Content uploaded by Shahidul Hassan
Author content
All content in this area was uploaded by Shahidul Hassan on Jun 30, 2020
Content may be subject to copyright.
Shahidul Hassan is assistant professor
of public management at the John Glenn
School of Public Affairs (The Ohio State
University). His research focuses on the role
that managerial practices play in improving
motivation, commitment, and performance
of public sector employees. His research
works have appeared in the Journal of
Managerial Psychology, Journal of
Leadership and Organization Studies,
International Public Management
Journal, and American Review of
Public Administration.
E-mail: hassan.125@osu.edu
716 Public Administration Review • September | October 2013
Public Administration Review,
Vol. 73, Iss. 5, pp. 716–725. © 2013 by
The American Society for Public Administration.
DOI: 10.1111/puar.12100.
Shahidul Hassan
The Ohio State University
is article examines how greater role clarifi cation
may be associated with increased work satisfaction and
decreased turnover rates in workgroups. ese linkages are
examined with the use of multivariate analysis of vari-
ance and hierarchical regression analysis for data collected
during two time periods from multiple sources: personnel
records and an organizational survey of 1,699 employees
working in 45 geographically distributed offi ces in a state
government agency. Results indicate that offi ces with a
high level of role clarifi cation had signifi cantly higher
levels of work satisfaction and lower rates of turnover.
Additionally, the eff ects of role clarifi cation on work sat-
isfaction and turnover behavior were mediated by overall
role clarity perceived in these offi ces. e implications of
these fi ndings for eff ective management of workgroups in
government agencies are discussed.
The institutional context in which public
organizations operate has important impli-
cations for their day-to-day operations and
performance (Perry and Rainey 1988; Rainey and
Steinbauer 1990; Wamsley and Zald 1973). Public
agencies deal with complex policy problems, pursue
value-laden goals, and provide services for which clear
performance criteria are not readily available (Allison
1983; Dahl and Lindblom 1953; Wildavsky 1979;
Wilson 1989). Public organizations also need to
attend to competing demands from various stakehold-
ers and interest groups (Lowi 1979; Rainey 1993;
Ring and Perry 1985). Public management scholar-
ship suggests that a lack of clear performance criteria,
policy complexity, and competing demands from
stakeholders lead to goal ambiguity (Chun and Rainey
2005), which, in turn, creates
substantial role ambiguity for
employees working in public
agencies (Wright 2004). Erera
(1989), for example, found that
vagueness and constant change
in state policies caused consid-
erable role ambiguity among
managers in a state agency.
Pandey and Wright (2006)
found that a lack of clarity in organizational goals
directly as well as indirectly (by increasing procedural
constraints) increased role ambiguity among managers
in health and human services agencies.
e consequences of high levels of role ambiguity
have important cost implications for public agen-
cies. While a certain level of role ambiguity is likely
to exist in all jobs and may even be benefi cial in
terms of increasing employee creativity and learning
(Savelsbergh et al. 2012), a high level of ambigu-
ity regarding job goals and performance expecta-
tions creates stress and frustration among employees
(Schaubroeck et al. 1993) and may infl uence them to
leave the organization (Jung 2011). Organizational
research indicates that work stressors, including role
ambiguity and role confl ict, contribute negatively to
employee mental health (Ganster and Schaubroeck
1991). Role ambiguity has also consistently been
shown to have a negative infl uence on a wide range of
benefi cial employee dispositions, including job satis-
faction, organizational commitment, and job involve-
ment (Fisher and Gitelson 1983; Jackson and Schuler
1985). More importantly, research indicates that
ambiguity about job goals and performance expecta-
tions lowers employee job performance (cf. Tubre and
Collins 2000).
Given the widely acknowledged costs associated with
role ambiguity, it is surprising that few studies in pub-
lic management have investigated how to enhance role
clarity in public organizations. Specifi cally, research
has yet to examine whether managerial practices can
attenuate the adverse eff ects
of role ambiguity on organiza-
tional outcomes. e present
study was designed to examine
how role clarifi cation may aff ect
perceived role clarity, work
satisfaction, and turnover rates
in distinct offi ces or workgroups
in a government agency. Based
on role theory (Graen 1976;
e Importance of Role Clarifi cation in Workgroups:
Eff ects on Perceived Role Clarity, Work Satisfaction,
and Turnover Rates
e present study was designed
to examine how role clarifi ca-
tion may aff ect perceived role
clarity, work satisfaction, and
turnover rates in distinct offi ces
or workgroups in a government
agency.
The Importance of Role Clarifi cation in Workgroups: Effects on Perceived Role Clarity, Work Satisfaction, and Turnover Rates 717
work practices can also increase role ambiguity (Jimmieson, Terry,
and Callan 2004; Lyons 1971). Additionally, lack of managerial
communication or poor communication may lead to increased role
ambiguity among employees (House and Mitchell 1974; House and
Rizzo 1972; Lyons 1971).
High levels of role ambiguity have detrimental eff ects on cooperative
attitudes and behaviors of employees (Jackson and Schuler 1985;
Tubre and Collins 2000). Role ambiguity increases stress because
concerns about how to perform job roles and obtain valued out-
comes (both material and social) often cause frustration and anxiety
among employees. It can also lead to, as previously noted, a lack of
employee commitment and involvement and diminished employee
performance (Jackson and Schuler 1985; Schaubroeck et al. 1993;
Tubre and Collins 2000). Additionally, incongruity between
received role requirements (i.e., role confl ict) can lower perceptions
of self-competence and engender frustration and dissatisfaction
among employees. Kahn and colleagues (1964) suggested that when
employees experience high levels of role ambiguity and role confl ict,
they will try to reduce the associated stress by avoiding the job situa-
tion through chronic absence or leaving the organization.
Impact of Role Clarifi cation on Perceived Role Clarity
in Workgroups
Role clarifi cation is a task-oriented leader behavior that is targeted
toward providing cognitive structures to subordinates about how
they can attain their job goals (House 1996; Yukl 2010). While
the main purpose of role clarifi cation is to guide and coordinate
subordinate work activities and make sure subordinates know what
they need to do, it also includes setting task objectives in work-
groups. Yukl and colleagues (Yukl 2010; Yukl, Gordon, and Taber
2002) noted that setting specifi c task objectives directs subordinates’
eff orts toward performance of important duties and responsibilities,
encourages search for effi cient ways of doing work, and facilitates
evaluation of performance by providing a benchmark against which
to compare it.
Role clarifi cation is a core component of initiating structure, one
of the two key leader behavior dimensions3 identifi ed in the Ohio
State leadership studies (Fleishman 1953; Fleishman and Harris
1962; Stogdill, Goode, and Day 1962). Role clarifi cation also is the
primary component of directive behavior in the path-goal theory
of leadership (House 1971; House and Mitchell 1974). Although
research on the consequences of using initiating structure was
inconclusive, studies on role clarifi cation found stronger results
(Fisher and Edwards 1988; Podsakoff et al. 1995; Woff ord and Liska
1993). Several studies showed that role clarifi cation is an important
determinant of managerial eff ectiveness (Kim and Yukl 1995; Yukl
and Van Fleet 1982). Additionally, laboratory and fi eld experiments
have consistently found that setting specifi c and challenging goals
results in higher levels of individual and group
performance (Locke and Latham 1990).
e eff ect of role clarifi cation on perceived role
clarity in workgroups may depend on several
situational factors, such as employee skill
level, experience, and job complexity (House
1996; House and Mitchell 1974). Workgroups
that are designed to accomplish complex and
Kahn et al. 1964; Katz and Kahn 1978) and path-goal theory of
leadership (House 1971, 1996; House and Mitchell 1974), role
clarifi cation in this study is anticipated to have an indirect positive
eff ect on work satisfaction and an indirect negative eff ect on turno-
ver behavior by improving perceived role clarity in workgroups.
ese linkages are examined with multivariate analysis of variance
(MANOVA) and hierarchical regression analysis with data that were
collected during two time periods using a survey and personnel
records of 1,699 employees working in 45 geographically distrib-
uted offi ces in an agency in state government. e following section
draws from the extant organizational research to develop theoreti-
cal arguments and a set of testable hypotheses about how greater
role clarifi cation may be associated with increased work satisfaction
and decreased turnover rates by improving perceived role clarity in
workgroups.
