Koman Wolff final JMD article 8 29 07 formatted

Design/ Methodology It is argued here that team leader’s emotional intelligence EI will influence the development of group level emotional intelligence GEI, which was measured by a team’

Elizabeth Stubbs Koman, Ph.D.

The Stubman Group, LLC

&

U.S Navy

2025 Tartar AveVirginia Beach, VA 23461-1924

* Research based on the dissertation research of Elizabeth Stubbs (Stubbs, 2005)

* The views expressed in this paper are those of the authors and not the U.S Navy

Purpose of Paper This research examines the relationships among team leader

emotional intelligence competencies, team level emotional intelligence, and team performance

Design/ Methodology It is argued here that team leader’s emotional intelligence

(EI) will influence the development of group level emotional intelligence (GEI), which was measured by a team’s

emotionally competent group norms (ECGN) Secondly, it is hypothesized that the presence of ECGNs will positively influence group effectiveness Data were collected from 422 respondents representing 81 teams in a military organization.Findings Results show that team leader emotional intelligence is

significantly related to the presence of emotionally competent group norms on the teams they lead, and that emotionally competent group norms are related to team performance

throughout the organizationWhat’s original/value of

paper

This research contributes to the field by offering support for the effects team leader’s emotional intelligence has on the teams they lead as well as by showing team level emotional intelligence affects team performance This study adds to the body of literature in what is considered a relatively new area

of study The four key contributions of this research are: (1) this research shows that leader’s behaviors are important at the team level, (2) this research further validates Wolff and Druskat (forthcoming) ECGN theory by lending support for the ECGNs as well as offers alternative clustering ideas for the norms, (3) ECGNs were shown to be related to

performance, and lastly (4) this research extends the knowledge base about emotions in groups

IntroductionEmotional Intelligence competencies have been shown to be significantly related

to individual performance (Boyatzis, 1982) both in cognitive tasks where the individual

is under stress and in tasks where individuals are interdependent on one another (see Druskat & Jordan, forthcoming, for a review) At the team level, the study of emotions and the effects of emotions on team performance is a relatively new avenue of research Since teamwork is an inherently social activity, emotions play an important role in team effectiveness Druskat and Wolff (1999, 2001a, 2001b) proposed a model of emotional intelligence at the group level Groups develop a set of behavioral norms labeled

emotionally competent group norms (ECGN) that guide the emotional experience in the group The degree to which a group develops these norms has been linked to team performance (Druskat, Messer, Koman & Wolff, 2003) Understanding the factors that lead to the development of ECGNs would be beneficial for team development

The purpose of this research is to assess the relation between team leader

emotional intelligence competencies and the emergence of emotionally competent norms

in a team Specifically, the present research examines the relationship between emotional intelligence of a leader, the group level emotional intelligence (GEI), and how both of these levels of emotional intelligence affect performance of the team

Definitions

Team/Group

Like Cohen and Bailey (1997), whose work is based on Hackman (1987) and Alderfer (1977), we define a team as “a collection of individuals who are interdependent

in their tasks, who share responsibility for outcomes, who see themselves and who are seen by others as an intact social entity embedded in one or more larger social systems and who manage their relationships across organizational boundaries” (Cohen & Bailey,

1997, p 241) The scope used in this research consisted of groups of individuals who worked together toward a common output, thus we refer to them as teams throughout the paper

Team effectiveness

Team effectiveness is a multidimensional construct (Goodman, 1979; Hackman, 1987; Sundstrom et al., 1990) The term team effectiveness entails both meeting customerspecifications and being able to work together effectively in the future (Hackman, 1987) This view ensures that the team is not focused on customer satisfaction to the exclusion

of concern with the well being of the team and its members, or vice versa

Defining Emotional Intelligence for the Present Study

While there are varying definitions of emotional intelligence, there is agreement

in the literature that EI includes an individual having an awareness of and an ability to regulate their emotions Salovey and Mayer’s theory of EI focuses on the emotional abilities that link emotion and individual cognition, where Goleman and Boyatzis’s theory focus on social and emotional competencies (Jordan & Troth, 2004; Matthews, Zeidner & Roberts, 2002) This study utilizes the emotional intelligence theory advanced

by Boyatzis and Goleman This EI theory has evolved into four overarching clusters of EIskills: self-awareness, self-management, social awareness, and relationship management (Goleman, 2001; Boyatzis, Goleman & Rhee, 1999) The four clusters represent a

recognition and regulation cluster for both the individual (self) and social competencies (other) Table 1 identifies the competencies in each cluster and their definitions

