In the field of communication studies the problem with identifying interactive communication styles has been dealt with on the one hand as an identification of a stable and recurring model of communication practices and on the other hand as an interpretation of the specific influence of individual personality characteristics on the verbal and non-verbal manifestations of people. Currently, the two approaches seem to be usefully converging toward an integrated solution of these two perspectives (Bakker-Pieper and De Vries, 2013; De Vries et al., 2013).

In defining communication style, researchers have for the most part used a description and a classification of the relational messages that individuals share in an interaction. Initially, Norton (1978, 1983) defined communication style as a relatively stable model of verbal and non-verbal interaction, associated to specific expectations of the role of the person. Besides identifying nine different styles of communication¹, he also added another construct referred to as communicator image, which measures the individual’s rating of how he/she perceives his/her own efficacy in communication. Various researchers have subsequently criticized the low level of internal reliability of Norton’s scale (Rubin and Martin, 1994; Horvath, 1998).

Hansford and Hattie (1987) while proceeding on the basis of Norton’s work, have instead conducted a factorial structure analysis and identified five dimensions of communicative style (dominant, expressive, relaxed, animated, attentive) and they have isolated two second-order factors on the scale: animated-dominant and supportive-attentive.

Other researchers from an organizational ambit (Bolton and Bolton, 1984) have, however, developed a more social model of communication styles, based on two fundamental dimensions of social behavior: assertion and reactivity. The first aspect measures how much dominance and self-confidence is manifested by the person during communication; the second expresses the person’s skills in conveying emotional reactions through communication. Bolton & Bolton also added another social dimension for versatility, that is, the individual’s capability of changing his/her style whenever he/she realizes that the situation is putting him/her under pressure. Recent contributions enhance the appropriateness of the communicative style for the achievement of the organizational goals (Crews et al., 2019; Yang et al., 2020).

Other studies on communication styles have considered the effect of the cultural component. Consequently, a distinction has been made between modes of communication that are little influenced by the cultural context of belonging and modes of communication that are strongly influenced and dependent on the cultural context of reference. In the first case individuals tend to send explicit and direct messages, while in the second case the messages are indirect and contextualized and always need to be adequately interpreted in order to avoid misunderstanding (Croucher et al., 2012).

Kapoor et al. (2003) in their research have shown that the first style of communication is typical of individualistic cultures, while the second is more frequently found in cultures similar to the collective model. The research of Gudykunst et al. (1996) was also oriented toward the cultural-comparative ambit and produced the Scale of Communication Styles (CSS) made up of eight main dimensions (ambiguous and indirect communication, interpersonal sensitivity, positive perception of silence – more frequent in cultures that are highly dependent on their context; inference of significance, use of emotion, dramatization, authenticity, precision – mainly associated with individualistic cultures). Successively, Leung and Bond (2001) in their analyses, extracted three second-order factors from these eight dimensions: verbal involvement, attention toward others, management of emotions and silence.

The recent contribution of Reinout De Vries et al. (2009) stands out among these studies on communication styles. He used a lexical type approach to describe the communicative behavior. First he looked for the underlying factors in an ample group of adjectives and verbs belonging to the communicative sphere. From this he extrapolated seven factors: precision, reflexion, expression, support, emotionality, kindness, threatening.

The development of the work of De Vries and his collaborators resulted in the construction of the Communication Styles Inventory (CSI), which presently constitutes, together with the Scale of Communication Styles (CSS) by Gudykunst, one of the few instruments available for the general measurement of communication styles adopted by an individual. However, in the literature we can find various instruments for the measurement of communication styles for specific ambits such as schools (Harkin et al., 1999), medical communication (Ong et al., 1995), leadership (Bechler and Johnson, 1995), and marketing (Dion and Notarantonio, 1992).

Beatty et al. (1998) is notable for having investigated the relationship between communicative behavior and personality traits. In this sense the communication style is considered to be the disposition of the person toward a certain kind of communication. Strictly speaking, communication styles are not personality traits but they emerge under the influence of these traits and the environment. Behavioral dispositions tend to generate stable relationships between the situations and the reactions of a person (Asendorpf, 2004).