Theory and Hypotheses
Organizational scholars have long suggested that the development
and maintenance of roles are critical to the socialization, perform-
ance, and well-being of employees in organizations (Graen 1976;
Kahn et al. 1964; Katz and Kahn 1978). Roles basically are a set of
activities or behaviors that are expected by relevant organizational
constituents from a person holding a particular position in an
organization (Graen 1976; Katz and Kahn 1978). Organizations
tend to divide complex tasks into specialized activities, assign the
activities to specifi c roles, and then integrate the outputs of those
activities into fi nal goods and services (Katz and Kahn 1978). e
formal specifi cation of role requirements is intended to provide
guidance and direction to employees about how to carry out their
work activities, as well as to hold them accountable for specifi c levels
of performance (Rizzo, House, and Lirtzman 1970).
Role theory suggests that role making is a dynamic process involving
an employee and his or her direct supervisor1 whereby the employee
acquires knowledge about the demands and constraints placed on
his or her behavior, receives feedback regarding his or her behavior
in the role, accepts a pattern of behavior, and modifi es it over time
(Graen 1976; Katz and Kahn 1978; Naylor, Pritchard, and Ilgen
1980). Roles are seldom fully specifi ed in advance for organiza-
tional members (Schaubroeck et al. 1993). In addition, role-making
processes can be complicated by poor communication between role
senders2 and role receivers, as well as by turbulence within an organi-
zation’s task environment that requires constant modifi cations in
roles (Katz and Kahn 1978; Schaubroeck et al. 1993). Role ambi-
guity occurs when roles are not suffi ciently articulated in terms of
domain, methods of fulfi llment, and consequences of role perform-
ance (Kahn et al. 1964; Schaubroeck et al. 1993).
Kahn and colleagues (1964) identifi ed three organizational factors
that may contribute to greater role ambiguity: (1) organizational
complexity, (2) organizational change, and
(3) managerial communication. Increased
organizational complexity in terms of higher
levels of centralization, formalization, and
goal ambiguity can lead to greater role ambi-
guity (House and Rizzo 1972; Morris, Steers,
and Koch 1979; Nicholson and Goh 1983;
Pandey and Wright 2006). Organizational
changes that require frequent restructuring of
Workgroups that are designed
to accomplish complex and
nonroutine tasks, for example,
may require more role clarifi ca-
tion than workgroups that carry
out routine or simple tasks.
718 Public Administration Review • September | October 2013
may also infl uence members to cognitively as well as behaviorally
disassociate from the workgroup. e relationships between role
ambiguity and job satisfaction and between role ambiguity and
turnover have been studied extensively in organizational research.
A meta-analysis by Abramis (1994) showed a negative correla-
tion between role ambiguity and job satisfaction. Another meta-
analysis found a negative relationship between role ambiguity and
turnover behavior (Griff eth, Hom, and Gaertner 2000). Studies
in public management also found a negative connection between
role ambiguity and job satisfaction (Kim and Wright 2007; Wright
and Kim 2004; Wright and Davis 2003) and a positive connection
between role ambiguity and turnover intention (Jung 2012; Kim
and Wright 2007) at the individual level of analysis. But very little
research in public management has examined the linkage between
role ambiguity and turnover behavior of employees. One study
found a positive association between role ambiguity and turnover
rates in federal agencies (Jung 2011). Based on these results and
arguments from role theory, the following two hypotheses are
tested in this study:
Hypothesis 2: Offi ces with a high level of role clarity will
have higher levels of work satisfaction than offi ces with a low
level of role clarity.
Hypothesis 3: Offi ces with a high level of role clarity will
have lower rates of turnover than offi ces with a low level of
role clarity.
Impacts of Role Clarifi cation on Perceived Work Satisfaction
and Turnover Behavior
Role clarifi cation is likely to enhance work satisfaction and reduce
turnover rates by increasing role clarity perceived in workgroups.
Specifi cally, the eff ects of role clarifi cation eff orts on work satisfac-
tion and turnover behavior in this study are expected to be medi-
ated by overall role clarity perceived in the offi ces. Two hypotheses
concerning these relationships were formulated based on the works
of House and colleagues (House 1971, 1996; House and Mitchell
1974; House and Rizzo 1972) that suggested that clarifying job
duties and performance expectations reduces stress, which, in turn,
enhances employee satisfaction and decreases employee work with-
drawal behaviors.
While the eff ects of role clarifi cation on work satisfaction and
turnover behavior have not been thoroughly investigated in previous
studies, a positive connection between initiating structure4 and work
satisfaction has been found in research in business and industry
settings (Judge, Piccolo, and Ilies 2004). Research has also shown a
negative relationship between instrumental or task-oriented mana-
gerial communication and employee turnover behavior (Griff eth,
Hom, and Gaertner 2000). Additionally, a recent public sector
study found a negative correlation between
a measure of eff ective leadership, which
comprised aspects of role clarifi cation, and
employee turnover behavior (Grissom 2012).
However, no previous study has investigated
whether role clarity mediates the eff ects of role
clarifi cation on work satisfaction and turnover
rates in workgroups. Hence, the following two
hypotheses are tested in this study:
nonroutine tasks, for example, may require more role clarifi cation
than workgroups that carry out routine or simple tasks. Additionally,
workgroups with highly experienced members may require less
role clarifi cation than workgroups with a large number of new or
inexperienced members. Nevertheless, role clarifi cation is likely to be
important in enhancing role clarity in workgroups especially when
there is substantial uncertainty about work objectives and perform-
ance expectations (Yukl 2010). Such conditions are likely to be more
prevalent in government work settings because public organizations
often need to deal with complex problems and provide services for
which there are no clear performance criteria (Allison 1983; Dahl
and Lindblom 1953; Wildavsky 1979; Wilson 1989).
Although the connection between role clarifi cation and role clarity
has not been thoroughly investigated in research in public manage-
ment, studies by Wright and colleagues (Kim and Wright 2007;
Wright 2004; Wright and Davis 2003) found that performance-
oriented feedback enhances employee role clarity in public organiza-
tions. Additionally, survey studies in business and industry settings
have consistently found a negative connection between role clarifi ca-
tion and role ambiguity at the individual level of analysis (Woff ord
and Liska 1993). Several fi eld experiments also showed that role
clarifi cation has a negative impact on role ambiguity (Quick 1979;
Schaubroeck et al. 1993). Following these results, a positive associa-
tion between role clarifi cation and role clarity at the workgroup level
is anticipated in this study.
Hypothesis 1: Offi ces with a high level of role clarifi cation
will have higher levels of role clarity than offi ces with a low
level of role clarifi cation.
