JOURNAL OF PUBLIC RELATIONS RESEARCH, 13(4), 303–320

Copyright © 2001, Lawrence Erlbaum Associates, Inc.

Public Relations Roles

and Media Choice
Tom Kelleher
School of Communications
University of Hawaii at Manoa

A national survey of Public Relations Society of America members (n = 267) was

conducted to examine the relationship between public relations roles and media
choice based on the integration of public relations theory and media richness theory.
Respondents were identified as either public relations managers or technicians using
confirmatory factor analysis, corroborating previous research. Managers reported
spending more time in oral communication than technicians, whereas technicians
spend more time using written communication. E-mail use in public relations and re-
lated communities is discussed, and areas for future research on new media and media
choices in public relations are explored.

Theory-building research from the field of organizational communication offers a

useful perspective from which to approach the question of how public relations
people accomplish the range of communication tasks that comprise their life’s
work. One provocative thread in this fabric of research, media richness theory, spe-
cifically addresses the topic of individual media choice for various communication
tasks. In the field of public relations, the range of communication tasks can be dis-
cussed in terms of the basic roles of public relations practitioners. Manager and
technician roles are among the most robust constructs in public relations research.
This study seeks to achieve two goals. The first is to expand the scope of current
public relations theory by connecting some of its central constructs with theory
from a related discipline. Public relations roles theory and media richness theory
are reviewed to develop hypotheses, which then are tested empirically. The second
goal is to explore a newer area of inquiry—the media choices of public relations
practitioners, including e-mail. To explore this area, preliminary research ques-
tions are asked, and the resulting data are used to advance the line of questioning.

Requests for reprints should be sent to Tom Kelleher, School of Communications, 2560 Campus Road,
George Hall 309, University of Hawaii at Manoa, Honolulu, HI 96822–5396. E-mail: [email protected]
304 KELLEHER

PUBLIC RELATIONS ROLES

Organizational communication theorists have examined the nature of communica-

tion tasks by comparing tasks based on job category, or “horizontal differentia-
tion,” and based on organizational level, or “vertical differentiation” (Rice &
Shook, 1990, p. 197). Individuals who work within a common job category “per-
form similar activities, are socialized in similar ways, and are confronted with simi-
lar information processing requirements” (p. 197). Exploratory research has sug-
gested that public relations practitioners do not share strong, common definitions of
their jobs compared to people in related fields such as journalism because of varia-
tions in workplace socialization across organizations (Berkowitz &
Hristodoulakis, 1999). Horizontal differentiation therefore seems better suited for
examining communication tasks within the context of organizations rather than
across the field of public relations as a whole.
Vertical differentiation, however, has been documented widely and offers a
useful framework for examining public relations roles and communication tasks
throughout the field. One of the most well-developed descriptive concepts in the
field of public relations is the vertical role distinction between those who pri-
marily perform management activities and those who primarily perform techni-
cal activities.
Dozier (1992) reviewed previous literature and empirical findings on public re-
lations roles beginning with the work of Broom and Smith (1978, 1979). Four roles
emerged: (a) expert prescriber, (b) communication facilitator, (c) problem-solving
process facilitator, and (d) communication technician. The expert prescriber role
describes public relations consultants who top management turn to for treatment of
public relations problems. The communication facilitator acts as a “go-between”
between an organization and its publics, working to ensure smooth dialogue and
accurate mutual interpretation. The problem-solving process facilitator works
closely with top management to handle public relations problems in a systematic,
process-oriented manner. Finally, the communication technician role describes the
large portion of public relations practitioners who primarily provide technical ser-
vices such as news release writing, event planning, and graphic design.
Dozier (1992) critiqued both the theoretical conceptualization of these four
roles and the empirical tests used to construct the roles operationally, and encour-
aged future researchers to use a simpler, two-factor conceptualization. That is, he
found it more useful to distinguish simply between managers and technicians:
“Managers make policy decisions and are held accountable for public relations
program outcomes,” whereas “technicians carry out the low-level mechanics of
generating communication products that implement policy decisions made by oth-
ers” (p. 333).
Dozier and Broom (1995) then tested the stability of the two-role conceptual-
ization by comparing results from a 1991 survey sample of practitioners from the
 PUBLIC RELATIONS ROLES AND MEDIA CHOICE 305

Public Relations Society of America (PRSA) to data from a 1979 PRSA survey
and other prior research. Using confirmatory factor analysis, they found the
two-role factor solution to be consistent over time.
In addition to these two roles, however, Toth, Serini, Wright, and Emig (1998)
identified an “agency profile” among a national sample of PRSA members. Toth et
al.’s agency role included the tasks of counseling, research, programming, com-
municating with clients and coworkers, and handling correspondence with media.
The agency role emerged from Toth et al.’s factor analysis as a third factor follow-
ing the manager and technician factors.
However, as all of the roles researchers cited here pointed out, manager and
technician roles refer only to the primary functions of a public relations practitio-
ner. That is, public relations people normally do not function only as managers or
only as technicians, but primarily as managers or primarily as technicians. Opera-
tionally, Dozier and Broom (1995) based the manager–technician distinction on an
orthogonal contrast in which factor scores on one role are independent of factor
scores on the other.
In theory, organizations with public relations practitioners functioning as man-
agers are more likely to practice two-way, open-systems models of public relations
(Grunig, J. E. 1992). One-way models of public relations—termed press
agentry–promotion and public information by J. E. Grunig and his colleagues—are
more likely to involve the technical tasks of producing and disseminating informa-
tion with less regard for feedback. Two-way models, on the other hand, require
public relations practitioners to monitor the environment, interpret feedback, and
participate in strategic decisions with management: “Conceptually, the role (man-
ager) and the functions (two-way asymmetric and symmetric models) go hand in
hand” (Dozier, 1992, p. 347). This study applies the parsimonious manager–tech-
nician conceptualization to examine media choices based on vertical differences in
public relations roles.

