Do you feel the connection? : a quantitative research on the effect of social identification in distributive negotiations

(1)

Master thesis

MSc Business Administration – Track Marketing

Do You Feel the Connection?

A Quantitative Research on the Effect of Social Identification in

Distributive Negotiations

Supervisor: Dr. A. Zerres

By: Nicoline Russel - 11141964 Date: 23-06-2016

(2)

2

Statement of originality

This document is written by Student Nicoline Russel 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

Acknowledgement

I would like to express my sincere gratitude to my supervisor Dr. A. Zerres. The completion of this master thesis could not have been possible without his support, advice and input. Additionally, I would like to thank all respondents who have filled in my experimental survey. Without them, I could not have finished my master thesis. Last but not least, I am most grateful to my family, who gave me the opportunity to finish my master’s degree. They have always supported me in my choices and were willing to help me whenever necessary. A big thanks to all of you!

(4)

4

Abstract

The present research investigated the effect of social identification on the intended bargaining strategy in distributive negotiations through an online experiment. As expected, a high level of social identification leads to a softline bargaining strategy, whereas a low level of social identification leads to a hardline bargaining strategy. This effect is partially meditated by negotiators’ bargaining goals they aim to achieve during the negotiation, which can either incline towards economic or socioemotional goals. Moreover, the moderator social value orientation is included to test whether this would influence the relationship between social identification and the bargaining strategy. Theoretical background has shown that prosocials aim for achieving equal outcomes and tend to behave cooperatively during negotiations. Proselves on the other hand tend to maximize their own outcome and tend to behave in a competitive way. Contrary to the predictions, social value orientation does not moderate the relationship between social identification and the intended bargaining strategy through the mediating presence of bargaining goals. To investigate these hypotheses, an online experiment is executed in which the level of social identification among participants is being manipulated. Based on the reported findings, needs for future research and theory building are identified and discussed.

Keywords: social identification; distributive negotiations; bargaining goals; economic goals; socioemotional goals; bargaining strategy; hardline strategy; softline strategy; social value orientation

(7)

7

1. Introduction

A distributive negotiation is a mutual decision-making process in which people decide how to allocate scarce resources. Several contradicting theories exist on how to negotiate most successfully, incorporating both economic and psychological factors as determinants (Kramer, Pommerenke, & Newton, 1993). One approach suggests a strict and competitive bargaining style to minimize own losses and to maximize economic outcomes, whereas the other suggests a cooperative style aiming for more socio-emotional outcomes. Several negotiation goals that individuals aim to achieve during the decision making process can be distinguished. In this study, two broad negotiation goals are defined with different bargaining strategies as a consequence. On the one hand economic goals lead to higher economic outcomes, while on the other, socioemotional goals lead to more social psychological outcomes (Hüffmeier, Freund, Zerres, Backhaus, & Hertel, 2014).

When negotiators aim to achieve socio-emotional goals during the bargaining process, it is plausible that they are more likely to act in a cooperative manner. They may not be merely concerned about their own outcome during the negotiation but also about the outcome obtained by their counterpart (Kramer et al., 1993). A softline bargaining strategy, also the inclination towards a cooperative bargaining style, is characterized as a moderate size of first offers and a frequent employment of substantive concessions. However, when negotiators wish to accomplish economic goals, it may be more likely for them act in a competitive way. When competitive goals are preferred during the bargaining process, negotiators are merely concerned about maximizing their own payoff. A hardline bargaining strategy, or the inclination towards of competitive bargaining style, is characterized as more aggressive offers and a minimum number and/or small sizes of concessions. Although a hardline strategy may lead to higher economical outcomes, a softline strategy is more likely to contribute to a positive relationship between both parties (Hüffmeier et al., 2014). Barley (1991) stated that:

(8)

8 “In daily life, persons are almost always members of groups whose values and beliefs shape their behavior and cognition. People typically dispute and bargain as members of families, communities, cliques, and organizations, not as isolated actors whose judgment is unfettered by social relationships" (as cited in Kramer et al., 1993, pp. 635). Therefore, social relationships between negotiating parties have an impact on the goals they wish to achieve during negotiations.

Several factors influence judgment and decision-making during negotiations. What defines which goals and as a consequence which bargaining strategy will be performed during distributive negotiations? One factor that may define the bargaining strategy pursued during a negotiation is the level of social identification one has with its opposing party. Social identification is as a sense of connectedness between an individual and an organization or between two individuals and is associated with one’s evaluative and emotional relationship with the group (Ahearne, Bhattacharya, & Gruen, 2005). The salience of a shared social identity influences how negotiators explain their interdependence with other negotiators. When a party feels socially identified with its opposing party during a negotiation, it may be more concerned about the outcome the other party obtains. The salience of a social identity during negotiations leads to a decreased focus on oneself. This enhances socio-emotional goals and a cooperative bargaining style during negotiations while diminishing economic goals and competitive approaches. Social identification can therefore influence judgment and decision-making, subsequently affecting negotiation processes and outcomes that decision makers try to achieve during negotiations (Kramer et al., 1993).

The effect of social identification on the bargaining strategy through bargaining goals might be dependent on an individual’s social value orientation, which can be an altruistic, cooperative, individual or competitive orientation. The former two orientations are usually combined to form a group of people with essentially prosocial orientations, concerned with

(9)

9 outcomes for both others and the self. The latter two orientations are primarily concerned with outcomes for the self, referred to as proselves1. Social value orientation is shaped during an individual’s lifetime, however social interactions, such as the behavior of an interdependent person involved, also play a role (Van Lange, De Bruin, Otten, & Joireman, 1997). Will a negotiators’ social value orientation influence the relationship between social identification and the bargaining strategy through bargaining goals? In other words: are prosocials more likely to be influenced by a high level of social identification than proselves? This study investigates whether social value orientation moderates the relationship between social identification and bargaining strategy, through bargaining goals as a mediator in distributive negotiations. Therefore the following research question can be formulated:

What is the effect of social identification on the intended bargaining strategy through bargaining goals in distributive negotiations, and does social value orientation influence this relationship?

