The purpose of this empirical study was to examine the interface
between two key models of communication style: social style and
communicator style. Social style is based on a two-by-two matrix
composed of two dimensions of observable patterns of behavior:
assertiveness and emotiveness. Communicator style was established in the
communication discipline and involves nine factors and one global
assessment of communicator image. The current study measured
self-assessments of social style and communicator style by 852
individuals to test the theoretical interface of the models. Factor
analysis confirmed the social style dimensions and some of the
communicator style dimensions. The resulting components were factored,
resulting in four dimensions of transactional style: emotive, assertive,
relaxed, and accurate. Regression results indicate three of the
dimensions are predictive of communicator self-image and explain nearly
36% of the variance.
Keywords: communication; leadership style; social style;
communication style; behavioral style
**********
It is by now axiomatic that effective communication is an essential
component of effective management and leadership. The search for ways to
understand complex communication behavior, however, is an evolving
process. One mechanism for understanding communication behavior has been
the notion of style. A number of models of style have been proposed in
various disciplines, including leadership style (Blake & Mouton,
1974; Fiedler, 1967; House & Mitchell, 1974), relational style
(e.g., Bales & Cohen, 1979; Borgatta, Cottrell, & Mann, 1958),
social style (Buchholz, 1976; Buchholz, Lashbrook, & Wenburg, 1976;
Merrill & Reid, 1981), and communicator style (Norton, 1978, 1983).
Although each model has distinct features, all have in common the notion
that some pattern of human behavior can be observed by others and will
have some impact on an individual's success in interpersonal
communication, organizational leadership, management, or other
interpersonal endeavors. Only two of these models--social style and
communicator style--are specifically concerned with general patterns of
communication behavior, and these are the focus of this investigation.
Although each model has been tested independently, to date the
models have rarely been examined together. The purpose of this research
is to examine the dimensions of communication behavior underlying each
model and to test the theoretical interface among them in hopes of
establishing a coherent, unified model of communication behavior.
Social Style Model
Over the past 30-plus years, social style has been used extensively
to train managers and sales personnel to increase behavioral versatility
through role shifting. This model has been the subject of extensive
research and application in the private sector, as well as the focus of
limited academic testing. A trainee is typically asked to have five
coworkers or regular co-interactants complete an instrument (Snavely,
1992) to indicate the trainee's other-perceived social style. The
trainee is taught to understand and accurately identify his or her
social style and those of others. Ultimately, the participant learns how
to alter his or her communication strategies to adapt to the styles of
others.
The result of learning such strategies appears to be positive.
Prince (1986) reported that "both formal and informal assessments
have found that more productive and satisfying work relationships have
resulted from the [social style] training" (p. 66).
The social style model was developed as a two-dimensional matrix
descriptive of human interactive behavior (Buchholz, 1976; Buchholz et
al., 1976; Merrill & Reid, 1981). V. J. Lashbrook and W. B.
Lashbrook (1980) and Snavely and Waiters (1983) reported four underlying
assumptions about human behavior that frame the concept of social style:
1. Humans develop relatively stable behavior patterns.
2. Humans form immediate impressions about others based on verbal
and nonverbal behavior.
3. The way individuals behave toward one another is largely
determined by their perceptions of each other.
4. The most prevalent and probably the most important perceptual
dimensions are assertiveness and responsiveness. (Snavely & Walters,
1983, p. 122)
Assertiveness
The assertiveness dimension of social style is defined as "the
degree to which a person is perceived as attempting to influence the
thoughts and actions of others" (Wenschlag, 1987, p. 22). The
assertive person is one "who shows a tendency to state opinions or
beliefs with assurance, confidence, or force" (Snavely, 1981, p.
133). Assertiveness is an action dimension often associated with an
orientation to task. Highly assertive people are perceived by others as
being confident, active, ambitious, opinionated, risk taking, fastpaced,
directive, and competitive. Low assertive individuals, however, are
perceived as being private, quiet, easygoing, submissive, risk avoiding,
supportive, and reserved (W. B. Lashbrook & V. J. Lashbrook, 1979).