According to Schulz von Thun (2003) these styles constitute a specific mode of contact, communication and management of relationships with other people. They are associated with need “flows,” states of mind and intentions that manifest their force through words and non-verbal signals. The various styles do not exclude each other but rather they come together in the person in a specific combination which as a whole expresses the individual’s preferred model of contact. The five-factor model of personality is currently a reference point for identifying the fundamental tendencies specific to each person, stable over time and mostly hereditary: neuroticism, extroversion, openness to experience, friendliness and conscientiousness (McCrae et al., 2000).

Taking this model into consideration in order to apply it to a communication theory means interpreting the communication style of an individual as a form of specific adaptation, relatively stable, of his/her behavioral models under the influence of his/her personality. This means that a person will develop a communication style that tends to be coherent with his/her fundamental personality traits. For example, a person with a prominent friendly trait will prefer to express himself in a kind and polite way rather than use aggressive expressions.

Cole and McCroskey (2000) examined the relationships between two personality models and the communication styles: the first model was the model with three factors (neuroticism, extroversion, psychosis) while the second was the now classic McCrae and John (1992). The results showed that half of the variances of the scores on the communication measurement scales were explained by the assertive dimension, while two thirds of the variances were due to the reactive dimension. At the moment these are the two main dimensions extrapolated and confirmed by diverse research on communication styles. Cole and McCroskey (2000) concluded their study by affirming that the communication style could be foreseen very clearly in the characteristics and personality traits of the individuals.

The limit of a model that refers to only two dimensions is without doubt that of an excessive simplification of reality. To contain this risk and at the same time safeguard the principles of precision and parsimony, other researchers have developed a model, defined as the “interpersonal circumplex” model, where the different modes of interaction are collocated in a circular area divided into eight sectors. The circular area is also defined by two main orthogonal, and therefore independent, dimensions (dominance and love). The sectors, each describing an interpersonal mode of behavior, are particular combinations of the two main dimensions: confident-arrogant; gregarious-extrovert; warm-kind; modest-naive; insecure-submissive; detached-introvert; cold-indifferent; arrogant-calculating. The adjacent sectors correlate positively with their adjacent sectors and negatively with their opposites; whereas there are no significant correlations with the orthogonal sectors (Leary, 1957; Wiggins, 1979; Wiggins et al., 1988, 1989).

There is general agreement that empirical research could find evidence of good collocations of communication styles inside an interpersonal circumplex model (Waldherr and Muck, 2011). However, although there has been some preliminary research orientated in this sense, to date there has not yet been a completely exhaustive study on this project (Sorenson and Savage, 1989; Leung and Bond, 2001; Muck, 2003; Riggio and Carney, 2003).

The current debate on the definition of communication style acknowledges the construct of adaptive behavior as an appropriate area where to include both the strictly personological aspects, and the social learning and cultural assimilation aspects that inevitably influence the communicative mode. The effects relating to the type of education received and the culture and environment in which the individual lives can significantly overlap with the personality trait (Liliweri, 2017). The communication style also appears to be closely related to the person’s concept of Self: individuals who perceive themselves as independent from others can communicate in a more assertive way than other individuals who present a concept of Self that is more interdependent (Kirtley and Weaver, 2010). Considering communication style as a specific form of adaptation of the person constitutes a way to adequately integrate the behavioral and personality perspectives (Waldherr and Muck, 2011; Ntoumanis et al., 2017; Tomášková, 2018).

De Vries has progressively perfected his inventory of communication styles through studies in which he tested the convergent validity of the instrument with the measurements of personality (De Vries et al., 2013). His analyses showed strong and medium level associations with the communication styles, thus further validating the integration between the perspectives of traits and communication styles.