In this study participants are confronted with a scenario in an online experiment, in which they negotiate a price for which they would sell their old mobile phone to their colleague. Participants, who have been manipulated by feeling either a low or a high level of social identification towards their opponent, are asked several questions about their bargaining goals and their intended2 strategy when selling their mobile phone. Also participants’ social value orientation is measured, serving as a moderator in this study. Based on this outcome the variables can be measured and the developed hypotheses can be tested.

1_{In the remaining of this study I continue to refer to the combination of social value orientations of either}

prosocial or proself

2_{As the hypotheses are tested with an online experiment, I cannot argue that participants will perform their}

intended bargaining strategy in real-life either as it is a prediction of their actual behavior. However, to avoid confusion the terms intended bargaining strategy and bargaining strategy are used interchangeably in this research.

(10)

10

Theoretical and practical relevance

Execution of this research will result in both theoretical and managerial contributions. The theoretical contribution of this study is that it adds to existing literature by investigating whether social identification influences the desired goals negotiators aim to achieve during the decision making process and whether this impacts their intended bargaining strategy. Although a considerable amount of research has been done on either the concept of social identification as well as on distributive negotiations, few of them analyzed the relationship between these two concepts. This research contributes to the current literature in the following ways. First, it investigates whether social identification influences the bargaining strategy during the negotiation process. Where Kramer et al. (1993) have focused on the effect of social identification on negotiators’ judgement and decision making, this study looks at the effect of social identification on the bargaining strategy through the mediating presence of bargaining goals in distributive negotiations. Second, this study investigates if the effect of social identification on the negotiators’ bargaining strategy through bargaining goals depends on its social value orientation. Although many studies have researched the concept of social identification, none of them investigated this concept when an individuals’ social value orientation is taken into account. Lastly, where some other studies have studied the impact of certain bargaining strategies on negotiation outcomes (e.g. Hüffmeier et al., 2014), this study does not focus on the outcome per se, but instead focuses on whether these bargaining strategies can be influenced by the concept of social identification.

This study is of practical value and relevance for firms and managers, as it can be very useful to know the extent to which social identification really influences the desired bargaining goals negotiators aim to achieve and whether a hard- or softline strategy will be exerted during the negotiation process. When it turns out that social identification indeed causes socioemotional negotiation goals and the performance of a softline bargaining strategy

(11)

11 as a consequence, managers should make sure they make advantage of this. Companies can manipulate and adapt the negotiation strategy employed by the opposing party in a beneficial way. When negotiators aim for a cooperative negotiation process, which is most often the case, they can make sure the opposing party is likely to identify itself with the focal party. The opening offer of its counterpart might be lower and the negotiation might result in a cooperative process, where frequent and substantive concessions are being done. Also when the focal party itself is very competitive and has high economic goals, it would rather have the opponent in a cooperative mood as it might be more willing to accept relatively high offers the focal party is doing. The focal party can make sure the opposing party feels a sense of connectedness, for example by building an emotional relationship with its counterpart. Trying to develop a bond together for example by sharing the same interests, norms and values can help by building this social identification. However, if the focal party is very competitive, has high economic goals and does not care about the relationship with its counterpart, it would rather wish the opponent to act competitively either to gain short term economic success. In this case the focal party should make sure the level of social identification from the opponent is not that high. When knowing in advance what the opponent’s social value orientation is, either prosocial or proself, this manipulation technique may even be targeted better. Employees should be briefed beforehand on how to manipulate the level of social identification of its counterpart and how to make use of this during the negotiation process in the most beneficial way. When making use of this negotiation tactic, firms may reach higher negotiation outcomes, eventually making them more profitable.

Thesis overview

This research is structured in the following chapters. First a review of the literature is given and the key concepts social identification, bargaining goals and bargaining strategies in distributive negotiations are described, as well as the concept of social value orientation.

(12)

12 Subsequently, the research design and methodology are proposed to explain how the research question is investigated. The results of the research will then be presented and analyzed in the next chapter. Finally the findings will be discussed after which the limitations, contributions and suggestions for future research are given.

2. Literature review

This chapter provides a comprehensive overview of the existing literature to explain the effect of social identification on the intended bargaining strategy in distributive negotiations, including the moderation and mediation effects. It will also explain the key concepts and the relationships studied in this research. First the term social identification is explained, followed by how social identification can lead to two different bargaining goals in distributive negotiations, namely socioemotional and economic goals. Then the two intended bargaining strategies, soft- and hardline strategies are explained and the main effect of social identification on these bargaining strategies in distributive negotiations is proposed. Lastly, the mediation effect of the bargaining goals and the moderation effect of social value orientation are explained.

2.1.Social identification in distributive negotiations

Social identity theory is a social-psychological theory that explains cognitions and interpersonal behavior in terms of individuals’ social group memberships (Kramer et al., 1993). A topic within social identity theory is the construct of social identification. According to Ashfort and Mael (1989, pp. 20) social identification: “(a) is a perception of oneness with a group of persons; (b) stems from the categorization of individuals, the distinctiveness and prestige of the group, the salience of outgroups, and the factors that traditionally are associated with group formation; and (c) leads to activities that are congruent with the identity, support for institutions that embody the identity, stereotypical perceptions of self and

(13)

13 others, and outcomes that traditionally are associated with group formation, and it reinforces the antecedents of identification”. Social identification is an individuals’ relationship towards a group as a perceived entity, which has a range of characteristics associated with its membership (Postmes, Haslam, & Jans, 2013). Identification with a group is similar to identification with an individual, like a person’s football hero, or a reciprocal role relationship, like mother and daughter. Social identification is not necessarily associated with specific behaviors or affective states, as an individual does not need to expend effort toward the goals of the group. For an individual to feel socially identified it only needs to perceive him- or herself as connected to an individual or a group (Ashfort & Mael, 1989).

