Description
One ofthe major objectives ofthe study was to gather information on the perceptions of the respondents on the importance of nonverbal feedback and their ability to decode nonverbal cues from their co-workers and managers.
Five hundred and five respondents, from a wide variety of business organizations,
were surveyed to gather information on their perceptions (^nonverbal communication.
Dividing the sample on the basis of perceived decoding ability and gender revealed
several differences between the groups. Nonverbal communication was more important
to self-rated good decoders than to other decoders. Better decoders relied most on facial
expressions for accurate information while less skilled decoders preferred iKHce level or
tone. Women, individually, rated themselves higher than men in decoding abUity and,
as a group, were perceived by both men and women to be better decoders and encoders
of nonverbal cues. Women working in the education field rated themselves higher in
decoding ability than any other group.
Recommendations for improved communication in businesses included paying more at-
tention to nonverbal cues, especially the facial expressions, engaging in more eye con-
tact, and probing for more information when verbal and nonverbal cues are
discrepant. Managers should be aware that most employees feel frustration and dis-
trust when receiving conflicting signals from their supervisors, and should try to
modify their behavior by being more honest when conimunicating their emotions.
The Impact of Nonverbal
Communication in Organizations:
A Survey of Perceptions
Gerald H. Graham
The WicWa State University
Jeanne Unruh
Natixmal Institute Jar Management
Paul Jennings
The Wichita State University
T
he plethora of literature on nonverbal communication includes many
studies of university students but few studies of persons in business
organizations. Moreover, much of the literature is comprised of ex-
perimental data collected in contrived situations with role-playing as a
major component. Very few studies have attempted to gain information
about self-rated abilities to decode nonverbal cues.
Perceptions are important because they make up or influence our
internal organization of dat a Further, evidence siaggests that a person's
perceived ability to accurately decode nonverbal cues is positively as-
sociated with the person's actual decoding skills (Graham, Unruh, &
Jennings, 1990; Zuckerman, DeFrank, Spiegel, & Larremce, 1982).
One ofthe major objectives ofthe study was to gather information on
the perceptions of the respondents on the importance of nonverbal
feedback and their ability to decode nonverbal cues from their co-workers
and managers. The other major objective was to compare those findings
with the existing body of knowledge on nonverbal communication and to
45
46 The Journal of Business Communication 28:l:Winter 1991
use these findings as a basis for recommendations on how to improve
nonverbal communication in business organizations.
Ninety-four percent ofthe respondents in this study felt that nonver-
bal communication in the business world was either somewhat or very
important. Some professionals might find this surprising. But, as Rem-
land (1981) asserts. There is little doubt that effective management
requires good communication." Communicating well is known to be a
critical success factor in such things as increasing productivity, improv-
ing employee satisfaction, and being recognized as an outstanding
leader.
REVIEW OF THE LITERATURE
The large volume of research data on nonverbal communication can
be, for the most part, categorized under several broad headings. These
headings include studies underlining tbe importance of nonverbal com-
munication, studies on encoding and decoding differences in relation to
types of nonverbal cues, studies exploring very discrepant verbal/non-
verbal cues, and studies investigating demographic (usually gender)
effects on nonverbal communication.
Importance of Nonverbal Communication
In 1967, Mehrabian was one of the first to draw attention to the
significance of nonverbal communication with his conclusion that non-
verbal channels such as facial expression, body movement, and voice tone
contribute 93 percent ofthe "attitudinal" message to the receiver. Sub-
sequent research (Hegstrom, 1979) has shown that it is not the sum of
the individual channels, but probably the blending of one channel with
another that contributes most to the total impact of the message.
However, Mehrabian's findings were so dramatic that they inspired
many researchers to investigate the power of nonverbal communication.
Many of these studies came to the same conclusion as Honeycutt,
Knapp, and Powers (1983): ' The availability of nonverbal signals has an
important bearing on accurate judgements of another's communication."
For example, Woodall and Folger (1981) speculated that when nonverbal
cues are not "in sync" with the rhythmic pattern of speech, decoders may
question the honesty and motivation of the speaker. Addington (1971)
also showed that decreases in vocal variations had a negative impact on
source credibility.
Many studies that underscore the importance of nonverbal com-
munication are focused on discrepant verbal/nonverbal cues. Burgoon,
Nonverbal Communication • Graham/Unruh/Jennings 47
Coker and Coker (1986) found that eye contact is important and that
even in the case of a "positive" verbal message, gaze aversion led the
receiver to perceive an overall negative message. Specifically, gaze
aversion conveyed nonaffection, superficiality, lack of trust and non-
receptivity.
Studies have also found that the effects of voice tone (especially
negative voice tone) make a disproportionately stronger impact on
decoders than verbal content (Mehrabian & Wiener, 1967; Bugental,
Daswan, & Love, 1970). Other studies demonstrate that nonverbal
behaviors can be used to predict truth or deceit in the sender (Mehrabian,
1971; Knapp, Hart, & Dennis, 1974; Ekman & Friesen, 1974; McClintock
& Hunt, 1975; Feldman, Devin-Sheehan, & Allen, 1978; O'Hair, Cody,
& McLaughlin, 1981). Apparently, deceitful communicators maintain
less eye contact, talk less, offer fewer specifics, either make more or fewer
body movements than usual, and either smile too much or too little.
In an experiment by Forbes and Jackson (1980) real job interviews
were observed and the nonverbal behaviors in accepted candidates were
compared to those in rejected candidates and candidates put on a reserve
list. The accepted candidates engaged in more direct eye contact, more
head movements, and more smiling than did the rejected or reserve
candidates.
All ofthe studies discussed so far underscore the power of nonverbal
communication and the influence it has on our interactions with others.
Types of Nonverbal Cues
Other studies are more focused on identifying which types of nonver-
bal cues are most accurate or informative. For example, considerable
research has produced data supporting the notion that facial expressions
are more influential and give more information than other nonverbal
channels such as voice or body movements (Mehrabian & Ferris, 1967;
Zaidel & Mehrabian, 1969; Bugental, Daswan, & Love, 1970; DePaulo,
Rosenthal, Eisenstat, Rogers, & Finkelstein, 1978; Rosenthal & De-
Paulo, 1979; Forbes & Jackson, 1980). Coker and Burgoon (1987) ex-
amined 59 tjnpes of nonverbal behaviors categorized into 21 groups. Five
ofthe 21 emerged as significant predictors of degree of conversational
involvement. At the top ofthe list was facial animation followed closely
by vocal warmth and then vocal pitch and body movements. According
to the cited studies facial characteristics seem to be the most accurate
predictors of attitudes and feelings and, therefore, of the true meaning
ofthe communication.
48 The Journal of Business Communication 28:l:Winter 1991
Extremely Discrepant Verbai/Nonverbai Cues
The data on very deceptive communication (when a person lies) tell a
slightly different stery. In DePaulo's study (DePaulo, Rosenthal,
Eisenstet, Rogers, & Finkelstein, 1978) when visual cues didn't agree
with audio cues, subjects were more influenced by visual than audio
except in extremely discrepant cases where they attended more te the
audio cues. Perhaps the best explanation for the perceived difference in
extremely discrepant cases can be found in two studies done by Zucker-
man (Zuckerman, Larrance, Spiegel, & Klorman, 1981; Zuckerman,
Amidon, Bishop, & Pomerantz, 1982). Zuckerman discovered that facial
cues can be either suppressed or exaggerated more easily than vocal
cues. Therefore, when a person lies, true feelings are more likely te be
revealed in the vocal channel than in the facial channel. The authors
concluded that when there is extreme discrepancy between vocal and
facial cues the decoder suspects the sender is lying and is, therefore, more
influenced by the tone of voice which is more likely te reveal the truth
than by the more controllable facial cues. However, in honest com-
munication, facial cues gave more information.
Demographic impact on Nonverbai Communication
Of all the demographic variables that have been investigated, gender
seems to have the most impact on ability to decode or encode nonverbal
cues. Women appear to be better than men at nonverbal communication
(Hall, 1978; Isenhart, 1980; Blanck, Rosenthal, Snodgrass, Depaulo, &
Zuckerman, 1981).
Hall (1978) surveyed 75 studies of gender's impact on decoding non-
verbal cues. She concluded that females have an advantage over men in
decoding nonverbal communication and that this gender effect did not
vary with the sex of the sender. In other words, women were better than
men at decoding nonverbal cues fi-om both male and female encoders.
Hall also concluded that the age of the encoder or decoder made no
difference. Females of all ages were better decoders than males of all
ages. Not only are females better decoders, they appear to be better
encoders as well (Buck, Miller, & Caul, 1974; Hall, 1978; LaFrance &
Mayo, 1979; and Rosenthal & DePaulo, 1979).
