely to have a high-
er initial trust for coworkers than are individuals with a low trusting stance.
Organizational Affiliation
Because of the burgeoning consultant workforce, it is becoming more and
more prevalent that project workers may be composed of individuals who are
either internal or external to the organization. The ability of an organization to
effectively develop and maintain strategic partnerships between internal employ-
ees and external employees has become a critical competence (Lewicki, McAl-
lister, & Bies, 1998). Yet little research has been done on initial trust levels
between internal and external employees in the workplace. Investigators could
speculate that an organization would give “insiders” a greater degree of initial
trust because the organization assumes that they operate under their rules and
norms but that the organization would consider “outsiders” as more unknown and
therefore possibly less trustworthy.
Kramer (1999) called this type of trust that the organization gives to insid-
ers category-based trust, which occurs when individuals from a specific group of
an organization might place high trust on each other simply because of their
shared membership within this group.
Hypothesis 2: Individuals will exhibit higher initial levels of trust for employees from
their own organization than for employees from a different organization.
312 The Journal of Social Psychology
Hierarchical Relationship
The hierarchical position that a person possesses within an organization can
influence the initial trust levels that a coworker has for that person. Kramer (1999)
stated that role-based trust is influenced by the knowledge that a person’s occu-
pation of a specific role can determine the trust level that one has for various
employees within organizations because knowledge of the role can substitute for
personalized knowledge of the individual. Some researchers (e.g., Butler, Cantrell,
& Flick, 1999; Wells & Kipnis, 2001; Whitener et al., 1998) have studied the
behaviors that enable coworkers and supervisors to earn each other’s trust, but no
researcher has experimentally tested initial differences between trusting a super-
visor and trusting a coworker. Wells and Kipnis reported a statistically significant
relationship between dependency and trust. Often there is more dependency on a
supervisor (e.g., for performance reviews, promotions, skill development, task
assignments, etc.) than on a coworker, supporting the following hypothesis.
Hypothesis 3: Individuals will exhibit higher levels of initial trust for employees who
are superiors than for employees who are peers.
Trustee Gender
The present study examined whether the gender of the trustee plays a role in
initial trust. With the advancement of women into professional positions, work
teams are more likely to include both male and female employees. Two studies
(Keller, 2001; Williams, 2001) have asserted that trust levels are higher within
the same gender, while another study (Wells & Kipnis, 2001) did not report a
gender difference. Tyler and Kramer (1996) discussed the idea that ethics and
morals are highly related to trust, and Kipnis (1996) noted that the terms ethical
and moral are used to describe trusting behaviors. Some research in the area of
ethics indicates that survey respondents perceived women to be more ethical or
trustworthy than men (Jones & Kavanagh, 1996; Paterson & Kim, 1991).
Hypothesis 4: Individuals will have higher levels of trust for employees who are
women than employees who are men.
Method
Participants
Participants were professional employees (N = 127) who were working reg-
ularly in an office environment of 19 major corporations in the northeast Unit-
ed States. According to self-reported demographics, the mean age of respon-
dents was 34 years (SD = 9.19). They had a mean of 12 years of work experience
(SD = 9.87). Of all participants, 46% were men, and 54% were women. Of all
Spector & Jones 313
participants, 76% were white, 11% were Asian, 8% were African American, 2%
were Hispanic, and 3% were of other ethnic origins. Participants represented a
variety of industries, including—in order of descending percentage—the phar-
maceutical industry (32.3%), information technology (20.5%), advertising
(18.9%), and consulting (18.1%). The remaining 10.3% were from finance, pub-
lishing, and insurance. Of the 32% of the participants who were consultants, 88%
had previously worked in a client-consultant relationship.
Procedure
Participation in the study was voluntary. We either approached participants
in person in their offices or e-mailed them the survey on a one-on-one basis. We
explained that the study was for research purposes only and involved the partic-
ipant’s attitudes and experiences regarding trust in the workplace. Study materi-
als with instructions were distributed to consenting participants with random
assignment to experimental conditions. Participants were told that their surveys
would be collected on a particular future date and that they could complete them
at their own leisure before that time.