Early research that tested the impact of assertiveness on other key
work-related perceptions indicated that assertive managers were seen as
being more powerful and competent than their low assertive counterparts
(Sullivan, 1977). W. B. Lashbrook, V. J. Lashbrook, Parsely, and Wenburg
(1977) found that assertive college students were trusted more and were
perceived as being more versatile than nonassertive students. In a study
of 375 adults, Snavely (1981) found assertiveness to be associated with
perceptions of trust, power, extroversion, versatility, and value
similarity. Knutson and Lashbrook (1976) found highly assertive
individuals as having lower communication apprehension when compared to
low assertive individuals. Snavely and Walters (1983) concluded that
assertiveness is associated with the managerial role, which calls for
directiveness and decisiveness. Bacon and Severson (1986) found that
assertiveness was a strong predictor of leadership emergence in small
groups.
Emotiveness
The second dimension of social style was originally termed
responsiveness. Unfortunately, the term responsiveness may not readily
convey the meaning of the dimension. All authors describe responsiveness
as involving emotive behavior. However, people may "respond"
to others in nonemotive ways. In fact, the dimension does not involve
degree of response but rather type of response. As a result, recent
research has instead used the term emotiveness.
Emotiveness can be defined as "the degree to which a person is
perceived as expressing feelings when relating to others"
(Wenschlag, 1987, p. 26). Snavely (1981) described the emotive person as
"someone who appears to express emotional states through verbal and
nonverbal behavior" (p. 133). A nonemotive person is more
controlled and is therefore harder to read. Emotiveness does not
indicate the amount of emotion that a person experiences but rather the
amount of emotion one demonstrates during interaction with others. More
important, the social style model posits that assertiveness and
emotiveness are independent dimensions. Highly emotive individuals are
perceived as emotional, people oriented, dramatic, approachable,
permissive, subjective, easygoing, open, and sociable. Low emotive
people are seen as independent, cool, rational, objective, cautious,
impersonal, and business-like (W. B. Lashbrook et al., 1977).
The research on emotiveness suggests that a number of positive
evaluations are associated with more emotive styles. Several studies
have found a correlation between emotiveness and perceptions of
versatility and trust (Buchholz, 1976; W. B. Lashbrook, Knutson,
Parsely, & Wenburg, 1976; W. B. Lashbrook et al., 1977; Snavely,
1981; Sullivan, 1977). Sullivan (1977) found emotiveness to be
associated with sociability, social attraction, interpersonal
solidarity, composure, and low dogmatism. Knutson and Lashbrook (1976)
found those with highly emotive styles to be less apprehensive about
communicating. Snavely and Walters (1983) found that emotiveness was
perceived as being more communicatively competent than nonemotiveness.
They concluded that
subordinates may find it easier to relate to "emoting"
behavior than to non-responsive, controlled behavior.
The latter style may be seen as too cold, detached
or uncaring. Because they do not tend to express
emotions in an overt manner, these individuals may
find it more difficult to convey to others whatever
empathy or versatility they feel. (p. 131)
[FIGURE 1 OMITTED]
Style Matrix
Combinations of the communication behavior dimensions of
assertiveness and emotiveness in a 2 x 2 matrix (see Figure 1) result in
four distinct styles. Those low in both assertiveness and emotiveness
are termed analyticals; those low in assertiveness and high in
emotiveness are termed amiables; those high in assertiveness and low in
emotiveness are termed drivers; and those high in both dimensions are
termed expressives.
It is interesting to note that individuals do not necessarily
perceive themselves in the same way that others perceive them.
Unfortunately, little research has been conducted to understand the
reasons or the extent of these perceptual differences. Prince (1991)
reports industry findings that managers are only 30% accurate in
estimating their own styles when compared to the perceptions of others.
People may know their own thoughts, feelings, and intents but can only
observe the enacted behavior in others.
May and Gueldenzoph (2006) examined how persons of the same or
different social styles worked together in groups. They found that group
members rated others of the same style higher than they did those with
any other social style. Those with opposite styles were not rated
differently than were those with adjacent styles.
Thompson and Klopf (1995) found cultural and sex differences in
perceptions of social style. Specifically, in their study, North
Americans and Korean men were found more assertive than Finnish and
Japanese men. Furthermore, men in each culture under study (North
America, Korea, Finland, Japan) tended to be more assertive, whereas
women tended to be more emotive. The results were discussed in terms of
cultural expectations--some cultures expect more assertiveness from men
than do others. Likewise, most cultures expect men to be more assertive
than women and for women to be more emotive than men. The researchers
also noted that there was variance within each group; so, gender and
culture were not deterministic of style. However, these results suggest
the importance of considering gender and nationality as variables in
style research.