The Communication Styles Inventory consists of 96 items which refer to the communicative behavior of the person. The items are equally divided among six scales (16 items per scale): Expressivity, Precision, Verbal Aggressiveness, Critical Spirit, Emotionality, Impression Manipulation. Each scale is composed of four sub-scales, each including four items. All the items must receive a reply on the Likert scale with an interval of five points (from 1 = complete disagreement to 5 = total agreement). The Cronbach reliability of the scale oscillates from 0.82 to 0.88 in the study where a non-student sample was used; from 0.83 to 0.87 with a sample composed of students (De Vries et al., 2013).

Considering the fact that the Italian context does not yet have a homologous instrument for measuring communication styles, we have seen fit to develop and present in this paper a preliminary validation study for an Italian version of the Communication Styles Inventory (CSI-B).

The verification of the assumptions of univariate and multivariate normality has been conducted using the procedure for the standardization of the variables, erasing the outlier cases with values greater than 3, then secondly, after calculating the Mahalanobis Distance, eliminating the multivariate outlier cases with D² greater than the critical value, calculated by considering chi-square as the reference distribution (level p < 0.001) with p degrees of liberty equal to the number of variables (Cattell and Vogelmann, 1977; Barbaranelli, 2006). The calculation of the Mardia Index (average of the squares of the Mahalanobis Distances) produced a coefficient (9285.81) lower than the limit value (9408). This selection of cases from the original matrix implied the elimination of 186 participants. Therefore, the rest of the validation procedure was carried out with 1,044 cases, 360 of which were males (34.5%) and 684 females (65.5%). The average age was 31.45 with DS = 10.37.

The evaluation of the metric properties of the scale was conducted both through an explorative factor analysis (EFA) and a confirming analysis (CFA) designed to test the goodness of the multidimensional model adopted by De Vries et al. (2013). The averages and standard deviations for the single items and those differentiated by gender are reported in Table 1.

An Exploratory Factor Analysis (EFA) with PA and estimation method of ML was conducted on the 96 items of the first version of the questionnaire, producing a Kaiser-Meyer-Olkin (KMO) index score of 0.815, a statistically significant Bartlett’s test (p < 0.001) a Chi-squared Test <0.001, RMSEA = 0.040, RMSEA TLI = 0.83. Considering Promax as rotation method and Cattell’s scree test indications (that only four main factors lay above the debris), 86 items resulted in the factor loadings Structure Matrix. For comparison purposes, a further PA with an estimation method of PAF was conducted. Once again, Cattell’s scree test indicated four factors above the debris, and 84 items resulted in the factor loadings Structure Matrix.

From the combination of the two estimation methods, before proceeding with the confirmatory analysis, four factors have therefore been reaffirmed and the 84 items common to both factor loadings SMs have been selected. This selection was subjected to a final EFA with ML and manual input of a maximum of four factors. Factor loadings indicated the elimination of eleven more items, arriving at a final number of 73 items. Kaiser-Meyer-Olkin (KMO) index score was 0.820, Bartlett’s test p < 0.001; Chi-squared Test <0.001; RMSEA = 0.055, RMSEA 90% confidence 0.04 - na; TLI = 0.66. Loadings pattern matrix is reported in Table 2.

Considering the saturations obtained and comparing with the subscales of the CSI items congruent with the content have been gathered on each component and reported in Table 3.

As can be observed in Table 3, the first factor comprised the subscales that De Vries indicated as Ingratiation, Charm and Authoritarianism. We referred to the underlying main factor that brought them together as Impression Manipulativeness. The second factor included the subscales of Worrisomeness, Sentimentality, Tension and Defensiveness. We indicated this underlying factor as Emotionality. The third factor presented the aggregation of Talkativeness, Conversational Dominance, Humor and Philosophicalness. The underlying factor that brought them together has been indicated as Expressiveness. The fourth factor presented the aggregation of Inquisitiveness, Thoughtfulness, Non-supportiveness, and Conciseness with an underlying factor which was called Preciseness.

In order to examine the validity of a 73-item CSI construct, a CFA was performed, using the Amos Graphics 18 software. The results obtained considering four factors did not show a good adaptation to the data. Therefore, the existence of a lower number of factors was verified. A further EFA with ML and manual input of maximum three factors was performed. Factor loadings indicated the elimination of thirty more items, arriving at a final number of 48 items. Kaiser-Meyer-Olkin (KMO) index score was 0.820, Bartlett’s test p < 0.001; Chi-squared Test < 0.001; RMSEA = 0.06; TLI = 0.58. Pattern matrix is reported in Table 4.