MEDIA RICHNESS THEORY

The basic idea behind media richness theory is that people match communication
tasks with the media they perceive to be most efficient for accomplishing the tasks.
More specifically (and in the language of media richness theory), people tend to
choose richer media to handle more equivocal communication tasks. Further dis-
cussion requires an explanation of two key variables: richness and equivocality.

Richness

According to the founders and main proponents of media richness theory, “media
can be characterized as high or low in ‘richness’ based on their capacity to facilitate
meaning” (Daft, Lengel, & Trevino, 1987, p. 358). This capacity results from a me-
306 KELLEHER

dium’s ability to convey (a) quick feedback, (b) personal focus, (c) multiple com-
munication cues, and (d) language variety. Theorists and empirical researchers
consistently have placed face-to-face communication as the high-end benchmark
for richness (e.g., Daft & Lengel, 1984; Daft et al., 1987; D’Ambra & Rice, 1994;
Trevino, Lengel, Bodensteiner, Gerloff, & Muir, 1990; Trevino, Lengel, & Daft,
1987). Telephone contact consistently falls immediately below face-to-face inter-
action in ability to convey rich information. Media found to be less rich than
face-to-face and telephone conversations include voice mail, e-mail, written mail
such as letters and memos, unaddressed letters and memos, fliers and bulletins, and
numeric computer reports.
Whereas face-to-face communication allows communicators to exchange imme-
diate feedback, vary their tone of voice, use body language and visual aids, and pre-
cisely tailor communication for specific individuals, other media are limited in at least
one, if not all, of these capacities. According to the basic tenets of media richness the-
ory, then, oral communication is generally richer than written communication.

Equivocality

Whereas theorists have used the concept of richness to describe communication

channels in terms of the richness of information they can convey, theorists have
used the concept of equivocality to describe the qualitative nature of situations re-
quiring communication (e.g., Daft et al., 1987; Trevino, Daft, & Lengel, 1990;
Weick, 1979).
People dealing with equivocal tasks often do not know the exact questions to
ask or the exact places to look for answers. Responding to equivocal situations in
organizations requires people to “exchange subjective interpretations and over-
come conflicting views in order to reach agreement about problem interpretation
and to enact a solution” (Trevino et al., 1990, pp. 177–178). Examples of equivo-
cal tasks used in this body of research include “you need clarification from a supe-
rior on a crucial issue affecting your zone/branch” (D’Ambra & Rice, 1994, p.
232) and “to get an explanation from a peer in another department of a complicated
technical matter in which you have little formal training or experience” (Daft et al.,
1987, p. 360). Equivocality has come to mean “the existence of multiple and con-
flicting interpretations about an organizational situation” (p. 357).
Given these conceptual definitions of richness and equivocality, the central hy-
pothesis of media richness theory becomes obvious. People are more likely to se-
lect richer media to handle information tasks that they perceive to be more
equivocal. Media richness theory has been applied as a rational choice model to
describe the decisions made by people who hold managerial positions in organiza-
tions. It also has been applied in a prescriptive manner, showing that those who
display a high degree of media sensitivity are those who earn higher performance
evaluations (Daft et al., 1987) and are more likely to advance to positions farther
 PUBLIC RELATIONS ROLES AND MEDIA CHOICE 307

up in organizational hierarchies (Rice & Shook, 1990). In these studies, media sen-
sitivity was measured by the level of agreement between the theory’s predictions
and the managers’ media choices. However, not all people use the same criteria to
choose media for various information tasks, and not all workers are managers.
Indeed, media richness theory has generated a fair amount of criticism. Markus
(1994) suggested two weaknesses. First, she suggested that media richness scales
may be inaccurate. Although Markus conceded (and her quantitative data corrobo-
rate) that empirical richness ratings of traditional channels such as face-to-face, tele-
phone, and memos are consistent with media richness theory, she challenged the
theory’s ability to explain media choices with newer media such as e-mail. Markus
also made the theoretical argument that richness scales “may be irrelevant, because
there are more important determinants of individual behavior than personal percep-
tions of media appropriateness as defined by information richness theory” (pp.
506–507). She cited research showing that managers also are influenced by situa-
tional constraints and social factors aside from any inherent media characteristics
they perceive (Fulk, Steinfield, Schmitz, & Power, 1987; Markus, 1987).
According to Markus, social factors play an especially prominent role when
universal access has not been achieved with a medium. Universal access means
that nearly everyone in a community with whom an individual wishes to commu-
nicate can be assumed to use the medium “on a regular basis” (Markus, 1994, p.
508). Markus (1987) defined a community as “a group of individuals with some
common interest and stronger communication flows within than across its bound-
aries” (p. 492) and proposed that the only stable states of interactive media use in a
community are “all or nothing” (p. 500).
In terms of social factors and situational constraints, this means that e-mail de-
cisions may be based more on socially perceived characteristics of media and ex-
pectations of reciprocity than on any objective characteristics of media such as the
richness scales proposed by media richness theorists. Markus (1994) suggested
that, compared to traditional media, “collective experience with electronic mail is
extremely limited … one cannot so readily expect that social definitions of elec-
tronic mail will be similar from one organization to the next” (p. 509). According
to Markus, e-mail was not a universal-access medium at the time she wrote her
1994 article. Therefore, she reasoned, media richness theory could not accurately
predict when people would use e-mail.
On the other hand, Markus (1994) found the fit between managers’ perceptions
of traditional media and predictions based on media richness theory to be “very
good” (pp. 514–515).1 She reasoned that media richness theory predicts managers’

