CAN I PROPOSE THIS IDEA?:
THE INFLUENCE OF INFORMAL HIERARCHY ON CREATIVITY MODERATED BY ORGANIZATIONAL STRUCTURE
Master Thesis
Master Human Resource Management
University of Groningen, Faculty of Economics and Business
July 15, 2016
Tim Leegte
Studentnumber: S2197502 Stadhouderslaan 47a
9717AJ, Groningen tel.: +31 636207504
e-mail: t.h.s.leegte@student.rug.nl
Supervisor
MSc. J. Oedzes
Second Assessor
Prof. Dr. O. Janssen
ABSTRACT
In working to understand the predictors of creativity in groups, this research focuses on informal hierarchy. Previous literature has found a negative relation between formal hierarchy and creativity. This research adds to the literature by looking at the negative effect of informal hierarchy on creativity. This research also looked at the moderating role of organizational structure on this relationship. To investigate this, data has been collected by means of
questionnaires, among a sample of 56 groups from Dutch organizations across a wide range of
industries. The results show that informal hierarchy has no significant negative effect on team
creativity. However, organizational structure does moderate this relationship in such a way
that creativity is highest under an organic structure with a low informal hierarchy.
INTRODUCTION
Numerous scholars have argued that improving creative performance is a step towards competitive advantage for organizations (Amabile, 1988; Devanna & Tichy, 1990; Kanter, 1983; Shalley, 1995). The initiation and implementation of new ideas enhances an
organization's ability to respond to opportunities and, thereby, to adapt, grow, and compete (Van de Ven, 1986). Amabile et al. (1996) argue that creativity is a starting point for innovation. (Oldham & Cummings, 1996). Therefore, creativity can be highly important within an organization.
A vast amount of research on the antecedents of group creativity show that organizational characteristics like organizational hierarchy predict group creativity. For instance, numerous academics have shown that creative tasks are often performed better by groups with less hierarchy (Anderson & Brown, 2010; Berdahl & Anderson, 2005; Bloom, 1999; Siegel & Hambrick, 2005). In the same vein, Caniëls & Rietzschel (2015) demonstrate that constraints have a negative impact on creativity. These examples illustrate how formal rules, and procedures have the ability to influence the way people cooperate and interact, and thus influence their ability to be creative.
However, Theoretical work by McEvily, Soda & Tortoriello (2014) argues that formal characteristics of teams are not the only factors that influence the manner of cooperation. The behavior and interactions of people are also influenced by the informal relations that exist between them. For instance, in teams that exhibit a great difference in the level of influence of individual members, less information will be shared amongst the team members, which would diminish creativity (Bunderson & Reagans, 2011). Many organizations have turned to team- based work systems (Mohrman, Cohen & Mohrman, 1995) in which informal relations are more frequent (Diefenbach & Sillince, 2011). Therefore, this research will focus on the effects of informal hierarchy on a team based level. Specifically, I argue that a high informal hierarchy has a negative effect on creativity.
Informal hierarchy, i.e. a hierarchy based on influence that emerges from social
interactions between group members (Magee & Galinsky, 2008), can have negative effects on
group creativity, in a similar way as formal hierarchical differences can. Informal hierarchy
differs from formal hierarchy in that informal hierarchy is a person dependent social relation
which emerges from social interaction. In a formal hierarchy members are clearly defined and
demarcated from each other. The rank order of all members is formalized (Diefenbach &
Sillince, 2011).
Not only hierarchy of the group determines the level of creativity of employees within a group. This creativity can also be limited by the freedom that the overall structure of an organization can give an employee (Anderson & Brown, 2010). Certain organizations are structured mechanistically, meaning that the organization has strict rules and procedures and tries to regulate the behavior of all employees. On the other hand, organizational structures can be organic, which means that the organization is less strict in its rules and procedures and gives the employee more freedom (Burns & Stalker, 1961).
In this study, I will focus on the combined effect of informal hierarchy and
organizational structure on team creativity. I expect informal hierarchy to have a negative effect on creativity and I expect this relation to be moderated by organizational structure in such a way that the level of creativity of group members within a team is high only when the organizational structure is organic and the informal hierarchy is low. I will investigate this proposition in a field study among Dutch organizational work teams. In doing so, I aim to make two theoretical contributions.
Former research has been done on whether formal group hierarchy influences performance and creativity within groups (Anderson & Brown, 2010; Blinder & Morgan, 2007; Haslam et al., 1998, Mccurdy & Eber, 1953; Mccurdy & Lambert, 1952). However, too little attention has been paid to informal hierarchy, even though such hierarchies become more and more important in today’s organizations where people work together on a day-to- day basis on relatively complex tasks (Mohrman, Cohen & Mohrman, 1995). The present study fills a gap in the literature by looking at informal hierarchy and investigating whether informal hierarchies have the same negative effect on creativity. Furthermore, I aim to contribute to the literature by researching whether the organizational structure moderates this relation.
