What is the effect of communication, physical distance and informal interaction on the speed of the post-merger integration phase?

(1)

1

Thesis

What is the effect of communication, physical distance and informal interaction

on the speed of the post-merger integration phase?

Student: Mariska Vernooij 10742166

Thesis supervisor: Bert Flier

(2)

2

Statement of Originality

This document is written by Mariska Vernooij who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

3

Abstract

The focus in this thesis is on the social integration of employees of the merged companies; when do employees of two companies feel they have become one. The extent to which employees identify with the new organization must be perceived a key factor in the success of a merger. Previous research shows that several factors can impact the speed of re-identification, e.g. the social integration of the two groups, with the new organization. Based on the presence of contact as explained by social identity theory, communication, physical distance and informal interaction were expected to have an impact on the speed of the post-merger integration phase.

Research was conducted among employees of three companies who experienced a merger. Within sample one, the average time to identify with the organization was about 20 months, with 20% of the sample not identifying (yet) with the company. The results of sample two show a lower speed of identification with the new organization, with 70% still not identifying with the organization after 1.5 years. Results further indicate physical distance does not influence the speed of the post-merger integration phase. Results indicate identification with the organization increases by approx. 4 months when management increases responsiveness. The impact of individuals characteristics on the speed of identification shows personality traits e.g. reciprocator or self-interested, did not provide a significant moderating impact. Results suggest change readiness is an independent variable.

(4)

4

1. Introduction

Since the ’90’s many companies have chosen to grow and create value by acquiring or merging with another company (Cartwright and Schoenberg, 2006). In 2004 30,000 mergers and acquisitions (M&A) were completed globally (Cartwright and Schoenberg, 2006) and 2014, 29,500 merger and acquisition deals were completed (Thompson Reuters, 2015). Mergers and acquisitions are intriguing phenomena which have been researched extensively.

An extensive body of research focused on the value creation and financial performance of mergers and acquisitions (Cartwright, 2005 in Cartwright and Schoenberg, 2006), concentrating on organizational and strategic fit. This research found that financial performance increases have shown to be positive in short term, but still seem ambiguous in the long run (Cartwright and Schoenberg, 2006). Here, mergers are defined as two companies getting, combining into one new company (Mastracchio et al, 2002).

More recently the focus shifted to human and psychological aspects of M&A. It is being acknowledged that strategic and organizational fit alone, determined during the pre-M&A stages, are necessary but insufficient conditions, as the realization of value creation depends on the guidance of management during the post-M&A integration process (Cartwright and Schoenberg, 2006; Birkinshaw et al., 2000). The integration of resources in the post-merger and acquisition stage does not only consist of the integration of systems and processes, but also of the integration of the organizational members of the newly formed company (Shrivastava, 1986). From a social identity perspective, a merger is the combination of two groups, becoming one new group (Van Knippenberg, 2002). A certain level of organizational change will inherently be the consequence following a merger and can have a major impact on organizational members. Studies show that inadequately coordinated change has a negative impact organizational commitment, lead to lower job satisfaction and higher turnover intentions. Organizational change was found to trigger identity threats during organizational

(6)

6 change, because change can affect the identity (Jacobs et al., 2008). The extent to which employees identify with the new organization must therefore be perceived a key factor in the success of a merger (Van Knippenberg et al., 2002).

Social identity theory

Social Identity Theory (SIT) explains how identities are shaped and also applies to the level of integration following a merger. The identity of people is built out of perceived memberships to groups. Part of the self-image derives from the social categories in which the person feels he belongs (Tajfel and Turner, 1979). When a social group changes, as it does during a merger, a “stratified society” can occur (Tajfel and Turner, 1979), which is a situation in which the objectives of the group are unevenly distributed. As a result of the unevenly distributed objectives dominant and subordinate groups are created. Both groups strive to maintain their own objectives. Each distinct group may start actions to maintain and justify its status quo (Tajfel and Turner, 1979). Employees related response to organizational change as a result of a merger or acquisition (Cartwright and Schoenberg, 2006) is likely to consist of negative behaviour as a result of not well coordinated integration during the post-merger integration phase. This inadequately coordinated integration phase negatively impacts employee attitude on collaboration and increases intention to leave as well as conflicts and decreases satisfaction (Marmenout, 2010). This will in turn have a negative impact on the financial performance and value creation of the company (Marmenout, 2010).

Communication

Communication is a widely established construct to influence the perception of employees during organizational change and help them cope with uncertainty (Bordia et al., 2004). However, as a construct in (re-)identification during the post-merger integration phase, it has only recently gained attention (Bartels, 2006). A positively evaluated communication

(7)

7 climate before the merger and the communication during stable, non-merger times, positively contributes to re-identification after a merger (Bartels, 2006). Furthermore, the quality of change communication, defined as the specific communication during and about the merger, also contributes to identification in the post-merger integration phase (Bartels et al., 2006).

Physical distance

Distance is defined as employees being located at different geographical distances, in different cities or countries (Mortensen et al., 2005). The difference between dispersed employees and colocated employees (employees physically at the same location), comes with limitations such as more conflict and lower cohesion and has an impact on group processes (Mortensen et al, 2001). Dispersed teams show a weaker sense of identity and are more susceptible to out-group feelings (Mortensen et al., 2002). SIT explains that people tend to move from one group to another to re-establish their positive self and their identity. The above-mentioned observations implicate these movements are hindered when employees are at distance.

Informal interaction

Informal interactions are related to informal, usually unplanned communication with an open and flexible character between members of the same level (Mortensen et al., 2001; Postmes et al., 2001; Kiesler and Cummings, 2002). This informal communication between employees is at the root of social identity, as social identities are largely established and sustained via communication (Gardner et al, 2001). Within the context of a merger, interaction among people decreases in-group favouritism, and increases positive perceptions and support for the merger (Gleibs, 2010). Frequency of interaction contributes positively to the social relationships and cooperation (Kiesler and Cummings, 2002), whereas cooperation is recognised to increase identity forming (Gaertner et al., 1990).

(8)

8

Speed of identification

In the strategy domain, speed is seen as a variable that predicts success, prosperity and advantage (Angwin, 2004). This concept has entered the merger and acquisition domain and is linked to the idea that if a company rapidly wraps up the post-merger integration phase, this is thought to positively contribute to performance (Angwin, 2004). Speed is a relatively new concept and social antecedents impacting speed in the post-merger integration phase have not been given much attention yet.