The survey consisted of three sections. First, we asked participants to read
and sign a consent agreement explaining the study and assuring anonymity. Sec-
ond, participants were asked to complete a questionnaire assessing their trusting
stance and to then play the role of an employee in a hypothetical work scenario
while reading one of eight organizational scenarios depicting trust issues. Third,
we asked participants to complete a questionnaire that we had designed to assess
reactions to the scenario, manipulation checks, and biographical information.
Once the surveys had been completed and collected, the participants were
thanked and debriefed regarding the purpose of the study.
Manipulations
General. The appendix shows the full text of the scenario. All participants were
asked to play the role of a coworker and read a scenario depicting the situation
of a new member of their project. The first paragraph of the scenario described
the organization. The second paragraph described a critical project that the
respondent (truster) had been asked to participate in. The deadline for the project
was fixed, and the current number of resources that was assigned to this project
prevented them from meeting the deadline. So, the organization brought in an
additional team member to assist the team with getting the project completed on
time. Finally, the new team member was described in the scenario as a real “take
charge” employee who felt that the critical deadline was no problem.
We embedded the three manipulations in the scenario. We manipulated orga-
nizational affiliation by stating whether the new employee was from the same
organization (internal condition) or came from an outside organization (external
314 The Journal of Social Psychology
condition). We depicted hierarchical relationship in the scenario by stating
whether the new employee was a coworker or a supervisor on the project team.
We manipulated trustee gender in the scenario by stating whether the new
employee was a man or a woman and by using the corresponding gender-based
pronouns (either “he” and “him” or “she” and “her”) throughout the scenario.
Manipulation checks. We used three items to check the manipulated variables.
The items were in multiple-choice form, including the correct response and two
distracters. The majority of respondents endorsed their correct manipulations for
the three variables: for organizational affiliation, 94% were correct, for hierar-
chical relationship, 93% were correct, and for trustee gender, 83% were correct.
Measures
Trusting stance. We measured trusting stance using the Interpersonal Mistrust-Trust
Measure (IMTM) that McLennan and Omodei (2000) had developed. Participants
rated 18 items on a 7-point Likert-type scale ranging from 1 = strongly disagree to
7 = strongly agree. Through the items, we asked, for example, whether the par-
ticipant believed that a stranger had sinister motives, believed that his or her
workmate had a headache, and believed that an attendant’s recommendations in
a department store was accurate.
The present researchers scored items in such a way that the higher scores
indicated a more trusting stance. McLennan and Omodei (2000) showed that the
measure exhibits reliability and construct validity. Cronbach’s coefficient alpha
reliability estimate for the scale in the present study was .76.
Initial trust level. We measured initial trust level in the workplace, the dependent
variable, using a 15-item scale in response to the workplace scenario presented.
We developed this scale for the present study. Participants rated the items on 7-
point Likert-type scales ranging from 1 = strongly disagree to 7 = strongly agree.
We scored items in such a way that the higher scores indicated a higher ini-
tial trust level. Examples of items are “Your first impression of this person is of
low trust,” “You could confide in everyone except this new person,” and “You
could depend on this new person for getting the job done.” Cronbach’s coeffi-
cient alpha reliability estimate for this scale was .93.
Design
We used a 2 × 2 × 2 experimental design to examine the effects of trusting
stance and the three manipulated variables on the participant’s initial trust level
for the new project team member. The three manipulated variables were (a) the
trustee’s organization membership (internal or external), (b) the hierarchical rela-
tionship to the participant (supervisor or peer), and (c) the gender of the trustee.