In short, social identification is the perception of belonging to a group, with the result that a person identifies with that group (Bhattacharya, Rao, & Glynn, 1995). According to Postmes et al. (2013) social identification is associated with one’s evaluative and emotional relationship with an individual or group. They capture social identification by the dimension of self-investment, which is closely related to three inter-correlated components, namely satisfaction, solidarity and centrality. The first component, satisfaction, is the degree a party feels satisfied with and has positive emotions and feelings towards the other group members. The second component, solidarity, is the degree of solidarity a party has with its group members, which is about having a strong attachment by feeling bonded and committed to the group. The last related component is centrality, which is the evaluation of the group as being more or less important to one’s sense of self. Postmes et al. (2013, pp. 599) provide the definition of social identification as being the “positive emotional valuation of the relationship between self and in-group’’. Individuals’ membership in groups may have serious implications for their experience and behavior (Leach et al., 2008).

Social identification is an important psychological process that may provide a basis for solving social dilemmas (De Cremer & Van Vugt, 1999). An example of these social

(14)

14 dilemmas is distributive negotiations, which mostly occur in the context of preexisting social ties and relations. Negotiations frequently involve decision makers who share membership in the same social group or organization (Kramer et al., 1993). Shared group identities or social identification have been shown to provide a basis for group behavior in a wide range of contexts and many different forms. Decision makers use their level of social identification as a reference point for their decision behavior. Highly socially identified individuals have been shown to invest more in public good dilemmas and exercise greater restraint in resource dilemmas than individuals who identify less with their group (De Cremer & Van Vugt, 1999). It is assumed that when a social identity is salient, an individuals’ personal identity tends to recede in importance, meaning both concepts are inversely related. For instance, when personal identity is salient, social identification tends to decease, resulting in individualistic and self-oriented emphases, whereas the salience of a social identity causes more concerns about the outcome of other group members (Kramer et al., 1993). The reason that social identification influences individuals’ attitude within distributive negotiations might be due to the fact that an increased level of social identification reduces the psychological distance between two parties so that they perceive each other as similar in terms of their goals and achievements. People may also become more motivated to achieve positive outcomes for their group rather than for themselves, which results in greater contributions to more equal outcomes (De Cremer & Van Vugt, 1999). Social identification is therefore likely to influence negotiators’ judgement and decision-making during negotiations, which may have an impact on negotiation processes and outcomes (Kramer et al., 1993).

2.2.The effect of social identification on negotiators’ bargaining goals

Distributive negotiation is a mixed motive task in which parties must cooperate to reach an agreement while also competing to fulfill their personal interests (Weingart, Brett, Olekalns, & Smith, 2007). Negotiation outcomes fall into two broad classes according to

(15)

15 Thompson (1990), namely economic and social psychological.Economic outcomes focus on the explicit terms and outcomes of the negotiation, for instance whether an agreement is reached, how much value or joint benefit is created and how resources are divided or claimed by both parties. Besides economic factors, people also value social psychological outcomes in negotiation. Social psychological measures do not solely focus on the explicit outcome of the bargaining process but also take the processes of the negotiation into consideration, which are based on the negotiator’s social perception (Thompson, 1990). There are several psychological factors that people value as the consequences of negotiations, which can be grouped into four broad factors according to Curhan, Elfenbein, and Xu (2006, pp. 507), namely: “feelings about the instrumental outcome (e.g., outcome satisfaction and distributional fairness), feelings about the self (e.g., saving face and living up to one’s own standards), feelings about the negotiation process (e.g., fairness and voice), and feelings about the relationship (good impressions and a solid foundation for the future’’.

To define the negotiator’s bargaining goals one should look at economic as well as social psychological factors, as these are different goals they aim to achieve in the domain of distributive negotiations. On the one extreme, decision makers may be inclined to focus primarily on economic outcomes and aim for their self-interest by trying to maximize their own expected utility during the bargaining process. When negotiators merely strive to maximize their own outcome they aim to focus on achieving economic goals. On the other extreme, negotiators can be concerned with other factors than only their economic outcome and also value social psychological factors, as a consequence of the bargaining process. When negotiators value social psychological outcomes during the negotiation, in particular feelings about the relationship, they aim to achieve socioemotional goals3 (Kramer et al., 1993).

3

In the remaining of this research I continue to refer to socioemotional and economic bargaining goals as this terminology is predominantly used in other research and thus tends to be more intuitively understandable.

(16)

16 The extent to which negotiators value either economic or socioemotional bargaining goals may be influenced by the level of social identification a party has with its counterpart during the decision making process. According to Ashfort and Mael (1989), social identification is seen as personally experiencing the successes and failures of the group. It seems that social identification is often present in situations where loss or suffering can occur, or when there is a chance on an expected failure, for example during distributive negotiations. The extent to which negotiators identify with the opposing party can be used as a reference point for their decision behavior. The level of social identification might have an influence on the bargaining goals decision makers prefer during the negotiation process. When negotiators experience the salience of a common or shared social identity this may heighten their concern about the other party’s outcomes. Negotiators may value social psychological outcomes as a consequence of the bargaining process and aim to achieve socioemotional goals. They may be concerned with either their own outcome as well as the outcome obtained by their counterpart and thus gain more importance to its future relationship. Social identification may lead to a more prosocial situational motivation, and negotiators may have a preference for securing socioemotional outcomes. The party that feels socially identified may prefer relatively more equal outcomes for both parties and will not primarily aim for its own economic outcomes. However, when no shared identity is salient negotiators may care less about the relationship with its counterpart. It is likely that the negotiator primarily values economic goals during the negotiation and cares less about the socioemotional goals (Kramer et al., 1993). Therefore, it is assumed that a high level of social identification causes the socially identified negotiator having socioemotional goals, whereas a low level of social identification causes negotiators aiming for economic goals during the bargaining process (Kramer et al., 1993).

(17)

17

2.2.1 The effect of social identification on socioemotional goals

Negotiators who strongly identify with their group have been shown to invest more in the collective outcome than low-identifying negotiators, who are merely concerned about their own outcome (De Cremer & Van Vugt, 1999). Negotiators who feel a high level of social identification towards its counterpart are likely to have socioemotional rather than economic goals and value social psychological outcomes as a consequence of the negotiation process. Social identification may blur the distinction between people’s personal outcome and the outcome of others. A high level of social identification reduces the psychological distance between the negotiating parties so that they perceive each other similarly in terms of their goals and achievements. Consequently, attraction among group members increases and people will become motivated to achieve socioemotional rather than economic goals. Parties may focus more on their future relationship and may reach agreements where they share a more equal division of resources, which results in greater efforts to the collective outcome of both parties (De Cremer & Van Vugt, 1999).