Insert Table 1 about here -

-To examine the relationship between a team leader’s emotional intelligence and the development of group-level emotional intelligence, we used the above mentioned EI competencies For a more robust discussion of the clusters and each competency see Stubbs (2005)

Defining Emotional Intelligence at the Team Level

Druskat & Wolff (1999) identified the existence of emotionally competent group norms (ECGNs) “that influence and manage the emotional process in a way that builds emotional capacity and develops social capital and leads to effectiveness” (p 9) These group norms are an indication of the group’s emotional intelligence and can help to determine if a group of individuals functions as a high-performing team (Goleman, Boyatzis & McKee, 2002) Wolff and Druskat state that each of the ECG norms is related

to the individual, group or cross-boundary (external) level Within each of the three levels, there is at least one norm that is an awareness norm and one that is a regulation norm (see Table 2 for a definition of the norms) In this study, we use the following definition for group-level emotional intelligence: The ability of a team to generate

operating norms that increase awareness of emotion and management of behavior in ways

that have positive emotional consequences

Insert Table 2 about here -

-The Relation between Team Leader Emotional Intelligence and Group Emotional

Intelligence

Team leaders are responsible for the success of the teams they lead As such, theyare not only responsible for their own emotions, but also for the emotions of the team they lead and the clients of the team (Rafaeli & Worline, 2001) To influence and move people, one must possess the knowledge and skills of emotional competencies (Boyatzis, Stubbs & Taylor, 2002) Boyatzis (1982) defines such competencies as “the underlying characteristics of a person that lead to or cause effective and outstanding performance.” With teams being social in nature, it is logical that emotional intelligence would be an important factor in team leader effectiveness; and it has been shown to be important for the success of managers and leaders (George, 2000; George & Bettenhausen, 1990; George, 1995; Gardner & Stoug, 2002)

Scholars have argued and shown that team leaders influence the processes,

behaviors, norms, and climate of the team they lead (Kimberly 1980; Schein 1992; Dickson, et al., 2001; Druskat & Wheeler, 2003) However, there has been a limited amount of research linking team leadership to performance The empirical work that has been conducted has found that leadership has effects on team motivation, efficacy, and performance (Sivasubramaniam et al., 2002; George, 2000; Schein, 1985; Dickson, et al., 2001); primarily through the development of the team’s climate (Piloa-Merlo et al., 2002)

Goleman (2001) and Williams (1994) suggested that leaders who are emotionally intelligent are essential to developing a climate where employees are encouraged to perform to the best of their ability When the leader is helping the team develop its norms,the climate that is developed maintains a consonance with the team leader’s individual personality (Dickson, et al., 2001, p 201) If the norms developed reflect the team

leader’s personality, it could be argued that the emotional intelligence norms developed

on the team would reflect the emotional intelligence competencies of the team leader This research will examine the effect that 18 EI competencies have on the presence of GEI, and the effect of GEI on team performance (see Figure 1)

Insert Figure 1 about here -

-Hypothesis 1: The level of team leader EI is positively related to the presence of Emotionally Competent Group Norms

Emotional Intelligence and Teams

Although there is a substantial body of literature on individual emotion and on emotional intelligence, there is mixed evidence regarding the effects of emotional intelligence in teams and work groups (Feyerherm & Rice, 2002; Jordan & Troth, 2004) Feyerherm & Rice (2002) found that the higher the team leader’s emotional intelligence, the worse the team performed, however, they did find a positive correlation between the team leader’s ability to understand emotion and the performance on the customer service metric (p.354) Whereas Jordan & Troth (2004) and Offerman, Bailey, Vasilopoulos, Seal

& Sass (2004) found that teams with higher levels of EI performed better than teams withlower levels of EI When assessing the team’s EI, all three research teams (Feyerherm & Rice; Jordan & Troth; Offermann et al.) used measures that assessed each individual team member’s emotional intelligence The current study varies from previous research

in that we used a team-level measure to assess the team’s overall emotional intelligence