As such, social style theorists have stressed that no style is
inherently the best style, although certainly, some behaviors may be
more or less appropriate for a given context. Wenschlag (1987) concludes
that style "isn't a trait or quality. So you should not look
at it as something that is either good or bad. It is just a way for you
to describe how you perceive ... behavior" (p. 22). Although that
may be true in a philosophical sense, it may also be true that there are
strengths to certain styles given a situational context.
Measurement of Social Style
A number of instruments exist for the measurement of social style.
Buchholz, Lashbrook, and Wenburg (1976), Merrill (1981), and Snavely
(1992) each used scales that are, in practice, given to five individuals
who regularly interact with the person whose style is being measured.
These scales determine social style by means of quartile splits on the
population. Style is conceived as a comparative, rather than absolute,
perception. As a result, approximately 25% of the population falls
within one of the four social styles, 25% in another style, and so on.
To be consistent with social style theory, a factor structure
representative of the assertiveness and emotiveness dimensions must be
orthogonal, and assertiveness and emotiveness should have a low and
statistically insignificant correlation. The most recent instrument
(Snavely, 1992) satisfies these requirements and consists of a series of
Likert-type items with high face validity and reported reliability.
Communicator Style Model
A number of other style systems have emerged in the social science
literature. Notable among these is communicator style, to which Norton
(1978, 1983) and his associates have conducted a number of studies on
its various aspects. Norton (1983) broadly conceives of communicator
style as "the way one verbally, nonverbally, and paraverbally
interacts to signal how literal meaning should be taken, interpreted,
filtered, or understood" (p. 19). He also suggests that
communicator style is a relatively enduring pattern of human
interaction. His model has been conceptualized and measured as a
self-perception of communication style. The relationship between self
and other-perceptions of communicator style is as weak as it is for
social style. Norton (1977) found that teachers viewed themselves as
better communicators and more effective teachers than did students.
Communicator Style Dimensions
Based on a number of studies, Norton's conclusion (1983) was
that there are several relevant dimensions of style: dominant, dramatic,
contentious, animated, impression leaving, relaxed, attentive, open,
precise, and friendly. The dependent variable in the model is
communicator image, which is the person's self-perception of his or
her effectiveness as a communicator. Norton suggested that the
dimensions of communicator style may be multicollinear (intercorrelated)
and mulifaceted (an individual may exhibit characteristic patterns
across the communicator style dimensions). Norton also suggests that
some dimensions would be better predictors of communicator image than
others. Most of the research on communicator style has been conducted in
the context of teacher effectiveness.
Dominant style is similar to the social style dimension of
assertiveness (Norton, 1983). Those who communicate in a dominant style
tend to be more confident, enthusiastic, forceful, competitive,
self-confident, self-assured, conceited, and business-like (Schereer,
London, & Wolf, 1973). A low dominant style has been associated with
teacher effectiveness in two empirical studies (Norton, 1977;
Sallinen-Kuparinen, Marttinen, Peramaki, & Porhola, 1987).
Dramatic style involves exaggerations, fantasies, stories,
metaphors, and other devices used to highlight or make more visible the
content of a message. In teaching contexts, the dramatic style has been
described as being more entertaining than the other styles. The dramatic
communicator is drawing attention to the message. Norton (1983) notes
that "the dramatic style often gives away true feelings" (p.
66), suggesting a link to emotiveness. Dramatic style has been found to
be associated with communication satisfaction among students (Andersen,
Norton, & Nussbaum, 1981; Norton & Nussbaum, 1980; Prisbell,
1994; Sallinen-Kuparinen, 1992; Sallinen-Kuparinen et al., 1987).
Contentious style is argumentative. Norton (1983) has suggested
that contentious style would be closely associated with dominant style.
Animated style uses nonverbal cues (eye contact, facial
expressions, and gestures) to communicate emotions. As such, animated
style and emotiveness should be closely related. Norton (1983) concludes
that this style "punctuates literal meaning, signals moods,
indicates theatrical emphases, increases or decreases intensity, and
filters qualitative content" (p. 68).