If speed is seen as a variable that predicts success and positive performance, it is interesting to start exploring from a social perspective the independent variables that influence the speed of social identification in post-merger integration. Especially the more soft factors on employee level seem underdeveloped. Organizational identification is defined by Ashforth and Mael (1992, p. 104) as “the perceived belongingness to the organization and the extent to which employees sees himself as a member of the group”.

In summary, previous research shows that several factors can impact the speed of re-identification, e.g. the social integration of the two groups, with the new organization. Based on the presence of contact as explained by social identity theory, communication, physical distance and employee interaction may have an impact on the speed of the post-merger and acquisition integration phase. The purpose of this study is therefore to determine the effect of physical distance, formal communication and informal interaction on the speed of the post-merger integration phase.

The research question of this paper is:

What is the effect of physical distance, formal communication and informal interaction on the speed of the post-merger integration phase?

(9)

9

2. Theoretical model on the speed of post-merger integration

Mergers and acquisitions are, simply put, two organizations getting together to form a new organization (Van Knippenberg et al., 2002). By way of merger, two companies merge into one as two parties being equal, combining into one new company (Mastracchio et al, 2002). When an acquisition is carried out, one new company is the outcome, as is the result of one company purchasing the other company (Mastracchio et al., 2002). Most merging companies are not equal parties. Generally there is one dominant party, exceeding the other party (Mastracchio et al., 2002; Van Knippenberg et al., 2002).

Both mergers and acquisitions usually involve significant organizational change (Seo et al., 2005). When forming one new organization, two main aspects need to be dealt with. First, a number of systems and procedures need to be decided upon, like the desired procedural integration, functional integration, legal and accounting integration (Shrivastava, 1986). Second, the employees of the two companies need to re-categorize to one new group (Van Knippenberg et al., 2002).

Integration will lead to a newly designed company, and is likely to impact the organization and its members (Seo et al., 2005). As this paper is related to the organizational change involved in the post-merger and post-acquisition phase, mergers and acquisitions will be treated the same. The focus in this research is on the speed re-identification with the company, meaning two groups have to become one, henceforth have to “merge”. Therefore,

merger will be used from here on, because the focus is on the combination and integration of

two or more companies into one new entity.

Seo et al. (2005) have further defined the post-merger integration phases. After thorough investigation of different descriptions of post-merger integration phases, Seo et al. (2005) created a framework and described four distinctive stages, ranging from pre-merger to

(10)

10 the full completion of the post-merger integration. Four stages are described (Seo et al., 2005):

a) The premerger stage starts when a company has the intention to merge and is finalized when the merger is announced.

b) The initial planning and formal combination stage relate to the moment the merger is legally established and top management of both organizations create the new vision and goals for the established organization. It also involves the decision making on the governance of the organization and re-staffing. At this stage, there are rumours in the organization and feelings of uncertainty can occur.

c) The operational combination stage is where the employees of the combined organization experience the changes. Until this stage, only top management has been involved in the merger and the remaining members of the organization have been informed, but not yet actually involved. This stage includes all employees in the change and as a result of the newly designed vision and governance, the employees only now realize the change in its full extent as the new processes and colleagues are introduced. In this stage conflicts and culture clashes between employees of the previously separate organizations may occur. Employees are likely to mythologize their old organization.

d) The final stage is the stabilization stage. In this stage the effects of the merger settle down and the organization no longer experiences the buzz, commotion is over and it is business as usual. The business as usual is however the new way of working.

Consolidation is achieved.

This framework will be used in this paper as the outline for the analysis. The operational

combination stage is the stage in which the employees experience the consequences of the

(11)

11 middle management to operational employees, are at best informed, but not involved in the pre-M&A phase.

Organizational change during the merger process can have a major impact on the organizational members. Studies have been conducted on the results of a merger, explaining employees’ attitudes on collaboration, intention to leave, conflicts and satisfaction. These are influenced by the way the merger is conducted (Marmenout, 2010). This is very much related to general results measured after organizational change, in which inadequately coordinated change results in lower organizational commitment, lower job satisfaction, higher turnover intentions and lower trust in management (Kernan and Hanges, 2002). As Lewin described in the early ‘50’s, mergers can be seen as episodic change (Weick and Quinn, 1999). This is a “sudden”, external change, following an equilibrium period and is experienced as dramatic. According to Lewin, change was a systems perspective of freeze, transition and refreeze (Weick and Quinn, 1999). Through the years, the original perspective of Lewin changed. The systems perspective evolved to a psychological phenomenon (Dent and Goldberg, 1999). Next to the systems and processes aspect of a merger, the human aspect of change has gradually gained attention. Organizational change was found to trigger identity threats during organizational change, because change can affect the identity (Jacobs et al., 2008).

The uncertainty of an employee about his or her own identity will likely occur during the initial planning and formal combination stage where top management will define the new vision and goals for the organization, thus implicitly formulating the new identity. Only in the operational combination stage employees are confronted with the implemented changes and experience how their identities are threatened.

In the remainder of this chapter, the variables and hypotheses of this thesis will be further explained.

(12)

12

Social Identity Theory

Social Identity Theory (SIT) was founded by Tajfel and Turner in 1979, as a social-psychological perspective. Social identity theory explains that the identity of people is built out of perceived memberships to groups, like gender, age, religion and organizational membership. Self-image partly derives from the social categories to which the person feels he belongs. As people join several groups, they have different identities and behaviours to align with a specific group. The theory predicts that intergroup behaviours are explained by the perceived group statuses, legitimacy and stability of these differences and the possibility to change groups (Tajfel and Turner, 1979). Tajfel and Turner explain three assumptions underlying SIT:

1) positive self-concept; people aim to sustain or enhance their self-esteem: they seek a positive self-concept.

2) Positive group perception: Social groups, or categories, and group memberships are linked to positive or negative emotional values. Hence, a social identity may be perceived positively or negatively. The perception is related to the valuations of the groups that together contribute to the persons’ individual social identity.

3) Individual mobility: Through social comparisons in terms of group characteristics and affective attributes, the individual evaluates his or her own group in comparison to other groups. People tend to leave a group they not perceive as positive and join a more favourable group.

Social identity theory is also applied to the organizational context. The organizational identity is one of the many identities a person holds. Being part of an organizational group is needed to have a positive personal self (Ashforth and Mael, 1989). In a situation of a merger, identities are often mixed and abandoning of the old identity in favour of a new identity is implied (Seo et al., 2005). This implies change of the social group, and it is likely that a “stratified society” occurs (Tajfel and Turner, 1979). This describes a situation in which the

(13)

13 objectives of the group are unevenly distributed. As a result from the unevenly distributed objectives dominant and subordinate groups are created, both groups striving to maintain their own objectives. The distinct groups may start actions to maintain and justify their status quo, which can lead to intergroup conflict. Both groups view themselves as the in-group (the favourable group) and the other group as the out-group. As both groups see each other as the out-group, the willingness of individual mobility declines, and therefore the emergence of a new identity slows down.