Hypothesis 1: A high level of social identification is positively related to socioemotional

goals compared to a low level of social identification in distributive negotiations.

2.2.2. The effect of social identification on economic goals

When a party experiences social identification with its counterpart it may be less concerned about its economic goals during the negotiation process. Negotiators will be less concerned by the outcome they obtain themselves and do not strive to maximize their economic goals. However, when negotiators perceive a low level of social identification and a personal rather than a social identity is salient, negotiators may strive for more individualistic and relatively self-interested orientations. Negotiators may be less likely to express their concern about the outcome the other party obtains and are less concerned about

(18)

18 the fairness of the other party’s outcome during the bargaining process. They will focus their attention primarily on their own economic outcomes and care less about the relationship with its counterpart (Kramer et al., 1993). People who have an economic orientation within negotiations tend to keep valuable information to themselves. Rather than supporting each other, people are motivated to impair the progress of others in an effort to gain positive advantage (Beersma et al., 2003).

Hypothesis 2: A low level of social identification is positively related to economic goals

compared to a high level of social identification in distributive negotiations.

2.3. Bargaining strategy

The bargaining strategy that is performed during negotiations can play a major impact on the outcome that will be reached. The bargaining strategy of a negotiator within distributive negotiations consists of a first offer and the degree to which the negotiator diverges from this offer during the negotiation process. In distributive negotiations, parties can incline to the performance of either a cooperative or a competitive bargaining strategy to reach an agreement. Two opposing theories exist on how to negotiate most successfully in the domain of distributive negotiations. The aspiration theory suggests a competitive bargaining style to gain the best outcome, whereas the tension-reduction model recommends a cooperative bargaining style where concessions need to be made in the beginning of the negotiation. When cooperating, there is a tendency to veer towards a softline strategy, which intends to evoke concessions from the opposite party by employing own concessions. When competing, negotiators tend to perform a hardline bargaining strategy, which is about making extreme first offers and minimize one’s own concessions (Hüffmeier et al., 2014). Although contradicting theories exist on how to bargain for reaching the best outcome, Hüffmeier et al. (2014) argue that softline bargaining strategies lead to higher socioemotional outcomes, whereas hardline bargaining strategies subsequently lead to higher economic outcomes. In the

(19)

19 following sections both bargaining strategies are explained further, as the focus of this research will be on either of these strategies.

2.3.1 Softline bargaining strategy

Hüffmeier et al. (2014, pp. 869) define softline bargaining as “strategies entailing any element or combination of moderate first offers, frequent concessions, substantive concessions, and gradually decreasing concessions.” The main goal of a cooperative bargaining style, as well as the inclination towards a softline strategy during distributive negotiations, is to create a cooperative context in which there is room for integrative bargaining. Amongst other things, this includes the sharing of information and jointly discussed goals. A cooperative bargaining strategy is characterized by a high amount of concessions employed, which is intended to demonstrate responsiveness to the counterpart’s needs (Allen, Donohue, & Stewart, 1990). Cooperative bargainers engage in problem-solving behavior as a way to satisfy both parties’ interests (Barry & Oliver, 1996).

A softline bargaining strategy has several characteristics. Softline strategists tend to give substantive, frequent and meaningful concessions, which are likely to induce more concessions from the opposing party. Concessions are gradually decreased rather than increased during the bargaining process. Moreover, the size of the opening offer is lower when compared to the hardline strategy. With these bargaining tactics the negotiators reduce tension and show commitment to the bargaining process (Hüffmeier et al., 2014). It should be noted however, that the opponent’s perceived reasons for these substantive and frequent employment of concessions depend on whether its counterpart will reciprocate the concessions. If the opponent party indeed perceives it as a sign of cooperation then the concessions are more likely to be reciprocated. The negotiator should not imply that the concessions are a sign of weakness but rather that they are based on an expectation that these will be matched by the other party or the concessions will not be reciprocated by the

(20)

20 opponent. These bargaining tactics can serve as a foundation for building a mutually trusting and long term, cooperative relationship with the other party. According to the cooperative bargainers, this strategy leads to the most beneficial outcome as it changes the bargaining relationship and the negotiation process (Allen et al., 1990).

2.3.2. Hardline bargaining strategy

Hardline bargaining is defined as “any element or combination of aggressive first offers, a minimum number of concessions, a minimum size of concessions, and gradually increasing rather than decreasing concessions” (Hüffmeier et al., 2014, pp. 869). A competitive bargaining style, or the inclination towards a hardline bargaining strategy, is about emphasizing competitive rather than collaborative goals during distributive negotiations. The main strategy within a competitive bargaining context is to minimize the total amount of loss during the bargaining process. The negotiators usually make extreme first offers to increase the likelihood of obtaining the most favorable outcome in the end. These extreme initial offers give the opposing party a comparison point from which it will judge the final negotiation outcome, working out to be favorable for the negotiator. Competitive negotiators will minimize their number and size of concessions and will not reciprocate every concession of the opponent party. Instead, competitive bargainers try to persuade their opponents to make concessions (Barry & Oliver, 1996). For example, this can be done by gradually increasing rather than decreasing concessions during the bargaining process (Hüffmeier et al., 2014). The negotiator maintains a high level of aspiration and its competitive behavior is very inflexible aimed at forcing concessions from the other party. Hardline strategists can also give frequent but very small concessions, giving the appearance of flexibility and openness when there is actually very little value (Clopton, 1984).

(21)

21

2.4. The effect of social identification on the intended bargaining strategy

It is assumed that a high level of social identification encourages negotiators to perform a soft- rather than a hardline bargaining strategy. When a party feels socially identified with its opponent, it may rather prefer to create a cooperative rather than a competitive bargaining context. Instead of only its own outcome, the socially identified negotiator may also be concerned about the outcome of its counterpart. The sense of connectedness a party feels with its counterpart may cause that party to do a moderate first offer4 and to employ a higher amount and a greater size of concessions. As the bargaining strategy consists out of three separate variables, namely the first offer, the amount and the size of concessions5, the third hypothesis is divided into three sub hypotheses.