Linkages between Role Clarity and Perceived Work
Satisfaction and Turnover Behavior
Rizzo, House, and Lirtzman (1970) suggested that when employees
are unaware of what is expected of them, they may hesitate to act,
show a lack of self-determination, and feel unable to make a diff er-
ence in achieving the organization’s goals. Yukl (2010) noted that
even highly competent and motivated employees may fail to achieve
a high level of performance if they are unsure about their work goals
and responsibilities. Katz and colleagues (1964) asserted that unclear
role expectations may increase employee work withdrawal behaviors,
including absenteeism and turnover.
Research on small groups indicates that role clarity is essential for
the eff ective functioning of workgroups (Bray and Brawley 2002).
As previously noted, roles are a set of prescriptions that defi ne the
behaviors that are expected from a member of a workgroup (Katz
and Kahn 1978). Clear knowledge about these behavioral expecta-
tions is important for a group member to eff ectively perform his or
her work, as well as to coordinate work activities within the group.
When group members do not clearly under-
stand their responsibilities, they may under-
estimate their ability to achieve their group’s
goals (Bandura 1997), leading to a low level
of group performance (Bray and Brawley
2002).
A high level of role ambiguity is likely to
increase dissatisfaction in workgroups. It
A high level of role ambiguity
is likely to increase dissatisfac-
tion in workgroups. It may also
infl uence members to cogni-
tively as well as behaviorally dis-
associate from the workgroup.
The Importance of Role Clarifi cation in Workgroups: Effects on Perceived Role Clarity, Work Satisfaction, and Turnover Rates 719
were expected to do in their job and understood which of their
job duties were more important than others. Role clarifi cation was
assessed with four items (α = .84) that captured the extent to which
managers (1) clearly expressed work and performance expecta-
tions to subordinates, (2) adequately instructed subordinates about
how to carry out their work activities, (3) informed subordinates
about organizational issues or changes, and (4) provided feedback
to subordinates when they performed their job well. Work satisfac-
tion was measured with two items (α = .79) that were derived from
the Michigan Organizational Assessment Questionnaire. ese
two items have been used to measure work satisfaction in several
recent studies (Hassan and Rohrbaugh 2011; Shim and Rohrbaugh,
forthcoming).
Offi ce-level employee turnover records for 2009 were collected
in 2010 from 45 of the 65 offi ce locations. e agency could not
provide disaggregated turnover data for employees in 19 offi ce loca-
tions. e unavailability of turnover data from 19 offi ce locations
resulted in a reduction of the sample size from 2,136 to 1,699, a
retention rate of almost 80 percent. e demographic characteristics
of these respondents were in no way diff erent from respondents
who participated in the survey. e numbers of employees in the
45 offi ces ranged from 3 to 167; 16 offi ces were described by fewer
than 10 employees, while 11 offi ces were described by more than 50
employees. Employee turnover rates were calculated by dividing the
sum of voluntary separations by the total number of employees in
each offi ce.
Research indicates that various individual and group characteristics
may infl uence employee satisfaction and turnover behavior (Cho
and Lewis 2012; Griff eth, Hom, and Gaertner 2000; Grissom 2012;
Kellough and Osuna 1995; Shaw et al. 1998). Hence, offi ce size
(i.e., total number of employees in each offi ce) and mean position
tenure were included as control variables in the data analyses. ese
two variables were included in the analyses because larger offi ces,
for instance, may experience more employee turnover than smaller
offi ces. Offi ces with more experienced employees may require less
role clarifi cation than offi ces with less experienced employees.
Additionally, because the level of role clarity in workgroups may
depend on employee skill level and job complexity, the percentage
of professional employees working in the offi ces was included as a
Hypothesis 4: e infl uence of role clarifi cation on work
satisfaction will be mediated by overall role clarity perceived
in offi ces.
Hypothesis 5: e infl uence of role clarifi cation on turn-
over rates will be mediated by overall role clarity perceived in
offi ces.
Method of the Study
Sample and Procedures
e fi ve research hypotheses were tested using data that were col-
lected during two time periods from two diff erent sources: person-
nel records and a survey of employees working in 65 geographically
dispersed offi ces of a government agency with 11 distinct divisions
of operation. e agency was responsible for maintaining and
administering the states’ accounting, payroll, and retirement systems
for public employees. e agency also was responsible for review-
ing states’ contracts and conducting audits of other state agencies
and public benefi ciaries, as well as overseeing the fi scal aff airs of
local governments (including one of the largest municipalities in
the United States). e survey was designed and distributed to a
population of 2,614 employees in the spring of 2008 to collect data
regarding their perceptions of managerial practices and their work
climate. Responsibility for the internal distribution and collection of
the survey was assigned to division managers. Prior to distributing
the survey, the research team and division managers clearly com-
municated the purpose of the study to all participants, the voluntary
nature of their participation, and the complete anonymity and
confi dentiality of their responses. Altogether, 2,136 usable ques-
tionnaires were returned, for an overall response rate of 82 percent;
response rates by division ranged from a low of 70 percent to a high
of 100 percent.
Approximately 87 percent of the survey respondents identifi ed
themselves as Caucasian, 8 percent as African American, 3 percent
as Asian, 2 percent as Hispanic, and 1 percent as Native-American;
these percentages matched exactly the distribution generated from
agency personnel records. Approximately 60 percent of the respond-
ents were female and 40 percent male, nearly matching agency
records of 59 percent female and 41 percent male. With regard to
position description, 31 percent of the respondents reported that
their job was best described as clerical and 66 percent as profes-
sional. e majority of the professional employees in the agency
were accountants, auditors, business analysts, lawyers, information
technology specialists, and managers. e mean age of the respond-
ents was 45.6 years; their mean tenures in their position, division,
and agency were 4.8 years, 9.8 years, and 11.6 years, respectively; all
three tenure distributions were skewed positively.
Measures
All of the survey items were measured either on a six-point (coded
1–6) strength of agreement (strongly disagree, generally disagree,
disagree a little, agree a little, generally agree, and strongly agree) or
on a fi ve-point (coded 0–4) frequency of occurrence (almost never/
never, rarely, sometimes, often, and almost always/always) scale (see
table 1 for a complete list of the survey items used in this study).
Steers’s (1975, 1976) Task-Goal Attribute Scales provided the basis
for the three-item measure of role clarity (α =.73). ese three items
captured the extent to which employees were clear about what they
Table 1 CFA Results: Standardized Factor Loadings
Items
Standardized
Factor Loading (λ)
Role clarifi cation
My supervisor clearly expresses work expectations to me. .78
My supervisor keeps me “in the loop” about issues that
affect my work. .76
I am told by my immediate supervisor when I do a good job. .70
My supervisor properly instructs me regarding how to do my
job. .83
Role clarity
I know exactly what I am supposed to do on my job. .66
I understand fully which of my job duties are more impor-
tant than others. .64
My responsibilities at work are very clear and specifi c. .78
Work satisfaction
I am very satisfi ed with the kind of work that I do. .80
At the end of the day, I feel good about the work that I do
here. .82
720 Public Administration Review • September | October 2013
group variances with an expected variance under the null hypoth-
esis of no agreement. Two main advantages in using the rwg index
are that (1) it does not depend on the between-groups variance
and thus is particularly useful when the group means are restricted
in their range, and (2) it provides an agreement measure for each
group rather than one summary for the entire sample (Cohen,
Doveh, and Eick 2001).7 Although there is no clear standard for
acceptable levels of inter-rater agreement, a value of .70 or higher
for the rwg index is considered desirable (Janz, Colquitt, and Noe
1997). e calculated median rwg values for role clarifi cation, role
clarity, and work satisfaction were .79, .86, and .70, respectively,
which indicated suffi cient within-group agreement in perceptions of
the three measures.