Other ways males and females differ in nonverbal communication
include the observation that females pay more attention to facial cues
than males do. In fact, one study found that women showed less supe-
riority in decoding nonverbal cues when heeding nonverbal channels
other than the face (Rosenthal & DePaulo, 1979). Tbe same study also
Nonverbal Communicatipn • Graham/Unruh/Jennings 49
found that women were markedly more accurate in readingnondeceptive
communication but less so when reading deceptive communication.
There could be two expleinations for this. One is that since the voice is
more revealing than the face during deception, tbe voice may be a better
source of accurate information than the face. Another possible explana-
tion is that those who are too good at decoding of nonverbal communica-
tion that "leaks" the sender's true feelings are less effective in their
interpersonal relationships.
The present study was an attempt to compare these literature find-
ings with the perceptions of respondents in business organizations.
Perceptions can influence behaviors and if the perceptions ofthe respon-
dents parallel the measurements of actual skill and behavior found in
experimental data, recommendations for improving nonverbal com-
munication in the work place can be made.
IMETHOD
The researchers identified the top 50 employers in a midwest city as
a possible sampling pool. The organizations were considered to be a
representative mix of employment in the region with regard to technol-
ogy, products/services offered, size, and industry. Managers in all 50
organizations were contacted and permission was obtained to survey
departments in 35 ofthe organizations. Department size ranged from 5
to 33 employees. In the five largest organizations, 3-5 departments
(totally unrelated to each other) were surveyed. Only 1-2 departments
were surveyed in the remaining 30 organizations. The researchers
trained a team of graduate students to distribute 1200 surveys to these
organizations. Seventy-five percent ofthe surveys went to organizations
with more than 1000 employees. The following industries were repre-
sented: manufacturing, health care, finance, retail, and government.
Five hundred and five usable surveys were retrieved yielding a return
rate of 42 percent. Subjects were 217 males and 288 females. The
majority of the sample (67%) was between the ages of 25 and 45.
Seventy-seven percent had at least some college education. Seventy-one
percent had at least 6 years full-time work experience. Seventy-nine
percent had an annual income of less than forty thousand dollars. Four
percent of the respondents categorized themselves as top level execu-
tives; 13 percent as middle managers; 29 percent as staff; 7 percent as
first-line supervisors; 40 percent as employees; and 7 percent as "other."
50 The Journal of Business Communication 28:l:Winter 1991
instrument
The survey was divided inte two parts. Part 1 consisted of questions
involving perceptions about the importance of, and the ability te decode
nonverbal feedback from others on the job. Part 2 was comprised of
questions designed to collect demographic information on the respon-
dents. The questions in part 1 were developed by the authors and given
te 10 managers and 30 employees as a pilot test. The questionnaire was
then revised based on feedback from the pilot subjects during a verbal
review session. Questions considered by the pilot subjects te be irrelevant
or ambiguous were either deleted or reworked.
Six of the questions required answers on a Likert scale. An example
of this type of question was: "When communicating vfith your immediate
manager, how often do you feel that his or her feedback is in agreement
with his or her nonverbal feedback?" Possible answers were: A. Always;
B. Usually; C. Sometimes; D. Usually Not; E. Never. On two of the
questions respondents were asked te rank order items. For example:
"Certain types of feedback are perceived te more accurately reflect the
true feelings of the sender than others. Rank the following from what
you perceive te be the most accurate to the least accurate." Listed items
were: A Verbal Content; B. Voice Tone or Level; C. Hand or Arm
Gestures; D. Facial Expressions; E. Posture or Stance; F. Head Move-
ments. Other questions asked the subject to identify an answer as shown
in this example: "When your supervisor's verbal communication te you
is not in agreement with his or her nonverbal feedback, which com-
munication do you perceive te be more accurate?" A. Verbal; B. Nonver-
bal; C. Unsure.
Means and standard deviations were tabulated for each question and
frequency data were obtained for the sample as a whole. An analysis was
conducted, using a correlation matrix, for identification of interrelation-
ships among the data. Analysis of variance (ANOVA) was used to
pinpoint demographic and other variables which had a significant im-
pact on the perceptions of the respondents.
RESULTS
Several variables had a significant impact on the outeome of this
study, the two most important being perceived decoding ability, and
gender. Differences in the sample based on decoding ability will he
addressed first.
Nonverbal Communication • Graham/Unruh/Jennings 51
Decoding Ability
Respondents were asked to rate their decoding ability on a scale of
one to five, one being very good, and five being very poor. Figure 1 shows
tbat 9 percent rated tbemselves to be very good, 45 percent indicated
above average, 43 percent indicated average, 2 percent below average,
and 1 percent rated themselves as very poor decoders. For reporting and
statistical purposes, tbe scale was collapsed into two groups which will
be referred to as good decoders (grouping very good and above average)
and average decoders (grouping average and lower ranks). Good and
average decoders comprised 55 percent and 45 percent respectively.
Six aspects of communication were ranked in order of perceived
accuracy in reflecting tbe true feelings and attitude of the sender, or
encoder. Table 1 sbows tbe rank order of tbe means for good and average
decoders, where analysis of variance was used to determine the sig-
nificance of tbe difference between tbe two decoding abilities. Those wbo
felt themselves to be superior decoders ranked facial expressions as the
most accurate followed by voice level and tone. Average decoders, on the
otber band, ranked voice level and tone first and facial expressions
second, switching the order of perceived accuracy. Both good and average
decoders ranked verbal content of feedback at the same relative level of
importance (3rd), but as noted by a significant difference in the means,
good decoders believed it to be less accurate than did average decoders.
VERY OOOD ABOVE
WE RAO E
BELOW
M/ERAGE
VERY POOR
Figure 1
Self-Rated Decoding Ability
52 The Journal of Business Communication 28:l:Wmter 1991
Relative to average decoders, good decoders relied more on the sender's
body posture and stance.
When considering facial expressions, 79 percent of the respondents
looked most closely at the eyes, followed by the mouth. Good and average
decoders, alike, ranked eyes as the most important facial feature when
reading facial expressions.
Table 1
Rating of Communication Types
According to Perceived Accuracy
Feedback
Facial Expressions
Voice Tbne
Verbal Content
Posture or Stance
Hand Gestures
Head Movements
Mean
(G)
4.93
4.84
3.35
3.08
2.84
1.13
Rank
1
2
3
4
5
6
Mean
(A)
4.67
4.99
3.74
2.73
2.92
1.08
Rank
2
1
3
5
4
6
F
6.31
1.76
5.43
6.27
0.66
0.00
P-
Value
0.10
0.19
0.02
0.01
0.42
0.97
Note: Good decoders (G); Average decoders (A)
Pertaining to group conversations, such as committee meetings and
department meetings, 92 percent agreed that nonverbal aspects were
either important or very important. Overall, respondents perceived the
nonverbal aspects to be more crucial in one-on-one settings than when
in a group. On a rating scale of 1 to 4 (4 = very important, 1 = very
unimportant), nonverbal importance on the job in a one-on-one setting,
had a mean rating of 3.5. In a group conversation the mean dropped to
2.3. A difference in opinion existed between good and average decoders
as to the importance of nonverbal aspects, both on the job and in a group:
(jtood decoders rated the importance of nonverbal cues higher than did
average decoders (note Table 2, variables 1 and 2).
2.23
3.38
2.38
2.87
2.19
2.07
3.27
4.30
4.91
5.54
2.44
3.19
0.039
0.005
0.002
0.001
0.050
0.042
Nonverbal Communication • GrahamAJnruh/Jennings 53
Table 2
Differences Based on Perceived Decoding Ability
Variable Mean Mean F p-
(G) (A) Value
1. Importance of NV communication
in group discussion 2.44
2. Importance of NV communication
on the job (one on one) 3.76
3. How frequently their memiagers
presented discrepant V/NV cues? 2.76
4. How frequently their managers
were in V/NV agreement? 2.55
5. Is verbal or nonverbal channel
more accurate? 2.40
6. Which channel are future actions
based on? 2.29
Note: Good decoders (G); Average decoders (A)
(variables 1 & 2) Scale range = 1 to 4, 1 = unimportant; 4 - important
(variables 3 & 4) Scale range = 1 to 4, 1 = never; 4 = fequently
(variables 5 & 6) Sale range = 1 to 3,1 = verbal; 2 = equal; 3 = nonverbal
In the context of the superior subordinate dyad two issues were
explored; first, how consistently a superior's verbal and nonverbal feed-
back were in agreement, and secondly, when the superior's verbal
communication was not in agreement with the nonverbal feedback,
which was perceived te be more accurate and which would have the
greater influence on future interaction with that superior?
The consistency of superiors' communication channels was measured
by asking how firequently respondents experienced confusion due to
confiicting signals. The mean rating for the entire sample was 2.5 on a
scale of 4 (frequently) te 1 (never). Grood decoders experienced confusion
more frequently than did average decoders. Correspondingly, average
decoders believed their superiors' verbal and nonverbal feedback to be
consistent more often than did good decoders. In other words, as decoding
skill rose, detection of inconsistent communication also rose, leading te
confusion as to the true meaning of the superiors' messages (see Table
2, variables 3 & 4).