Hypothesis 3. A high level of social identification is more positively related to a softline

bargaining strategy than a low level of social identification.

Hypothesis 3a. A high level of social identification is positively related to lower6 first offers than a low level of social identification

Hypothesis 3b. A high level of social identification is positively related to a greater

amount of concessions than a low level of social identification

Hypothesis 3c. A high level of social identification is positively related to a greater size of

concessions than a low level of social identification

2.5. Mediating effect of bargaining goals

Goals are well known to influence behaviors as they are clearly related to people’s responses across a variety of situations. Negotiation goals can have a tremendous effect on people’s actions, behaviors and performances (Sanna, Parks, & Chang, 2003). Also Deutsch (as cited in Beersma et al., 2003) has argued a long time ago that individuals’ beliefs about

5

Also the gradually increasing or decreasing of concessions is part of the bargaining strategy as discussed before. This is not taken into account in the proposed hypotheses. However, it is incorporated in the analysis of the results.

6_{In this research this effect is investigated with the sale of a product, so a lower first offer means in this case a}

moderate and not an aggressive first offers, meaning the push towards a softline strategy. We continue to refer to lower rather than moderate first offers to make it more intuitively understandable

(22)

22 how their goals are related, determine the way in which they interact. This in turn affects their performance and the group’s cohesiveness during social interactions. People differ in the way in which they approach interdependent others. Some people are inclined to approach their counterpart cooperatively, whereas others behave in a less cooperative and a more competitive manner (Van Lange et al., 1997). Therefore bargaining goals, which may either incline towards socioemotional or economic goals, may mediate the relationship between social identification and negotiators’ bargaining strategy during distributive negotiations. The bargaining strategy is the main driver of a dynamic and interdependent bargaining process (Hüffmeier et al., 2014).

It is plausible that a high level of social identification induce negotiators having socioemotional goals that lead to the performance of a softline bargaining strategy. A low level of social identification may lead to economic goals negotiators aim to achieve, causing negotiators to behave in a more competitive manner and engage in a hardline bargaining strategy. Therefore it is assumed that bargaining goals mediate the effect of social identification and the bargaining strategy, which may push towards a soft- rather than a hardline bargaining strategy during distributive negotiations.

Hypothesis 4: Bargaining goals mediate the effect of social identification and the intended

bargaining strategy such that a high level of social identification leads to asoftline bargaining strategy through socioemotional bargaining goals compared to a low level of social

identification.

As mentioned before, an individual’s bargaining strategy consists of a first offer, a certain amount of concessions, and the size of concessions. Therefore the fourth hypothesis can be divided into the following three sub hypotheses.

Hypothesis 4a: Bargaining goals mediate the effect of social identification and a first offer

such that a high level of social identification leads to a lower first offer through socioemotional bargaining goals compared to a low level of social identification.

(23)

23

Hypothesis 4b: Bargaining goals mediate the effect of social identification and the amount of

concessions such that a high level of social identification leads to a higher amount of concessions through socioemotional bargaining goals compared to a low level of social identification.

Hypothesis 4c: Bargaining goals mediate the effect of social identification and the size of

concessions such that a high level of social identification leads to a greater size of concessions through socioemotional bargaining goals compared to a low level of social identification.

2.6. Moderating effect of social value orientation

Several studies have indicated that people behave differently in the manner they approach interdependent others. Where one person acts very cooperatively to an interdependent other, another may approach someone else in a more competitive way. The differences in the way individuals prefer certain patterns of outcomes for oneself and others is related to the concept of social value orientation (Van Lange et al., 1997). Social value orientation is related to behavioral differences among people in situations where social interaction plays a role. Van Lange et al. (1997, pp. 733) define social value orientation as “stable preferences for certain patterns and outcomes of oneself and others.” They distinguish three types of orientations: prosocial, individualistic and competitive orientations. Prosocials tend to maximize the outcome for either themselves as well as the other and to minimize differences between outcomes of both parties. Individualists tend to maximize their own outcomes with less or no attention for the outcome of the opponent’s party. Lastly the competitors, who aim to maximize their own outcomes relative to its opponent’s outcome and seek relative advantage over the other party. In this study individualists and competitors are combined and are called proselves, which ultimately results in two types of social value orientations: prosocials and proselves. Where prosocials behave in cooperative manner, proselves tend to maximize their own gain and act in a competitive way. Social value orientations are predictive of behavior in several social dilemma situations. Examples of this are several behavioral situations such as judgements of everyday life incidents of cooperation and competition, decisions and

(24)

24 judgements regarding commuting choices, and willingness to sacrifice in relationships (Van Lange et al., 1997). In this study the focus is on individuals’ (bargaining) behavior during distributive negotiations.

Van Lange et al. (1997) investigated where these social value orientations come from and which factors shape the orientations of individuals. The study found that different orientations are partially rooted in different patterns of social interaction as experienced during individuals’ early childhood to their young adulthood. Orientations are further shaped by different patterns of social interaction as experienced further along in the individuals’ lives. For example another person’s behavior might affect the tendency in which an individual behaves, either cooperatively or competitively. Thus, social interaction experiences within a situation where two or more people are involved form the basis for the development of relatively stable social value orientations of individuals (Van Lange et al., 1997). As an individual’s social value orientation is partially shaped by different patterns of social interaction, it may be likely that social value orientation influences a negotiator’s behavior during the negotiation process. As prosocials are known by their cooperative behavior, they may incline towards the performance of a softline rather than a hardline bargaining strategy during negotiations. Proselves, who prefer competition over cooperation, will incline towards the performance of a hardline rather than a softline bargaining strategy. As discussed earlier, social identification may influence the individual’s intended bargaining strategy through the mediating presence of bargaining goals during distributive negotiations. An individuals’ behavior with interdependent others is related to its social value orientation, which is shaped by different patterns of social interaction during its life. Therefore it is plausible that the impact of social identification on the bargaining strategy is depended on an individuals’ social value orientation. A prosocial and a proself may respond differently to their level of social identification he or she perceives towards its counterpart. It is assumed

(25)

25 that prosocials tend to incline towards more socioemotional bargaining goals, and perform a softline bargaining strategy when perceiving a high level of social identification, whereas for proselves this effect may be less strong as they care less about this identification. It is assumed that social value orientation moderates the effect of social identification on the intended bargaining strategy through bargaining goals. It is expected that a high level of social identification leads to a softline bargaining strategy and that this effect is stronger for prosocials than for proselves. Based on these assumptions, a fifth hypothesis can be formulated.