1Markus (1994) tested information richness theory’s “core” prediction that managers’ media choices

would “agree” with the theory based on the content of information tasks (e.g., managers will choose
face-to-face interaction for information tasks containing highly equivocal information). She also tested
extensions of the theory based on situational and symbolic factors rather than content factors, but found
little support for the extensions.
308 KELLEHER

decisions to use traditional forms of interaction such as face-to-face meetings, tele-

phones, and written memos because these ways of communicating have long met
the criteria of universal access:

It seems, then, that given sufficient social experience with a communication technol-
ogy, one would expect to find considerable similarities in perceptions and use patterns
across organizations and considerable correspondence between the perceptions and
uses of a technology and its material characteristics. This suggests that the richness
scale of Daft, Lengel, and colleagues might reflect not only the functional capabilities
and limitations of traditional media of business communication like the telephone and
the interoffice memorandum, but also the converged social definitions [emphasis in
original] of these media shared widely among managers today. (p. 509)

The scope of media richness theory may be limited to explanations of media

choices involving traditional media in which social definitions of media have
converged with the description of each medium offered by the theory. And,
based on Markus’ propositions, universal access is a prerequisite to converged
social definitions.

ROLES, RICHNESS, AND E-MAIL

The concepts of public relations roles and media richness both are born from orga-
nizational communication theory that suggests that the communication structures
of organizations are developed and adapted in response to external forces. Man-
agers are more likely to handle decisions and tasks concerning interpretations of ex-
ternal forces. Nonmanagers are more likely to spend their time implementing the
strategies developed by managers. Rice and Shook (1990) reviewed literature on
job categories and organizational level and concluded that “higher organizational
levels involve reducing equivocality” and “lower-level job categories are more
likely to involve operational and technical matters” (p. 198). Dozier and Broom
(1995) added that such categories “provide a useful way to link individual attributes
of practitioners and characteristics of the public relations unit to the function’s par-
ticipation in the management of organizations” (p. 6). Media richness theory sug-
gests that those facing equivocal situations are more likely to use rich communica-
tion channels like face-to-face meetings than those facing lean, technical tasks,
who are more likely to choose lean media such as written memos.
Rice and Shook (1990) conducted a meta-analysis of studies linking organiza-
tional level with use of communication channels and found that higher level man-
agers spent more time communicating via oral media and less time communicating
via written media than did lower level employees. That is, the total amount of time
spent in oral communication (both face-to-face and telephone) was positively and
significantly (p < .001) correlated with organizational level. And, the total amount
 PUBLIC RELATIONS ROLES AND MEDIA CHOICE 309

of time spent communicating via textual media (reading and writing) correlated
negatively and significantly (p < .001) with organizational level.
Individual media choice, then, can be linked to public relations roles with the
following hypotheses (Hs).

H1: Public relations managers will report spending more time communicating
with oral communication channels than technicians.
H2: Public relations technicians will report spending more time communicating
with traditional written channels (e.g., letters, memos, news releases) than
managers.

Although media richness theorists place telephone contact directly below

face-to-face interaction in terms of richness, it is unclear whether public relations
managers or technicians use this mode of communication more. In the aforemen-
tioned meta-analysis, Rice and Shook (1990) found that telephone use correlated
negatively and significantly (p < .01) with organizational level. People lower in or-
ganizational hierarchies reported more telephone use than people in higher man-
agement positions. Whether public relations managers spend more time at work on
the telephone specifically than public relations technicians is phrased better as a
research question (RQ) than as a hypothesis:

RQ1: Do public relations managers or technicians spend more time communicat-

ing on the telephone?

Likewise, the current dearth of data on e-mail use among public relations prac-
titioners invites the next research question:

RQ2: Do public relations managers or technicians spend more time communicat-

ing via e-mail?