Fifty-two percent of the respondents felt that when there was dis-
crepant verbal/nonverbal communication from their supervisors, the
nonverbal channel communicated more accurate information than the
verbal channel. Twenty-one percent were not sure which channel would
54 The Journal of Business Communication 28:l:Winter 1991
be more accurate, and only 27 percent ofthe respondents perceived that
the verbal channel gave more accurate information. (3ood decoders
perceived the nonverbal channel to be more accurate than did average
decoders. Similarly, given conflicting channels, respondents reported
that the nonverbal channel of the message would have more influence
on the subordinates' future interaction with their superiors. There was
a strong relationship (correlation coefficient of 0.67 significant at .0001)
between the channel perceived more accurate and the channel that
would infiuence future interactions more. For example, if the nonverbal
channel was perceived to be more accurate, future interactions with the
superior would be based on the nonverbal message (see Table 2, variables
5&6).
Ninety-four percent of the sample experienced negative emotions
when the verbal and the nonverbal channels did not agree, (jood and
average decoders did not differ on this point. Specifically, 36 percent felt
distrustful and 33 percent reported feeling irritated or frustrated.
Demographic Impact
Stepwise multiple regression was run to determine whether any
combination of demographic variables could be used to identify a profile
that would help predict perceived decoding ability. Seven variables were
tested for their power in predicting decoding ability: gender, age, level of
formal education, industry sector, years of work experience, position, and
income. The optimal prediction equation contained two demographic
variables, gender and the education industry sector (Table 3). That is to
say, women and individuals who work in education consistently ranked
their decoding ability significantly higher than did any other group of
respondents. More specifically, women who worked in education per-
ceived themselves higher in decoding ability than did any other group.
Other demographic variables were not significant predictors.
Nonverbal Communication • Graham/Unrub/Jenninga 55
Table 3
Demographics of Good Decoders
(Optimal Prediction Equation)
Variable
1.
2.
3.
Gender
(1 = male, 2 - female)
Education
(0 = other.
1 = education)
Constant
Beta
0.1298
0.2374
2.4010
t-Statistic
2.100
2.030
23.729
t-Probability
0.036
0.044
0.000
Note: Significance: F = 4.596
F-Probability = 0.0105
Dependent variable = decoding ability
1 = below average, 2 = average, 3 = above average, 4 = very good
Not only did women rate themselves more highly than did men, women
rated women as better decoders and men agreed that women have
greater nonverbal decoding ability (Figure 2). Fifty-seven percent of all
respondents regarded women to be the most accurate decoders. Analysis
of variance was used te determine if men and women differed in their
perception of which gender was the better decoder. Althou^ women
were considered to be better decoders, women believed this te be true
more than did men (F=24.33, p=0.0001). When asked which gender was
easier to correctly decode, 48 percent said nonverbals from women were
easier to decode tban nonverbals from men. Furthermore, women felt
more strongly than did men that womens' nonverbal cues were easier to
decode (F=7.676, p=.OO6).
Respondents in management or supervisory positions tended to be
older males having higher educations, more years of work experience,
and higher incomes. Managers believed that age improved decoding
ability more than did employees and staff. As noted earlier, the age of
tbe respondent was not found to have a significant impact in the way
respondents rated their own decoding ability. In fact as the age of the
respondent rose, the nonverbal channel was relied upon less in deter-
mining how they would interact with that person in the future. This was
also the case witb seniority. Individuals with more years of work ex-
perience tended to rely less on nonverbal cues and more on verbal
content.
56
The Journal of Business Communication 28:l:Winter 1991
NODIFFERENCE
I Male reapondenta ^ ^ ^ Female roapondenta
Figure 2
Respondents' Perceptions of
Gender Effect on Decoding Skills
DISCUSSION
The results of this study were , for the most part, consistent with the
the literature. However, there were several surprising findings. As
perceived decoding ability increased, the importance of nonverbal com-
munication increased. Better decoders believed that they were more
capable, than poorer decoders, of detecting when their supervisor's
verbal and nonverbal communication were incongruent.
Perhaps people who considered themselves to be average decoders did
not pay as much attention to nonverbal cues as people who rated
themselves above average. This does not mean, however, that average
decoders were less emotionally affected when confronted with obviously
discrepant cues. Indeed, they reported that they were just as frustrated
and distrustful as the better decoders.
Almost half (43%) of the respondents rated themselves average in
decoding ability, and another 3 percent rated themselves below average.
Since effective communication is critical to success, consider that more
than 45 percent of those surveyed claim that they are either frequently
or occasionally confused by inconsistent verbal and nonverbal cues from
their supervisors. In addition, more than 94 percent ofthe entire sample
felt frustrated or distrustful when confronted with discrepant verbal/non-
verbal communications. It should follow that organizational productivity
cannot be optimal with the current amount of miscommunication and
frustration generated by verbal/nonverbal discrepancy. The authors
surmise that both morale and productivity could be significantly im-
Nonverbal Communication • Graham/Unruh/Jennings 57
proved if employees and managers were better informed about nonverbal
communication and if they felt more confident of their encoding and
decoding skills.
Since evidence supports the proposition that actual encoding and
decoding ability is positively correlated with perceived encoding and
decoding ability (Graham, Unruh, & Jennings, 1990; Zuckerman, De-
Frank, Spiegel, & Larrance, 1982) and since the current study identified
several perceived problems related to nonverbal communication skills in
a business setting, the authors offer the following recommendations.
Decoding Sidlls
In this study, the major difference between self-rated good and
average decoders was how the two groups ranked types of nonverbal
communication. Good decoders felt that facial expressions gave the most
accurate information about the true meaning of the message while
average and below decoders relied more on voice level or tone. This
finding is consistent with the large body of literature that reports the
face as the best provider of accurate nonverbal communication.
Assuming that respondents who perceive themselves as better
decoders are more accurate, the data suggests that those who wish to
improve their nonverbal decoding skills could do so by paying more
attention to nonverbal cues and concentrating mainly on facial expres-
sions. This means engaging in increased eye contact, since this study and
many others have shown that eye contact is, by far, the most important
facial characteristic in nonverbal communication.
Self-rated better decoders also tended to rely on nonverbal cues more
than on verbal cues when communicating with a person in the future.
Reliance upon nonverbal cues makes good sense since nonverbal com-
munication has been shown (in the review of literature) to more accurate-
ly reflect the true feelings of the encoder. Decoding nonverbal cues
successfully is not an end unto itself. Employees or managers who decode
nonverbal cues accurately but fail to act on the information can face the
same consequences—e.g., negative emotions or faulty decision making—
as those who are unsuccessful at decoding nonverbal cues. Most impor-
tantly, instead of ignoring nonverbal communication, one should gently
probe for more information when faced with discrepant verbal/nonverbal
cues.
58 The Journal of Business Communication 28:l:Winter 1991
Encoding Sitiiis
Supervisors and managers should take special note: More than 50
percent ofthe respondents felt their supervisors' verbal and nonverbal
communication was occasionally or frequently in conflict. As Remland
(1981) pointed out, "Much of what a manager says may be contradicted
by what he or she does." Keeping in mind that such discrepancy can cause
miscommunication, distrust, and frustration managers should become
more cognizant of this problem and make real efforts to keep their verbal
and nonverbal communication consistent with each other.
Unfortunately, many mangers have felt that they must not show their
emotions to their subordinates. When angry or upset they may try to
hide or disguise their feelings. But, what they are really doing is sending
a mass of confiicting signals which confuse employees and erode trust.
The authors recommend a more honest approach. Numerous studies
imply that people who don't overreact, but who appropriately frown,
smile, maintain eye contact, and use proper voice inflections come across
as more caring, trustworthy, and honest They are also much more likely
to communicate what they intend to communicate (Mehrabian &
Wiener, 1967; Burgental et al., 1970; Feldman et al., 1978; Forbes &
Jackson, 1980; Garrison, 1984). This appears to be especially applicable
to managers.
Gender Differences
Gender was the other main variable which impacted the results of
this study. Women rated themselves higher in decoding ability than men
rated themselves. And, interestingly, both men and women, as a whole,
reported that women are not only better decoders, but also better
encoders of nonverbal cues. These data are consistent with the literature
(Hall, 1978).
A study by Rosenthal and DePaulo (1979) found, by observation, that
women pay significantly more attention to the face than do men. The
current study showed no difference between men and women in this
respect. This is an interesting discrepancy and one which merits some
speculation. This study has clearly shown that those who rated tbemsel-
ves high in decoding ability favored facial cues over any otber nonverbal
cues. Since experimental evidence also consistently shows that the face
is the best source of accurate information, it may be that many men are
laboring under the impression that they do pay enough attention to facial
cues and that they do maintain sufficient eye contact, when, in fact, they
do not.