Descriptive Statistics and Results of Correlation Analyses
Table 2 provides descriptive statistics and correlation coeffi cients
of the measures included in the present study. e mean position
tenure of employees in the offi ces was close to fi ve years, and the
standard deviation was 3.49 years. On average, each offi ce had
approximately 37 employees, and 60 percent of the employees were
professionals. e minimum rate of annual turnover in the offi ces
was zero; the maximum was .33. On average, 11 percent of employ-
ees from each offi ce had voluntarily left the organization by the end
of 2009. As indicated in table 2, role clarifi cation had a positive
correlation with role clarity perceived in the offi ces (r = .49, p <
.01). Additionally, measures for role clarifi cation, role clarity, and
work satisfaction had signifi cant negative correlations with turnover
rates in the offi ces (rs = –.33, –.49, –.32, respectively, p < .05). e
results also showed that role clarifi cation scores were lower in larger
offi ces (r = –.34, p < .05) and higher in offi ces with a larger number
of professional employees (r = .27, p < .05).
Tests of the Research Hypotheses
A three-step analytical procedure was undertaken to test the hypoth-
eses identifi ed for the present study. Initially, the 45 offi ces were
divided into three groups with low (turnover rates = .00 to .07, n =
15 offi ces), moderate (turnover rates = .08 to .14, n = 17 offi ces) and
high (turnover rates = .15 to .33, n = 13 offi ces) rates of turnover.8
A multivariate analysis of variance (MANOVA) test was performed
to assess direct connections between role clarifi cation, role clarity,
and work satisfaction and turnover rates in the offi ces. Next, assum-
ing a positive result from the fi rst step, univariate F statistics were
examined for the three measures with respect to low, moderate, and
high rates of turnover. Finally, hierarchical regression analyses were
performed to examine direct and indirect eff ects of role clarifi cation
on perceived role clarity and work satisfaction and turnover rates in
the offi ces.
control variable as well. e rationale was that offi ces with a high
number of professional employees were likely to perform more com-
plex or nonroutine tasks and therefore might have less role clarity
than offi ces with a high number of clerical or support employees.
Results
Psychometric Properties of the Measures
Confi rmatory factor analyses (CFA) were conducted to assess the
discriminant validity of the study measures. is study relied on one
measure of parsimony fi t—root mean square error of approximation
(RMSEA)—and two measures of relative fi t—comparative fi t index
(CFI) and Tucker Lewis fi t index (TLI)—to assess validity of the
measures. A reasonable fi t is indicated by CFI and TLI values of .90
or greater and RMSEA value of .08 or less (Hu and Bentler 1995).
e CFA results indicated that the three-factor measurement model
provided a good fi t to the data (χ2 = 221.20 [24], p < .05, RMSEA =
.07, CFI = .97, TLI = .94). Standardized factor loadings (lambdas)
for all items were .60 or greater (see table 1), and factor intercorrela-
tions were below .65. Additionally, a sequential chi-square diff erence
test was performed to examine whether the proposed three-factor
measurement model provided a better fi t to the data than alterna-
tive one-factor and two-factor models. e results indicated that the
one-factor model fi t the data signifi cantly worse than the proposed
three-factor model (Δχ2 = 1052.8, p < .05, RMSEA = .17, CFI =
.79, TLI = .64). e three-factor model also fi t the data signifi cantly
better than the two-factor model, in which the correlation between
role clarifi cation and role clarity was set to one (Δχ2 = 525.8, p <
.05, RMSEA = .13, CFI = .87, TLI = .79). ese results indicated
suffi cient discriminant validities for the three measures. Moreover,
poor fi t of the one-factor model suggested that common method
bias did not materially aff ect the data collected through survey in
the fi rst phase of this study.5
Data Aggregation for Offi ce-level Analyses
e compatibility principle regarding unit of analysis suggests
that when the outcome variable is measured at the group level, the
predictor variables should also be measured at the same level (Ajzen
2005; Pugh and Dietz 2008). Because turnover behavior in this
study was measured at the workgroup or offi ce level,6 scale scores for
role clarifi cation, role clarity, and work satisfaction were aggregated
at the offi ce level to test the fi ve research hypotheses. Several recent
studies in public management have also relied on this approach
to assess relationships at the group level (Fernandez, Cho, and
Perry 2010; Hassan and Rohrbaugh 2012; Jung 2011; Rubin and
Kellough 2012; Shim and Rohrbaugh, forthcoming).
Methodologists have suggested that suffi cient between-group vari-
ability and within-group agreement must be demonstrated before
any study of relationships between constructs at the group level
(Hoff man 2002). Between-group diff erences typically are assessed
with an examination of F statistics from one-way analysis of vari-
ance (ANOVA) across the intended groups. In the present study,
there was an ANOVA result for each of the three measures across 45
offi ces. All F statistics were signifi cant (p < .05), indicating that the
three measures varied appropriately by offi ce location. In addition,
within-group agreement in the three measures was assessed by exam-
ining values for the rwg index developed by James, Demaree, and
Wolf (1984). e rwg index was originally developed as a measure of
inter-rater reliability. It is estimated by comparing observed within
Table 2 Means, Standard Deviations, and Correlation Coeffi cients
Measures Mean SD 1 2 3 4 5 6
Percentage of
professionals
.60 .24 —
Mean position
tenure
4.69 3.49 .02
Offi ce size 37.09 45.94 –.06 .03
Role clarifi cation 3.88 .38 .27* –.19 –.34*
Role clarity 4.50 .28 –.01 .09 –.24 .49**
Work satisfaction 4.88 .40 .06 –.18 –.31* .61** .70**
Turnover rates .11 .09 .09 –.12 –.06 –.33* –.49** –.32*
**p < .01; *p < .05; N = 45.
The Importance of Role Clarifi cation in Workgroups: Effects on Perceived Role Clarity, Work Satisfaction, and Turnover Rates 721
Scheff e’s post hoc comparisons also revealed that offi ces with a high
rate of turnover had a signifi cantly lower level of work satisfaction
than offi ces with a low rate of turnover.