Whereas the answer to the telephone question (RQ1) likely will offer a stable
indicator of use among public relations people, the e-mail question (RQ2) may
yield an answer that only reflects a single point in time in a changing media envi-
ronment. Whether the current state of e-mail use will last depends on whether
e-mail has become a universal-access medium. Markus (1987), in discussing her
theory of critical mass for communication media, defined universal access in the
context of communities. Markus’s definition of a community is a group of people
with some common interest and stronger communication ties within than outside
the group. The question of universal access, then, is framed best in terms of various
potential communities related to work in public relations such as the organization
where the person is employed, clients, the media, the field of public relations, and
people in the geographical community in which the person works:
310 KELLEHER

RQ3: What portion of public relations people’s coworkers, clients outside their
own organization, media contacts, public relations contacts outside their or-
ganization, and local (nonmedia) contacts have easy access to e-mail?
RQ4: What portion of public relations people’s coworkers, clients outside their own
organization, media contacts, public relations contacts outside their organiza-
tion, and local (nonmedia) contacts respond to e-mail on a regular basis?
RQ5: In general, to whom do public relations people use e-mail to communicate
with most often?

METHOD

A mail questionnaire was pretested with 47 respondents from the Hawaii chapter of
PRSA. Based on feedback from pretest respondents, a revised questionnaire was
designed.
The sampling frame for the main study was a February 2000 national database
of PRSA members. A random sample of 800 address labels was computer gener-
ated for the revised mail survey. The goal for this sampling method was to survey
enough respondents to conduct analyses comparable to the confirmatory factor
analyses of major public relations roles conducted by Dozier and Broom (1995),
who sampled PRSA membership in 1979 (n = 458) and again in 1991 (n = 207).
The actual mailing included 793 questionnaires because 7 international mailing
labels were excluded due to cost and underrepresentation of international practi-
tioners. Three were returned for incorrect addresses, and 8 were returned incom-
plete by retirees. In all, then, the sample included 782 possible respondents. Two
hundred sixty-seven questionnaires were returned with usable data, for a response
rate of 34%.2

Role Measures

To measure public relations roles, the questionnaire included a set of 7-point,

Likert-type items composed of 15 manager items (e.g., “I plan and recommend
courses of action for solving public relations problems”) and 6 technician items
(e.g., “I handle the technical aspects of producing public relations materials”) from
Dozier and Broom (1995, pp. 11–12).
A principal-components factor analysis using varimax rotation was performed
on these 21 items seeking a two-factor solution. The results of this confirmatory

2Budget and time constraints prevented a follow-up mailing, which likely would have increased the

response rate. E-mail and fax follow-ups were ruled out due to the potential bias in media choice re-
sponses among those who may only respond to requests via these media.
 PUBLIC RELATIONS ROLES AND MEDIA CHOICE 311

factor analysis are consistent with prior research. As in previous studies, the first
factor to emerge was the manager factor followed by the technician factor (Table
1). A reliability test for the manager scale yielded an alpha of .89. Reliability anal-
ysis for the six-item technician set yielded an alpha of .73. Manager and technician
factor scores were generated for each respondent. Those with higher scores on the
manager factor than on the technician factor are designated as managers and those
with higher scores on the technician factor are designated as technicians in the re-
mainder of the analyses. (See Dozier & Broom, 1995, for a more detailed discus-
sion of this method of defining respondents’ roles relative to other members of the
same sample.)
Replicating Dozier and Broom’s (1995) confirmatory factor analysis on practi-
tioner roles serves three purposes. First, it tests the stability of the two-factor solu-
tion found in 1979 and confirmed in 1991 with yet another test from the same
general population in 2000. Second, the confirmatory factor analysis divides the
current group of respondents into two groups—managers and technicians—for the
sake of hypothesis testing. Third, using these well-established constructs for the
main manager–technician comparisons provides some generalizability to data
from studies with larger response rates.
Men in this study reported an average of 16.7 years of experience in public rela-
tions and women reported an average of 11.3 years experience. These data parallel
the findings of Dozier and Broom’s 1991 data set, in which men and women re-
ported averages of 16.9 and 11.0 years, respectively. Furthermore, Toth et al.
(1998) reported that men in their 1995 national sample of PRSA members aver-
aged 17 years of experience and women averaged 11 years of experience. These
nearly identical figures indicate a favorable comparison between the current sam-
ple and national population of PRSA members in recent years.
Dozier and Broom (1995) also tracked the difference between women and men
on manager role factor scores. In the 1979 data, they found men significantly more
likely to enact the manager role, even after controlling for years of experience in
public relations. In the 1991 data, they again found that men were more likely to
enact the manager role, but this difference was not statistically significant after
controlling for years of experience in public relations and respondents’ years with
their current employers. In the 1979 data set, 28% of women and 57% of men pri-
marily enacted the manager role. In the 1991 data set, 39% of women and 55% of
men primarily enacted the manager role. In this study’s 2000 data set, 42% of
women and 60% of men primarily enacted the manager role. The difference be-
tween men and women is significant (p < .01) before controlling for years of expe-
rience in public relations, but this difference is insignificant after respondents’
years of experience in public relations are accounted for.3 Table 2 lists effect sizes

3See Toth et al. (1998) for a more thorough discussion of gender-related trends in public relations roles.
 TABLE 1
Factor Loadings for Role Items of the 1979, 1991, and 2000 Public Relations Society
of America Surveys