Speed of identification

According to the Van Dale (2015) dictionary, speed is “rapidity in moving, traveling, proceeding or performing, and can also be described as the relative rapidity in moving or going”. In the strategy domain, speed is seen as a variable that predicts success, prosperity and advantage (Angwin, 2004; Epstein, 2004). From a merger perspective, a quick integration phase is thought to positively contribute to performance (Angwin, 2004), as the realization of benefits can be reaped earlier (Epstein, 2004). Epstein (2004) furthermore explains that success is determined by immediate planning and design following the merger agreement.

The concept of speed in the merger domain is relatively new and focusses mainly on the impact of changes in the first 100 days as a measure of overall financial successfulness. However, speeding on the post-merger acquisition phase was indicated inadvisable (Angwin, 2004). The ‘correct’ pace of speed was also examined, depending on factors of strategic and cultural (mis)fit (Bauer and Matzler, 2004), but not quantified. Furthermore it is explained that the faster the integration phase is completed, the shorter the uncertainty and the shorter the dysfunctional behaviour (Angwin, 2004; Homburg and Bucerius, 2005) as stability is re-established.

The component of speed as an independent variable only emerged recently. The first relationships between speed and performance are established, but specifically the more soft factors on employee level seem underdeveloped.

(14)

14 The focus in this paper is on the “human” drivers that influence the speed of integration. From a social identity perspective, a merger is the combination of two groups, becoming one new group.

According to Ashforth and Mael (1992, p. 104) social identification “is the perception of belongingness to a group classification”. Organizational identification is “the perceived belongingness with an organization and the extent to which the employee perceives him or herself as an actual or symbolic member of the group” (Ashforth and Mael, 1992, p. 104).

For a new group to be formed, they have to both identify themselves with the post-merger organization (Van Knippenberg et al., 2002). This generally implies, to a greater or lesser extent, a change of identity for both groups (Van Knippenberg et al., 2002). The achievement of re-identification in this context analogizes integration.

According to social identity theory, positive in-group comparison has to take place in order to identify with new people. When positive in-group comparison takes place, individual mobility occurs and new groups form (Tajfel and Turner, 1979).

Intergroup comparison takes place by interaction between people (Tajfel and Turner, 1979; Gioia et al., 2000). It is therefore expected that the degree of formal communication, physical distance and employee interaction impact the speed of identification and accordingly the duration of the post-merger integration phase.

Communication

Communication is a widely established construct to influence the perception of employees during organizational change. The quality of information, received by employees from management during the reorganization process, should be helpful and consist of timely, accurate information (Kernan and Hanges, 2002). However, communication as a variable in (re-)identification in the post-merger integration phase has only recently gained attention (Bartels et al., 2006). Bartels et al. (2006) analysed recent studies and found that only a limited number of studies researched communication as an important factor in successful

(15)

15 identification in the post-merger phase. Bartels et al. (2006) concluded communication is an important factor, but the influence on integration in the post-merger integration phase is relatively unknown.

Communication also is an important factor in managing uncertainty (Bastien, 1987). It is explained that quality and congruence as well as the quantity, especially communication towards the acquired organization, is a determinant of the degree of turn over and levels of productivity (Bastien, 1987).

These constructs are defined as “communication climate” by Bartels et al. (2006). The communication climate consists of the quality, honesty and support of top management during stable, non-merger, times (Bartels et al., 2006). Postmes et al. (2001) found that the more positive employees evaluate the communication from top management, the more they identify with the organization. The communication climate in general was found to contribute strongly to the identification with the organization (Bartels et al., 2009). Communication about the strategy of the organization, and providing sufficient information in general before the merger, is suggested to contribute to make employees identify more strongly with the organization as a whole, after a merger (Bartels et al., 2006).

Top management usually communicates about the merger. It has been confirmed communication specifically about the merger reduces uncertainty and ultimately stress and turnover intensions (Bordia et al., 2004). Poor change communication often results in rumours and negative perceptions about the change and is likely to lead to resistance to the change (Bordia et al., 2004) and can ultimately lead to destructive behaviour (Schweiger and DeNisi, 1991). Applied to the merger context, Bartels et al. (2006) found that the more positive employees evaluated the quality of communication about the merger, the stronger they identify with the post-merger organization.

(16)

16 It is therefore expected that communication provided by top management positively influences the speed of the post-merger integration phase.

Hypothesis 1: The quality of communication positively influences the speed of the

post-merger integration phase.

Physical distance

Geographically dispersed teams are a rapidly increasing phenomenon, partly as a result of development of telecommunications (Mortensen et al., 2001). Besides the technological innovations contributing to the increased emergence of distant teams, there are more types of organizations which have large amount of geographically dispersed teams, such as consultancy firms.

This research will investigate whether physical distance, by definition “team members who are located at significant geographical distances from one another, in different cities or countries” (Mortensen et al., 2005) impacts the speed of the post-merger integration phase. In turn, colocated teams are teams who are geographically at the same location, meaning they are physically present (colocated) in the same building or site (Mortensen et al., 2001, p. 213).

In literature, distance comes with limitations and has impact on group processes (Mortensen et al., 2001). Distance is related to more conflict (Mortensen et al., 2001), within distant teams employees show lower levels of cohesion and seem to like each other less than members of face to face teams (Mortensen et al, 2005). Employees in dispersed teams also tend to have more harsh preconceptions about their colleagues, because of the inadequate interpretations of the behaviours of their colleagues (Mortensen et al., 2001).

The identity of people is built out of perceived memberships to groups (Tajfal and Turner, 1979). It is expected that physical distance between employees decreases the development of shared identity, because observation and interaction are less adequate. The

(17)

17 other group cannot be perceived to a sufficient extent and therefore the comparison of positive group perception is absent. This is supported by the research performed by Mortensen et al. (2001) that colocated teams, because of their more adequate observation and interaction, tend to be more open to include members than dispersed teams and dispersed teams have a tendency to drop, especially distant members (Mortensen et al., 2001). Dispersed teams show a weaker sense of identity and are more susceptible to experience outgroup feelings (Mortensen et al., 2002).