A four-step procedure proposed by method-
ologists (Baron and Kenny 1986; Judd and
Kenny 1981; MacKinnon, Fairchild, and
Fritz 2007) was followed to examine direct
and indirect eff ects of role clarifi cation on
work satisfaction and turnover behavior in
offi ces.9 Table 4 presents the results of media-
tion analyses. Figure 1 provides a graphical
representation of the results of hierarchical
regression analyses.10 As indicated in table 4, role clarifi cation was
found to be a signifi cant predictor of perceived role clarity in the
offi ces (β = .54, p < .01; R2 = .22, F = 4.45, p < .01). Role clarifi ca-
tion also had a positive eff ect on work satisfaction (β = .59, p < .01;
R2 = .35, F = 6.90, p < .01). When role clarity was included in the
second step of regression analyses, the coeffi cient for the direct eff ect
of role clarifi cation on work satisfaction became smaller in magni-
tude (β = .28, p < .05), which indicated that role clarity partially
mediated the eff ect of role clarifi cation on work satisfaction. Role
clarity explained an additional 22 percent of the total variance in
work satisfaction (ΔF = 23.45, p < .01). As shown in table 4, role
clarifi cation was found to have a signifi cant negative impact on
turnover rates in the offi ces (β = –.49, p < .01; R2 = .13, F = 2.74,
p < .05). Role clarity had a signifi cant negative eff ect on turnover
rates in the offi ces, as well (β = –.39, p < .05). When role clarity was
included in the second step of regression analyses, the infl uence of
role clarifi cation on turnover rates became nonsignifi cant, which
indicated that role clarity fully mediated the infl uence of role clarifi -
cation on turnover rates. Role clarity explained an additional 11 per-
cent of the total variance in turnover rates in the offi ces (ΔF = 6.31,
p < .01). ese results provided support for all fi ve hypotheses.11
Discussion
Limitations of the Study
Before discussing the implications of this study, it is important to
note its limitations. Because data for role clarifi cation, role clarity,
and work satisfaction were collected from the same source and at
MANOVA results supported the anticipated connections between
role clarifi cation, role clarity, work satisfaction, and turnover
behavior in offi ces. Hotelling’s trace (T = .40, F = 2.66, p < .05)
and Wilks’ lambda (λ= .70, F = 2.69, p < .05) were both statistically
signifi cant for the combination of the three
measures. is result led appropriately to an
examination of the three univariate F statis-
tics, as shown in table 3. All three F statistics
were statistically signifi cant (p < .05). Scheff e’s
post hoc comparisons suggested that manag-
ers in offi ces with low and moderate rates of
turnover were signifi cantly more active in
clarifying subordinates’ roles and responsibili-
ties than those in offi ces with a high rate of turnover. In addition,
offi ces that experienced a low rate of turnover had signifi cantly
higher levels of role clarity than offi ces with a high rate of turnover.
Table 4 Results of Hierarchical Regression Analyses
Role Clarity Work Satisfaction Turnover Rates
Measures βtβtβt
Step 1 Percentage of professionals –.17 –1.24 –.10 –.81 .22 1.52
Offi ce size –.08 –.69 –.13 –.98 –.18 –1.20
Mean position tenure .20 1.49 –.06 –.48 –.21 –1.50
Role clarifi cation .54** 3.70 .59** 4.35 –.48** –3.08
Adjusted R2.22 .35 .13
F-ratio 4.45** 6.90** 2.74*
Step 2 Percentage of professionals –.01 –.06 .15 1.10
Offi ce size –.07 –.70 –.24 –1.54
Mean position tenure –.17 –1.68 –.14 –1.00
Role clarifi cation .28* 2.26 –.26 –1.59
Role clarity .56** 4.85 –.39* –2.51
Adjusted R2.57 .23
F-ratio 13.20** 3.72*
ΔR2.22 .10
ΔF23.45** 6.31**
** p < .01; *p < .05; N = 45.
Turnover
Rates
Role
Clarification
Role
Clarity
Work
Satisfaction
–.39**
.56**
.54**
.28*
R2= .22
R2= .57
R2= .23
Figure 1 Summary of Hierarchical Regression Analyses Results
**p < .01; *p < .05; N = 45.
Offi ces that experienced a low
rate of turnover had signifi -
cantly higher levels of role clar-
ity than offi ces with a high rate
of turnover.
Table 3 Univariate F Tests from MANOVA: Different Rates of Turnover in Offi ces
Measures
Low
(n = 15)
Moderate
(n = 17)
High
(n = 13) F-ratio
Role clarifi cation 3.97a3.97a3.60b3.99*
Role clarity 4.64a4.51ab 4.34b4.98**
Work satisfaction 5.08a4.84ab 4.71b3.24*
Different subscripts indicate signifi cant differences in turnover rates at p < .05.
**p < .01; *p < .05.
722 Public Administration Review • September | October 2013
clarifi cation eff orts can attenuate the adverse eff ects of role ambigu-
ity on work satisfaction and turnover behavior in workgroups. e
results also are promising because the fi ve research hypotheses were
tested with data that were collected from multiple sources and in
diff erent time periods.
e fi ndings also contribute to the emerging literature on mana-
gerial leadership in public organizations. Although managerial
leadership is considered important in improving performance of
public agencies (Rainey and Steinbauer 1999; Terry 1995; Van
Wart 2003), there has been limited research on how specifi c leader
behavior aff ects organizational outcomes. Recent studies have
investigated how transformational leadership practices may lead to
positive outcomes in public organizations (Moynihan, Pandey, and
Wright 2012; Wright, Moynihan, and Pandey 2012; Vigoda-Gadot
and Beeri 2012), but limited research has examined how task- and
relations-oriented leader behaviors aff ect eff ectiveness of workgroups
in public agencies. e current study, therefore, extends the litera-
ture by examining how role clarifi cation infl uences work satisfaction
and turnover rates in workgroups in a public agency.
Much of previous research on employee turnover in public agen-
cies focused on intended rather than actual turnover of employees.
e few studies that examined turnover behavior focused primarily
on the impact that demographic factors have on turnover rates in
federal agencies (Cho and Lewis 2012; Kellough and Osuna 1995;
Lewis 1991; Lewis and Park 1989). e present study indicated that
managerial practices may play an important role in reducing turno-
ver of public employees. Specifi cally, the results indicated that an
increase in role clarifi cation by one standard deviation resulted in a
decrease in turnover rates by almost half a standard deviation. Given
that the average rate of turnover in the offi ces was 11 percent, the
size of this eff ect may appear relatively small. Nevertheless, even a
slight reduction in turnover could conceivably mean a large savings
in costs because when a skilled employee leaves, an agency not only
loses a source of valuable institutional knowledge but also needs to
spend resources on training a new employee that otherwise could be
used for an important public program (Kim 2005; Moynihan and
Landuyt 2008; Moynihan and Pandey 2008).
Finally, the results of this study have important implications for
public managers’ practice. e fi ndings suggest that reducing role
ambiguity is particularly important for increasing satisfaction and
retention of employees. Managers can take a number of steps to
reduce role ambiguity in their workgroups. Specifi cally, managers
can reduce role ambiguity in workgroups by setting specifi c task
objectives, communicating clearly about performance expecta-
tions, and providing clear direction about
how to accomplish work activities. Keeping
subordinates informed about changes in the
organizational environment that may have
an impact on their work and providing them
with periodic feedback about their perform-
ance are also likely to improve role clarity
in workgroups. While setting task objec-
tives and providing directions and feedback
are important for improving role clarity in
workgroups, it is also important for managers
to recognize that subordinates may vary in
the same time, causal connections between these three variables
could not be ascertained in this study. Relationships between these
variables may involve reciprocal causality over time rather than uni-
directional causality. Future research should examine causal relation-
ships between these variables with an experimental design and use of
longitudinal data.
e possibility that common source bias infl uenced the research
fi ndings could not be fully ruled out, even though data for employee
turnover were collected from personnel records in the agency. But
steps were introduced in the design of this study to reduce the
potential of common method bias in the results. For example, ano-
nymity of response was assured repeatedly in all correspondence and
evidenced in every aspect of data collection. e particular items
used in the present study were interspersed widely over seven pages
and nine distinct sections in the questionnaire. ese items also
were almost evenly split in the directionality of their wording and
varied response formats were used. Because these steps were taken,
problems of common method bias were attenuated, if not fully
eliminated. e CFA results also provided some assurance that such
problems did not materially aff ect the results.