The CFA bore out that the model with three related factors presented overall good indices of adaptation to data (see Table 5 and Figure 1). Overall, the model included 18 items. The first factor measures Impression Manipulativeness (6 items) and includes ingratiation and charm. The second factor measures the Emotionality (6 items) and includes sentimentality, worrisomeness and defensiveness. The third factor measures the Expressiveness (6 items) and includes talkativeness, conversational dominance and inquisitiveness.

Finally, a further EFA with ML on the list of the 18 items. Table 6 shows the model matrix with saturations on the three identified factors, McDonald’s ω and Crombach’s Alpha values, Guttman Split-Half Coefficients, Corrected item/total correlations.

Concurrent validity was tested by examining the significance of correlation coefficients between Communication Styles Inventory factors and the five factors that make up the MPP personality measurement model, relating to a second administration carried out on a sample of 518 individuals (228 males and 290 females) with an average age of 26.40 and SD = 9.50. In relation to the results of this association, three hypotheses have been formulated: (1) the higher the Impression Manipulativeness, the higher the Cynicism and Impression Management would have been; (2) the higher the Emotionality in Conversation, the higher the Social-Emotional Intelligence would have been and the lower the Ability to Cope; (3) the higher the Expressiveness, the higher the Agentivity and the Self-Regulation would have been. As shown in Table 7, the results have confirmed the assumed directions of correlation; therefore, the measure proved good convergent validity with Multidimensional Personality Profile (MPP) by Caprara et al. (2006) and consequently its usefulness in describing communication styles in relation to personality traits. McDonald’s ω and Alpha coefficients for this second administration were 0.80 and 0.80 (Impression Manipulativeness), 0.80 and 0.79 (Emotionality), 0.74 and 0.74 (Expressiveness), respectively.

The following Table 8 reports English and Italian versions of the CSI-B, and the grouping of the items on respective factors.

As far as the scoring of the instrument is concerned, the 18 items in total are distributed over three factors, each comprising six items. Every item has a scoring range from 1 (completely disagree) to 5 (completely agree). The scoring calculation will produce, through a summation of the scores of the component items, separate measurements for each factor. Therefore, Impression Manipulativeness: 1 + 4 + 7 + 10 + 13 + 16; Emotionality: 2 + 5 + 8 + 11 + 14 + 17; Expressiveness: 3 + 6 + 9 + 12 + 15 + 18. Each factor can have a total score range from 6 to 30. Based on the distribution of the scores obtained from the normative sample, the cut-off criteria, differentiated by gender, have been identified and reported in the following Table 9.

The analyses carried out led to the definition of a scale composed of a total of 18 items, aggregated into eight subscales, which in turn converge separately on three factors. The first factor measures the ability of the person to exercise an effective Impression Manipulativeness during the conversation. Two components were also identified in this case: Charm and Ingratiation. Charm is a mixture of attractive appearance, sensuality and inner security. When in a communication one of the interlocutors appeals to one’s charisma or aesthetic attractiveness, one speaks of arguing through one’s personal charm. We usually measure the success of a conversation by how fascinated we are by what the other person has told us. The person who has charm attracts attention and involves people, despite their own will. In all of this, the voice is an important vehicle of suggestion that can positively influence the interlocutor and deepen the relationship, creating an empathic atmosphere, fundamental for the successful outcome of the conversation. The expert speaker succeeds in exerting a sometimes hypnotic charm on his listeners, to the extent that they hang on to his voice and his words. The ingratiation in the conversation refers to the idea of a conscious expressive choice, of a rational strategy aimed at achieving goals, and which consists in captivating the interlocutor, almost appeasing or seducing him. This mode of communication is based on the pleasure of others, so that they are well prepared to accept the requests that are addressed to them. Individuals need to keep other people’s impressions in line with the perceptions they wish to convey, so as to determine important consequences on the perception and evaluation of others and influence interpersonal dynamics.