It is therefore expected employees within dispersed teams, such as consultants working within a consultancy firm, identify slower with the organization after a merger. Social identity theory argues that individuals engage in a process of categorization relative to those around them. Due to individual mobility, people tend to move to another group to re-establish their positive self and identity (Tajfal and Turner, 1979). Distant employees are less often at headquarters than their colocated colleagues, and are therefore expected to lead to less adequate perceptions about their co-workers and less positive group perceptions.

Hypothesis 2: The more distant employees are from headquarters, the slower the

post-merger integration phase evolves.

Informal interaction

Social psychology literature suggests that informal communication drives the creation of social processes, like perceptions about people (Kraut et al., 1990). Informal communication is defined in many ways. It is termed spontaneous communication (Mortensen et al, 2005), casual encounters and face-to-face contact (Kiesler and Cummings, 2002), intergroup contact (Gleibs, 2010), informal interactions (Mortensen et al., 2001), interpersonal communication and horizontal communication (Postmes et al., 2001). This type of communication is open and flexible, and allows for rumours to be discussed (Mortensen et

(18)

18 al., 2001). They all refer to the interaction among members of approximately the same level (equality) and are usually unplanned and informal. From here on, informal interaction is referred to.

It was already suggested in the 1950’s by Festinger et al. that informal interaction strengthens social ties and builds interpersonal bonds between distant workers, which is in line with social identity theory (Mortensen et al., 2001). Social identity has an evident link with informal interaction, because social identities are largely established and maintained via communication (Gardner et al., 2001). The categorization process underlying identification (Hogg and Turner, 1987), demands for communication cues to develop social categories and perceptual elements by which they characterize and thereby identify with the group (Fiol et al., 2005)

It was also found that lack of interaction disrupts the development and maintenance of a shared identity (Mortensen et al., 2005). Absence of informal interaction may result in a decrease of connectedness and community (Sarbaugh-Thompson, 1998 in Mortensen et al., 2005). With the lack of informal interaction, parts of information are incomplete and can lead to lower mutual awareness and unshared context. These factors will likely create task and affective conflict (Mortensen et al., 2001).

When employees interact, they do not only do this as individuals, but as members of the same organization as well (Gleibs, 2010). Intergroup contact promotes the development of positive intergroup relations, also in the context of a merger (Gleibs, 2010). The contact has a positive impact on reducing intergroup biases and reduces conflict. The group identity and the intergroup contact are crucial to reduce biases among and between groups (Gleibs, 2010). When employees did contact a merger partner, it was revealed that in-group favouritism decreased and did so for members of both organizations (Gleibs, 2010). This led to faster

(19)

19 integration. Gleibs (2010) explained that group identity and intergroup contact are crucial to in-group favouritism and their perceptions and support for the merger.

Heide and Miner (1992) explained that the frequency of interaction is positively related to the level of cooperation. Kiesler and Cummings (2002) also note that people are mostly influenced by the people they interact frequently with. They further found that the frequency of interaction has a strong relationship with social ties and the identification process (Kiesler and Cummings, 2002; Fiol et al., 2005).

It is therefore expected that employee interaction positively impacts the speed of the post-merger integration phase.

Hypothesis 3: The more employees interact, the higher the speed of the post-merger

integration phase.

From a social identity perspective, a merger is the combination of two groups, becoming one new group. For a new group to occur, they must both identify with the post-merger organization (Van Knippenberg et al., 2002). This generally implies, to a greater or lesser extent, a change of identity for both groups (Van Knippenberg et al., 2002). The extent to which employees identify with the new organization must be perceived a key factor in the success of a merger (Van Knippenberg et al., 2002). This all takes place in the post-merger

operational combination stage, in which employees experience the change; get to know each

other and are expected to work together as the result of top management’s vision (Seo et al., 2005).

The purpose of this thesis is to study to what extent the factors communication, physical distance and informal interaction impact the speed of re-identification, e.g. the social integration of the two groups, with the new organization.

(20)

20 The outlined construct can be organized in the following conceptual framework as outlined in figure 1:

H3 H2 Pre merger stage Initial planning

and formal combination

Operational combination stage stabilization

Merger and integration stages (Seo et al., 2005)

Phys i cal Distance Forma l Communication

Speed

-Freeze Transition Refreeze

+

Informal i nteraction

+

H1

(21)

21

3. Methodology

This paper aims to study if and how the speed of the post-merger integration is impacted by communication, physical distance and informal interaction. The focus in this study is on the social integration of employees of the merged companies; when do employees of two companies feel they have become one. We concentrate on the “operational combination stage”, that is, the specific integration stage in which the integration of the operational units of the former companies are to become one. In this stage people get to know each other, and are expected to work together as the vision of the top management is implemented in the newly established organization.

In order to measure whether communication, physical distance and informal interaction impact the speed of the post-merger operational combination stage a questionnaire was distributed among employees who experienced a merger situation. In the paragraphs below, the research design is explained in detail, starting with the reason for choosing a questionnaire, the sample and data collection, and description of the measures.

3.1 Research method: questionnaire

A questionnaire is a widespread and familiar method of collecting data and people find it easy to understand (Saunders and Lewis, 2012). A questionnaire is especially suitable for questions related to “where”, “how much” and “what” (Saunders and Lewis, 2012). As this thesis seeks to explain to what extent speed is impacted by communication, physical distance and informal interaction, the explanatory nature of a questionnaire collects data about the relationships between variables provides the information for answering the research question (Saunders et al., 2012). Furthermore, an anonymized questionnaire increases the likeliness of answering honestly, not providing socially desirable answers (Saunders et al., 2012). In this research, encountered negative experiences among participants during the post-merger integration phase were expected to slow down the integration. As persons might be hesitant to answer honestly in person about negative experiences, an anonymous questionnaire increases

(22)

22 the likeliness of honest responses, because it ensures privacy and is less intrusive (De Leeuw, 2008).

On the other hand, a questionnaire has shortcomings. As the questionnaire at its nature creates distance between the researcher and participant, no interviewer support nor flexibility in the questions is possible, thereby decreasing response rates as well as the validity of the questionnaire (De Leeuw, 2008), because questions can be misinterpreted or misunderstood (Bowling, 1997). The closed questions used in a questionnaire are at the expense of the depth of the responses and risks the quality of the data being incomplete or diminished (Bowling, 1997).