While offi ce size, mean job tenure, and percentage of professional
employees were included as control variables in the regression analy-
ses, none of these variables had any infl uence on work satisfaction
and turnover rates in the offi ces. Research indicates that a variety of
individual, job, and organization related factors including employee
satisfaction with pay, training, career advancement opportunity,
job characteristics, quality of manager–employee relationships,
workgroup cohesion and diversity, as well as labor market condi-
tions may infl uence employee satisfaction and turnover behavior
(Griff eth, Hom, and Gaertner 2000; Grissom 2012; Jung 2011;
Selden and Moynihan 2000; Shaw et al. 1998). However, these
variables were not included in the analyses because of unavailability
of data. is particular limitation suggests that the indirect eff ects
of role clarifi cation on work satisfaction and turnover behavior that
were found in this study should be interpreted with caution and
viewed as not fully conclusive. A fuller understanding of how role
clarifi cation aff ects employee satisfaction and retention requires
additional empirical tests of the hypothesized indirect relationships
with a larger number of workgroups from a broader range of public
organizations. Further, given that the sample included 45 offi ces
from a public agency, the regression analyses provided a conserva-
tive test of the hypothesized relationships and the results may have
limited generalizability.
Implications for Research and Practice
Despite the limitations, the fi ndings can be
considered noteworthy in several ways. e
results provided new insight about how role
clarifi cation may lead to benefi cial outcomes
in workgroups. Specifi cally, the results
indicated that clarifying work objectives and
performance expectations improved overall
role clarity perceived in the offi ces, which, in
turn, enhanced work satisfaction and reduced
turnover rates in those offi ces. ese fi ndings
are noteworthy, as no previous study in public
management has investigated whether role
Specifi cally, the results indicated
that clarifying work objectives
and performance expectations
improved overall role clarity
perceived in the offi ces, which,
in turn, enhanced work satisfac-
tion and reduced turnover rates
in those offi ces.
The Importance of Role Clarifi cation in Workgroups: Effects on Perceived Role Clarity, Work Satisfaction, and Turnover Rates 723
behavior in the offi ces, but the results indicated no evidence in support of an
interaction eff ect.
References
Abramis, David J. 1994. Work Role Ambiguity, Job Satisfaction, and Job
Performance: Meta-Analyses and Review. Psychological Reports 74(3): 1411–33.
Allison, Graham T. 1983. Public and Private Management: Are ey Fundamentally
Alike in All Unimportant Respects? In Public Management: Public and Private
Perspectives, edited by James L. Perry and Kenneth L. Kraemer, 72–92. Palo Alto,
CA: Mayfield.
Ajzen, Icek. 2005. Attitudes, Personality, and Behavior. 2nd ed. Milton Keynes, UK:
Open University Press.
Bandura, Albert. 1997. Self-Effi cacy: e Exercise of Control. New York: W. H.
Freeman.
Baron, Reuben M., and David A. Kenny. 1986. e Moderator-Mediator Variable
Distinction in Social Psychological Research: Conceptual, Strategic, and
Statistical Considerations. Journal of Personality and Social Psychology 51(6):
1173–82.
Bray, Steven R., and Lawrence R. Brawley. 2002. Role Effi cacy, Role Clarity, and
Role Performance Eff ectiveness. Small Group Research 33(2): 233–53.
Cho, Yoon Jik, and Gregory B. Lewis. 2012. Turnover Intention and Turnover
Behavior: Implications for Retaining Federal Employees. Review of Public
Personnel Administration 32(1): 4–23.
Chun, Young H., and Hal G. Rainey. 2005. Goal Ambiguity and Organizational
Performance in U.S. Federal Agencies. Journal of Public Administration Research
and eory 15(4): 529–57.
Cohen, Ayala, Etti Doveh, and Uri Eick. 2001. Statistical Properties of the rwg Index
of Agreement. Psychological Methods 6(3): 297–310.
Dahl, Robert A., and Charles E. Lindblom. 1953. Politics, Economics, and Welfare:
Planning and Politico-Economic Systems Resolved into Basic Social Processes. New
York: Harper & Bros.
Erera, Irit P. 1989. Role Ambiguity in Public Welfare Organizations. Administration
in Social Work 13(2): 67–83.
Fernandez, Sergio, Yoon J. Cho, and James L. Perry. 2010. Exploring the Link
between Integrated Leadership and Public Sector Performance. Leadership
Quarterly 21(2): 308–23.
Fisher, Bruce M., and Jack E. Edwards. 1988. Consideration and Initiating Structure
and eir Relationships with Leader Eff ectiveness: A Meta-Analysis. Academy of
Management Best Papers Proceedings, August, 201–5.
Fisher, Cynthia D., and Richard Gitelson. 1983. A Meta-Analysis of the Correlates of
Role Confl ict and Ambiguity. Journal of Applied Psychology 68(2): 320–33.
Fleishman, Edwin A. 1953. e Description of Supervisory Behavior. Journal of
Applied Psychology 37(1): 1–6.
Fleishman, Edwin A., and Edwin F. Harris. 1962. Patterns of Leadership Behavior
Related to Employee Grievances and Turnover. Personnel Psychology 15(1): 43–56.
Ganster, Daniel C., and John Schaubroeck. 1991. Work Stress and Employee Health.
Journal of Management 17(2): 235–71.
Graen, George B. 1976. Role-Making Processes within Complex Organizations.
In Handbook of Industrial and Organizational Psychology, edited by Marvin D.
Dunnette, 1201–45, Chicago: Rand-McNally.
Griff eth, Rodger W., Peter W. Hom, and Stefan Gaertner. 2000. A Meta-Analysis of
Antecedents and Correlates of Employee Turnover: Update, Moderator Tests,
and Research Implications for the Next Millennium. Journal of Management
26(3): 463–88.
Grissom, Jason A. 2012. Revisiting the Impact of Participative Decision Making on
Public Employee Retention: e Moderating Infl uence of Eff ective Managers.
American Review of Public Administration 42(4): 400–418.
Hassan, Shahidul, and John Rohrbaugh. 2011. e Role of Psychological Climate
on Public Sector Employees’ Organizational Commitment: An Empirical
their need for role clarifi cation. Experienced employees are likely
to feel less uncertain about their work goals and thus may need
only general directions, whereas new or inexperienced employees
may require more elaborate and specifi c instructions in carrying
out their work activities. erefore, it is important that public
managers adapt their role clarifi cation eff orts to the needs of each
subordinate.
Acknowledgments
e author would like to thank John Rohrbaugh (State University
of New York, Albany), Anand Desai ( e Ohio State University),
and the three anonymous reviewers for their insightful comments
and suggestions on this article.
Notes
1. Relevant peers and subordinates can also be members of an employees’ role
set (see Graen 1976 for more information about role-making processes in
organizations).
2. Role senders are members of a focal employee’s role set, which may include his or
her manager, subordinates, and relevant peers (Katz and Kahn 1978).
3. e other key behavior dimension in the Ohio State leadership studies was
consideration.
4. Role clarifi cation is a key component of initiating structure in the Ohio State
leadership studies (Fleishman 1953; Fleishman and Harris 1962; Stogdill,
Goode, and Day 1962).
5. A test of Harman’s single factor also was done to check for the presence of com-
mon method bias in the data. ree factors emerged from the analysis, which
accounted for 68 percent of the total item variance. e largest factor did not
account for a majority of the variance (44 percent), which suggested that same
source bias did not adversely aff ect the data.
6. e research team collected offi ce- rather than individual-level turnover records
because the survey did not have any code/identifi er (to protect the anonymity of
the respondents). Specifi cally, it was not possible to link survey responses with
individual turnover records.