This research will also add to practice. Results will show that if an organization’s aim
is to increase the level of creativity among their employees, they should organize their
company in such a way that informal hierarchy is low and that the organizational structure is
an organic one. Organizations can implement an organic structure by limiting the amount of
rules and procedures and by making the hierarchical structure within the whole company flat
(Burns & Stalker, 1961). An organization can implement a low informal hierarchy by creating
teams where every actor has the same social status. Organizations could also maintain that equality by instructing teams that the teams should remain egalitarian.
THEORY
Informal team hierarchy can be defined as “person-dependent social relationships of dominance and subordination which emerge from social interaction and become persistent over time through repeated social processes ” (Diefenbach & Sillince, 2011, p 1517). Whereas formal hierarchy is an official system where the official roles and positions of all team
members are clearly defined and demarcated from each other (Mousnier, 1973), informal hierarchy is less well-defined and is a form of vertical social relations that is formed by unofficial mechanisms, such as interactions between group members. In this paper, informal hierarchy will be measured by the level of centralization of influence in a team. According to Bunderson et al. (2015), centralization is an approach that views hierarchy as the
concentration of power, status, or privilege in one member or in a small subset of the full membership of a social group. Centralization is maximized when one actor scores at the maximum and all other actors score at the minimum on some dimension, and is minimized when all actors have the same score. (Bunderson et al., 2015)
Creativity is defined as “the creation of a valuable, useful new product, service, idea, procedure, or process by individuals working together in a complex social system”
(Woodman, 1993, p. 293). It has been heralded as one of the keys for enduring advantage (Amabile, Barsade, Mueller & Staw, 2005). A study by Amabile et al. (1996) assumes that the social environment can influence creative behavior. However, little empirical work has systematically examined the possibility that characteristics of the social environment
contribute significantly to creative performance at work (Amabile, 1988; Shalley, 1991; Staw, 1990) (Oldham & Cummings, 1996).
Previous studies have reported that a larger difference between actors on influence
diminishes creativity in a number of ways. Paulus (2000) argues that sharing ideas within a
team increases the chances of creating novel ideas; this requires the group members to attend
and listen to one another. However, these activities are interpersonally risky, since actors
become vulnerable to criticism, judgment, sanction and disapproval (Detert & Edmondson,
2011; Edmondson, 1999). Groups with a larger difference amongst group members share
fewer ideas within the group and listen less to each other. For instance, Bunderson and Reagans (2011) report that differences in power and status can create an environment in which lower-ranking actors do not feel safe enough to engage in openly sharing information and perspectives, to consider different ideas, and to experiment with unproven approaches (Bunderson & Reagans, 2011). Brooks (1994) found that members of R&D teams did not feel free enough to engage in group reflection when centralization was high. In addition, Keltner et al. (2003) concluded that a power disadvantage prompts an “inhibition” response pattern.
This basic proposition has been supported by a growing body of evidence: individuals in position of lower power experience more negative emotions (Langer & Keltner, 2008), act in more situationally constrained ways (Galinsky et al., 2008), are less likely to take risks during social interactions (Magee et al. 2007), and are less likely to take initiative (Galinsky et al.
2003). Furthermore, past research suggests that the perspectives and insights of higher- ranking members are given disproportionate weight, whereas the contributions of lower- ranking members are overlooked (Bunderson, 2003; Pitcher & Smith, 2001; Westphal &
Zajac, 1995) and high-power subjects were less likely to adopt the perspective of others (Galinsky et al., 2006; Lammers et al., 2008). A more centralized hierarchy causes group members with low power to interact less and causes group members with high power to listen less, which decreases the chance of creating novel ideas (Paulus, 2000). Furthermore,
Anderson & Brown (2010) argue that when groups work on tasks that benefit from a broader range of opinions and perspectives (e.g. creative tasks), flatter structures should be more advantageous. A flat hierarchical structure is less controlling and thus better for creativity than a steep one. In addition, West (2002) said that participation in teams could, under
appropriate conditions, lead to high levels of creativity. King and Anderson (1990) stated that the probability of creative outcomes may be highest when leadership is democratic and collaborative. Structures of communication could constrain the ability to listen to everyone’s ideas and it could constrain participation within teams. According to the “intrinsic
motivation” perspective (Amabile, 1988), the context in which an individual performs a task influences his or her intrinsic motivation, which in turn affects creative achievement.