Ideally, data collection would be longitudinal. This research is cross-sectional due to time constraints and based on retrospective information recall, because it examines past events; the mergers carried out by T&P. Recall of information relies on the memory of the individual, and people find it difficult to remember events accurately as the mental processes are often imperfect and can be unreliable (Hassan, 2005). Research investigated 20% of critical details of an event will be reported incomplete after one year (Hassan, 2005). However, a standardized and well-structured questionnaire can minimize the risk of this bias (Hassan, 2005). Furthermore, this research relies on self-report data, which might increase common method bias as three out of four independent variables and the dependent variable are answered by the participant (Podsakoff et al., 2003), therefore increasing the risk of this systematic error. Especially the consistency effect might occur, as respondents are questioned about retrospective emotions and perceptions (Podsakoff et al., 2003). As it is not possible to fully tackle this bias up front, post hoc statistical measures will be undertaken to minimize the potential effects.

According to Social Identity theory the level of integration always is an individual measurement (Knippenberg et al., 2003) and the questionnaire was therefore sent to

(23)

23 individual employees, to measure at the individual level. Another important advantage of a questionnaire is the fact that a larger sample can be reached than with qualitative research. The questionnaire was distributed to 145 employees, which could not be interviewed in person due to time constraints. A questionnaire therefore seems the most appropriate method, as the explanatory nature suits this research, a large group of respondents was reached and privacy guaranteed.

3.2 Sample and data collection

The research about the speed of the post-merger integration phase will be held among employees of the former companies Numerando, Match&More and Triple A. Talent&Pro (T&P) is an employment agency in the insurance and banking industry, founded in 1999. T&P provides advice, project members and interim managers in the banking, insurance and pensions industry and serves all large financial institutions in the Netherlands. Since the new owners acquired T&P in 2009, the company has expanded significantly, from 300 employees in 2009 to 1100 employees per 2016. This growth is mainly established by the acquisition of five independent companies between 2011 and 2015. Of these five companies, three companies are suitable for this research. The relevant companies are:

Numerando Group. This was an employment agency in the pension, insurance and

pharmacy industry. T&P explained they acquired Numerando to enhance their position, and become an employment agent market leader within the pensions segment. The executives of T&P felt this was necessary to cope with the highly dynamic environment within the pensions industry. The acquired company, Numerando, was on the break of financial bankruptcy, and although they had a firm position and reputation, it was an easy target to acquire.

Match and More was a consultancy company specialized in interim employees for the

financial services industry and a preferred supplier for the Rabobank. The acquisition of Match&More can be defined as a take-over. The takeover of Match&More by Talent&Pro

(24)

24 was communicated as a further enhancement of T&P’s position as an employment agency and consultancy firm within the financial industry. By this merger T&P aimed to further establish their position as biggest employment agency within the financial services industry. Match&More was a successful company and financially healthy and therefore an attractive acquisition for T&P.

Triple A – Risk and Finance, is a consultancy firm specialized in the actuarial

department and risk management. The merger with Triple A is slightly different. The merger with Triple A was communicated as a step to further strengthen their position in the dynamic market. Talent&Pro acknowledges and values Triple A’s expertise in the Finance and Risk markets. Furthermore, T&P acknowledges the trend in the market towards “total solutions”; the take-over of the entire problem and solving the solution on behalf of the client. Along with the merger with Triple A, the governance structure of T&P was changed. Redmore Group was established and Talent&Pro became a separate business unit as well as Triple A, thereby also confirming that Triple A was not to be integrated within T&P as were Numerando and Match&More. The former and current organization charts are included in appendix 1.

Data collection

The study will be conducted by use of a questionnaire among employees. The total sample consists of 151 employees, which have been employed during one of the aforementioned mergers. Probability sampling is applied here as acquaintance with the company is believed to increase response rates. The sample will be stratified as the employees will be divided into two strata. These are: 1) Former Numerando employees and former Match & More employees, 2) Former Triple A Risk Finance employees. These strata are

(25)

25 applied because the identification with the overall organization is different. For former Numerando and Match&More employees, the identification is towards Talent&Pro, whereas the identification of Numerando is related to Redmore. For further information, see the organization chart in appendix 1.

After distribution of the questionnaire by e-mail, reminders were sent to the personal mail of the participants. As this research has the interest of the company’s employees, the personal network within T&P was used to increase response rates by posting an engaging news item on the intranet site and active stimulations to fill in the survey by the people managers of T&P. The survey is written in Dutch, as all participants have Dutch nationality and is believed to enhance the understanding of the response rates.

Due to time constraints, the data collection will be cross-sectional. The mergers investigated are a few years ago, and the establishment of new identity is expected to have taken place, especially for Numerando and Match&More. The merger of Triple A Risk finance is more recent and if the data explain the process of reestablishment is not finished yet, it will provide information about the process of re-establishment of an identity after the merger. Repeated measurement might provide additional information in this case, which might be subject to further research.

3.3 Measures

3.3.1 Dependent variable: Speed

Speed is the dependent variable in this thesis. Speed is measured as the time in months required to re-establish the group identity, e.g. the employee considers him or herself part of the group and feels he or she has formed a psychological bond (Van Leeuwen et al., 2003). To measure the speed, it is first established whether the employee identifies with T&P by the “group identification measure” , based on the group identification measure as used by Doosje et al. (1995) using four items. This identification measure is applied to both the current

(26)

26 situation and the former situation, in which the employee was not yet part of T&P and added to assess whether integration is perceived completed . Subsequently, the employee is asked to indicate at what point in time he or she felt part of the organization.

3.3.2 Independent variables: Communication

Communication has been established as an important determinant for identification with the organization (Bartels et al., 2006). Good quality of communication is important for successful identification after a merger (Bartels et al., 2006). Therefore communication as such is measured by using a 16-item scale developed by Postmes et al. (2001), which includes both the quality and quantity of communication, based on social identity theory. To investigate specifically the support of communication during the merger, questions developed by Bordia et al. (2004) about the quality of change communication are also included.

3.3.3 Independent variables: Physical distance

Distance in this research will be measured on the basis of “colocated” versus “dispersed”, in which “closeness” is measured (Mortensen et al., 2001). This principle is applied to this thesis by first measuring whether employees are internal employees or external employees. In the companies subject to this study, external employees are employed full-time at a consultancy assignment and internal employees are employed at headquarters and therefore expected to be present more frequent.

To examine closeness more thoroughly, hours spend at headquarters are also surveyed as an equivalent of being in each other’s physical presence and to test whether the degree of closeness influences the speed of the post-merger integration phase.

3.3.4 Independent variables: Informal interaction

Interaction between colleagues is contributing to social identification (Gleibs, 2010; Kiesler and Cummings, 2002). In the context of a merger, specifically, it reduces in group

(27)

27 favouritism, which increases the positive perceptions and contributes to merger success (Gleibs, 2010). The frequency of interaction has a strong relationship with social ties and the identification process (Kiesler and Cummings, 2002; Fiol et al., 2005). Frequency, here, will be measured by hours during a year an employee is having interaction with colleagues. Furthermore, the type of interaction is also measured, by asking what they discuss and who they discuss it with.