The second factor, indicated as Emotionality, refers to the emotional activation produced in the individual as a result of verbal interaction. Three distinct components were identified. The first (Sentimentality) indicates the emotional transport that tends to accompany the person’s conversation, which with difficulty contains their emotions, both when the subject is current and when it relates to stories that refer to the person’s past. The individual has a pronounced empathic capacity, so that the intense emotional states of others do not leave him/her indifferent, rather he/she tends to try to identify with the emotions of others. The individual is aware of being particularly “sensitive” toward certain topics in particular which emotionally touch him/her more closely. The second component, called the Defensiveness, indicates the level of vulnerability perceived by the person following criticisms or verbal attacks received from the interlocutors. Negative comments can cause concern and an immediate closing reaction. The person feels hurt and is aware that he/she will need time to overcome the attacks and criticisms received. The most sensitive individuals tend to assume an acquiescent role in the conversation, in order to avoid situations of blatant opposition, or they prefer to listen in silence, minimizing participation in the speech. The third component was called Worrisomeness and refers specifically to the incapacity of the person to mask his/her emotional state in verbal interaction; therefore, the individual fears that every negative emotional state will be immediately grasped by the interlocutor.

The third factor, referred to as Expressiveness, refers to the individual’s ability to be effective in conversation, by monitoring and balancing the elements of communication, such as the quality of the topic illustrated, supported with an adequate amount and variety of data and sources; the clear and consistent presentation of the topic to keep interest and attention alive; the enhancement of non-verbal resources, such as posture, gestures, eye contact, pauses and silences; the ability to reposition the speech if it deviates and not least a shrewd management of the time available. Three specific characteristics that define this ability were identified: the disposition or tendency to talk a lot and willingly; choosing and addressing the topics of the conversation; involving the interlocutor with many questions aimed at testing the validity of his opinions.

The positive loquacity of an expressive style serves to better explain concepts and ideas, and is typical of the eloquent person, that is, able to express himself with verbal richness and linguistic knowledge and illustrate even complex concepts with his own words. The above mentioned loquacity can be a “fixed” characteristic of a person, something that is fixed in character, or an occasional behavior: for example, some who becomes loquacious suddenly, because he feels uncomfortable, or instrumentally, because he thinks that in a certain context such behavior is beneficial to him. Those who show a low level of loquacity, on the other hand, might show good listening skills, but also have difficulty expressing their ideas, expressing their desires and intervening in discussions.

Expressive assertiveness is also manifested as conversational dominance, exercising control over the arguments brought into discussion and with the ability to impose one’s own point of view. In this case, the person activates moves that initiate the sequence, i.e., questions, statements through which the subsequent development of the interaction is determined. He therefore exercises control over the topics under discussion, their articulation into sub-themes, their development and their conclusion. In cases where the participants’ agreement on the definition of the current situation seems to be lacking, he has the power to re-establish the interactional order through meta-communicative comments that redefine the contextual framework and the type of activity in which one is involved. Those who instead have low levels of dominance in the conversation tend to play a passive role and adapt to the arguments proposed by others.

The active and reflective attitude of the person during the conversation is also manifested through the tendency to ask the interlocutor precise questions about the content of the conversation, even if they are strict about the truthfulness and plausibility of the conclusions presented. The attention required of the interlocutor to the thematic focus is always high. This may reveal a tendency to not get lost in generic, superficial issues, not considered functional to the objectives of the conversation. A low level of inquisitiveness, on the other hand, may indicate a limit in the critical and reflective capacity of the person, who is substantially disengaged and little involved in the topic of discussion.