GEI has been shown to be significantly related to performance (Stubbs & Messer,2002; Druskat, Messer, Koman & Wolff, 2003) This research will further validate the findings that GEI effects team performance through the testing of the relationship

between the ECGNs and team performance (Stubbs & Messer, 2002; Druskat, Messer, Koman & Wolff, 2003;)

Hypothesis 2: There is a positive relationship between ECGN presence and team

effectiveness

MethodThe objective of this research was to assess the relationship between individual emotional intelligence competencies, team level emotional intelligence, and team

effectiveness This field study was a cross sectional examination of the EI and GEI norms that were present in teams and team leaders in a military organization

Sample & Procedure

A military sample was used In each of two commands, both aircrew teams and maintenance teams participated The maintenance teams served as direct support for the aircrew teams A total of 349 aircrew and maintenance team members participated representing 81 aircrew and maintenance teams Additionally, 70 team leaders and 73

managers (team leaders supervisors) rated team leaders’ emotional intelligence, 13% of the team leaders were women and 70% were men1 About 600 team members were asked

to participate with 349 (58%) completing surveys All participation was voluntary and everyone was given the opportunity to decline participation

Final team level data analysis was completed on a sample of 275 men and 50 women There were 55 officers and 294 enlisted personnel who participated On average,participants had been members of their teams for 14.6 months, and a member of the military for 74.8 months Each team had a mean of 6.8 team members (Range = 3-11; Median = 6.5) Any team where the number of respondents was less then 50% of the total number of team members was not included in the analysis The final sample

consisted of 64 teams where the response rate represented at least 50% of the team members

Aircrew team leaders were officers (n = 9)2 and maintenance team leaders were senior enlisted personnel (n= 49) Team leaders had been involved with their teams for 1-

24 months, with the average being 10.6 months Average military tenure was 210 months (17.5 yrs; range of 60-300 months)

All participants in this study were either a team leader or member on a

functioning military team This population was used because the nature of military work requires the use of teams (Prapavessis & Albert, 1997; Orasanu & Backer, 1996; Zaccaro,Gualtieri & Minionis, 1995) The teams that participated in this study were direct missionsupport aircrew teams as well as maintenance teams that support the aircrew teams The

1 Twelve participants did not indicate their gender

2 Not all participants indicated their rank

aircrew teams are tasked to perform operational military flights on a regular basis The teams are composed of a mission commander, two pilots, and individuals performing avionics type tasks The maintenance teams are tasked to support the aircrew teams by performing needed maintenance on the aircraft The team of maintainers are specialized mechanics who work together on one specific portion of the aircraft, ie, engine,

propellers, avionics, etc Participant’s lives depend on their team members, and their support teams All teams that participated have a direct role in mission accomplishment, and are able to see the results of their teamwork

Measures

Team Leader Emotional Intelligence

To assess team leader emotional intelligence, the emotional competence inventory(ECI-2) was administered The ECI-2 uses 360-degree feedback methodology to assess the emotional competencies of individuals (Wolff, 2006) The ECI-2 was developed by Richard Boyatzis and Daniel Goleman with the help of the Hay Group (McClelland, 1973; Boyatzis, 1982; Goleman, 1998; Boyatzis and Sala, 2004) The ECI-2 has an overall average internal consistency coefficient of 0.78 and the self-ratings have an overall average internal consistency coefficient of 0.63 (Wolff, 2006) The ECI-2 has been used in various venues, in particular, in assessing the relationship between an individual’s emotional intelligence and their leadership behaviors (McClelland &

Boyatzis, 1982; Cavallo & Brienza, 2002; Boyatzis & Sala, 2004)

The ECI-2 consists of 72 questions that assess 18 EI competencies Participants rated frequency of use of each item on a one-to-five Likert scale ranging from never (1)

to consistently (5) If an item was not applicable or the respondent did not feel they could accurately assess the ratee on a particular item, there was a space marked “don’t know”

so participants were not forced into an answer The ECI-2 is a proprietary instrument; information on the scales can be obtained from the Hay Group Each team leader in the final sample had 2 -14 raters rate their behaviors, with an average of 4.34 ratings

completed for every team leader, excluding the self-rating

Cronbach’s alpha internal reliabilities were assessed for each scale Most scales had acceptable reliabilities, with their alphas close to or above 0.70 (see Vogt, 1999), withthe exceptions of emotional self-control, initiative, and conflict management Reliabilitieswere: accurate self assessment, α = 75; emotional self awareness, α = 75; self-

confidence, α = 73; achievement orientation, α = 69; adaptability, α = 77; emotional self-control, α = 42; initiative, α = 16; optimism, α = 78; transparency, α = 70;

empathy, α = 83; organizational awareness, α = 63; service orientation, α = 83; change catalyst, α = 60; conflict management, α = 33; developing others, α = 83; influence, α