Hypothesis 5: Social value orientation moderates the effect of social identification on

bargaining strategy through bargaining goals during distributive negotiations such that the effect of social identification on a softline bargaining strategy through socioemotional bargaining goals is stronger for prosocials than for proselves

Figure 1 provides an overview of the conceptual framework of this study. This model has been developed to test the proposed hypotheses.

(26)

26

3. Research design and methodology

To be able to test the relationship between social identification and the intended bargaining strategy, with bargaining goals as a mediation and social value orientation as a moderation effect, quantitative data is collected in order to test the proposed hypotheses. This chapter provides a description about the data collection and the research methods and techniques that are used to answer the research question and the proposed hypotheses. First the sample of this study is illustrated. Subsequently the research design is described which includes a description of the experimental survey and the measurement variables. Finally the procedure of the data collection is clarified.

3.1.Sample

The population of this study are Dutch individuals who can be confronted with distributive negotiations in their lives, which may actually be everyone who is able to communicate properly. Examples of negotiations can be a salary negotiation between employer and employee or a negotiation about a buyer and a seller in any kind of occasion. Individuals between 20 and 65 years are the testing population, as this age group is most likely to be confronted with distributive negotiations. Non-probability sampling is used for the sample selection, because no sampling frame could be retrieved for this large research population. The non-probability sampling technique is snowball sampling and is distributed via e-mail and social media channels. The main goal for data collection is to achieve a sample as large as possible, to increase the chance of having a representative sample and to be able to generalize conclusions over the population.

On April 26 2016 the experimental survey was distributed and on May 10 2016 it was deactivated. In total 323 questionnaires have been filled in, subsequently 129 of these have been deleted due to insufficient or non-answered questions (further explained in the results section). Thus, in total 194 questionnaires were used during the analysis. From the 194

(27)

27 participants, 41 are male (21.1%) and 153 are female (78.9%)7. Most of the participants are aged between 20 and 30 with a mean age of 25.09. The youngest participant is 20 years old and the eldest participant is 63 years old. The high proportion of women and students makes the sample skewed towards female and younger aged individuals. The majority of the participants are currently enrolled, or already finished their Master’s degree. This implies highly educated individuals participated in the experiment, which is a positive outcome since the experiment requires a certain level of understanding of the situation and of the English language. An overview of all demographic statistics is provided in table 1 below. The participants were randomly distributed to either the low or the high level of social identification. 94 respondents (48.5%) participated in the low and 100 respondents (51.5%) participated in the high social identification condition.

Table 1. Demographical statistics

3.2.Research design

To be able to test the relationship between social identification and the intended bargaining strategy, and to test whether bargaining goals mediate and whether social value orientation moderates this effect, quantitative data is collected in order to test the proposed hypotheses. A quantitative research through an online experiment is done to collect the data.

7

Due to the high proportion of female in the sample, I controlled for gender in my analysis. Shown is that the high proportion of female did not affect the data.

Demographical statistics (N=194)

Gender Male 41 21.1%

Female 153 78.9%

Age 20-29 180 92.8%

30-63 14 7.2%

Education Secondary school 2 1.0%

Bachelor's degree 61 31.4% Master's degree 128 66.0%

Doctorate degree 2 1.0%

(28)

28 The reason for choosing an online experiment is because it offers easy access to a diverse demographic participant population. It also increases the external validity of the study as a large percentage of participants will remain in their familiar situation, such as for example at their computer at home or at work (Reips, 2000). In addition, statistical analysis can be done to compare a diversified and large sample of respondents (Saunders & Lewis, 2014). The independent variable in this study is social identification and the dependent variable is the intended bargaining strategy, which can be either a softline or a hardline bargaining strategy. The mediating variable is bargaining goals, which can be either economic or socioemotional, and the moderating variable is social value orientation, which is either prosocial or proself. Participants are assigned randomly to either a low or a high level of social identification, after which their bargaining goals, their social value orientation and their intended bargaining strategy is tested.

3.3.Experimental survey and procedure

The experimental survey consists of three parts. The first part is executed to manipulate the participant’s level of social identification. The second part is executed to measure the participant’s bargaining goals and strategy in a distributive negotiation context. The third part consists of a decomposed game in which individuals’ social value orientation is measured. The participants are assigned to either one of two groups: one group in which the participants do not feel identified with its opponent party, and another group in which participants do feel identified with its counterpart (please refer to appendix II for the complete experimental survey).

In the first part, the level of social identification is manipulated. An existing reliable and valid scale is used of Postmes et al. (2012), which measures the level of social identification with a ten item scale. In the low social identification condition, participants need to take in mind one colleague with whom he or she has not a lot in common with. They are asked to list

(29)

29 all of the ways in which they are different from this colleague and to list attributes, interests, beliefs and values that they believe distinguishes them from their colleague. In the high social identification condition, participants need to take in mind one colleague with whom he or she has a lot in common with and need to answer the same questions as in the low social identification condition. However, here the questions are rephrased and respondents need to answer questions related to their shared similarities with this colleague rather than its differences. The first condition is designed to increase the salience of the participants’ unique and distinctive personal identities and decrease the level of social identification, whereas the second is intended to increase the salience of their social identity and to increase their level of social identification. This manipulation is tested in a pre-test before to check whether the desired effect is achieved (please refer to the results section for the analysis of the pre-rest).

