Bachelor Thesis Social Psychology
The Effect of Perceived Future Cooperation on Talent
Attribution
Name: Jos Schreij
Student number: 10804439
Professors: Noah Millman & Simon Columbus Educational Institution: Universiteit van Amsterdam Datum: 31-05-2016
2 Abstract
In this thesis we researched the effect of future cooperation on talent attribution. 82 participants with an average age of 29.73 participated in our experiment. The participants were randomly assigned to either group “A” or group “B” after which they were shown six paintings. All the paintings had a vignette that shown if the painter was from an in-group, an out-group or an unknown-group. The participants were asked to rate the talent of the painters. The next part of the experiment existed out of filling out two small questionnaires that measured the moderator’s future interdependence and group based self-esteem. The final part of the experiment was a cooperation game in which the participants could win some money. Our main hypothesis expected a higher rating of talent for the in-group painters compared to the out-group/unknown-group painters. This hypothesis was supported, which means that future cooperation has a positive effect on talent attribution. There was no significant effect found of the moderator’s future interdependence and group based self-esteem on talent attribution. However because there was no difference in talent attribution of the out-group compared to the unknown-group this thesis supports interdependence based theories rather ten social identity based theories to lie at the foundation of in-group favoritism.
3 In high school we all wanted to get picked first in PE; children who get picked first are usually the ones who are the most talented at sports. From the moment you are placed in the same team, you and your teammates are all working towards the same goal: winning. In this way, your team members become part of your ‘in-group’, and because you see your team members as your in-group, you also start seeing them as more favorable. This effect is called in-group favoritism (Turner, Brown and Tajfel, 1979).
Although there has already been a lot of research on in-group favoritism, there are still some areas that have been left relatively untouched. Current theories have shown that in-group favoritism makes you judge your in-in-group in more positive light (Brewer, 1979). Research has shown that people tend reward (Tajfel et al., 1971) and offer more help towards people from their in-group (Fu et al., 2012). People have also been shown to work harder to accomplish in-group goals (Ellemers, De Gilder and Haslam, 2004). It seems therefore logical that there could be in-group favoritism for talent as well, however this is yet to be researched.
In-group favoritism is defined as any tendency to favor the in-group over the out-group (Turner, Brown and Tajfel, 1979). There are multiple theories explaining why in-out-group favoritism occurs, the two most famous are the Social Identity Theory by Tajfel and Turner (1979) and Interdependence-based theories such as described by Rabbie and Wilkens (1971) and Yamagi-shi, Jin, & Kiyonari (1999). Both theories state that in-group favoritism will occur when you are perceived to be working with someone in the future (Turner, Brown and Tajfel, 1979; Rabbie and Wilkens, 1971).
However, these two theories are inconsistent and therefore inconclusive about how in-group favoritism works. The contradiction between the theories leads to an inconsistency in the literature. This inconsistency needs to be resolved in order to get a better understanding
4 about in-group favoritism. Therefore, resolving this inconsistency in the literature will be one of the goals of this thesis. This will offer an advantage to all future research on the subject of in-group favoritism. If the effect of in-group favoritism on talent attribution will be found, this research would be very interesting for companies. This research would then have direct implications for hiring processes. For example, perhaps companies would be willing to invest more in people when they know they will be working with someone in the future.
The aim of this study was to look at the effects of perceived future cooperation on talent attribution. To do so we will first look at the existing literature on in-group favoritism. We will discuss both previously-discussed theories that try to explain in-group favoritism and see what the differences and similarities are between these theories. After this literature review, the methods and results of the study will be elaborated on. We will then discuss our findings, and the implications of our results.
Literature Review
In-group favoritism is any tendency to favor the in-group (Turner, Brown and Tajfel, 1979). For example, this could be by evaluating in-group members as more positive (Brewer, 1979) or by showing other positive biases towards your in-group (Balliet, Wu and De Dreu, 2014). We can conclude from this that because in-group favoritism works on many different levels, there might be in-group favoritism in talent attribution as well. However, there is lack of literature about how in-group favoritism affects talent attribution. Talent attribution is defined as how talented someone perceives someone else to be. It is good to make a clear distinction that talent attribution is not about how much someone likes a painting, but solely about how much talent someone attributes to the painter.