Impression-leaving style, when exhibited, results in
individuals' being remembered because of their communicative
behavior. It is not clear exactly what behaviors are involved in this
dimension. Impression leaving has been linked to teacher effectiveness
(Andersen et al., 1981; Norton, 1977; Sallinen-Kuparinen et al., 1987).
Additionally, Duran and Zakahi (1987) found impression leaving to be
strongly related to communication satisfaction in interpersonal
interactions.
Relaxed style is related to the concept of social anxiety. Those
with a relaxed style do not manifest signs of anxiety in their
communication patterns. Relaxed style projects calm, serenity,
confidence, and comfortableness. Relaxed style has been related to
teaching effectiveness (Andersen et al., 1981; Prisbell, 1994;
Sallinen-Kuparinen et al., 1987).
Attentive style involves empathy and listening skills and should
not correlate positively with dominant or dramatic style. Attentive
style has been found to be associated with communication satisfaction
(Duran & Zakahi, 1987; Inglis, 1993; Norton, 1977; Prisbell, 1994;
Sallinen-Kuparinen et al., 1987).
Open style is associated with gregarious, unreserved, nonsecretive,
and self-disclosive behavior. Those with an open style show more of
their feelings and emotions than do those of other styles. As such, it
logically relates to emotiveness. Those with open style tend to be
perceived as being socially attractive and trustworthy (Giffin, 1967)
and as effective teachers (Andersen et al., 1981).
Friendly style ranges from a basic lack of hostility to deep levels
of intimacy. The friendly communicator is one who confirms others and
positively recognizes co-interactants. This involves emotive behavior
and sociability. Prisbell (1994) found friendliness to be the strongest
predictor of student satisfaction with instructors' communication
in the classroom. Andersen et al. (1981) also found a link to teacher
effectiveness. Duran and Zakahi (1987) found other-perceived friendly
style to be a primary predictor of communication satisfaction.
Precise style has to do with being able to unambiguously explain
concepts. Those with a precise style are perceived as being in control
of the content. Precise style was shown (Norton, 1983) to be associated
with effective teaching.
Other Communicator Style Research
Although most research has focused on teachers, some work in an
organizational setting has been conducted. Norton (1983) suggested that
communicator styles may cluster to form other styles of communication
behavior. One such cluster that has been studied is the affirming style,
which is a combination of high attentive, friendly, and relaxed styles.
Norton concluded that social attraction was predicted best by this
style. Infante and Gorden (1989) and Infante, Rancer, and Jordan (1996)
found that arguments were perceived more positively with an affirming
style than with a nonaffirming style. Infante, Anderson, Martin,
Herington, and Kim (1993) also found subordinates' satisfaction was
higher with superiors using an affirming style.
Bednar (1982) found objective measures of managerial performance to
be linked to perceptions of communicator style. Subordinate satisfaction
was linked to leaders' styles that were open, friendly, calm,
relaxed, and attentive (Baker & Ganster, 1985) and, in another
study, to impression leaving, open, friendly, and flexible styles
(Honeycutt & Worobey, 1987).
Measurement of Communicator Style
The Communicator Style Measure (CSM) is a self-report Likert-type
measure developed by Norton (1978). Most of the research on this concept
has used this measure, and it consists of 51 items, of which 45 are
scored. Six items are filler items, which are intended to be ignored.
Norton also developed measures to fully explore the dramatic, open, and
attentive styles. Norton demonstrated construct and content validity as
well as reliability for the CSM.
Model Comparison
Social style differs from Norton's communicator style
construct (1983) in a number of respects. Although both concentrate on
communication behavior, social style research emphasizes patterns of
behavior across communication behavior dimensions--or style--and style
correlates. Communicator style research focuses primarily on dimensions
of communication behavior and their correlates. Snavely and Waiters
(1983) noted that
these independent variables [dimensions] can be
viewed as possessing more or less of each social style
dimension. For example, the dominant and animated
communicator styles are both high in assertiveness;
the open and friendly styles are both high in
[emotiveness]. (p. 121)
As a result, communicator style and social style have been
conceptualized as complementary, rather than competing, style systems.
The social style model may be seen as a macro model of style, whereas
communicator style involves micro-behavioral strategies or habits that
are used to implement various social styles. For example, whereas
emotive social style is a general habit, there are many ways to be
emotive, including being dramatic, open, and friendly.