To collect data about the variables communication, physical distance, informal interaction and speed a questionnaire was used. Despite the limitations like hindsight bias, a questionnaire is perceived a suitable method to collect data as it provides anonymity, privacy and enables collection of larger data sample.

In the next chapter the data collected among employees from the companies Numerando, Match&More and Triple A, is analysed. It is tested whether the variables communication, physical distance and informal interaction impact the speed of integration. First, relationships will be confirmed by testing for correlations. Next, follow up tests are performed to find relationships regarding the degree of impact on speed of integration. Depending on the distribution and type of the data, regression analysis will be performed on continuous data, and the ANOVA or independent samples t-test on the categorical data.

(28)

28

4. Results

This paper aims to explain whether the speed of the post-merger integration phase is affected by communication, physical distance and informal interaction. In this research, the focus of attention is on the “operational combination stage”, the specific integration stage in which the integration of the operational units of the former companies are to become one (Seo et al., 2005) This stage is most focused on the integration of the employees within the companies. At this point people get to know each other, and are “forced” to work together (Seo et al., 2005).

It is expected that with a lack of communication, physical distance, and informal interaction the development of the organizations’ new identity will be slower. The research question of this thesis is “What is the effect of communication, physical distance, and informal

interaction on the speed of the post-merger integration phase?”

4.1 Analytical strategy

Sample – former Numerando and Match&More employees

The questionnaire was distributed to 56 employees. In total 37 people participated in this study, of which 15 former Numerando employees and 22 former Match&More employees. The response rate was 66%. The average years of service among former Numerando and Match&More employees was 7,67 years.

The data was first checked for missing values and seven non-completed questionnaires were removed. List wise exclusion was applied, because in these cases respondents did only open the questionnaire and not answer questions. Mergers subject to this research took place in 2011 and 2013; all employees with an employment date after 2011 and respectively 2013

(29)

29 were excluded from the dataset. A sample of 12 former Numerando and 19 former Match&More employees remain.

The total dataset consists of 29 external employees and 2 internal employees. Due to the low number of internal employees, it is decided to only analyse data about external employees. This is the final dataset to test the hypotheses.

Sample – Triple A employees

The questionnaire was distributed to 89 employees In total 35 employees participated in the survey, a response rate of 39%. Five employees did not complete the questionnaire and were removed by list wise exclusion. These five respondents only opened the questionnaire and only the first answers were completed. The answers did not provide any data to analyse. The merger between Triple A and Redmore took place in 2014, employees with employment data after December 2014 were removed from the sample. A sample of 30 Triple A employees remain.

Four employees are staff (internal employees), 26 employees are consultant. The average years of service among Triple A employees was 5.4 years.

4.2 Descriptive statistics

Former Numerando and Match&More

Next, descriptive statistics, including skewness, kurtosis and normality tests were performed on the related questions for communication, distance, informal interaction and speed of identification. The Shapiro-Wilk test was performed on items to further indicate (non-)normality of a distribution, and is often used for small datasets (< 50) (Field, 2013).

(30)

30 The 7 items of Quantity of Communication show a moderate negative skewness, ranging from -1.07 to -.07 (SE = .42). The kurtosis is ranging from -1.21 to -.20, indicating normal kurtosis. Furthermore, the Shapiro-Wilk test also confirms a non-normal distribution as all items have a p-value of p < 0.05.

The 3 items for Management Responsiveness show a moderate negative skewness, as skewness is between -1.09 and -.57 (SE = .42). The kurtosis of these items is between -.20 and .45. This indicates a slightly leptokurtic peak in distributions. Shapiro-Wilk test further confirms the non-normal distribution as all items have a p-value of p < 0.05.

The 7 items of Quality of Change Communication are normally distributed. Skewness for these items was between -0.68 and 0.015 (SE = .42). The Kurtosis is between -.53 and 1.18 (SE = .82), which indicates items are normally distributed.

Informal interaction was measured by asking questions about the number of hours

they made conversation (during a year) with colleagues within Talent&Pro and consists of one item. Skewness is 1.86 (SE = .42) and kurtosis is 2.70, indicating substantial positive skewness and a leptokurtic peak. Shapiro-Wilk test further confirms non-normal distribution with a p-value of p < 0.001.

The conversation type in hours was also measured, by questions on the number of hours per year an employee engaged in social talk, conversations about operational matters or conversations of strategic nature. These sub-items all showed skewness from 2.68 to 3.34 and a kurtosis between 6.95 and 10.54, indicating extreme positive skewness and positive kurtosis, showing a sharp peak in distribution. The distribution is non-normal.

Lastly, type of colleagues an employee interacted with was also surveyed, questioning the number of hours per year on conversations with their people manager/account manager with direct and indirect colleagues and colleagues at the consultancy assignment. Skewness

(31)

31 reports between 2.45 and 3.68 (SE = .42) and kurtosis between 4.63 and 12.52 (SE = .82) indicating extremely positive skewness and positive kurtosis, showing a sharp peak in distribution. These distributions are non-normal.

Data about distance is measured in two ways. According to Mortensen et al. (2001) measurement is about being internal or external. In this dataset 2 employees reported being an internal employee and 29 reported being an external employee. This leads to extreme skewness of -3.73 and kurtosis of 12.71. Shapiro-Wilk test was performed on this data and shows a significant p < 0.001, further confirming nonmorality of this data. Exclusion of the 2 internal employees decreases the non-normality of data, but did not make distribution normal.

Data about distance was also gathered by the question ”presence at headquarters, per year, in hours”. The hours employees were at the headquarters ranged from 0 to 1872 hours a year (M = 114.9, SD = 374.39), with skewness of 4.2 (SE = .42) and kurtosis of 17.96 (SD = .82) and the conclusion a non-normal distribution. Descriptives excluding internal employees, which usually are more often at headquarters, shows a skewness of 2.10 (SE = .43) and kurtosis of 5.21 (SE = .85). Distribution is still extremely positively skewed and leptokurtic.

Speed consists of one item. The speed of identification of participants ranged from 0 -

3 months to “I do not identify (yet) with Talent&Pro” (M = 7.62, about 20 months, SD = 4.35). Speed was normally distributed, with skewness of -.51 (SE = .43) and kurtosis of -1.31.