Nonverbal Communication • Graham/Unruh/Jennings 59
Predicting Nonverbai Skiiis
In the authors' attempt to identify a demographic profile which could
predict self-rated decoding ability, two variables were found that fit the
equation. The highest individual predictor was, interestingly, occupation
in education. Respondents working in the education industry were likely
to rate their decoding skills very highly. The second highest individual
predictor was gender. When combined with education, these two vari-
ables provide even more information. Women who work in the education
field rated their decoding ability higher than any other group. Adding in
more variables reduced the predictability of the equation and did not
contribute to the overall information.
These findings seem reasonable. Evidence indicates women are better
than men at nonverbal communication. Women in the teaching profes-
sion would seem, logically, to have an even greater advantage because
of the nature of the occupation. Teaching demands an intense com-
munication effort—probably more so than any other profession.
Perception Differences in Managers and Employees
The authors predicted differences between the perceptions of
employees and those of managers. However, only one difference
emerged. Those in management believed that decoding ability improves
with increasing age. Respondents who held management positions
tended to be older than those respondents who were employees. Older
respondents, however, did not judge themselves to be better decoders
than younger respondents. Furthermore, some studies have shown that
older decoders (sixties and older) are less skilled at decoding nonverbal
cues than younger persons (Parham & Feldman, 1981).
SUMMARY
In summary, the results of this study showed that the perceptions of
505 respondents in the business world agree fairly well with the ex-
perimental evidence reported in the literature. Nonverbal communica-
tion was important to all surveyed, and most respondents agreed that
nonverbal communication would influence their interactions with people
more than would verbal content. Better decoders relied most on facial
expressions for accurate information while less skilled decoders
preferred voice level or tone.
Females, individually, rated themselves higher than men in decoding
ability, and as a group were perceived by both men and women to be both
better decoders and encoders of nonverbal cues. Women working in the
60 The Journal of Business Communication 28:l:Winter 1991
education field rated themselves higher in decoding ability than other
group.
Recommendations for improving nonverbal skills included paying
more attention te nonverbal cues, especially the facial expressions,
engaging in more eye contact, and probing for more information when
verbal and nonverbal cues are discrepant. Managers should be aware
that most employees feel frustration and distrust when receiving con-
flicting signals from their supervisors, and should try to modify their
behavior by being more honest in communicating their emotions.
REFERENCES
Addington, D. W. (1971). The effect of vocal variations on ratings of source
credibility. Speech Monographs, 38, 242-247.
Blanck, P. D., Rosenthal, R., Snodgrass, S., DePaulo, B., & Zuckerman, M.
(1981). Sex differences in eavesdropping on nonverbal cues: Developmental
changes. Journal of Personality and Social Psychology, 41, 391-396.
Buck, R., Miller, R. E., & Caul, W. F. (1974). Sex, personality and physiological
variables in the communication of emotion via facial expression. Journal of
Personality and Social Psychology, 30, 587-596.
Bugenthal, D., Daswan, J., & Love, L. (1970). Perception of contradictory
meanings conveyed by verbal and nonverbal channels. Personality and
Social Psychology, 16, 647-655.
BuUer, D., & Burgoon, J. (1986). The effects of vocalics and nonverbal sen-
sitivity on compliance: Areplication and extension. Human Communication
Research, 13, 126-144.
Burgoon, J., Coker, D., & Coker, R. (1986). Communicative explanations.
Human Communication Research, 12, 495-524.
Cody, M., & O'Hair, D. (1983). Nonverbal communication and deception:
Differences in deception cues due to gender and communicator dominance.
Com.munication Monograph, 50, 175-192.
Coker, D., & Burgoon, J. (1987). The nature of conversational involvement and
nonverbal encoding patterns. Human Communication Research, 13, 463-
494.
Depaulo, B., Rosenthal, R., Eisenstat, R., Rogers, P., & Finkelstein, S. (1978).
Decoding discrepant nonverbal cues. Journal of Personal and Social
Psycholc^y, 36, 313-323.
Ekman, P., & Friesen, W. (1974). Defecting deception from the body or face.
Journal of Personality and Social Psychology, 19, 288-298.
Feldman, R., Devin-Sheehan, L., & Allen, V. (1978). Nonverbal cues on in-
dicators of verbal ^ssembMnQ.AmericanEducationalResearch Journal, 15,
217-231.
Nonverbal Communication • Graham/Unruh/Jennings 61
Forbes, R., & Jackson, P. (1980). Nonverbal behavior and the outcome of
selection interviews. Journal of Occupational Psychology, 53(1), 65-72.
Garrison, L. L. (1984). Communication—The nonverbal way. Journal of Busi-
ness Education, 190-192.
Graham, G. H., Unruh, J., & Jennings, P. (1990). Demographic impact on
nonverbal decoding and encoding skills of genuine emotion in facial expres-
sions. Unpublished manuscript.
Hall, J. (1980). Voice tone and persuasion. Journal of Personality and Social
Psychology, 38, 924-934.
Hall, J. (1978). Gender effects in decoding nonverbal cues. Psychological
Bulletin, 68, 845-857.
Hegstrom, T. (1979). Message impact: What p)ercentage is nonverbal? Western
Journal of Speech Communication, 43(2), 134-142.
Honeycutt, J., Knapp, M., & Powers, W. (1983). On knowing others and
predicting what they say. Western Journal of Speech Communication, 47(2),
157-174.
Isenhart, M. (1980). An investigation ofthe relationships of sex and sex role
to the ability to decode nonverbal cues. Human Communication Research,
6, 309-317.
Knapp, M., Hart, R., & Dennis, H. (1974). An exploration of deception as a
communication construct. Human Communication Research, 7,15-20.
La France, M., & Mayo, C. (1979). A review of nonverbal behaviors of women
and men. Western Journal of Speech Communication, 43(2), 96-107.
Leffler, A., Gillespie, D., & Conaty, J. (1982). The effects of status differentia-
tion on nonverbal behavior. Social Psychology Quarterly, 45(3), 153-161.
McClintock, C, & Hunt, R. (1975). Nonverbal indicators of affect amd decep-
tion in an interview setting. Journal of Applied Social Psychology, 5(1),
65-67.
Mehrabian, A. (1971). Nonverbal betrayal of feeling. Journal of Experimental
Research on Personality, 5(1), 64-73.
Mehrabian, A., & Wiener, M. (1967). Decoding of inconsistent communications.
Journal of Personality and Social Psychology, 6(1), 109-114.
Mehrabian, A., & Ferris, D. (1967). Inference of attitudes from nonverbal
communication in two channels. Journal of Consulting Psychology, 31,
248-252.
Mulac, A., Studley, OL., Wiemann, J., & Bradac, J. (1987). Male/female gaze
in same-sex and mixed-sex dyads. Human Communication Research, 13,
323-343.
Neal, W. (1982). Focusing on nonverbal communication in business. Journal
of Business Education, 58(2), 78-80.
O'Hair, H., Cody, M., & McLaughlin, M. (1981). Prepared lies, spontaneous lies,
Machiavellianism, and nonverbal communication. Human Communication
Research, 7, 325-339.
62 The Joumai of Business Communication 28:l:Winter 1991
Parham, I., Feldman, R., Oster, G., & Popoola, O. (1981). Intergenerational
differences in nonverbal disclosure of deception. Journal of Social Psychol-
ogy, 113, 261-269.
Remland, M. (1981). Developing leadership skills in nonverbal communica-
tion: A situational perspective. The Journal of Business Communication,
18(3), 17-29.
Remland, M. (1984). Leadership impressions and nonverbal communication
in a superior-subordinate interaction. Communication Quarterly, 32(1),
41-48.
Rosenthal, R., & DePaulo, B. (1979). Sex diferences in eavesdropping on
nonverbal cues. Journal of Personality and Social Psychology, 37, 273-285.
St. John, W. D. (1985). You are what you communicate. Personnel Journal,
72(3), 40-43.
Woodall, W., & Folger, J. (1981). Encoding specificity and nonverbal cue
context: An expansion of episodic memory research. Communication
Monographs, 48, 39-53.
Zaidel, S., & Mehrabian, A. (1969). The ability to communicate and infer
positive and negative attitudes facially and vocally. Journal ofExperimental
Research in Personality, 3, 233-241.
Zaremba, A. J.( 1987). Beyond body language. BMsmessMar^fmg, 72(3), 132+.
Zuekerman, M., Amidon, M., Bishop, S., & Pomerantz, S. (1982). Face and tone
of voice in the communication of deception. Journal of Personality and
Social Psychology, 43, 347-357.
Zuekerman, M., DeFrank, R., Spiegel, N., & Larrance, D. (1982). Masculinity-
femininity and encoding of nonverbal cues. Journal of Personality and
Social Psychology, 42, 548-556.
Zuekerman, M., Larrance, D., Spiegel, N., & Klorman, R. (1981). Controlling
nonverbal displays: Facial expressions and tone of voice. Journal of Ex-
perimental Social Psychology, 17, 506-524.