Spector & Jones 315
Results
The mean score for trusting stance was M = 5.04 (SD = .68), indicating that,
on average, this sample was toward the trusting end of the continuum. The mean
score for the dependent variable, initial trust level, was M = 3.82 (SD = 1.06),
which is just below the midpoint of the 7-point scale. The correlation between
the two was r = .35 (p < .001), supporting Hypothesis 1. Table 1 shows the means
and standard deviations for initial trust across the manipulated variables. We did
a univariate analysis of variance (ANOVA) to test Hypotheses 2–4 as shown in
Table 1. We found no main effects for trustee’s organizational membership, par-
ticipant’s hierarchical relationship, or gender of the trustee, thus demonstrating
no support for Hypotheses 2–4.
We found an interaction effect between respondent gender and trustee gender.
Trust levels varied among male respondents, who indicated low initial trust levels
316 The Journal of Social Psychology
TABLE 1. Means and Standard Deviations of Initial Trust Across Levels of
the Manipulated Variables (Summated Scales), Univariate F(15, 111)
Statistics, and Effect Sizes
Effect
Independent Univariate size
variables MSDNFtest p (η
2
)
Organization .72 .20 .01
Internal 58.62 16.59 63
External 56.03 15.18 64
Relationship .33 .29 .00
Peer 56.17 13.72 63
Superior 58.44 17.80 64
Trustee gender 1.66 .10 .01
Male 59.37 15.15 64
Female 55.22 16.45 63
Respondent gender .40 .27 .00
Male 58.86 16.35 58
Female 55.48 15.24 69
Respondent gender ×
Trustee gender 4.04 .02 .04
Respondent gender–male 6.42 .01 .10
Trustee gender–male 61.25 15.43 24
Trustee gender–female 51.41 13.93 34
Respondent gender–female .13 .36 .00
Trustee gender–male 58.25 15.06 40
Trustee gender–female 59.69 18.23 29
Note. R = 399. Higher (summated scale) means indicate higher levels of trust.
for a new female employee on the team and higher initial trust levels for a new
male employee on the team. There were no effects among female respondents.
Discussion
Our results indicate that trusting stance is positively related to initial trust
level in the workplace, thus supporting prior theory (McKnight et al., 1998).
Hence, the more an individual has a propensity toward trusting others, the more
likely he or she will attribute trustworthiness to a new work associate.
We found no main effects for the manipulated variables. However, one inter-
action among the variables was found: that men possess a higher initial trust level
for other men than they do for women. Women, however, showed no differences
in trust level across gender. Finding higher trust levels for members of an indi-
vidual’s own demographic group is consistent with prior theory (Creed & Miles,
1996; Keller, 2001; Williams, 2001). Williams asserted that social categorizations
are the mechanisms through which group membership influences trust develop-
ment and that perceptions of trustworthiness are based on cognitive shortcuts
such as stereotyping. In studies of cross-functional groups (by age, gender, and
ethnicity), Keller found that demographic similarities among group members
tend to increase the formation of interpersonal trust and relations. Both Keller
and Williams have asserted that future research is needed on what particular mix
of functional backgrounds and demographics specifically affects trust develop-
ment. Although prior research supports an interaction effect—in that men will
trust men, and women will trust women, more than they will people of the oppo-
site gender—the present study found this difference to be the case for men only,
yielding a main effect for one gender only.
Interestingly, we found no effects for two of the manipulated variables, orga-
nizational affiliation and hierarchical relationship. These results indicate that
whether the new team member is from another, external organization or from with-
in the organization does not affect the initial trust level. Moreover, trust is not
affected by whether the new team member is a peer or a superior. Possibly the rea-
sons for attributing trust to an external team member are different from those to
an internal team member, and possibly the reasons for attributing trust to a peer
are different from those to a supervisor. For example, one might trust an internal
team member because of kinship or camaraderie but also trust an external team
member because of the attribution of higher (specialized) expertise and compe-
tence. Investigators need further research to examine these possible differences.