Item Description 1979 1991 2000

Manager
I diagnose public relations problems and explain them to others in
the organization. .80 .85 .83
Because of my experience and training, others consider me the
organization’s expert in solving public relations problems. .82 .80 .81
I make the communication policy decisions. .70 .70 .77
In meetings with management, I point out the need to follow a
systematic public relations planning process. .82 .76 .76
I keep management informed of public reactions to organizational
policies, procedures, and/or actions. .76 .74 .75
I operate as a catalyst in management’s decision making. .79 .83 .75
I plan and recommend courses of action for solving public relations
problems. .83 .84 .74
When working with managers on public relations, I outline
alternative approaches for solving problems. .73 .71 .73
I take responsibility for the success or failure of my organization’s
public relations program. .77 .80 .72
I create opportunities for management to hear the views of various
internal and external publics. .61 .53 .72
I observe that others in the organization hold me accountable for the
success or failure of public relations programs. .75 .74 .69
I encourage management participation when making the important
public relations decisions. .74 .71 .68
I report public opinion survey results to keep management informed
on the opinions of various publics. .61 .67 .56
I conduct communication audits to identify communication
problems between the organization and various publics. .56 .57 .54
I work with managers to increase their skills in solving and/or
avoiding public relations problems. .73 .73 .41
Technician
I handle the technical aspects of producing public relations
materials. .82 .83 .78
I produce brochures, pamphlets, and other publications. .77 .70 .72
I do photography and graphics for public relations materials. .61 .65 .71
I edit and/or rewrite for grammar and spelling the materials written
by others in the organization. .50 .59 .60
I maintain media contacts and place press releases. .72 .55 .53
I am the person who writes public relations materials presenting
information on issues important to the organization. .67 .58 .53

Note. Factor data from 1979 and 1991 are from Dozier and Broom (1995).

312
 PUBLIC RELATIONS ROLES AND MEDIA CHOICE 313

TABLE 2
Effect Sizes (r)

Manager–Technician Manager Role Technician Role

Variable Dichotomy Factor Score Factor Score

% of day in oral communication .23*** .22** –.28***

% of day in written (non–e-mail)
communication –.23*** –.10 .28***
% of day in face-to-face
communication .17** .17** –.24***
% of day on telephone .17** .15* –.15*
% of day on e-mail .03 –.11 –.07
Number of face-to-face meetings .26*** .22*** –.19**
Number of e-mails sent .28*** .25*** –.15*
Number of e-mails read .32*** .23*** –.18**
Female–male dichotomy –.17** –.07 .20**
Female–male dichotomy
(controlling for years of
experience) –.06 .02 .13*

*p < .05. **p < .01. ***p < .001.

based on the manager–technician dichotomy, manager role factor scores, and tech-
nician role factor scores.

Communication Channels

Two types of questions were used to determine how much time respondents spend
communicating via various oral and written channels. The simplest set of questions
asked respondents how many hours they work in an average workday and how
much of that time was spent in oral communication and in written communication.
Many responses indicated trouble with the validity of this set of questions. Only
52% reported oral and written communication hours that summed to equal the num-
ber of hours in their total workday.4
A somewhat more precise set of questions asked respondents to estimate the
percentage of their “time communicating in an average working day” spent on spe-
cific response options developed in the pretest. Response options included
face-to-face communication, telephone calls, e-mail communication, written
memos (writing or reading), written letters (writing or reading), bulletins or news-

4Aside from the difficulty of estimating such broad categorizations of time, respondents apparently
differed on what they considered communication. For example, some might have considered reading as
time spent in written communication whereas others did not think of reading as communication at all in
the context of this survey.
314 KELLEHER

letters (writing or reading), written press releases (writing or reading), other writ-
ten communication, and other electronic communication.
Of the 262 valid responses to this item set, 91% added to exactly 100% and 99%
were within 10 percentage points of 100%. Although responses to these questions
cannot be interpreted as much more than rough estimates, this set of items forced
respondents to consider the amount of time they spend in specific communication
tasks relative to other communication tasks. Therefore, the item set serves this
study’s hypothesis tests and comparison-based research questions well.
Regarding e-mail specifically, respondents were first asked if they had easy ac-
cess to e-mail for work purposes. The next set of questions asked them to report
how many e-mail messages they sent, how many messages they received and did
not read, and how many messages they received and read in their most recent full
day of work. Eighty-four percent were able to check e-mail records to confirm
their responses to these items. Another question asked respondents to report the
number of face-to-face meetings they had in the previous day.5

E-Mail Use in Public Relations-Related Communities

To explore the adoption of e-mail in various communities, the survey asked respon-
dents what portion of their coworkers, clients outside their own organization, me-
dia contacts, public relations contacts outside their organization, and local
(nonmedia) contacts had easy access to e-mail. Respondents also were asked to re-
port the portion of those with easy access to e-mail in each of the groups listed who
“respond regularly to your e-mail.” At the end of the questionnaire, respondents
rank-ordered these six groups according to whom they correspond with most often
via e-mail and via any medium.