Sorenson and Savage (1989) examined the relationship between
communicator style and relational style. Their factor analysis of the
CSM revealed two main factors: Dominance and Supportiveness. These
factors are conceptually similar to the social style dimensions of
assertiveness and emotiveness. The Dominance factor included items from
contentious, precise, and dominant communicator styles. The
Supportiveness factor included items from open, dramatic, friendly,
attentive, and animated communicator styles.
However, beyond Kearney's work (1984) on teaching style, to
date there has been little or no research to examine the empirical
relationship between communicator style and social style. The goal of
this research is to begin this examination. An understanding of the
structure of the domain of communication behavior represented by the
communicator style and social style models is helpful to the theoretical
conceptualization of style and the eventual construction of a single
style model.
Research Questions
This study revolves around three research questions:
1. To what extent are the dimensions of social style and
communicator style unique constructs?
2. Can a combined model of style be derived to best characterize
the nature of communicative behavior?
3. What are the best predictors of communicator image?
Method
Sample
Social style data were collected from 1,456 North American
undergraduate and graduate students. Forty-eight percent (n = 692) of
the respondents were female, and 52% (n = 764) were male. The mean age
of the sample was 24, with a range from 19 to 47 years. Of this sample,
852 also completed the communicator style instrument.
Measures and Procedures
Social style. Responses of 1,456 students to the Social Style
Questionnaire (Snavely, 1992) were factor analyzed. Initial components
were extracted by a principal components analysis. Consistent with
previous research and the conceptualization of social style as having
orthogonal dimensions, components with an eigenvalue equal to or greater
than 1 were retained and subjected to a varimax rotation. The adequacy
and interpretability of the rotated factor pattern were determined by
the following criteria: A given variable was considered meaningful if it
had a loading of .40 or greater; variables with meaningful loadings on
more than one factor were eliminated; a factor must be defined by a
minimum of three meaningful variables; variables loading on a factor
share the same conceptual meaning; different factors have distinct
conceptual meanings; and the factor pattern demonstrated simple
structure. An item analysis was performed on the variables defining each
factor. Items were dropped from a scale if their removal increased the
reliability (coefficient alpha) of the scale.
Communicator style. Responses of 852 students to the CSM (Norton,
1983) were factor analyzed. Initial components were extracted by a
principal components analysis. Consistent with previous research and the
conceptualization of communicator style as having oblique relationships,
components with an eigenvalue equal to or greater than 1 were retained
and subjected to a promax rotation. The adequacy and interpretability of
the rotated factor pattern were determined by the same criteria used for
the Social Style Questionnaire. An item analysis was performed on the
variables defining each factor. Items were dropped from a scale if their
removal increased the reliability (coefficient alpha) of the scale.
Uniqueness of the constructs. The uniqueness of the social style
and communicator style constructs was examined by factor analyzing the
scales developed in the preceding analyses. Initial components were
extracted by a principal components analysis. Components with an
eigenvalue equal to or greater than one were retained and subjected to a
promax rotation. The adequacy and interpretability of the rotated factor
pattern were determined by the same criteria used for the questionnaire
items.
Communicator image. The relationship between communication behavior
and self-perceptions of communication effectiveness (communicator image)
was tested using multiple regression. Factor scores were computed for
the factors derived from the factor analysis of scales, and communicator
image was regressed on the factor scores using stepwise multiple
regression.
Results
The results of the factor analysis of social style items are
summarized in Table 1. Consistent with past research, two
factors--Assertive and Emotive, accounting for 58% of the total
variance--emerged from the analysis. Variables defining each factor are
in italics. Both scales achieved acceptable levels of reliability as
indicated by Cronbach's coefficient alpha--.86 for Assertive and
.83 for Emotive.
Results of the factor analysis of communicator style items are
presented in Table 2. Eight factors accounting for 51% of the total
variance were found--Dramatic, Image, Contentious, Relaxed, Attentive,
Impression Leaving, Open, and Accurate. Variables defining each factor
are in italics. Nonitalicized variables did not achieve meaningful
loadings on any factor. Scale reliability ranged from .62 (Accurate) to
.82 (Open).