Accepted by NLR, 5/14/90
doc_437580152.pdf
One ofthe major objectives ofthe study was to gather information on the perceptions of the respondents on the importance of nonverbal feedback and their ability to decode nonverbal cues from their co-workers and managers.
Five hundred and five respondents, from a wide variety of business organizations,
were surveyed to gather information on their perceptions (^nonverbal communication.
Dividing the sample on the basis of perceived decoding ability and gender revealed
several differences between the groups. Nonverbal communication was more important
to self-rated good decoders than to other decoders. Better decoders relied most on facial
expressions for accurate information while less skilled decoders preferred iKHce level or
tone. Women, individually, rated themselves higher than men in decoding abUity and,
as a group, were perceived by both men and women to be better decoders and encoders
of nonverbal cues. Women working in the education field rated themselves higher in
decoding ability than any other group.
Recommendations for improved communication in businesses included paying more at-
tention to nonverbal cues, especially the facial expressions, engaging in more eye con-
tact, and probing for more information when verbal and nonverbal cues are
discrepant. Managers should be aware that most employees feel frustration and dis-
trust when receiving conflicting signals from their supervisors, and should try to
modify their behavior by being more honest when conimunicating their emotions.
The Impact of Nonverbal
Communication in Organizations:
A Survey of Perceptions
Gerald H. Graham
The WicWa State University
Jeanne Unruh
Natixmal Institute Jar Management
Paul Jennings
The Wichita State University
T
he plethora of literature on nonverbal communication includes many
studies of university students but few studies of persons in business
organizations. Moreover, much of the literature is comprised of ex-
perimental data collected in contrived situations with role-playing as a
major component. Very few studies have attempted to gain information
about self-rated abilities to decode nonverbal cues.
Perceptions are important because they make up or influence our
internal organization of dat a Further, evidence siaggests that a person's
perceived ability to accurately decode nonverbal cues is positively as-
sociated with the person's actual decoding skills (Graham, Unruh, &
Jennings, 1990; Zuckerman, DeFrank, Spiegel, & Larremce, 1982).
One ofthe major objectives ofthe study was to gather information on
the perceptions of the respondents on the importance of nonverbal
feedback and their ability to decode nonverbal cues from their co-workers
and managers. The other major objective was to compare those findings
with the existing body of knowledge on nonverbal communication and to
45
46 The Journal of Business Communication 28:l:Winter 1991
use these findings as a basis for recommendations on how to improve
nonverbal communication in business organizations.
Ninety-four percent ofthe respondents in this study felt that nonver-
bal communication in the business world was either somewhat or very
important. Some professionals might find this surprising. But, as Rem-
land (1981) asserts. There is little doubt that effective management
requires good communication." Communicating well is known to be a
critical success factor in such things as increasing productivity, improv-
ing employee satisfaction, and being recognized as an outstanding
leader.
REVIEW OF THE LITERATURE
The large volume of research data on nonverbal communication can
be, for the most part, categorized under several broad headings. These
headings include studies underlining tbe importance of nonverbal com-
munication, studies on encoding and decoding differences in relation to
types of nonverbal cues, studies exploring very discrepant verbal/non-
verbal cues, and studies investigating demographic (usually gender)
effects on nonverbal communication.
Importance of Nonverbal Communication
In 1967, Mehrabian was one of the first to draw attention to the
significance of nonverbal communication with his conclusion that non-
verbal channels such as facial expression, body movement, and voice tone
contribute 93 percent ofthe "attitudinal" message to the receiver. Sub-
sequent research (Hegstrom, 1979) has shown that it is not the sum of
the individual channels, but probably the blending of one channel with
another that contributes most to the total impact of the message.
However, Mehrabian's findings were so dramatic that they inspired
many researchers to investigate the power of nonverbal communication.
Many of these studies came to the same conclusion as Honeycutt,
Knapp, and Powers (1983): ' The availability of nonverbal signals has an
important bearing on accurate judgements of another's communication."
For example, Woodall and Folger (1981) speculated that when nonverbal
cues are not "in sync" with the rhythmic pattern of speech, decoders may
question the honesty and motivation of the speaker. Addington (1971)
also showed that decreases in vocal variations had a negative impact on
source credibility.
Many studies that underscore the importance of nonverbal com-
munication are focused on discrepant verbal/nonverbal cues. Burgoon,
Nonverbal Communication • Graham/Unruh/Jennings 47
Coker and Coker (1986) found that eye contact is important and that
even in the case of a "positive" verbal message, gaze aversion led the
receiver to perceive an overall negative message. Specifically, gaze
aversion conveyed nonaffection, superficiality, lack of trust and non-
receptivity.
Studies have also found that the effects of voice tone (especially
negative voice tone) make a disproportionately stronger impact on
decoders than verbal content (Mehrabian & Wiener, 1967; Bugental,
Daswan, & Love, 1970). Other studies demonstrate that nonverbal
behaviors can be used to predict truth or deceit in the sender (Mehrabian,
1971; Knapp, Hart, & Dennis, 1974; Ekman & Friesen, 1974; McClintock
& Hunt, 1975; Feldman, Devin-Sheehan, & Allen, 1978; O'Hair, Cody,
& McLaughlin, 1981). Apparently, deceitful communicators maintain
less eye contact, talk less, offer fewer specifics, either make more or fewer
body movements than usual, and either smile too much or too little.
In an experiment by Forbes and Jackson (1980) real job interviews
were observed and the nonverbal behaviors in accepted candidates were
compared to those in rejected candidates and candidates put on a reserve
list. The accepted candidates engaged in more direct eye contact, more
head movements, and more smiling than did the rejected or reserve
candidates.
All ofthe studies discussed so far underscore the power of nonverbal
communication and the influence it has on our interactions with others.
Types of Nonverbal Cues
Other studies are more focused on identifying which types of nonver-
bal cues are most accurate or informative. For example, considerable
research has produced data supporting the notion that facial expressions
are more influential and give more information than other nonverbal
channels such as voice or body movements (Mehrabian & Ferris, 1967;
Zaidel & Mehrabian, 1969; Bugental, Daswan, & Love, 1970; DePaulo,
Rosenthal, Eisenstat, Rogers, & Finkelstein, 1978; Rosenthal & De-
Paulo, 1979; Forbes & Jackson, 1980). Coker and Burgoon (1987) ex-
amined 59 tjnpes of nonverbal behaviors categorized into 21 groups. Five
ofthe 21 emerged as significant predictors of degree of conversational
involvement. At the top ofthe list was facial animation followed closely
by vocal warmth and then vocal pitch and body movements. According
to the cited studies facial characteristics seem to be the most accurate
predictors of attitudes and feelings and, therefore, of the true meaning
ofthe communication.
48 The Journal of Business Communication 28:l:Winter 1991
Extremely Discrepant Verbai/Nonverbai Cues
The data on very deceptive communication (when a person lies) tell a
slightly different stery. In DePaulo's study (DePaulo, Rosenthal,
Eisenstet, Rogers, & Finkelstein, 1978) when visual cues didn't agree
with audio cues, subjects were more influenced by visual than audio
except in extremely discrepant cases where they attended more te the
audio cues. Perhaps the best explanation for the perceived difference in
extremely discrepant cases can be found in two studies done by Zucker-
man (Zuckerman, Larrance, Spiegel, & Klorman, 1981; Zuckerman,
Amidon, Bishop, & Pomerantz, 1982). Zuckerman discovered that facial
cues can be either suppressed or exaggerated more easily than vocal
cues. Therefore, when a person lies, true feelings are more likely te be
revealed in the vocal channel than in the facial channel. The authors
concluded that when there is extreme discrepancy between vocal and
facial cues the decoder suspects the sender is lying and is, therefore, more
influenced by the tone of voice which is more likely te reveal the truth
than by the more controllable facial cues. However, in honest com-
munication, facial cues gave more information.
Demographic impact on Nonverbai Communication
Of all the demographic variables that have been investigated, gender
seems to have the most impact on ability to decode or encode nonverbal
cues. Women appear to be better than men at nonverbal communication
(Hall, 1978; Isenhart, 1980; Blanck, Rosenthal, Snodgrass, Depaulo, &
Zuckerman, 1981).
Hall (1978) surveyed 75 studies of gender's impact on decoding non-
verbal cues. She concluded that females have an advantage over men in
decoding nonverbal communication and that this gender effect did not
vary with the sex of the sender. In other words, women were better than
men at decoding nonverbal cues fi-om both male and female encoders.
Hall also concluded that the age of the encoder or decoder made no
difference. Females of all ages were better decoders than males of all
ages. Not only are females better decoders, they appear to be better
encoders as well (Buck, Miller, & Caul, 1974; Hall, 1978; LaFrance &
Mayo, 1979; and Rosenthal & DePaulo, 1979).