FINDINGS

Role Differences

The first hypothesis, “public relations managers will report spending more time
communicating with oral communication channels than technicians,” was sup-
ported. The mean percentage of time that managers estimated spending on
face-to-face and telephone communication activities combined was 41.91 (SD =
17.23). For technicians, the mean percentage was 34.07 (SD = 15.35). A t test re-
vealed this difference was significant at p < .001, t(236) = 3.71. Managers estimated
a larger percentage of time for face-to-face communication specifically (M = 23.17,
SD = 14.35) than technicians (M = 18.77, SD = 11.32), p < .01, t(236) = 2.63. A sig-

5Answers to the e-mail and face-to-face questions that were reported as a range (e.g., “10–15” e-mail

messages) were coded as an average (e.g., 12.5 messages).

 PUBLIC RELATIONS ROLES AND MEDIA CHOICE 315

nificant difference also showed up in the number of face-to-face meetings manag-

ers reported having the day before completing the survey (M = 3.34, SD = 2.60),
compared to technicians (M = 2.22, SD = 1.43), p < .001, t(234) = 4.11. (See Table 2
for effect sizes based on the manager–technician dichotomy, manager role factor
scores, and technician role factor scores.)
The second hypothesis, “public relations technicians will report spending more
time communicating with traditional written channels (e.g., letters, memos, news
releases) than managers,” also was supported at p < .001, t(236) = 3.65. The esti-
mated percentage of time spent on nonoral, non–e-mail communication averaged
45.91 (SD = 20.60) for technicians and 36.88 (SD = 17.37) for managers. Table 3
displays the mean estimated percentages for each type of communication.
RQ1 asked whether managers or technicians spend more time communicating
on the phone. As displayed in Table 3, managers estimated an average of 19% of
their working day was spent on the phone, whereas technicians estimated 15%. A
two-tailed test of this difference is significant at p < .01, t(236) = 2.66.

E-Mail

RQ2 asked whether managers or technicians spend more time communicating via
e-mail. No significant differences in terms of time on e-mail were found. Because
managers reported longer workdays (M = 9.18 hr, SD = 1.86) compared to techni-
cians (M = 8.69 hr, SD = 1.17), two-tailed p < .05, t(226) = 2.44, the e-mail compari-
son was made again after multiplying the percentage-of-workday estimates by the
number of hours estimated to account for the difference in workday length. Still, no
significant differences were found. However, when asked how may e-mail mes-

TABLE 3
Percentage of Day Spent in Communication Activities as Estimated
by Managers and Technicians

Managers Technicians

Channels of Communication M SD M SD

Face to face 23.17 14.35 18.77 11.32

E-mail 20.63 13.46 19.69 14.42
Telephone 18.74 9.88 15.30 10.06
Written press releases (writing or
reading) 8.09 9.66 9.41 9.13
Written memos (writing or reading) 7.20 6.62 6.26 5.08
Written letters (writing or reading) 5.34 6.57 5.43 5.36
Bulletins or newsletters (writing or
reading) 4.81 6.30 10.64 11.53
Other written communication 7.97 9.55 9.71 11.65
Other electronic communication 3.48 4.67 4.46 5.21
316 KELLEHER

sages they received and how many they sent, managers reported a significantly
larger volume of messages coming across their monitors. Managers reported send-
ing an average of 25.14 (SD = 22.91) e-mail messages, compared to technicians’
average of 14.49 (SD = 12.32), two-tailed p < .001, t(236) = 4.51. And managers re-
ported reading an average of 33.25 (SD =30.24) e-mail messages during the day be-
fore completing the survey compared to technicians’ average of 17.74 (SD =
13.00), two-tailed p < .001, t(235) = 5.20. Almost all (99%) of the 267 respondents
reported having easy access to e-mail for work purposes; 2 reported no access, and
4 did not answer the question.
Table 4 addresses RQ3, regarding access to e-mail among the communities in
which public relations people work. Likewise, Table 5 covers RQ4 regarding re-
sponsiveness to e-mail.
The final research question asked, “To whom do public relations people use
e-mail to communicate with most often?” Seventy-one percent of respondents said
they communicate with co-workers via e-mail, 18% reported clients as their main
e-mail contacts, 5% reported they communicate most with other PR people via
e-mail, 3% said media, and less than 2% said local, non-media contacts were their
most common e-mail contacts. These answers mirrored responses to the question

TABLE 4
Percentages of Public Relations-Related Communities Estimated
to Have Easy Access to E-Mail

None (0) Few (1) Half (2) Most (3) All (4) N/Aa M SD

Coworkers 0.0 3.4 4.5 19.1 69.3 3.7 3.60 .74

Other public relations people 0.0 0.4 3.0 50.2 41.6 4.8 3.40 .57
Clients 0.0 1.1 7.5 52.1 30.3 9.0 3.23 .65
Media 0.0 4.1 11.6 52.8 23.2 8.3 3.04 .75
Other business contacts 0.0 1.9 15.0 68.9 11.2 2.9 2.92 .59
aIncludes “not applicable” responses and missing data.