Supervision that is experienced as controlling undermines intrinsic motivation and shifts an
employee's focus of attention away from work activities and toward external concerns (Deci
et al., 1989; Deci & Ryan, 1987). Thus, hierarchy will have a negative effect on creativity by
affecting these processes.
Although the previous research links formal hierarchy to creativity. In this study, I argue that the same effect will be seen for informal hierarchy. I argue that creativity is most likely when informal hierarchy centralization is low. This is the case when all group members have approximately equal positions in terms of influence.
The first hypothesis will therefore be:
Hypothesis 1: Informal hierarchy has a negative effect on the creativity within a team This hypothesized relation will not be the same amongst various types of
organizations. The effect of hierarchy will depend on the type of organization (Burns &
Stalker 1961; Katz & Kahn, 1966; Scott 1998). Burns and Stalker (1961) suggested that there are two kinds of organizational systems, the mechanistic and the organic. A mechanistic management system is a contractual, hierarchical form of organization. Here work is distributed among specialist roles within a clearly defined hierarchy. A mechanistic
organization is viewed as more suitable for static environments. The organic form is less rule based, and can be seen more as loosely coupled networks of workers. Mechanistic
organizations have lower complexity, higher formalization and centralization, lower internal and external communication, and higher vertical differentiation than organic organizations.
Burns and Stalker indicated that the organic structure was more suited for change and thus more conducive for creative outcomes (Aiken, 1971).
Evidence suggests that mechanistic organizational designs constrain the ability of the system to produce creative outcomes. In a meta-analysis of 23 empirical studies, 21 articles and two books, Damanpour (1991) looked at the characteristics of an organic organization, e.g. high complexity, high internal and external communication and low vertical
differentiation, and found that all characteristics of an organic organization, except vertical differentiation, correlated with creative outcomes. Similarly, creativity is enhanced by adaptive, flexible organizational structures such as network designs and parallel or collateral structures (Damanpour, 1991; King, 1990; Rubinstein & Woodman, 1984; Zaltman, Duncan,
& Holbek, 1973). In addition, King and Anderson (1990) also listed group structure as an
antecedent of creativity. They argued that the probability of creative outcomes is higher in an
organic structure than in a mechanistic one, since an organic structure does not constrain the
group members.
This paper argues that both informal hierarchy and organizational structure are important for creativity. If the organization has a mechanistic design, it will matter less whether informal hierarchy is low or not. In both cases, creativity will remain low, since the characteristics of a mechanistic organization, e.g. low internal and external communication, high formalization and high vertical differentiation, constrain the ability of group members to be creative. This paper expects that the combination of an organic organizational design and a low informal hierarchy among employees creates the best environment for creativity.
This research argues that organizational design will moderate the previously discussed relation between informal hierarchy and creativity in such a way that under an organic
structure and a low informal hierarchy, creativity will be high, and, under a mechanistic structure, regardless of the level of informal hierarchy, creativity will be low. Therefore, the second hypothesis will be:
Hypothesis 2: Organizational structure moderates the negative relation between informal hierarchy and creativity, such that creativity within a team is highest under an organic structure with a low informal hierarchy
METHOD
Data and Sample
To test the hypotheses, data has been collected among a sample of 56 groups from
Dutch organizations across a wide range of industries. Groups had consist of at least 4
members, interact often, share resources and information and combine efforts to meet
common goals (Kozlowski & Bell, 2003). A questionnaire was send to 314 employees of
which 56 were leaders. A total of 205 employees have completed the survey. To be able to
receive the questionnaire, the leader of the team had to fill in a team description form with the
email addresses of every individual team member. Every member of the team then got a
unique link to their questionnaire. The measures and questions were in both English and
Dutch. Teams that were more proficient in English received the English survey and teams that
were more proficient in Dutch received the Dutch survey. The survey was not conducted
anonymously. For the researchers, it was retrievable how high the specific teams scored on
certain dimensions. Finally, we had 160 employees and 45 leaders who have filled in the
questionnaire. This resulted in 38 teams that could be used for statistical analysis. For the
employee sample, 41.3 % were males, the average age was 36.19 (SD = 12.4) years, and the
average organizational tenure and the average team tenure were 9.37 (SD = 10.7) and 4.11 (SD = 5.2) years, respectively. Most of the teams said they were situated in education
(15.6%), healthcare (13.3%) or business or professional services (11.1%), although the largest amount of teams said they could be placed in a category not listed by us, such as sports and administrative services (37.8%). Of the people who filled in the questionnaire, 42% was male and 58% was female, the mean age was 36.19 years, and the size of the group varied from 4 to 12 members (M = 5.47, SD = 2.16).