7. e rwg index is calculated using the following formula:
rwg = J * [1 − (savg
2/σ2)]/ [J * [1 – (savg
2/σ2)] + savg
2/σ2],
where savg
2 is the average of the observed variances on the J items and σ2 is the
variance of a null distribution corresponding to some null response pattern.
8. Offi ces with turnover rates below one-half of one standard deviation from the
mean were categorized as low, above one-half of one standard deviation from the
mean were categorized as high, and within one-half of one standard deviation
from the mean were categorized as moderate.
9. Kenny and colleagues (Baron and Kenny 1986; Judd and Kenny 1981) suggested
that to establish mediation, fi rst, the independent variable (IV) must have a
signifi cant direct eff ect on the dependent variable (DV). Second, the IV should
also have a signifi cant direct eff ect on the mediator variable (MV). ird, the MV
must be signifi cantly related to the DV when both the IV and MV are predic-
tors of the DV. Fourth, the coeffi cient relating the IV to the DV in the regression
model in which both the IV and MV are predictors should become nonsignifi -
cant (i.e., full mediation) or become smaller in magnitude (i.e., partial mediation)
than the initial regression coeffi cient estimated in step one. Recently, MacKinnon,
Fairchild, and Fritz (2007) suggested that the fi rst step identifi ed by Kenny and
colleagues (Baron and Kenny 1986; Judd and Kenny 1981) is not always neces-
sary to establish mediation, especially when the goal is to examine indirect eff ects.
10. Because of the small sample size, the regression analyses did not control for the
eff ects of 11 divisions in which the 45 offi ces were located.
11. Supplemental analysis also was conducted to assess whether role clarifi cation
moderated the eff ects of role clarifi cation on work satisfaction and turnover
724 Public Administration Review • September | October 2013
Kim, Soonhee, and Bradley E. Wright. 2007. Information Technology Employee
Work Exhaustion: Toward an Integrated Model of Antecedents and
Consequences. Review of Public Personnel Administration 27(2): 147–70.
Lewis, Gregory B. 1991. Turnover and the Quiet Crisis in the Federal Civil Service.
Public Administration Review 51(2): 145–55.
Lewis, Gregory B., and Kyungho Park. 1989. Turnover Rates in Federal White-Collar
Employment: Are Women More Likely to Quit than Men? American Review of
Public Administration 19(1): 13–28.
Locke, Edwin A., and Gary P. Latham. 1990. A eory of Goal Setting and Task
Performance. Englewood Cliff s, NJ: Prentice Hall.
Lowi, eodore J. 1979. e End of Liberalism: e Second Republic of the United
States. New York: W. W. Norton.
Lyons, omas F. 1971. Role Clarity, Need for Clarity, Satisfaction, Tension, and
Withdrawal. Organizational Behavior and Human Performance 6: 99–110.
MacKinnon, David P., Amanda J. Fairchild, and Matthew S. Fritz. 2007. Mediation
Analysis. Annual Review of Psychology 58: 593–614.
Morris, James H., Richard M. Steers, and James L. Koch. 1979. Infl uence of
Organization Structure on Role Confl ict and Role Ambiguity for ree
Occupational Groupings. Academy of Management Journal 22(1): 58–71.
Moynihan, Donald P., and Noel Landuyt. 2008. Explaining Turnover Intention in
State Government: Examining the Roles of Gender, Life Cycle, and Loyalty.
Review of Public Personnel Administration 28(2): 120–43.
Moynihan, Donald P., and Sanjay K. Pandey. 2008. e Ties at Bind: Social
Networks, Person–Organization Value Fit, and Turnover Intention. Journal of
Public Administration Research and eory 18(2): 205–27.
Moynihan, Donald P., Sanjay K. Pandey, and Bradley E. Wright. 2012. Setting the
Table: How Transformational Leadership Fosters Performance Information Use.
Journal of Public Administration Research and eory 22(1): 143–64.
Naylor, James C., Robert D. Pritchard, and Daniel R. Ilgen. 1980. A eory of
Behavior in Organizations. New York: Academic Press.
Nicholson, Peter J., and Sewe C. Goh. 1983. e Relationship of Organization
Structure and Interpersonal Attitudes to Role Confl ict and Ambiguity in
Diff erent Work Environments. Academy of Management Journal 26(1): 148–55.
Pandey, Sanjay K., and Bradley E. Wright. 2006. Connecting the Dots in Public
Management: Political Environment, Organizational Goal Ambiguity, and the
Public Manager’s Role Ambiguity. Journal of Public Administration Research and
eory 16(4): 511–32.
Perry, James L., and Hal G. Rainey. 1988. e Public–Private Distinction in
Organizational eory: A Critique and Research Strategy. Academy of
Management Review 13(2): 182–201.
Podsakoff , Philip M., Scott B. MacKenzie, Michael Ahearne, and William H.
Bommer. 1995. Searching for a Needle in a Haystack: Trying to Identify Illusive
Moderators of Leadership Behaviors. Journal of Management 21(3): 423–70.
Pugh, S. Douglas, and Joerg Dietz. 2008. Employee Engagement at the Organizational
Level of Analysis. Industrial and Organizational Psychology 1(1): 45–48.
Quick, James C. 1979. Dyadic Goal Setting and Role Stress: A Field Study. Academy
of Management Journal 22(2): 241–52.
Rainey, Hal G. 1993. A
eory of Goal Ambiguity in Public Organizations. In
Research in Public Administration, edited by James L. Perry, 121–66. Greenwich,
CT: JAI Press.
Rainey, Hal G., and Paul Steinbauer. 1999. Galloping Elephants: Developing
Elements of a eory of Eff ective Government Organizations. Journal of Public
Administration Research and eory 9(1): 1–32.
Ring, Peter Smith, and James L. Perry. 1985. Strategic Management in Public and
Private Organizations: Implications of Distinctive Contexts and Constraints.
Academy of Management Review 10(2): 276–86.
Rizzo, John, Robert J. House, and Sidney I. Lirtzman. 1970. Role Confl ict and
Ambiguity in Complex Organizations. Administrative Science Quarterly 15(2):
150–63.
Assessment of ree Occupational Groups. International Public Management
Journal 14(1): 27–62.
———. 2012. Variability in the Organizational Climate of Government Offi ces and
Aff ective Organizational Commitment. Public Management Review 14(5): 563–84.
Hoff man, David A. 2002. Issues in Multilevel Research: eory Development,
Measurement, and Analysis. In Handbook of Research Methods in Industrial and
Organizational Psychology, edited by Steven G. Rogelberg, 467–511. Malden,
MA: Blackwell.
House, Robert J. 1971. A Path-Goal eory of Leader Eff ectiveness. Administrative
Science Quarterly 16(3): 321–38.
———. 1996. Path-Goal eory of Leadership: Lessons, Legacy and a Reformulated
eory. Leadership Quarterly 7(3): 323–52.
House, Robert J., and Terence R. Mitchell. 1974. Path-Goal eory of Leadership.
Journal of Contemporary Business 3(4): 1–97.
House, Robert J., and John R. Rizzo. 1972. Role Confl ict and Ambiguity as Critical
Variables in a Model of Organizational Behavior. Organizational Behavior and
Human Performance 7(3): 467–505.
Hu, Li-tze, and Peter M. Bentler. 1995. Evaluating Model Fit. In Structural Equation
Modeling: Concepts, Issues, and Applications, edited by Richard H. Hoyle, 76–99.
London: Sage Publications.