After answering these questions, participants of both groups continued to the second part of the online experimental study. They needed to imagine they are in a bargaining situation in which they need to sell their one year old IPhone 6, to their colleague they had in mind during the first part, with whom they feel either socially identified with or not. After reading a scenario, the bargaining goals and the bargaining strategy of the participants are measured via several questions participants needed to answer and statements they needed to rate. The goals incline towards either socioemotional or economic goals and the strategy inclines towards either a hardline or a softline bargaining strategy.

Lastly participants were confronted with a decomposed game in which they had to choose among three options, regarding the division of points between its colleague and itself. As a result participants’ social value orientation was measured, which could be either prosocial or proself.

(30)

30

3.4.Measurement

3.4.1. Social identification

The influence of social identification on the participants’ bargaining goals and strategy is measured by different questions and statements respondents answered. To manipulate the participants’ level of social identification towards their opponent bargaining party, the independent variable Social Identification is manipulated first. Respondents had to fill out three open questions about their colleague, with whom they either have a lot in common or not. To check whether this manipulation had the desired effect, it is tested in a pre-study before the actual survey took place (see chapter 3.5).

3.4.2. Bargaining strategy

The dependent variable, Intended Bargaining Strategy which can either incline to a hard- or softline strategy, is measured by several questions about the participant’s bargaining behavior, using an existing scale by Hüffmeier et al. (2014). The scale measures the participants’ first offer, the amount of concessions and the size of concessions they are willing to employ. It also measures whether participants gradually increase rather than decrease concessions. First, participants had to give their opening offer, an amount between 0 and 600 euros8, and this way the size of their first offer is being measured. Subsequently, participants had the option to lower their first offer. If they were not willing to lower their selling price at all, which means participants engage in a hardline strategy, the remaining questions about the Bargaining Strategy were skipped and participants continued to the questions about the next variable. However, if participants were willing to do a lower offer (to employ concessions), they had to choose for a lower selling price and they got again the option to lower their price for a second time etc., which continued this way until a maximum

8

A maximum amount of €600,- is chosen, as the IPhone is worth new €600,- so I assume no one would offer more than €600,- for a one year old IPhone.

(31)

31 of three new offers. This way the amount as well as the size of concessions is being measured. When moderate first offers and frequent and substantive concessions are being done, and participants gradually decrease rather than increase concessions, a softline strategy is used. A hardline strategy is used when participants do aggressive first offers, a minimum number and size of concessions, and gradually increase rather than decrease concessions (Hüffmeier et al., 2014). To measure this properly, three different variables are created that cover Bargaining Strategy, namely ‘First Offer’, ‘’Amount of concessions’’ and ‘’Size of concessions’’910

. Therefore, all tests have been performed three times separately to test the hypotheses regarding the independent variable Bargaining Strategy.

3.4.3. Bargaining Goals

The mediating variable, Bargaining Goals, is measured by an existing scale of Curhan et al. (2006) via a ten-item scale about the participants’ bargaining goals, which were rated on a 7-point Likert scale. Some statements were phrased a bit different to fit this research better. Example items of the scale are: ‘’I want to benefit the most from the outcome, regardless the outcome of my colleague’’ and ‘’I want to be satisfied with the balance between my own outcome and my colleagues’ outcome’’. The outcome can incline towards either economic or socioemotional goals, or somewhere in between. The results were measured on a 7-point Likert scale anchored by ‘1= strongly disagree and 7= strongly agree’, with Cronbach’s Alpha = .968.

3.4.4. Social Value Orientation

Lastly participants were confronted with a social dilemma task, in which their Social Value Orientation, the moderator in this study, is measured. To measure Social Value

9_{For the size of concessions, the absolute rather than the average size is taken.} 10

The fourth element that forms an individuals’ bargaining strategy, the gradually decreasing (softline strategy) or increasing (hardline strategy) concessions is taken into account during the analysis, however, no separate variable is made for this as the ignorance of this extra variable did not affect the results.

(32)

32 Orientation, an existing decomposed game by Van Lange et al. (1997) is used. As can been seen in the appendix, an example of a decomposed game is the choice among three options, regarding the division of points between the participant’s colleague and itself: Option A, 480 points for self and 80 points for the colleague; Option B, 540 points for self and 280 points for the colleague; and Option C, 480 points for self and 480 points for the colleague. In this example, Option A represents the competitive choice, because it provides a larger difference between one's own and the colleague’s outcomes (480 - 80 = 400) than does either Option B (540 - 280 = 260) or Option C (480 - 480 = 0). Option B represents the individualistic choice, because one's own outcomes are larger (540) than are those in Option A (480) or Option C (480). Finally, Option C represents the prosocial choice, because it provides a larger joint outcome (480 + 480 = 960) than does either Option A (480 + 80 = 560) or Option B (540 + 280 = 820); also, Option C represents a smaller discrepancy between one's own and other's outcomes (480 - 480 = 0) than does either Option A (480 - 80 = 400) or Option B (540 - 280 = 260). Participants were classified as either competitor, individualist or prosocial if at least six choices were consistent with one of these social value orientations (Van Lange et al., 1997). However, as this study only makes a distinction between prosocials and proselves, the competitors and individualists are combined and referred to as ‘’Proself’’ and the prosocials as ‘’Prosocial’’. This measure has generally revealed good internal consistency and reliability (Van Lange et al., 1997).

3.5.Pre-test

Prior to the main experimental survey, a pre-test is performed to assure that the manipulating variable is having the desired effect so that the level of social identification indeed is being manipulated. It is executed to test if participants would indeed feel either a low or a high level of social identification when being confronted with the different scenarios. The pre-test is executed between the 13th till the 19th of April among 58

(33)

33 participants. Half of the participants were in the low and the other half were in the high social identification condition. After deleting 26 participants who did not finish the pre-test, 32 participants were left. To measure the level of social identification a single item measure of this variable is used from an existing scale by Postmes et al. (2012), which is reported reliable. There was no need to recode items.

The social identification scale has high reliability, with Cronbach’s Alpha = .962. The corrected item-total correlations indicate that all the items have a good correlation with the total score of the scale (all above .30). Also, none of the items would substantially affect reliability if they were deleted. A one sample t-test was conducted to analyze the means of the variables. The means of both conditions significantly differ with a MeanLSI of t=7.125 and MeanHSI of t=23.450, which is significantly larger than 7.125 (p=.000) (please refer to output 1 in appendix III). This means the level of social identification is indeed being manipulated, so this manipulation technique can be used in the main study as well.