Descriptive statistics Triple A

The same analysis was applied to the Triple A data sample. The questions “quality of change communication” and “quantity of communication” were moderately negatively skewed, indicating non normal distribution, although kurtosis is normal. Management responsiveness was substantially negatively skewed and showed leptokurtic kurtosis. Questions of interactions, here too, consisted of one item. Both “conversation type” and “type

(32)

32 of people”, were both not normally distributed, with significant positive skewness. The questions on “distance”, were alike the T&P dataset, non-normally distributed with moderate positive skewness. Speed within the Triple A dataset was not normally distributed, with substantial leptokurtic kurtosis and extreme negative skewness. The speed of identification with Triple A ranged from 0-3 months to “I do not identify (yet) with Redmore” (M = 9.23, about 28 months, SD = 4.68). In total, 22 out of 30 employees indicated they do not identify (yet) with Redmore.

Non-normal distribution will be taken into account when testing the hypotheses.

Reliability tests

Reliability tests for Quality of change communication, management responsiveness and quantity of communication were performed. Cronbach’s alpha on all items are above 0.7, which means the scale confirms internal consistency. The Corrected Item-Total Correlation tests on all items were all above 0.3, also indicating good correlation with the total score of the scale. Table 1 and table 2 provide an overview of the Cronbach’s Alpha of both data samples. Removal of any items was not executed, as the Cronbach’s Alpha will not increase by > .10 and thus not improve correlation with the total score of the scale. On remaining items no Cronbach’s Alpha was performed, because items were not measured using scaling.

Table 1: Cronbach's Alpha - Former Numerando and Match&More

Variable Cronbach's Alpha

Q15. Quality of Change Communication .79

Q12.Management responsiveness .85

Q11. Quantity of communication .87

Table 2: Cronbach's Alpha - Triple A

Variable Cronbach's Alpha

Q15. Quality of Change Communication .94

Q12.Management responsiveness .76

(33)

33 As a final step, means in both datasets were calculated regarding the questions Quality of change communication, Management responsiveness and Quantity of communication for hypothesis testing. The variable speed was recoded as well, to match the question sequence in line with the remaining questions which are all grading from negative to positive. Speed was originally surveyed as positive to negative.

4.3 Hypothesis testing

A correlation matrix was assembled of all variables involved. In table 3, a correlation matrix on variables is provided on the former Numerando and Match&More data. In table 4, on page 34, the correlation matrix on Triple A data is shown.

The correlation matrix indicates management responsiveness positively correlates with speed, Pearson’s r(29) = .443, p = 0.016, but not with the remaining communication-related questions; quality of change communication, Pearson’s r(29) = .32, p = .07 and quantity of communication, Pearson’s r(29) = .32, p = .09.

To measure the relationship between physical distance and speed of integration, “presence at headquarters” was included in the survey. Presence at headquarters, e.g. being

Table 3: Means, Standard deviations and correlations - Former Numerando and Match&More

Variables M SD 1 2 3 4 5 6 6a 6b 6c 7 8 9 10

1. Speed 7.621 4.354

-2. Quality of change communiation 4.158 .886 .341 (.79) 3. Quantity of communication 4.228 1.310 .320 .764** (.85) 4. Mangement responsiveness 4.506 1.373 .443* .606** .594** (.87) 5. Presence at Head Quarters 22.41 22.098 -.127 .051 -.065 .018 -6. Total interaction 81.03 126.277 -.250 -.142 -.467* -.263 .205 -6a. Social interaction (what) 54.345 105.324 -.370* -.162 -474** -.298 .281 .884** -6b. Operational interaction (what) 19.655 33.594 .104 -.033 -.216 -.062 -.032 .624** .192 -6c. Strategic interaction (what) 7.035 26.932 .144 .008 -.066 .012 -.048 .454* -.007 .927**

-7. Type: interaction with PM/AC (who) 26.900 19.875 -.430* .265 .234 .258 -.073 -.220 -.155 -.240 -.126 -8. Type: contact with direct collegues (who) 20.480 45.850 -.288 -.036 -.099 -.209 .206 .338 .307 .133 .221 .222 -9. Type: contact with indirect collegues (who) 12.000 29.218 -.347 -.248 -.531** -.360 .279 .675** .760** .134 .026 -.044 .470* -10. Type: contact with collegues on consultancy

assignment (who) 108.000 337.984 -.420* -.179 -.399* -.263 .269 .708** .778** .152 .086 .063 .671** .900** -** Correl a ti on i s s i gni fi ca nt a t the 0.01 l evel (2-tai l ed)

(34)

34 close by, does not correlate with speed of integration Pearson’s r(29) = -.13, p = .19, implying distance does not impact the speed of integration.

Total Interaction is not significantly related to speed, Pearson’s r(29) = -.25, p = 0.191, but sub-item Social interaction is significant, Pearson’s r(29) = -.37, p = .04. No relationship appears on employee discussions on operational, Pearson’s r(29) = .10, p = .59 or strategic matters, Pearson’s r(29) = .14, p = .46 and speed of integration. The questions who one talks to (social interaction), a subset of questions on interactions, shows that interaction between colleagues on the consultancy assignment negatively correlates with the speed of integration, Pearson’s r(29) = -.42, p = .04.

Triple A

Correlation indicates that the quality of change communication, Pearson’s r(26) = .606, p = .001 and quantity of communication, Pearson’s r(26) = .568, p = .002 are positively related to speed, for Triple A employees.

Presence at headquarters, as a measure for distance, appeared not to be related to speed, with Pearson’s r(26) = .047, p = .82.

In the Triple A dataset, total interaction shows a positive relationship with speed, with a Pearson’s r(26) = 207, p = .008. The sub-items social interaction and operational interaction

Table 4: Means, Standard deviations and correlations - Triple A

Variables M SD 1 2 3 4 5 6 6a 6b 6c 7 8 9 10

1. Speed 4.192 4.899

-2. Quality of change communiation 4.253 1.116 .606** (.94) 3. Quantity of communication 5.008 1.141 .568** .752** (.86) 4. Mangement responsiveness 5.744 .9116 .264 .548** .569** (.76) 5. Presence at Head Quarters 477.500 475.031 .047 -.202 -.157 -.404* -6. Total interaction 524.308 497.729 .207** .367 .394* .216 .149 -6a. Social interaction (what) 102.000 77.935 .516** .335 .278 .058 .201 .795** -6b. Operational interaction (what) 345.231 374.497 .404* .291 .332 .166 .133 .934** .731** -6c. Strategic interaction (what) 77.077 164.121 .371 .291 .305 .248 .053 .524** .267 .203

-7. Type: interaction with PM/AC (who) 6.404 11.747 .268 .227 .277 .212 -.203 .460* .102 .220 .845** -8. Type: contact with direct collegues (who) 40.192 33.614 .612** .516** .402* .352 .137 .581** .596** .580** .156 .011 -9. Type: contact with indirect collegues (who) 17.269 12.873 .216 .210 .348 .378 .142 .612** .517** .531** .398* .197 .347 -10.