Jackson, Susan E., and Randall S. Schuler. 1985. A Meta-Analysis and Conceptual
Critique of Research on Role Ambiguity and Role Confl ict in Work Settings.
Organizational Behavior and Human Decision Processes 36(1): 16–78.
James, Lawrence R., Robert G. Demaree, and Gerrit Wolf. 1984. Estimating Within-
Group Interrater Reliability With and Without Response Bias. Journal of Applied
Psychology 69(1): 85–98.
Janz, Brian D., Jason A. Colquitt, and Raymond A. Noe. 1997. Knowledge
Worker Team Eff ectiveness: e Role of Autonomy, Interdependence, Team
Development, and Contextual Support Variables. Personnel Psychology 50(4):
877–904.
Jimmieson, Nerina L., Deborah J. Terry, and Victor J. Callan. 2004. A Longitudinal
Study of Employee Adaptation to Organizational Change: e Role of Change-
Related Information and Change-Related Self-Effi cacy. Journal of Occupational
Health Psychology 9(1): 11–27.
Judd, Charles M., and David A. Kenny. 1981. Process Analysis: Estimating
Mediation in Treatment Analysis. Evaluation Review 5(5): 602–19.
Judge, Timothy A., Ronald F. Piccolo, and Remus Ilies. 2004. e Forgotten Ones?
e Validity of Consideration and Initiating Structure in Leadership Research.
Journal of Applied Psychology 89(1): 36–51.
Jung, Chan S. 2011. Predicting Organizational Actual Turnover Rates in U.S. Federal
Government. International Public Management Journal 13(3): 297–317.
———. 2012. Why Are Goals Important in the Public Sector? Exploring the
Benefi ts of Goal Clarity for Reducing Turnover Intention. Journal of Public
Administration Research and eory. Published electronically on December 10.
doi:10.1093/jopart/mus058.
Kahn, Robert L., Donald M. Wolfe, Robert P. Quinn, J. Diedrick Snoek, and Robert
A. Rosenthal. 1964. Organizational Stress: Studies in Role Confl ict and Ambiguity.
New York: Wiley.
Katz, Daniel, and Robert L. Kahn. 1978. e Social Psychology of Organizations. 2nd
ed. New York: Wiley.
Kellough, J. Edward, and Will Osuna. 1995. Cross-Agency Comparisons of Quit
Rates in the Federal Service: Another Look at the Evidence. Review of Public
Personnel Administration 15(4): 58–68.
Kim, Helen, and Gary Yukl. 1995. Relationships of Managerial Eff ectiveness and
Advancement to Self-Reported and Subordinate-Reported Leadership Behaviors
from the Multiple-Linkage Mode. Leadership Quarterly 6(3): 361–77.
Kim, Soonhee. 2005. Factors Aff ecting State Government Information Technology
Employee Turnover Intentions. American Review of Public Administration 35(2):
137–56.
The Role of Role Clarity Remains Unclear 725
Rubin, Ellen V., and J. Edward Kellough. 2012. Does Civil Service Reform Aff ect
Behavior? Linking Alternative Personnel Systems, Perceptions of Procedural
Justice, and Complaints. Journal of Public Administration Research and eory
22(1): 121–41.
Savelsbergh, Chantal, Josette M. P. Gevers, Beatrice I. J. M. van der Heijden, and Rob F.
Poell. 2012. Team Role Stress: Relationships with Team Learning and Performance
in Project Teams. Group and Organization Management, 37(1): 67–100.
Schaubroeck, John, Daniel Ganster, Wesley E. Sime, and David Ditman. 1993. A
Field Experiment Testing Supervisor Role Clarifi cation. Personnel Psychology
46(1): 1–25.
Selden, Sally Coleman, and Donald P. Moynihan. 2000. A Model of Voluntary
Turnover in State Government. Review of Public Personnel Administration 20(2):
63–74.
Shaw, Jason, John Delery, Douglas Jenkins, and Nina Gupta. 1998. An
Organization-Level Analysis of Voluntary and Involuntary Turnover. Academy of
Management Journal 41(5): 511–25.
Shim, Dong Chul, and John Rohrbaugh. Forthcoming. An Explanation of
Diff erences between Government Offi ces in Employees’ Organizational
Citizenship Behavior. Public Management Review.
Steers, Richard M. 1975. Task-Goal Attributes, n Achievement, and Supervisory
Performance. Organizational Behavior and Human Decision Processes 13(3):
392–403.
———. 1976. Factors Aff ecting Job Attitudes in a Goal-Setting Environment.
Academy of Management Journal 19(1): 6–19.
Stogdill, Ralph M., Omar S. Goode, and David R. Day. 1962. New Leader Behavior
Description Subscales. Journal of Psychology 54(2): 259–69.
Terry, Larry D. 1995. Leadership of Public Bureaucracies: e Administrator as
Conservator. ousand Oaks, CA: Sage Publications.
Tubre, Travis C., and Judith M. Collins. 2000. Jackson and Schuler (1985) Revisited:
A Meta-Analysis of the Relationships between Role Ambiguity, Role Confl ict,
and Job Performance. Journal of Management 26(1): 155–69.
Van Wart, Montgomery. 2003. Public-Sector Leadership eory: An Assessment.
Public Administration Review 63(2): 213–28.
Vigoda-Gadot, Eran, and Ira Beeri. 2012. Change-Oriented Organizational
Citizenship Behavior in Public Administration: e Power of Leadership and
the Cost of Organizational Politics. Journal of Public Administration Research and
eory 22(3): 573–96.
Wamsley, Gary A., and Mayer N. Zald. 1973. e Political Economy of Public
Organizations: A Critique and Approach to the Study of Public Administration.
Lexington, MA: Lexington Books.
Wildavsky, Aaron. 1979. Speaking Truth to Power: e Art and Craft of Policy Analysis.
Boston: Little, Brown.
Wilson, James Q. 1989. Bureaucracy: What Government Agencies Do and Why ey Do
It. New York: Basic Books.
Woff ord, J. C., and Laurie Z. Liska. 1993. Path-Goal eories of Leadership: A
Meta-Analysis. Journal of Management 19(4): 857–76.
Wright, Bradley E. 2004. e Role of Work Context in Work Motivation: A Public
Sector Application of Goal and Social Cognitive eories. Journal of Public
Administration Research and eory 14(1): 59–78.
Wright, Bradley E., and Brian S. Davis. 2003. Job Satisfaction in the Public Sector:
e Role of the Work Environment. American Review of Public Administration
33(1): 70–90.
Wright, Bradley E., and Soonhee Kim. 2004. Participation’s Infl uence on Job
Satisfaction: e Importance of Job Characteristics. Review of Public Personnel
Administration 24(1): 18–40.
Wright, Bradley E., Donald P. Moynihan, and Sanjay K. Pandey. 2012. Pulling the
Levers: Transformational Leadership, Public Service Motivation, and Mission
Valence. Public Administration Review 72(2): 206–15.
Yukl, Gary. 2010. Leadership in Organizations. 7th ed. Upper Saddle River, NJ:
Prentice Hall.
Yukl, Gary, Angela Gordon, and Tom Taber. 2002. A Hierarchical Taxonomy of
Leadership Behavior: Integrating Half a Century of Leadership Research. Journal
of Leadership and Organization Studies 9(1): 15–32.
Yukl, Gary, and David Van Fleet. 1982. Cross-Situational, Multi-Method Research
on Military Leader Eff ectiveness. Organizational Behavior and Human
Performance 30(1): 87–108.