3.6.Main study

The experimental online survey is administered on April 26, 2016 through social media, E-mail and LinkedIn. The experimental survey is distributed by using convenience snowball sampling (asking participants to spread the questionnaire in their own social network). Participation was completely voluntary and all responses received were kept strictly confidential. Confidentiality had been clearly stated in the experimental survey. The experimental survey was online for a total of fourteen days, on May 10, 2016 the survey was deactivated and no more responses could be recorded. In total 323 participants have started the survey of which 194 were usable for analysis.

(34)

34

4. Results and analysis

4.1.Preliminary analysis

First the data is cleaned from inadequate results. Since there were 121 participants who did not finish the survey, they were excluded from the study. Eight other respondents were deleted, as they gave very unreliable answers to the questions about their bargaining strategy. Some answered for example 0 or 2 to the question about the first offer, which I do not perceive as very reliable for the sale of an IPhone.11 After excluding these participants, 194 were left and no more missing values were present in the dataset. During the survey, respondents were forced to answer all questions before they could proceed with the survey. Therefore, no missing values were noted on these questions for the people who finished the experiment. In all scale items several counter-indicative items were present, which were recoded first. As there were no negative Cronbach Alpha’s in one of the scales, no items were forget to recode (for the reliability analysis please refer to next subchapter). There are two conditions in the questionnaire, the manipulation of a low (LSI) and high (HSI) level of social identification. Therefore all questions were displayed twice in SPSS, which were converted into condition. A new variable called ‘SI_cond’, is made to distinguish between the low and the high level of social identification and are valued as 1 (LSI) and 2 (HSI) in SPSS.

4.2.Reliability and scale means

For this study to be able to achieve meaningful answers, averages of variables with a multi-item scale had to be computed, however prior to this the reliability of the scales needed to be checked. The Cronbach’s Alpha, had been tested to verify if all the items in one scale measure the same. It is the extent to which data collection techniques or analysis procedures yield consistent findings (Saunders et al., 2009). In this study there were three variables with

11

When respondents gave a first offer which was less than 10 euros, they were excluded from the analysis. However, a check has been done whether this removal would affect the results and it did not make a significant difference.

(35)

35 scale items, namely the mediator variable Bargaining Goals, the moderator variable Social Value Orientation and the dependent variable Intended Bargaining Strategy.

The variable Bargaining Goals had a Cronbach’s α = .968 which indicate a high level of internal consistency.Also, the corrected item-total correlations of the variable indicate that all the items have a good correlation with the total score of the scale (all above .30). Also, none of the items would substantially affect reliability if they were deleted (please refer to output 2 in appendix III).

For the variable Social Value Orientation, an existing and common measurement scale is used. To cite Van Lange et al. (1997, pp. 736): ‘’These measures of social value orientation have generally revealed good internal consistency and test-retest reliability over a period ranging from 2 months to 6 months (e.g., Kuhlman, Camac, & Cunha, 1986; Van Lange & Semin-Goossens, 1997)’’. Therefore no reliability analysis is done by myself as all choices were randomized and it would make no sense to recode them. Therefore, I checked for all respondents separately, whether they gave six or more consistent choices and named them either, prosocial, individualist or competitor. However, as this study does only distinguishes between prosocials and proselves, the individualists and competitors were combined and named ‘Proself’ (valued 2 in SPSS), and the prosocials stay ‘’Prosocial’’ (valued 1 in SPSS). If less than six consistent choices were given, the values were reported as missing.

For the variable Bargaining Strategy, three separate variables are made, namely; First Offer, Amount of Concessions and Size of Concessions. The First Offer is the amount that is offered the first time, the Amount of Concessions is about how many times participants were willing to do a next (lower) offer, and the Size of Concessions is the absolute value of the last offer participants did.

(36)

36 Since the reliability of the variables were acceptable, for the hypotheses testing two new variables were computed as a function of existing variables. The first new variable was the calculated mean of the nine scale items that were used to measure Bargaining Goals, named Mean_BG, which can incline towards either economic (inclines towards 1) or socioemotional (inclines towards 7) goals. For the variable Social Value Orientation first was checked whether respondents gave six or more consistent answers (from the total of nine questions), as then these respondents could be classified as either prosocial or proself (valued respectively 1 and 2 in SPSS). The ones who could not be classified were reported as missing. After the classification a new variable is made for Social Value Orientation called SVOtot.

4.3.Correlational analysis

The relationship between the variables is examined through a correlation analysis. The results of the correlation analysis, illustrated in table 2 below, clarify the intensity and meaning between two or more variables and exhibit the means and standard deviations.

Table 2. Correlation analysis

The results suggest there is a significant positive relationship between age and education of r=.249. The variable Social Identification has significant correlations with either the mediator variable Bargaining Goals (r=.578), as well as the moderator variable Social Value Orientation (r=-.362). Social Identification also correlates significantly with all three

Means, Standard deviations, Correlations

Variables M SD 1 2 3 4 5 6 7 8 9 1. Gender 1.79 0.409 _ 2. Age 25.08 6.583 -0.059 _ 3. Education 2.69 0.538 0.026 .249** _ 4. Social Identification 1.52 0.501 0.13 0.058 0.105 _ 5. Bargaining Goals 4.81 1.59 -0.001 -0.051 -0.026 .578** 0.968 6. Social Value Orientation 1.36 0.515 -0.029 -0.028 -0.087 -.362** -.489** _

7. First offer 276.07 75.73 -.164* 0.076 -0.081 -.454** -.223** 0.135 _

8. Amount of concessions 1.14 1.046 0.045 -0.046 0.069 .376** .552** -.237** 0.08 _

9. Size of concessions 244.58 77.316 -.199** 0.091 -0.098 -.577** -.413** .230** .906** -.277** _ ** Correlation is significant at the 0.01 level (2-tailed).

Do you feel the connection? : a quantitative research on the effect of social identification in distributive negotiations