Other ways males and females differ in nonverbal communication
include the observation that females pay more attention to facial cues
than males do. In fact, one study found that women showed less supe-
riority in decoding nonverbal cues when heeding nonverbal channels
other than the face (Rosenthal & DePaulo, 1979). Tbe same study also
Nonverbal Communicatipn • Graham/Unruh/Jennings 49
found that women were markedly more accurate in readingnondeceptive
communication but less so when reading deceptive communication.
There could be two expleinations for this. One is that since the voice is
more revealing than the face during deception, tbe voice may be a better
source of accurate information than the face. Another possible explana-
tion is that those who are too good at decoding of nonverbal communica-
tion that "leaks" the sender's true feelings are less effective in their
interpersonal relationships.
The present study was an attempt to compare these literature find-
ings with the perceptions of respondents in business organizations.
Perceptions can influence behaviors and if the perceptions ofthe respon-
dents parallel the measurements of actual skill and behavior found in
experimental data, recommendations for improving nonverbal com-
munication in the work place can be made.
IMETHOD
The researchers identified the top 50 employers in a midwest city as
a possible sampling pool. The organizations were considered to be a
representative mix of employment in the region with regard to technol-
ogy, products/services offered, size, and industry. Managers in all 50
organizations were contacted and permission was obtained to survey
departments in 35 ofthe organizations. Department size ranged from 5
to 33 employees. In the five largest organizations, 3-5 departments
(totally unrelated to each other) were surveyed. Only 1-2 departments
were surveyed in the remaining 30 organizations. The researchers
trained a team of graduate students to distribute 1200 surveys to these
organizations. Seventy-five percent ofthe surveys went to organizations
with more than 1000 employees. The following industries were repre-
sented: manufacturing, health care, finance, retail, and government.
Five hundred and five usable surveys were retrieved yielding a return
rate of 42 percent. Subjects were 217 males and 288 females. The
majority of the sample (67%) was between the ages of 25 and 45.
Seventy-seven percent had at least some college education. Seventy-one
percent had at least 6 years full-time work experience. Seventy-nine
percent had an annual income of less than forty thousand dollars. Four
percent of the respondents categorized themselves as top level execu-
tives; 13 percent as middle managers; 29 percent as staff; 7 percent as
first-line supervisors; 40 percent as employees; and 7 percent as "other."
50 The Journal of Business Communication 28:l:Winter 1991
instrument
The survey was divided inte two parts. Part 1 consisted of questions
involving perceptions about the importance of, and the ability te decode
nonverbal feedback from others on the job. Part 2 was comprised of
questions designed to collect demographic information on the respon-
dents. The questions in part 1 were developed by the authors and given
te 10 managers and 30 employees as a pilot test. The questionnaire was
then revised based on feedback from the pilot subjects during a verbal
review session. Questions considered by the pilot subjects te be irrelevant
or ambiguous were either deleted or reworked.
Six of the questions required answers on a Likert scale. An example
of this type of question was: "When communicating vfith your immediate
manager, how often do you feel that his or her feedback is in agreement
with his or her nonverbal feedback?" Possible answers were: A. Always;
B. Usually; C. Sometimes; D. Usually Not; E. Never. On two of the
questions respondents were asked te rank order items. For example:
"Certain types of feedback are perceived te more accurately reflect the
true feelings of the sender than others. Rank the following from what
you perceive te be the most accurate to the least accurate." Listed items
were: A Verbal Content; B. Voice Tone or Level; C. Hand or Arm
Gestures; D. Facial Expressions; E. Posture or Stance; F. Head Move-
ments. Other questions asked the subject to identify an answer as shown
in this example: "When your supervisor's verbal communication te you
is not in agreement with his or her nonverbal feedback, which com-
munication do you perceive te be more accurate?" A. Verbal; B. Nonver-
bal; C. Unsure.
Means and standard deviations were tabulated for each question and
frequency data were obtained for the sample as a whole. An analysis was
conducted, using a correlation matrix, for identification of interrelation-
ships among the data. Analysis of variance (ANOVA) was used to
pinpoint demographic and other variables which had a significant im-
pact on the perceptions of the respondents.
RESULTS
Several variables had a significant impact on the outeome of this
study, the two most important being perceived decoding ability, and
gender. Differences in the sample based on decoding ability will he
addressed first.
Nonverbal Communication • Graham/Unruh/Jennings 51
Decoding Ability
Respondents were asked to rate their decoding ability on a scale of
one to five, one being very good, and five being very poor. Figure 1 shows
tbat 9 percent rated tbemselves to be very good, 45 percent indicated
above average, 43 percent indicated average, 2 percent below average,
and 1 percent rated themselves as very poor decoders. For reporting and
statistical purposes, tbe scale was collapsed into two groups which will
be referred to as good decoders (grouping very good and above average)
and average decoders (grouping average and lower ranks). Good and
average decoders comprised 55 percent and 45 percent respectively.
Six aspects of communication were ranked in order of perceived
accuracy in reflecting tbe true feelings and attitude of the sender, or
encoder. Table 1 sbows tbe rank order of tbe means for good and average
decoders, where analysis of variance was used to determine the sig-
nificance of tbe difference between tbe two decoding abilities. Those wbo
felt themselves to be superior decoders ranked facial expressions as the
most accurate followed by voice level and tone. Average decoders, on the
otber band, ranked voice level and tone first and facial expressions
second, switching the order of perceived accuracy. Both good and average
decoders ranked verbal content of feedback at the same relative level of
importance (3rd), but as noted by a significant difference in the means,
good decoders believed it to be less accurate than did average decoders.
VERY OOOD ABOVE
WE RAO E
BELOW
M/ERAGE
VERY POOR
Figure 1
Self-Rated Decoding Ability
52 The Journal of Business Communication 28:l:Wmter 1991
Relative to average decoders, good decoders relied more on the sender's
body posture and stance.
When considering facial expressions, 79 percent of the respondents
looked most closely at the eyes, followed by the mouth. Good and average
decoders, alike, ranked eyes as the most important facial feature when
reading facial expressions.
Table 1
Rating of Communication Types
According to Perceived Accuracy
Feedback
Facial Expressions
Voice Tbne
Verbal Content
Posture or Stance
Hand Gestures
Head Movements
Mean
(G)
4.93
4.84
3.35
3.08
2.84
1.13
Rank
1
2
3
4
5
6
Mean
(A)
4.67
4.99
3.74
2.73
2.92
1.08
Rank
2
1
3
5
4
6
F
6.31
1.76
5.43
6.27
0.66
0.00
P-
Value
0.10
0.19
0.02
0.01
0.42
0.97
Note: Good decoders (G); Average decoders (A)
Pertaining to group conversations, such as committee meetings and
department meetings, 92 percent agreed that nonverbal aspects were
either important or very important. Overall, respondents perceived the
nonverbal aspects to be more crucial in one-on-one settings than when
in a group. On a rating scale of 1 to 4 (4 = very important, 1 = very
unimportant), nonverbal importance on the job in a one-on-one setting,
had a mean rating of 3.5. In a group conversation the mean dropped to
2.3. A difference in opinion existed between good and average decoders
as to the importance of nonverbal aspects, both on the job and in a group:
(jtood decoders rated the importance of nonverbal cues higher than did
average decoders (note Table 2, variables 1 and 2).
2.23
3.38
2.38
2.87
2.19
2.07
3.27
4.30
4.91
5.54
2.44
3.19
0.039
0.005
0.002
0.001
0.050
0.042
Nonverbal Communication • GrahamAJnruh/Jennings 53
Table 2
Differences Based on Perceived Decoding Ability
Variable Mean Mean F p-
(G) (A) Value
1. Importance of NV communication
in group discussion 2.44
2. Importance of NV communication
on the job (one on one) 3.76
3. How frequently their memiagers
presented discrepant V/NV cues? 2.76
4. How frequently their managers
were in V/NV agreement? 2.55
5. Is verbal or nonverbal channel
more accurate? 2.40
6. Which channel are future actions
based on? 2.29
Note: Good decoders (G); Average decoders (A)
(variables 1 & 2) Scale range = 1 to 4, 1 = unimportant; 4 - important
(variables 3 & 4) Scale range = 1 to 4, 1 = never; 4 = fequently
(variables 5 & 6) Sale range = 1 to 3,1 = verbal; 2 = equal; 3 = nonverbal
In the context of the superior subordinate dyad two issues were
explored; first, how consistently a superior's verbal and nonverbal feed-
back were in agreement, and secondly, when the superior's verbal
communication was not in agreement with the nonverbal feedback,
which was perceived te be more accurate and which would have the
greater influence on future interaction with that superior?
The consistency of superiors' communication channels was measured
by asking how firequently respondents experienced confusion due to
confiicting signals. The mean rating for the entire sample was 2.5 on a
scale of 4 (frequently) te 1 (never). Grood decoders experienced confusion
more frequently than did average decoders. Correspondingly, average
decoders believed their superiors' verbal and nonverbal feedback to be
consistent more often than did good decoders. In other words, as decoding
skill rose, detection of inconsistent communication also rose, leading te
confusion as to the true meaning of the superiors' messages (see Table
2, variables 3 & 4).