The Dramatic factor contains items from Norton's Dramatic and
Animated scales (1983). The Image factor combines items from the
Communicator Image and Dominant scales. Three items from Norton's
Contentious Scale define the Contentious Scale in this analysis. The
Relaxed Scale is identical to Norton's. The Attentive Scale
contains items from Norton's Attentive and Friendly scales. The
Impression Leaving scale is the same as Norton's. The Open scale is
defined by three items from Norton's Open scale. The Accurate scale
contains two items from Norton's Precise scale.
The factor analysis of the resulting scales developed in the
preceding analyses is presented in Tables 3 and 4. Four factors
accounting for 69% of the total variance were found--Assertive, Emotive,
Relaxed, and Accurate. Scales defining each factor in italics. The
factors are relatively independent; the average correlation among the
four factors is .138.
The stepwise multiple regression is reported in Tables 5, 6, and 7.
The Emotive, Assertive, and Relaxed factors all entered the model. The
Accurate factor did not enter the stepwise model. A total of 35.6% of
the variance in the dependent variable (image) was explained by these
style factors.
Discussion
The goal of this research was to test the theoretical interface
between the communicator style and social style constructs. The
following discussion considers each of the three research questions,
limitations of this study, and future directions for style research.
The first question asks to what extent the two models were unique
constructs. The results of the factor analysis indicate that the
constructs overlap. Neither model adequately maps the domain of
communication behavior underlying the two models. Four factors emerged
from the analysis: Assertive, Emotive, Relaxed, and Accurate.
Strong support for the factor structure and reliability of social
style (Snavely, 1992) was found. Both the assertive social style (alpha
= .86) and the emotive social style (alpha = .83) were replicated. The
dimensions of communicator style did not remain intact. The dramatic and
animated styles collapsed into one factor (Dramatic), as did friendly
and attentive (Attentive). The combined factors achieved acceptable
reliability for basic research--alphas ranged from .62 to .82 (Nunnally,
1967). The reliability of the Attentive and Accurate scales was
problematic and suggests that additional scale development may be
needed. After the item analysis, only two items formed the Accurate
scale. The dominant style did not emerge in this data as a separate
factor. Thus, the communicator style dimensions in this study were
dramatic, open, contentious, impression leaving, relaxed, accurate, and
attentive. As Norton (1983) suggested, these dimensions may be
multifaceted. Our results confirm his suggestion.
The factor analysis of scale scores resulted in four factors. The
first factor, Assertive, was composed of 14 items (alpha = .88) from the
Assertive, Contentious, and Impression Leaving scales. Because all
involve an action dimension of communicative behavior, this could be
expected. Again, we suggest that assertiveness is best viewed as the
macro dimension of style. Although being contentious and leaving an
impression are clearly assertive aspects of behavior, assertiveness is
not, for example, necessarily contentious.
Sixteen items (alpha = .83) from the Emotive, Dramatic, Open, and
Communicator Image scales formed the second factor, labeled Emotive.
This is not too surprising, given that being dramatic, animated, and
open all involve different ways of expressing emotion while
communicating. Examination of the items of each factor confirms that
although Emotive is a general factor involving the expression of
emotion, the communicator style factors involve specific ways of
communicating emotion. To that extent, this supports the idea that
emotiveness may be a macro component of style, and the dimensions of
communicator style may be micro aspects of style.
The third factor was Relaxed, and it contains all four items from
the Relaxed scale. This appears to be a factor of style relatively
independent from either social style dimension. The correlations between
relaxed style and the social style dimensions are fairly low (.18 with
Emotive and .24 with Assertive). It is also conceptually a unique
construct. A person can be relaxed or tense with any social style.
Snavely and Waiters (1983) discussed a related concept of social anxiety
as a key dimension of communicative competence. This research suggests
that it may more appropriately be considered an evaluative element of
style that affects communication competence.
The fourth factor, Accurate (alpha = .68), was composed of six
items from the accurate and attentive styles. Five of the items came
from the Attentive scale and one item from the Accurate scale. Both
factors have to do with accuracy in communication. The attentive items
have to do with accurate listening, active listening, and being
attentive to others. The accurate items are directly tied to precision
and accuracy in encoding messages. This factor represents another
evaluative style component, which is relatively independent of the other
dimensions (average r = .13). Because only o