= 74; inspirational leadership, α = 87; teamwork and collaboration, α = 67 The

majority of the theorized scales were found to have acceptable reliabilities Since overall individual emotional intelligence is being examined, we decided to drop the three

competencies with low reliabilities This leaves 15 competencies, which still provides a good indication of overall emotional intelligence that will allow us to test our hypotheses

We next tested the theoretical factor structure of the EI scales using AMOS 6 Weexamined each theoretical cluster (self-awareness, self-management, social awareness, and relationship management) separately due to the large number of variables The

results showed that the theorized factor structure did not produce a good fitting model Although all values of RMSEA were acceptable (.07 - 085), all but one NFI and RFI were below the acceptable 9 level (.73-.88)

Group Emotional Intelligence

Team level emotional intelligence was assessed using the Group Emotional Intelligence measure developed by Druskat and Wolff and later refined based on work by Hamme (2003) Team member participants self rated their team’s behavior according to each of the nine ECG norms measured by the instrument The ECGN scales are

comprised of 57 questions, representing nine team norms The nine scales were

comprised of 5-8 questions, with one to three items in each scale reversed scored

Respondents rated each item on a one-to-seven Likert scale ranging from very inaccurate (1) to very accurate (7) Interpersonal understanding was measured with six items, e.g.,

"On our team we make an effort to understand one another’s attitudes and views."

Confronting members who break norms was measured with six items, e.g., "In our group,

we let members know if they do something considered unacceptable." Caring behavior was measured with eight items, e.g., “We let members know that we value their

contributions.” Team self-evaluation was measured with seven items, e.g., "On our team

we often discuss what is helping or hurting our performance.” Creating resources for working with emotion was measured by six items, e.g., “When there is tension in our group, we acknowledge or talk about it” Creating an affirmative environment was measured with five items, e.g., “When something goes wrong, we look at it as a challengerather than an obstacle” Proactive problem solving was measured with six items, e.g.,

"In our team we work hard to anticipate problems that might occur.” Organizational

understanding was measured with seven items, e.g., "We understand how our work contributes to the company’s goals." Building external relationships was measured with six items, e.g., "We build relationships with teams that can help make a difference in our performance."

Cronbach’s alpha internal consistency reliabilities were assessed for each GEI

scale All scales had acceptable reliabilities, with their alphas close to or above 0.70 (see

Vogt, 1999): interpersonal understanding, α = 83; confronting members who break norms, α = 67; caring behavior, α = 82; team self-evaluation, α = 75; creating

resources for working with emotion, α = 75; creating an affirmative environment, α = 70; proactive problem solving, α = 75; organizational understanding, α = 73; building external relationships, α = 71

To test the proposed relationships, individual participant responses were

aggregated to create a mean score for the entire team The questions were written at the group level in order to capture group-level constructs, not individual attributes (Earley, 1999; Rousseau, 1985) This enabled the aggregation of the data to assess the norms present at the group level (Earley, 1999; Langfred, 2000) For all norms, intraclass correlations (James, 1982) indicated that within group variance was less than the betweengroup variance; with the variance being significantly greater for seven of the nine norms Building external relationships and proactivity in problem solving were near significant (Stubbs, 2005)

Using AMOS 6, we tested the theoretical factor structure of the GEI norms looking at each level separately due to the large number of variables Although previous

analyses (see Stubbs, 2005) showed the factor structure to produce good-fitting models for each factor, analysis with AMOS 6, which uses stricter standards for calculating fit indices (see Mayer, Panter, Salovey, Caurso & Sitarenios, 2005), showed that the

theoretical factors (individual, group, and cross-boundary) did not produce a good-fitting model Although RMSEA was in an acceptable range (.06 to 1) for the three factors, the NFI and RFI were below an acceptable 9 (ranging from 7 to 85)