The tool altogether allows to evaluate three crucial dimensions that concur jointly with an adaptive exercise of the communicative interaction: the control of the Other (measure of impression manipulativeness); control of oneself (measure of emotionality); control of the expressive means (measure of expressiveness). Concurrent validity testing showed associations worthy of further investigation: first of all, the Impression Manipulation factor has shown a significant correlation with the MPP sub-scale of cynicism, which highlights the instrumental aspect of manipulation, the tendency to put one’s own interests before those of others, and to bend others and circumstances to one’s own ends and objectives. The further negative correlation with Innovation is consistent with a profile that is not very open to values, principles and behavioral habits that are different from one’s own, and which is above all concerned with concreteness in situations. Finally, the positive correlation with the MPP subscale of impression management confirms the tendency to use communication tools, using charm and ingratiating to obtain appreciation, esteem and trust.

The predictive measure of Emotionality could provide in the first place useful inferences on the dimension of Emotional Intelligence of the person, by virtue of the positive association found, i.e., on the individual abilities that manifest themselves in knowing how to put themselves in other people’s shoes and in knowing how to recognize and express emotions, in knowing how to take into account the needs and requirements of others, in knowing how to put others at ease. At the same time, the inverse relationship with Ability to Cope would allow to infer the measure of the vulnerability of the person in maintaining calm and tranquility in the presence of difficulties and stressful situations, in experiencing states of discomfort and general mood swings, in showing difficulties in recovering after failures, tending to brood over what happened.

The Expressiveness factor showed strong positive correlations with the dimensions of Agentivity, Self-regulation and Emotional Intelligence. It would be useful to firstly note the predictive value of the measure of this style of communication on those of Agentivity and Self-Regulation. In the first case, it would mean obtaining useful inferences on the person’s ability to assert his or her own opinions, to set ambitious goals, to know how to guide and motivate others and to conduct one’s activities with vigor and promptness. In the second case, the inferences would concern the individual’s abilities with regard to planning and persistence in achieving a goal, in self-discipline understood as a self-reflexive ability to organize, as well as tenacity in success, love of order and accuracy, which transpires in the desire to do things well. A further positive correlation with emotional intelligence indicates the predictive value of the measure of expressiveness also with respect to the person’s ability to actively listen, to feel at ease with others, and to put others at ease through one’s own behavior. The expressiveness factor therefore demonstrates an important predictive relationship with the assertive capacity and effectiveness in communication.

The tool appears agile and versatile, prefiguring an application during the assessment in multiple contexts where it could be important to obtain information on the communicative attitudes of the person: companies, institutions, staff selection, individual, and group profile analysis, coaching, psychotherapy, counseling, career guidance. Several recent contributions have emphasized the importance of taking into account communication styles in education, health and in the dissemination of news by the media (Ashton et al., 2004; Minnema, 2014; Fourie, 2018; Chlopicki and Laineste, 2019; Labreche et al., 2019; Lõrincz, 2019; Ramachandiran and Mahmud, 2019; Trant et al., 2019; Iskandarova and Griffin, 2020).

This first Italian validation study of the Communication Styles Inventory turned out to be a significantly reduced version compared to the original instrument of De Vries et al. (2013) which presented 16 scales and six main domains (expressiveness, precision, verbal aggression, critical spirit, emotionality, manipulation of the impression). De Vries’ main objective was to obtain a tool capable of understanding most of the main lexical dimensions together with the behavioral components of the communicative style. The result was a particularly large and composite instrument (96 items). Instead, in our study, starting from the original instrument and through a subsequent analysis of factorial refinement and confirmatory verification, we have achieved a brief tool that displays good fit indices which are suitable for Italians. Future studies could test the adequacy of the model on specific samples of the population and explore the possibility of recovering the dimensions previously excluded in this first work.

The datasets generated for this study are available on request to the corresponding author.

The studies involving human participants were reviewed and approved by the Institutional Review Board (IRB_SUSS) of the University of Cassino and Southern Lazio. The patients/participants provided their written informed consent to participate in this study.

PD, GV, and SM designed the study, analyzed the data, and discussed the results. PD, GV, and AG drafted the manuscript. SM and AG revised the manuscript. All authors approved the final manuscript. Finally, the authors have agreed to be accountable for all aspects of the manuscript in ensuring that questions related to the accuracy or integrity of any part of it are appropriately investigated and resolved.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.