Type: contact with collegues on consultancy

assignment (who) 60.192 155.786 -.081 .311 .103 .038 -.161 -.072 -.017 -.062 -.069 -.108 -.091 .107 -** Correl a ti on i s s i gni fi ca nt a t the 0.01 l evel (2-tai l ed)

(35)

35 appear to explain this positive relationship, with Pearson’s r(26) = .516, p = .007 and Pearson’s r(26) = 404, p = .041 respectively. Operational interaction shows a lower significance than social interaction.

Who one talks to, is correlated as well. Within Triple A, the contact with direct colleagues shows a significant positive relationship, with Pearson’s r(26) = .612, p = .001.

Communication

Communication has been identified as a construct to influence the perceptions of employees (Bordia et al., 2004; Kernan and Hanges, 2002; Bartels et al., 2006). In this study, focus is on the impact of communication climate in general (Bartels et al., 2006) and the quality of change communication (Bordia et al., 2004). Both have been identified as contributors to identification after a merger (Bartels et al., 2006) and are expected to positively impact the speed of the post-merger integration phase.

Hypothesis 1: communication positively influences the speed of the post-merger

integration phase.

As shown in the correlation matrix on page 33, Management responsiveness positively correlates with speed of integration. The independent variables Quantity of communication and Quality of change communication do not show a relationship with speed.

To further analyse the relationship between management responsiveness and speed, a simple linear regression is performed. It provides more information about the causality of the relationship.

During the analysis of the descriptive statistics it was established the variables are not normally distributed. With moderate or larger sample size, the test is expected to give

(36)

36 genuinely accurate p-values despite non-normality (Green and Salkind, 2002). With smaller sample sizes, like this sample size (N = 29), the p values may be invalid (Green and Salkind, 2002). According to Field (2013) the sample size should at least be 55 when a regression with medium effect is run. Unfortunately, this dataset consists of 29 samples. To correct for the normality assumption violation and small sample size, bootstrapping is applied to this sample. Table 5 shows the outcome on the regression analysis.

The variation of speed can be explained by management responsiveness, with F(1,27) = 6.61; p = 0.016, indicating the statistical significance of this regression model. The R2 of the overall model is 0.197, explaining 20% of the speed is related to management responsiveness. If management responsiveness increases by one point, speed of integration increases with 1.41.

Triple A

Preliminary analysis shows a positive relationship between quantity of communication and quality of change communication. To further analyse the relationship and causality, a linear regression is executed. Table 6 provides an overview of the outcomes of the test.

The outcome of the regression analysis, with bootstrap (5000 bootstrap resamples) to correct for the normality assumption violation and small sample size, shows no statistical

Table 5 - Testing the effect of management responsiveness on speed

Variable R R2 R2 change B Std. Error P Lower Upper

Model .443 .197 .167 -4.154 5.831

Management responsiveness 1.406 .542 .011 .397 2.560

Note: N=29. BCa: bias corrected and accelerated; 5000 bootstrap resample’s

Bca 95% CI

Table 6 - Testing the effect of communication variables on speed

Variable R R2 R2 change B Std. Error P Lower Upper

Model .630 .397 .344 .003

Quantity of communication 1.111 .926 .132 -.165 3.664

Quality of change communication 1.807 1.119 .102 -.505 3.923

Note: N=26. BCa: bias corrected and accelerated; 5000 bootstrap resample’s

(37)

37 significance. The overall model was statistically significant, with F(2,23) = 7.557, p = .003. However, the coefficients for both quantity of communication (p = .132) and quality of change communication (p = .102) show no statistical significant linear dependence of the means from quality of change communication and quantity of communication on speed of identification was found.

Distance

In previous research dispersed teams have shown weaker sense of identity and were more susceptive to out group feelings (Mortensen et al., 2002). Dispersed employees are less often at headquarters within Talent&Pro than their colocated co-workers, which was explained to lead to less adequate perceptions about their co-workers and less positive group perceptions (Mortensen et al., 2001). It is therefore analysed whether employees within dispersed teams, such as consultants working within a consultancy firm, will identify slower with the organization after a merger.

Hypothesis 2: The more distant employees are from HQ, the slower the post-merger

integration phase evolves.

Correlation already provided the information that “Presence at headquarters” is not correlated to speed of integration Pearson’s r(29) = .127, p = .510, indicating hypothesis 2 has to be rejected.

Distance was also measured by binary distinction; colocated versus dispersed (Mortensen et al., 2001). This is similar to the difference of internal versus external employees. However, due to the limited number of internal employees in the sample (N = 2), these were excluded from the final dataset. Although this sample is too small to make an actual inference about the outcome, for the purpose of the hypothesis an attempt is made by analysing data including the internal employees.

(38)

38 A one-way ANOVA is performed, however bootstrapping to correct for the normality assumption violation and the small sample size is not applied. Field (2013) explains bootstrap only is applied to confidence intervals around the mean and differences between the mean. It will not bootstrap the main test and it is advised not to bootstrap this test at all, especially with a small dataset (Field, 2013). The performed ANOVA for the variable “internal/external” does not show a significant effect of being an internal or external employee on the speed of integration, F(1,29) = 1.29, p = .26.

With no overall significance on either presence at headquarters or collocation (Mortensen et al., 2001), hypothesis 2 is (totally) rejected.

Triple A

Within the Triple A dataset, presence at headquarters was not correlated with speed either, Pearson’s r(26) = .047, p = .82. Indicating Hypothesis 2 must be rejected here, as well.

Like the dataset on Talent&Pro data, the number of staff employees was limited (N = 4). To be able to compare data, the staff employees were also removed from this dataset. In line with Talent&Pro dataset, the data will now be analysed including staff employees.

A one-way ANOVA is performed, without bootstrapping. The performed ANOVA for the variable “staff/consultant” does not show a significant effect of being an internal or external employee on the speed of integration, F(1,28) = 1.65, p = .210.

With no overall significance on either presence at headquarters or collocation (Mortensen et al., 2001), hypothesis 2 is totally rejected from Triple A perspective as well.

Frequency of interaction

Within the context of a merger, interaction among people decreases in-group favouritism, and increases positive perceptions and support for the merger (Gleibs, 2010). The

What is the effect of communication, physical distance and informal interaction on the speed of the post-merger integration phase?

Thesis