Team Performance

As previously discussed, team effectiveness is defined as multidimensional; therefore, to assess team effectiveness, both objective and subjective measures were used.Subjective performance measures were gathered from upper level officers who had observed multiple teams within the command over time This typically was an individual

at least two levels above the team Participants were asked to evaluate each of the teams under their management using a 7-point Likert scale The subjective performance

measure consisted of a 5-item questionnaire developed and tested by Druskat, Messer, Koman and Wolff (2003) The following criteria were evaluated: efficiency in getting things done, quality of work, ability to be self-directed, performance against other teams that perform similar work, and ability to continue working together in the future The responses to each question were totaled to produce a subjective rating of each team’s effectiveness

Each command provided the researchers with metrics used in their management process to measure team performance A criterion for participation in the study was that effectiveness measures were recorded monthly and measured multiple criteria, thus each team had performance metrics calculated on multiple dimensions Examples of the

measures include: percentage of raw material waste; number of accidents; and percentage

of flight objectives met The objective performance rating was calculated from the percentage of goals attained by each team on each measure The researchers collected objective performance data for the period preceding and following data collection Performance measures were collected for a period of 2-8 months depending on

availability of data

As previously discussed, team effectiveness is defined as multidimensional with both objective and subjective measures used The measures were moderately positively correlated with one another, r = 08, p = 24 Subjective and objective performance scoreswere standardized within team type to minimize variations in reporting The combined performance score was used for hypothesis testing

Results

Exploratory Factor Analysis

Because the confirmatory factor analyses failed to confirm the theoretical factor structure of either EI or GEI we performed an exploratory factor analysis We first did a principal components analysis using an oblimin rotation with the 15 EI and 9 GEI

competencies/norms entered simultaneously The analysis showed two factors with all EIcompetencies loading on one factor and all GEI competencies loading on the other We next examined EI and GEI competencies separately This resulted in two EI factors (see Table 3) and two GEI factors (see Table 4) The two GEI factors can be interpreted as theawareness norms and the regulation norms Caring Behavior is a regulation norm that loaded highest on the awareness factor (.58) but the loading on the regulation factor was

comparable (.44) Since it theoretically should be included with the regulation norms, weplace it with the regulation factor in further analyses It was difficult to interpret the two

EI factors so we decided to use an overall mean value of the competencies for a measure

of overall emotional intelligence The final model we used to test our hypotheses is shown in Figure 2

Insert Tables 3 and 4 about here, Insert Figure 2 about here -

-SEM Model Testing Hypotheses

The model in Figure 2 was run with all team data using AMOS 6 and was not a good fit Based on theoretical considerations the model was modified as shown in Figure

3 Specifically, Druskat & Wolff (2001b) argue that awareness precedes behavior, thus, it

is reasonable to assume that the awareness norms will lead to regulation norms, which then affect performance This modification highlights the fact that performance

ultimately depends on behavior Also, the link between EI and the Regulation Norms wasnot significant and was deleted from the model All parameters are significant (one-tailed) and the model is approaching a good fitting model Although NFI and RFI are less than the recommended 9 (Byrne, 2001), CFI is approaching an acceptable value of

95 (Byrne, 2001) and RMSEA is at the recommended 1 cutoff (Tabachnick, B G., Fidell,

L S 2001)

Insert Figure 3 about here -

-To further understand the relationships between team leader’s emotional

intelligence and the development of ECGNs, Table 5 shows the correlations between the two We also ran a series of stepwise regressions with each ECGN as a dependent variable and all team leader emotional intelligence competencies as independent

variables The results of these analyses are shown in Table 6

Insert Table5 and Table 6 about here -

-DiscussionThis study showed that a team leader’s emotional intelligence affects team level emotional competence and team performance through the development of ECGNs This study also supported that the emotionally competent group norms affect team

performance

The four key contributions of this research are: (1) this research validated that leader’s behaviors are important in the development of team norms, (2) this research further validates Wolff and Druskat (forthcoming) ECGN theory by lending support for the ECGNs as well as offers alternative clustering ideas for the norms, (3) ECGNs were shown to be related to performance, and lastly (4) this research extends the knowledge base about emotions in teams

Previous research has shown that leaders who are emotionally competent are better performers, more successful, are able to handle relationships successfully, and demonstrate a myriad of other traits (George, 2000; Goleman, 1998; Gabriel & Griffiths,