Limitations of the present study exist. Although data were collected from
adults working in a variety of organizations, the participants were reacting to
a simulated role-play rather than to a real situation. Investigators have accept-
ed role-play as a surrogate in experimental research (Greenberg & Folger,
1988) on attitudes of sensitive topics such as relationship situations in the work-
place (e.g., Jones, 1999). Moreover, the participants’lack of actual involvement
Spector & Jones 317
in the situation most likely lessens the effects that investigators would find in
a real witnessed situation (Stone & Kotch, 1989). Investigators who add more
realism while still manipulating and controlling experimental variables, possi-
bly using videotape or virtual-reality technology, could observe more accu-
rately the causal relationships that the present study indicates (Pierce & Agui-
nis, 1997). Very promising but technically sophisticated tools for the future
study of trust are Microworlds (DiFonzo, Hantula, & Bordia, 1998), which are
computer-generated environments that simulate real-world conditions.
An additional limitation of the present study, which is associated with its
nature and methodology, is due to possible demand characteristics or social desir-
ability. These might have caused errors in our measures of any main effects that
we found, such as those for trusting stance, but are less likely to have caused
errors in our measures of interaction effects.
Finally, although the sample of the present study included adults working in
a variety of industries, all were in white-collar jobs in the northeastern part of the
United States. Generalizability of these results is limited to this population.
Understanding initial trust levels is important for the improvement of team
functioning. Work groups are more commonly composed of both genders than
before, and team members of each gender need to be aware of a possible ten-
dency to be less trusting of the opposite gender, so that they can overcome these
biases. While investigators can argue that some degree of skepticism is healthy,
Keller (2001) has found that trust levels tend to stay in their initial direction,
either upward or downward. Thus, initial trust levels are highly important because
they may start a negative trend.
REFERENCES
Butler, J. K., Cantrell, R. S., & Flick, R. J. (1999). Transformation leadership behaviors,
upward trust, and satisfaction in self-managed work teams. Organizational Develop-
ment Journal, 17, 13–28.
Creed, W. E. D., & Miles, R. E. (1996). Trust in organizations: Linking organizational
forms, managerial philosophies, and the opportunity costs of controls. In R. M. Kramer
& T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research. Thou-
sand Oaks, CA: Sage.
DiFonzo, N., Hantula, D. A., & Bordia, P. (1998). Microworlds for experimental research:
Having your (control and collections) cake, and realism too. Behavior Research Meth-
ods, Instruments & Computers, 30, 278–286.
Greenberg, J., & Folger, R. (1988). Controversial issues in social research methods. New
York: Springer-Verlag.
Jones, G. E. (1999). Hierarchical workplace romance: An experimental examination of
team member perceptions. Journal of Organizational Behavior, 20, 1057–1072.
Jones, G. E., & Kavanagh, M. J. (1996). An experimental examination of the effects of
individual and situational factors on unethical behavioral intentions in the workplace.
Journal of Business Ethics, 15, 511–523.
Karl, K. A. (2000). Trust and betrayal in the workplace: Building effective relationships
in your organization. Academy of Management Executive, 14, 133–135.
318 The Journal of Social Psychology
Keller, R. T. (2001). Cross-functional project groups in research and new product devel-
opment: Diversity, communications, job stress, and outcomes. Academy of Management
Journal, 44, 547–555.
Kipnis, D. (1996). Trust and technology. In R. M. Kramer & T. R. Tyler (Eds.), Trust in
organizations: Frontiers of theory and research. Thousand Oaks, CA: Sage.
Kramer, R. M. (1999). Trust and distrust in organizations: Emerging perspectives, endur-
ing questions. Annual Review of Psychology, 50, 537–567.
Lewicki, R. J., McAllister, D. J., & Bies, R. J. (1998). Trust and distrust: New relation-
ships and realities. Academy of Management Review, 23, 438–458.
McKnight, D. H., Cummings, L. L., & Chervany, N. L. (1998). Initial trust formation in
new organizational relationships. Academy of Management Review, 23, 473–490.
McLennan, J., & Omodei, M. M. (2000). Conceptualizing and measuring global inter-
personal mistrust-trust. The Journal of Social Psychology, 140, 279–294.