TABLE 5
Percentages of Public Relations-Related Communities Estimated
to Respond Regularly to E-Mail

None (0) Few (1) Half (2) Most (3) All (4) N/Aa M SD

Coworkers 0.0 1.9 3.0 44.9 44.6 5.6 3.40 .65

Other public relations people 0.0 1.5 4.9 61.8 24.0 7.8 3.12 .59
Clients 0.0 2.2 7.1 55.1 27.7 7.8 3.12 .67
Media 1.9 13.1 19.9 36.7 11.6 16.8 2.52 .99
Other business contacts 0.4 3.4 7.5 65.9 15.7 7.1 3.00 .66
aIncludes “not applicable” responses and missing data.
 PUBLIC RELATIONS ROLES AND MEDIA CHOICE 317

of which group PR people correspond with most often via any channel of
ocmmunication: 74% ranked co-workers first, 18% ranked clients first and 4%
ranked media first among the six groups. The remaining groups of contacts were
each ranked first by less than 2% of respondents.

LIMITATIONS

Sample and Response Rate

Although the national sample of PRSA members used in this study is comparable to
previous studies, the smaller response rate weakens its generalizability. Compari-
sons in which statistically significant differences were not found are more subject
to Type II errors in drawing conclusions than if the response rate were higher. In
addition, PRSA members may differ from public relations practitioners in other
professional organizations or public relations practitioners with no professional af-
filiation. Nonetheless, the major hypothesis tests and comparisons in this study are
based on the distinction between those who primarily enact the public relations
manager role relative to those in the same sample who primarily enact the techni-
cian role.
Although the focal comparisons made in this study are based on vertical differ-
entiation in public relations roles, future studies might benefit from data on the
horizontal variation in job descriptions within and across organizations. For exam-
ple, Toth et al.’s (1998) research on the “agency profile” suggests the importance
of identifying public relations agency employees for comparison with those who
work in corporate or nonprofit settings. Media choices also may be affected by
variables such as whether the communication is internal or external, and whether
the worker is telecommuting or working in a traditional office setting.

Self-Report Data

Although questions were developed carefully with a pretest, self-report media

choice measures are based on respondent estimates and may not be as precise as
other methods such as direct observation. Given limited resources, however, some
degree of precision was necessarily sacrificed to obtain a broader sample.

CONCLUSIONS AND DIRECTIONS

FOR FUTURE RESEARCH

Pasadeos, Renfro, and Hanily (1999) identified the most influential authors and
works in recent public relations literature. Not surprising, names like J. E.
Grunig, L. A. Grunig, Dozier, and Broom were found at the core of the main
318 KELLEHER

clusters of scholarly networks in our field. And, “public relations roles” emerged
as the “most co-cited category of works” (Pasadeos et al., p. 29). This study
demonstrates some of the reasons why research on public relations roles is so
central to our field. The manager and technician roles are built on sound logic.
They also are supported empirically.
Data on gender and the manager role offer clear evidence of the utility of the
role constructs. Whereas the construct validity of the manager role remains intact
after more than 20 years, the percentage of women enacting the manager role
seems to be increasing among PRSA membership, albeit very slowly. Examining
critical, long-term trends like this in our field is made easier by the availability of
lasting constructs.
However, at the center of Pasadeos et al.’s (1999) conclusions is a call to ex-
tend our collective research agenda to areas of research beyond the well-estab-
lished core clusters. This study takes a step in that direction. It reaches across a
gray boundary to connect with theory from the related but separate field of orga-
nizational communication.
The literature review makes a clear link between public relations roles and me-
dia richness theory. The operationalization of manager and technician roles, in
which each of the items associated with the technician factor involves written
communication, underscores the overlap in theory. Not surprising, the prediction
that technicians would spend more time writing and reading traditional textual ma-
terial than managers was accurate. Likewise, the prediction that managers, by the
very nature of their jobs, will spend more time in oral communication than techni-
cians was supported. Phrasing these predictions as hypotheses and testing them
empirically confirmed what some might have considered obvious in light of the lit-
erature reviewed. Such is the process of scientific inquiry. However, this study of-
fers more than simple replication in that it reveals a connection between two
related disciplines that has received little scholarly attention to date.
Perhaps the most important outcome of this study, however, is the new line of
questioning it forwards. Existing literature and research are unclear on questions
that probe further into the topic of media choice by public relations people, leaving
interesting areas of inquiry unexplored. For example, managers from other fields
have been found to use the telephone less than their counterparts lower in organiza-
tional hierarchies. Yet, in this study, public relations managers reported spending
significantly more time on the phone than technicians. Markus’s work (e.g., 1994)
suggests that media choice with traditional channels is based on the convergence of
the channel’s material characteristics with shared social definitions within a field.
Data from this study suggest that the telephone is defined among public relations
practitioners as an acceptable channel for the equivocal work of managers. How-
ever, what are the actual tasks for which managers choose the telephone?
 PUBLIC RELATIONS ROLES AND MEDIA CHOICE 319