Fifty-two percent of the respondents felt that when there was dis-
crepant verbal/nonverbal communication from their supervisors, the
nonverbal channel communicated more accurate information than the
verbal channel. Twenty-one percent were not sure which channel would
54 The Journal of Business Communication 28:l:Winter 1991
be more accurate, and only 27 percent ofthe respondents perceived that
the verbal channel gave more accurate information. (3ood decoders
perceived the nonverbal channel to be more accurate than did average
decoders. Similarly, given conflicting channels, respondents reported
that the nonverbal channel of the message would have more influence
on the subordinates' future interaction with their superiors. There was
a strong relationship (correlation coefficient of 0.67 significant at .0001)
between the channel perceived more accurate and the channel that
would infiuence future interactions more. For example, if the nonverbal
channel was perceived to be more accurate, future interactions with the
superior would be based on the nonverbal message (see Table 2, variables
5&6).
Ninety-four percent of the sample experienced negative emotions
when the verbal and the nonverbal channels did not agree, (jood and
average decoders did not differ on this point. Specifically, 36 percent felt
distrustful and 33 percent reported feeling irritated or frustrated.
Demographic Impact
Stepwise multiple regression was run to determine whether any
combination of demographic variables could be used to identify a profile
that would help predict perceived decoding ability. Seven variables were
tested for their power in predicting decoding ability: gender, age, level of
formal education, industry sector, years of work experience, position, and
income. The optimal prediction equation contained two demographic
variables, gender and the education industry sector (Table 3). That is to
say, women and individuals who work in education consistently ranked
their decoding ability significantly higher than did any other group of
respondents. More specifically, women who worked in education per-
ceived themselves higher in decoding ability than did any other group.
Other demographic variables were not significant predictors.
Nonverbal Communication • Graham/Unrub/Jenninga 55
Table 3
Demographics of Good Decoders
(Optimal Prediction Equation)
Variable
1.
2.
3.
Gender
(1 = male, 2 - female)
Education
(0 = other.
1 = education)
Constant
Beta
0.1298
0.2374
2.4010
t-Statistic
2.100
2.030
23.729
t-Probability
0.036
0.044
0.000
Note: Significance: F = 4.596
F-Probability = 0.0105
Dependent variable = decoding ability
1 = below average, 2 = average, 3 = above average, 4 = very good
Not only did women rate themselves more highly than did men, women
rated women as better decoders and men agreed that women have
greater nonverbal decoding ability (Figure 2). Fifty-seven percent of all
respondents regarded women to be the most accurate decoders. Analysis
of variance was used te determine if men and women differed in their
perception of which gender was the better decoder. Althou^ women
were considered to be better decoders, women believed this te be true
more than did men (F=24.33, p=0.0001). When asked which gender was
easier to correctly decode, 48 percent said nonverbals from women were
easier to decode tban nonverbals from men. Furthermore, women felt
more strongly than did men that womens' nonverbal cues were easier to
decode (F=7.676, p=.OO6).
Respondents in management or supervisory positions tended to be
older males having higher educations, more years of work experience,
and higher incomes. Managers believed that age improved decoding
ability more than did employees and staff. As noted earlier, the age of
tbe respondent was not found to have a significant impact in the way
respondents rated their own decoding ability. In fact as the age of the
respondent rose, the nonverbal channel was relied upon less in deter-
mining how they would interact with that person in the future. This was
also the case witb seniority. Individuals with more years of work ex-
perience tended to rely less on nonverbal cues and more on verbal
content.
56
The Journal of Business Communication 28:l:Winter 1991
NODIFFERENCE
I Male reapondenta ^ ^ ^ Female roapondenta
Figure 2
Respondents' Perceptions of
Gender Effect on Decoding Skills
DISCUSSION
The results of this study were , for the most part, consistent with the
the literature. However, there were several surprising findings. As
perceived decoding ability increased, the importance of nonverbal com-
munication increased. Better decoders believed that they were more
capable, than poorer decoders, of detecting when their supervisor's
verbal and nonverbal communication were incongruent.
Perhaps people who considered themselves to be average decoders did
not pay as much attention to nonverbal cues as people who rated
themselves above average. This does not mean, however, that average
decoders were less emotionally affected when confronted with obviously
discrepant cues. Indeed, they reported that they were just as frustrated
and distrustful as the better decoders.
Almost half (43%) of the respondents rated themselves average in
decoding ability, and another 3 percent rated themselves below average.
Since effective communication is critical to success, consider that more
than 45 percent of those surveyed claim that they are either frequently
or occasionally confused by inconsistent verbal and nonverbal cues from
their supervisors. In addition, more than 94 percent ofthe entire sample
felt frustrated or distrustful when confronted with discrepant verbal/non-
verbal communications. It should follow that organizational productivity
cannot be optimal with the current amount of miscommunication and
frustration generated by verbal/nonverbal discrepancy. The authors
surmise that both morale and productivity could be significantly im-
Nonverbal Communication • Graham/Unruh/Jennings 57
proved if employees and managers were better informed about nonverbal
communication and if they felt more confident of their encoding and
decoding skills.
Since evidence supports the proposition that actual encoding and
decoding ability is positively correlated with perceived encoding and
decoding ability (Graham, Unruh, & Jennings, 1990; Zuckerman, De-
Frank, Spiegel, & Larrance, 1982) and since the current study identified
several perceived problems related to nonverbal communication skills in
a business setting, the authors offer the following recommendations.
Decoding Sidlls
In this study, the major difference between self-rated good and
average decoders was how the two groups ranked types of nonverbal
communication. Good decoders felt that facial expressions gave the most
accurate information about the true meaning of the message while
average and below decoders relied more on voice level or tone. This
finding is consistent with the large body of literature that reports the
face as the best provider of accurate nonverbal communication.
Assuming that respondents who perceive themselves as better
decoders are more accurate, the data suggests that those who wish to
improve their nonverbal decoding skills could do so by paying more
attention to nonverbal cues and concentrating mainly on facial expres-
sions. This means engaging in increased eye contact, since this study and
many others have shown that eye contact is, by far, the most important
facial characteristic in nonverbal communication.
Self-rated better decoders also tended to rely on nonverbal cues more
than on verbal cues when communicating with a person in the future.
Reliance upon nonverbal cues makes good sense since nonverbal com-
munication has been shown (in the review of literature) to more accurate-
ly reflect the true feelings of the encoder. Decoding nonverbal cues
successfully is not an end unto itself. Employees or managers who decode
nonverbal cues accurately but fail to act on the information can face the
same consequences—e.g., negative emotions or faulty decision making—
as those who are unsuccessful at decoding nonverbal cues. Most impor-
tantly, instead of ignoring nonverbal communication, one should gently
probe for more information when faced with discrepant verbal/nonverbal
cues.
58 The Journal of Business Communication 28:l:Winter 1991
Encoding Sitiiis
Supervisors and managers should take special note: More than 50
percent ofthe respondents felt their supervisors' verbal and nonverbal
communication was occasionally or frequently in conflict. As Remland
(1981) pointed out, "Much of what a manager says may be contradicted
by what he or she does." Keeping in mind that such discrepancy can cause
miscommunication, distrust, and frustration managers should become
more cognizant of this problem and make real efforts to keep their verbal
and nonverbal communication consistent with each other.
Unfortunately, many mangers have felt that they must not show their
emotions to their subordinates. When angry or upset they may try to
hide or disguise their feelings. But, what they are really doing is sending
a mass of confiicting signals which confuse employees and erode trust.
The authors recommend a more honest approach. Numerous studies
imply that people who don't overreact, but who appropriately frown,
smile, maintain eye contact, and use proper voice inflections come across
as more caring, trustworthy, and honest They are also much more likely
to communicate what they intend to communicate (Mehrabian &
Wiener, 1967; Burgental et al., 1970; Feldman et al., 1978; Forbes &
Jackson, 1980; Garrison, 1984). This appears to be especially applicable
to managers.
Gender Differences
Gender was the other main variable which impacted the results of
this study. Women rated themselves higher in decoding ability than men
rated themselves. And, interestingly, both men and women, as a whole,
reported that women are not only better decoders, but also better
encoders of nonverbal cues. These data are consistent with the literature
(Hall, 1978).
A study by Rosenthal and DePaulo (1979) found, by observation, that
women pay significantly more attention to the face than do men. The
current study showed no difference between men and women in this
respect. This is an interesting discrepancy and one which merits some
speculation. This study has clearly shown that those who rated tbemsel-
ves high in decoding ability favored facial cues over any otber nonverbal
cues. Since experimental evidence also consistently shows that the face
is the best source of accurate information, it may be that many men are
laboring under the impression that they do pay enough attention to facial
cues and that they do maintain sufficient eye contact, when, in fact, they
do not.