Morris, L. (1995). Creating and maintaining trust. Training & Development, 49(12),
52–54.
Patterson, J., & Kim, P. (1991). The day America told the truth. New York: Prentice Hall.
Pierce, C. A., & Aguinis, H. (1997). Using virtual reality technology in organizational
behavior research. Journal of Organizational Behavior, 18, 407–410.
Stone, D. L., & Kotch, D. A. (1989). Individuals’attitudes toward organizational drug test-
ing policies and practices. Journal of Applied Psychology, 74, 518–521.
Tyler,T. R., & Kramer, R. M. (1996). Whither trust? In R. M. Kramer & T. R. Tyler (Eds.),
Trust in organizations: Frontiers of theory and research. Thousand Oaks, CA: Sage.
Wells, C. V., & Kipnis, D. (2001). Trust, dependency, and control in the contemporary
organization. Journal of Business and Psychology, 15, 593–603.
Whitener, E. M., Brodt, S. E., Korsgaard, M. A., & Werner, J. M. (1998). Managers as ini-
tiators of trust:An exchange relationship framework for understanding managerial trust-
worthy behavior. Academy of Management Review, 23, 513–530.
Williams, M. (2001). In whom we trust: Group membership as an affective context for
trust development. Academy of Management Review, 26, 377–396.
Spector & Jones 319
320 The Journal of Social Psychology
APPENDIX
Scenario Text
The text of the scenario that we used in the present study follows. The terms “AND,” “OR,”
“<,” and “>” indicate the alternatives that we used to vary the scenario.
Please read the following carefully and picture yourself in the situation described.
Ace Inc. is a typical mid-size corporation where you have been working for over 4 years.
Your department consists of 20 individuals with diverse and varied backgrounds. The com-
petition is fierce and fast-paced, therefore, the team must be highly creative, competent,
and work together well.
Earlier in the week, you and eight other coworkers were selected to work on a new project
with a critical deadline. A memo went out yesterday stating officially that this project must
be completed by the end of this upcoming July, which is only three months from now. Yes-
terday, you and your new team members got together to discuss this new project. Your
team felt that meeting this deadline with only eight people seemed impossible. Your team
presented this information to your Project Sponsor to justify a request for two to three
additional members on your team. The Project Sponsor said that he would see what he
could do.
A few days pass, and you and your team are beginning to worry because you need the
additional team members soon so they can be brought up to speed and begin working.
This morning, you were introduced to a man/woman that was selected to join you and your
team on the project.
<This man/woman has been with your company for a few years, yet you have never
worked with him/her>
OR
<This man/woman is from an outside firm and was brought in to work on this
project>
AND
<You will be working with this man/woman on the project as coworkers both re-
porting to the same manager.>
OR
<You will be directly reporting to this man/woman on the project>
Later that afternoon, a kickoff meeting was held for the project team to discuss the pro-
ject milestones and deadlines. The Project Sponsor opened up the meeting by having
everyone in the room introduce themselves and say a few words.
The new man/woman that just joined the team today had a very take-charge attitude.
He/She stated that his/her prior experiences should be a good asset to the success of the
project. Furthermore, he/she emphasized that he/she felt that the deadline was no problem
with the current total of nine team members. The Project Sponsor then thanked him/her
for his/her insightful comments and opened up the floor to the rest of the team to get their
Spector & Jones 321
input. Nobody said anything to the contrary (although yesterday the general consensus
from the team was that we needed two to three additional people–not just one person).
After the meeting, some of your team members gathered around the watercooler to dis-
cuss the meeting that they just attended. The team starting discussing their opinions of this
new person and what that person said regarding that the deadline was “no problem.” You
didn’t say anything and left the kitchen with your coffee in hand, and began thinking about
your feelings about this new person. On your way out, you saw this new person whisper-
ing in a corner to the Project Sponsor and you wondered what they were talking about.
Received June 4, 2001
Accepted July 28, 2003