This study shows that e-mail is essentially a universal-access medium for

PRSA members. Yet, have practitioners’ social definitions of e-mail converged
with its objective characteristics as a medium? For which tasks do practitioners
choose e-mail? How are these decisions related to public relations roles and the
functioning of public relations departments in general?
Based on these findings, public relations people basically are using e-mail to
communicate most with the people whom they already interact with
most—namely, coworkers and clients. Is the content of these communications un-
changed as well? What, if anything, does choosing newer media technologies
communicate beyond the textual–verbal content transmitted?
Although managers and technicians do not seem to differ in the amount of time
they spend with e-mail, managers report a much larger volume of e-mail traffic than
do technicians. This finding is consistent with a recent study by Napoli, Taylor, and
Powers (1999). They found that senior public relations practitioners perform about
the same number of writing tasks in a workday as their younger colleagues but do so
in a significantly smaller percentage of their workday. Napoli et al. suggested this
might be due to “higher levels of writing efficiency” (p. 369). Another possible ex-
planation for this finding is that technicians spend more time researching and plan-
ning for outgoing mail and acting on incoming e-mail, whereas managers may be
more likely to use the forward and “cc” functions of their e-mail programs. This
would allow managers to deal with a larger volume of e-mail than technicians in the
same amount of time. This is an area for future research that may be informed by lit-
erature outside the core clusters of recent public relations research.
In general, future research should explore further the context of media
choices in public relations and the specific content of the resulting communica-
tions. Direct observation, content analysis of communications, interviewing, and
case studies are among the methods that should be considered along with sur-
veys in follow-up research.

ACKNOWLEDGMENT

An earlier version of this article was presented as a Top Three paper to the Public
Relations Division at the annual Association for Education in Journalism and Mass
Communication (AEJMC) conference in Phoenix, Arizona, August 2000.

REFERENCES

Berkowitz, D., & Hristodoulakis, I. (1999). Practitioner roles, public relations education and profes-
sional socialization: An exploratory study. Journal of Public Relations Research, 11, 91–103.
Broom, G. M., & Smith, G. D. (1978, August). Toward an understanding of public relations roles: An
empirical test of five role models’ impact on clients. Paper presented to the public relations division
at the annual convention of the Association for Education in Journalism, Seattle, WA.
320 KELLEHER

Broom, G. M., & Smith, G. D. (1979). Testing the practitioner’s impact on clients. Public Relations Re-
view, 5(3), 47–59.
Daft, R. L., & Lengel, R. H. (1984). Information richness: A new approach to managerial information
processing and organizational design. In L. L. Cummings & B. M. Staw (Eds.), Research in organi-
zational behavior (pp. 191–233). Greenwich, CT: JAI.
Daft, R. L., Lengel, R. H., & Trevino, L. K. (1987). Message equivocality, media selection, and manager
performance: Implications for information systems. MIS Quarterly, 11, 355–366.
D’Ambra, J., & Rice, R. E. (1994). Multimethod approaches for the study of computer-mediated com-
munication, equivocality and media selection. IEEE Transactions on Professional Communication,
37, 231–239.
Dozier, D. M. (1992). The organizational roles of communications and public relations practitioners. In
J. E. Grunig (Ed.), Excellence in public relations and communication management (pp. 71–94).
Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.
Dozier, D. M., & Broom, G. M. (1995). Evolution of the manager role in public relations practice. Jour-
nal of Public Relations Research, 7, 3–26.
Fulk, J., Steinfield, C., Schmitz, J., & Power, G. (1987). A social information processing model of media
use in organizations. Communication Research, 14, 529–552.
Grunig, J. E. (1992). Excellence in public relations and communication management. Hillsdale, NJ:
Lawrence Erlbaum Associates, Inc.
Markus, M. L. (1987). Towards a “critical mass” theory in interactive media. Communication Research,
14, 491– 511.
Markus, M. L. (1994). Electronic mail as the medium of managerial choice. Organization Science, 5,
502–527.
Napoli, P. M., Taylor, M., & Powers, G. (1999). Writing activities of public relations practitioners: The
relationship between experience and writing tasks. Public Relations Review, 25, 369–380.
Pasadeos, Y., Renfro, R. B., & Hanily, M. L. (1999). Influential authors and works of public relations
scholarly literature: A network of recent research. Journal of Public Relations Research, 11, 29–52.
Rice, R., & Shook, D. (1990). Relationships of job categories and organizational levels to use of commu-
nication channels, including electronic mail: A meta-analysis and extension. Journal of Manage-
ment Studies, 27, 195–229.
Toth, E. L., Serini, S. A., Wright, D. K., & Emig, A. G. (1998). Trends in public relations roles:
1990–1995. Public Relations Review, 24, 145–163.
Trevino, L. K., Daft, R. L., & Lengel, R. H. (1990). Understanding managers’ media choices: A sym-
bolic interactionist perspective. In J. Fulk & C. W. Steinfield (Eds.), Organizations and communica-
tion technology . Newbury Park, CA: Sage.
Trevino, L. K., Lengel, R. H., Bodensteiner, W., Gerloff, E., & Muir, N. (1990). The richness imperative
and cognitive style: The role of individual differences in media choice behaviour. Management
Communication Quarterly, 4, 176–197.
Trevino, L. K., Lengel, R. H., & Daft, R. L. (1987). Media symbolism, media richness, and media choice
in organizations: A symbolic interactionist perspective. Communication Research, 14, 553–574.
Weick, K. E. (1979). The social psychology of organizing (2nd ed.). Reading, MA: Addison-Wesley.