The leaders of the teams had to fill in a different survey in which they were asked about the general features of the team and the organization. All the employees had to fill in a questionnaire that was more from their point of view.
Measures
Informal hierarchy. For this study, the informal hierarchy was measured by asking every team member to rank every other team member on three questions regarding influence, e.g. influences me, asks me for advice, and I give this person advice, on a three-point scale (1=never, 2=sometimes, 3= often). This was converted to a team centralization measure by using Freeman’s centralization index (Freeman, 1978):
Centralization = Σ (I
max– I
i) / (n-1)
Where I
maxis the maximum individual influence score within a group, I
iis the influence score of an individual member i, and n is the total number of group members. This quantifies the distance between the highest scoring actor and all the other actors. It calculates the sum in terms of differences in centrality between the most central actor and all the other actors and divides this quantity by the theoretically largest such sum of differences in any network of the same size.
Organizational structure. Organizational structure was measured through seven
questions of the organization structure scale (Covin & Slevin, 1988). An example item
includes “either my organization has a focus on a uniform management style throughout
organization”, or “it is allowed that the management style of executives may vary from very
formal to very informal”. These questions were then scored on a seven-point scale (1 highly
favoring the former choice which would be mechanistic and 7 highly favoring the latter
choice indicating an organic design). Cronbach’s alpha for this scale was .86.
Creativity. Creativity within the teams was measured by asking the leaders to grade the separate members on their individual creativity on a seven-point scale (1=not creative, 7=highly creative). Based on the individual scores a mean was composed for every team. This mean was used as the score for group creativity. A Cronbach’s alpha was not calculated for the individual creativity scores, since it is possible that a highly creative team has a member that is less creative and vice versa.
Control variables. On the demographic page of the questionnaire, the participants provided information about their highest level of education (1, ‘primary school,’ 2, ‘high school,’ 3, ‘technical secondary school,’ 4, ‘university of applied sciences’, 5,’ university’ 6,
‘doctor’s degree’) and their income. These socio-demographics have been entered as control variables in the statistical analyses reported below because they correlated with creativity and team centralization. It might be possible that these variables correlate with creativity, since creative jobs are often done by highly educated people (Shaheen, 2010) and highly educated people often earn a higher income.
Analysis
For the analysis of the questionnaire responses, SPSS was used. A dataset was created with the answers on the questionnaires. The internal reliability was measured with the
“Cronbach’s α”. Then, the descriptive statistics for each variable was computed. Furthermore, the Pearson’s correlation coefficient was calculated to determine whether our variables were correlated to each other. . Based on the rankings made by the team members the steepness, centralization and linearity of every team has been calculated. To test the hypothesis, the process macro from Hayes has been used (Hayes, 2012).
RESULTS Descriptive Statistics and Correlations
Table 1 presents the means, standard deviations, and zero-order Pearson correlations
among all variables in this study. The only variable that is correlating with other variable is
education level. All the other variables did not significantly correlate with each other.
**=Correlation is significant at the 0.01 level (2-tailed).**
*=Correlation is significant at the 0.05 level (2-tailed).*
Tests of Hypotheses Hypothesis 1
Team hierarchy has a negative effect on creativity within a team
In order to analyze whether or not a lower team hierarchy leads to a higher creativity, a regression analysis has been performed. The results of this regression, R2 = 0.050, B= 0.86, t(38)= 1.383, p= .175, show that there is no significant relationship between informal
hierarchy and creativity. Therefore, a lower informal hierarchy does not lead to a significantly higher creativity.
Hypothesis 2
Organizational structure moderates the negative relation between team hierarchy and creativity within a team
In order to analyze hypothesis 2 and to find out whether or not organizational structure moderates the relation between team hierarchy and creativity, Hayes’ process macro has been performed. The results are summarized in table 2.
Table 1 M SD 1 2 3 4 5
1. Team hierarchy 0.49 1.99
2. Organizational Structure 4.24 1.18 -.077
3. Creativity 4.98 0.75 .225 .266
Control variables
4. Education Level 3.88 0.72 .409** .335* .446**
5. Income 4.70 2.09 .155 -.215 .208 -.125
Table 2 Moderation Effects of organizational structure on informal hierarchy in Predicting Creativity (N=38)
Step Variable r B SE b Beta
1 Informal
hierarchy
.225 .086 .062 .225
2 Organizational structure
.266 .168* .099 .266
3 Informal
Hierarchy x Organizational Structure
-.256** .108
(Constant) 3.89*** .78
*p<0.1, **p<0.05, ***p<0.01