2002) With leaders having direct influence over their subordinates (Rafaeli & Worline, 2001), it is easy to see how a team leader’s EI influences the development of ECG norms

in the team they lead While earlier research established that an individual’s EI affects their performance, the present research now shows that a team leader’s EI affects team performance through the ECGNs that are established on the team they lead

Another contribution of this research is the validation it provides to the Wolff and Druskat ECGN theory Results partially support the relationship between ECGN

presence and team effectiveness Specifically, it was found that awareness norms

contribute to the development of regulation norms and that the regulation norms are related to performance While this was not what was originally hypothesized in this research, the reason why this relationship is present is clear On the individual level, one cannot have a regulation of emotions without having an awareness of their emotions first

It is logical that the same relationship would be present with the group level emotional intelligence; the group has an awareness of the emotion, which leads to the ability to regulate it

consequences, which is a less prevalent condition in industry, the results may not

generalize to an industrial setting

The sample used in this research consisted of two types of teams, aircrews and maintenance teams These types of teams are very different, and neither are highly interdependent The Wolff and Druskat GEI theory is based on interdependent teams While Stubbs and Messer (2002) found that task interdependence was not a moderating factor for GEI, the lack of interdependence on the teams in this study could have been a contributing factor to them not being more strongly related to performance

Implications for future theory and research

This research provided an initial examination of the relationship between

individual and group level emotional intelligence and their effects on group effectiveness.This research has sparked numerous additional questions to be addressed by future research

One area for future research to examine is that of the organizational culture and organizational climate We would suspect that an organization’s climate would greatly influence GEI; the climate might directly affect the development of team norms, or the climate might influence the leader’s EI behaviors, which in turn would affect the team EI.The context in which a leader and a team are placed have been theorized and found to impact their success (Wageman, 1999; Hackman, 1999) Similarly, Ashkanasy, Wilderom and Peterson’s (2000) meta analysis on organizational culture and climate indicates that organizational climate does impact work processes With context having an effect on

groups and organizational climate affecting various work processes, it is reasonable to assume that organizational climate would affect ECGN development and sustainment.

Current research is focusing on examining the presence and effects of GEI, research should also examine the prerequisites and conditions that are necessary for normdevelopment The development of team norms takes time; the amount of time necessary for a team to develop ECGNs has never been examined Nor have the individual level traits necessary for ECGN development When team members have worked together over time, they begin to develop relationships with one another and build trust in one another Time on the team might be an important component for the development of group emotional intelligence, but it should be examined in conjunction with relationship and trust scales to assess the amount of variance each of these items account for

Additionally, another area of particular interest was the stepwise regression findings (Table 6) Results indicated that the individual EI competencies optimism, and

organizational awareness were both frequent predictors for the ECGNs There is a logical connection between optimism and the norms it is a predictor for; without

optimism an individual most likely would not confront members who break norms or create resources for working with emotion as there would not be any perceived benefit While the deductive reasoning on why these two competencies are frequent predictors forECGNs can produce compelling arguments, this demonstrates another area for further examination

With results indicating that team leader’s EI affects GEI development, it could be reasoned that individual group member EI would also effect overall group EI

Conversely, since team leader EI affects GEI development, research should examine if

leader/mentor EI affects individual EI development Another aspect of GEI development that should be researched is how team training on norm development impacts a group’s ability to develop emotionally competent group norms This research took a snapshot view of the presence of team leader EI and team GEI; both of these variables could potentially be impacted with training

Lastly, this research was conducted on a military sample While the military has been utilized in research for centuries, the teams in the military cannot be assumed to be representative of teams in industry Future research should examine the relationship between team leader EI, group EI, and group effectiveness with a sample comprised of industry teams

Implications for practice

The implications of the findings presented in this study are also important for practice Information that will help corporations improve performance is always desired This research provides implications for practice in three primary areas: 1) development and sustainment of emotionally intelligent managers and leaders, 2) development and sustainment of emotionally intelligent work groups, and 3) establishment of

organizational leaders at all levels to foster and support emotional competence throughoutthe organization

Development of emotionally competent managers

Up to now, it has been widely known that individual EI affected individual

performance, this research has shown that individual EI also affects team performance through the development of emotionally competent group norms In light of these