Nonverbal Communication • Graham/Unruh/Jennings 59
Predicting Nonverbai Skiiis
In the authors' attempt to identify a demographic profile which could
predict self-rated decoding ability, two variables were found that fit the
equation. The highest individual predictor was, interestingly, occupation
in education. Respondents working in the education industry were likely
to rate their decoding skills very highly. The second highest individual
predictor was gender. When combined with education, these two vari-
ables provide even more information. Women who work in the education
field rated their decoding ability higher than any other group. Adding in
more variables reduced the predictability of the equation and did not
contribute to the overall information.
These findings seem reasonable. Evidence indicates women are better
than men at nonverbal communication. Women in the teaching profes-
sion would seem, logically, to have an even greater advantage because
of the nature of the occupation. Teaching demands an intense com-
munication effort—probably more so than any other profession.
Perception Differences in Managers and Employees
The authors predicted differences between the perceptions of
employees and those of managers. However, only one difference
emerged. Those in management believed that decoding ability improves
with increasing age. Respondents who held management positions
tended to be older than those respondents who were employees. Older
respondents, however, did not judge themselves to be better decoders
than younger respondents. Furthermore, some studies have shown that
older decoders (sixties and older) are less skilled at decoding nonverbal
cues than younger persons (Parham & Feldman, 1981).
SUMMARY
In summary, the results of this study showed that the perceptions of
505 respondents in the business world agree fairly well with the ex-
perimental evidence reported in the literature. Nonverbal communica-
tion was important to all surveyed, and most respondents agreed that
nonverbal communication would influence their interactions with people
more than would verbal content. Better decoders relied most on facial
expressions for accurate information while less skilled decoders
preferred voice level or tone.
Females, individually, rated themselves higher than men in decoding
ability, and as a group were perceived by both men and women to be both
better decoders and encoders of nonverbal cues. Women working in the
60 The Journal of Business Communication 28:l:Winter 1991
education field rated themselves higher in decoding ability than other
group.
Recommendations for improving nonverbal skills included paying
more attention te nonverbal cues, especially the facial expressions,
engaging in more eye contact, and probing for more information when
verbal and nonverbal cues are discrepant. Managers should be aware
that most employees feel frustration and distrust when receiving con-
flicting signals from their supervisors, and should try to modify their
behavior by being more honest in communicating their emotions.
REFERENCES
Addington, D. W. (1971). The effect of vocal variations on ratings of source
credibility. Speech Monographs, 38, 242-247.
Blanck, P. D., Rosenthal, R., Snodgrass, S., DePaulo, B., & Zuckerman, M.
(1981). Sex differences in eavesdropping on nonverbal cues: Developmental
changes. Journal of Personality and Social Psychology, 41, 391-396.
Buck, R., Miller, R. E., & Caul, W. F. (1974). Sex, personality and physiological
variables in the communication of emotion via facial expression. Journal of
Personality and Social Psychology, 30, 587-596.
Bugenthal, D., Daswan, J., & Love, L. (1970). Perception of contradictory
meanings conveyed by verbal and nonverbal channels. Personality and
Social Psychology, 16, 647-655.
BuUer, D., & Burgoon, J. (1986). The effects of vocalics and nonverbal sen-
sitivity on compliance: Areplication and extension. Human Communication
Research, 13, 126-144.
Burgoon, J., Coker, D., & Coker, R. (1986). Communicative explanations.
Human Communication Research, 12, 495-524.
Cody, M., & O'Hair, D. (1983). Nonverbal communication and deception:
Differences in deception cues due to gender and communicator dominance.
Com.munication Monograph, 50, 175-192.
Coker, D., & Burgoon, J. (1987). The nature of conversational involvement and
nonverbal encoding patterns. Human Communication Research, 13, 463-
494.
Depaulo, B., Rosenthal, R., Eisenstat, R., Rogers, P., & Finkelstein, S. (1978).
Decoding discrepant nonverbal cues. Journal of Personal and Social
Psycholc^y, 36, 313-323.
Ekman, P., & Friesen, W. (1974). Defecting deception from the body or face.
Journal of Personality and Social Psychology, 19, 288-298.
Feldman, R., Devin-Sheehan, L., & Allen, V. (1978). Nonverbal cues on in-
dicators of verbal ^ssembMnQ.AmericanEducationalResearch Journal, 15,
217-231.
Nonverbal Communication • Graham/Unruh/Jennings 61
Forbes, R., & Jackson, P. (1980). Nonverbal behavior and the outcome of
selection interviews. Journal of Occupational Psychology, 53(1), 65-72.
Garrison, L. L. (1984). Communication—The nonverbal way. Journal of Busi-
ness Education, 190-192.
Graham, G. H., Unruh, J., & Jennings, P. (1990). Demographic impact on
nonverbal decoding and encoding skills of genuine emotion in facial expres-
sions. Unpublished manuscript.
Hall, J. (1980). Voice tone and persuasion. Journal of Personality and Social
Psychology, 38, 924-934.
Hall, J. (1978). Gender effects in decoding nonverbal cues. Psychological
Bulletin, 68, 845-857.
Hegstrom, T. (1979). Message impact: What p)ercentage is nonverbal? Western
Journal of Speech Communication, 43(2), 134-142.
Honeycutt, J., Knapp, M., & Powers, W. (1983). On knowing others and
predicting what they say. Western Journal of Speech Communication, 47(2),
157-174.
Isenhart, M. (1980). An investigation ofthe relationships of sex and sex role
to the ability to decode nonverbal cues. Human Communication Research,
6, 309-317.
Knapp, M., Hart, R., & Dennis, H. (1974). An exploration of deception as a
communication construct. Human Communication Research, 7,15-20.
La France, M., & Mayo, C. (1979). A review of nonverbal behaviors of women
and men. Western Journal of Speech Communication, 43(2), 96-107.
Leffler, A., Gillespie, D., & Conaty, J. (1982). The effects of status differentia-
tion on nonverbal behavior. Social Psychology Quarterly, 45(3), 153-161.
McClintock, C, & Hunt, R. (1975). Nonverbal indicators of affect amd decep-
tion in an interview setting. Journal of Applied Social Psychology, 5(1),
65-67.
Mehrabian, A. (1971). Nonverbal betrayal of feeling. Journal of Experimental
Research on Personality, 5(1), 64-73.
Mehrabian, A., & Wiener, M. (1967). Decoding of inconsistent communications.
Journal of Personality and Social Psychology, 6(1), 109-114.
Mehrabian, A., & Ferris, D. (1967). Inference of attitudes from nonverbal
communication in two channels. Journal of Consulting Psychology, 31,
248-252.
Mulac, A., Studley, OL., Wiemann, J., & Bradac, J. (1987). Male/female gaze
in same-sex and mixed-sex dyads. Human Communication Research, 13,
323-343.
Neal, W. (1982). Focusing on nonverbal communication in business. Journal
of Business Education, 58(2), 78-80.
O'Hair, H., Cody, M., & McLaughlin, M. (1981). Prepared lies, spontaneous lies,
Machiavellianism, and nonverbal communication. Human Communication
Research, 7, 325-339.
62 The Joumai of Business Communication 28:l:Winter 1991
Parham, I., Feldman, R., Oster, G., & Popoola, O. (1981). Intergenerational
differences in nonverbal disclosure of deception. Journal of Social Psychol-
ogy, 113, 261-269.
Remland, M. (1981). Developing leadership skills in nonverbal communica-
tion: A situational perspective. The Journal of Business Communication,
18(3), 17-29.
Remland, M. (1984). Leadership impressions and nonverbal communication
in a superior-subordinate interaction. Communication Quarterly, 32(1),
41-48.
Rosenthal, R., & DePaulo, B. (1979). Sex diferences in eavesdropping on
nonverbal cues. Journal of Personality and Social Psychology, 37, 273-285.
St. John, W. D. (1985). You are what you communicate. Personnel Journal,
72(3), 40-43.
Woodall, W., & Folger, J. (1981). Encoding specificity and nonverbal cue
context: An expansion of episodic memory research. Communication
Monographs, 48, 39-53.
Zaidel, S., & Mehrabian, A. (1969). The ability to communicate and infer
positive and negative attitudes facially and vocally. Journal ofExperimental
Research in Personality, 3, 233-241.
Zaremba, A. J.( 1987). Beyond body language. BMsmessMar^fmg, 72(3), 132+.
Zuekerman, M., Amidon, M., Bishop, S., & Pomerantz, S. (1982). Face and tone
of voice in the communication of deception. Journal of Personality and
Social Psychology, 43, 347-357.
Zuekerman, M., DeFrank, R., Spiegel, N., & Larrance, D. (1982). Masculinity-
femininity and encoding of nonverbal cues. Journal of Personality and
Social Psychology, 42, 548-556.
Zuekerman, M., Larrance, D., Spiegel, N., & Klorman, R. (1981). Controlling
nonverbal displays: Facial expressions and tone of voice. Journal of Ex-
perimental Social Psychology, 17, 506-524.
Accepted by NLR, 5/14/90
doc_437580152.pdf