To see if there actually is an in-group favoritism effect on talent attribution, it is important to first understand how in-group favoritism works. In the literature, there are two
5 dominating but somewhat contradicting groups of theories that try to explain how in-group favoritism occurs (Balliet, Wu and De Dreu, 2014). These two groups of theories are: interdependence based theories and social identity based theories. We will review in-group favoritism from both streams of theories and in the end, will lend support to one them depending on the results.
Interdependence is one of the two leading theories that attempts to explain in-group favoritism. It states that people are always interdependent in some form, in all types of social interaction (Gerpott, Bailliet and de Vries, 2016). It does not matter if a person is greeting another person on the street or if you have to work with someone else on a group project, whenever there is social interaction, there is interdependence. The behavior of someone greeting another person on the street, despite how big or small, affects the outcome of the interaction for both people. This is interdependence (Gerpott, Bailliet and de Vries, 2016).
In-group favoritism occurs from interdependence because there is a chance for
reciprocity. Individuals will treat in-group members as more favorable because they expect to get the same favorable treatment from other in-group members (Yamagi-shi, Jin, & Kiyonari, 1999; Lodewijkx, Rabbie and Visser, 2006). As it turns out, people are even able to detect a form of interdependence in situations. When this happens, individuals will react in a way that benefits them (Gerpott, Bailliet and de Vries, 2016). For example, subjects tend to reward people more who they believe will later have to determine their pay (Yamagi-shi, Jin, & Kiyonari, 1999). By displaying in-group favoritism to another person, the chance that they will also treat you more favorably increases.
Social identity theory, on the other hand, states that self-esteem, not interdependence, lays at the foundation of in-group favoritism. Social identity theory states that people are always looking to have a positive self-esteem (Gleitman, Reisberg and Gross, 2011). A
6 person’s self-esteem is partly derived from the identity with the group that the individuals are in (Gilovich, Keltner and Nisbett, 2006; Yamagi-shi, Jin, & Kiyonari, 1999). Rating our in-group as more positive, compared to the out-in-group, will lead to higher levels of self-esteem (Tajfel & Turner, 1979). Therefore by having in-group favoritism, we raise our own self-esteem.
A recent meta-analysis compared both theories to see which of them has the most plausible explanation for group favoritism. This meta-analysis showed that looking at in-group favoritism from an interdependent point of view is the more likely explanation of the effect (Balliet, Wu and De Dreu, 2014). However, both theories give a plausible explanations for in-group favoritism, as described above, and the meta-analysis states that this is not conclusive, and even states that they might be working at the same time since they share some similarities.
One of the similarities is that both theories state that perceiving that you will work with someone in the future is enough to begin the effect of in-group favoritism (Turner, Brown and Tajfel, 1979; Rabbie and Wilkens, 1971). From an interdependence point of view, this would happen because if you are perceiving to be working with someone in the future, you perceive yourself as being dependent on them in the future, and this perceived future interdependence will lead to in-group favoritism (Rabbie and Wilkens, 1971). Social identity theory, on the other hand, says that perceived future cooperation will make someone part of your in-group. Therefore, an individual you are perceived to be working with in the future will have an effect on your self-esteem. By showing in-group favoritism, you will
unconsciously raise your group-based self-esteem and hereby also raise your own self-esteem (Turner, Brown and Tajfel, 1979). To check which theory will lead to in-group favoritism from perceived future cooperation we have added both future interdependence and group based self-esteem as mediators to our experiment.
7 The theories do differ though, in how you would rate the out-group. Interdependence based theories predict that we will rate the out-group the same as the control-group. This is the case because interdependence based theories state that we are trying to get utilitarian outcomes (Yamagi-shi, Jin, & Kiyonari, 1999). The reason for in-group favoritism is to get maximum benefit for the in-group, while not saying anything about the out-group, and
therefore not derogating it. Social Identity theory, on the other hand, predicts that subject will rate the out-group as lower than the control-group. They will derogate the out-group because, following Social identity theory, you want to maximize the differences between the in-group and the out-group (Brewer, 1979). By derogating the out-group, you will make the relative difference bigger and therefore gain a more positive feeling about your group. However, for this effect to happen the out-group must be salient for the in-group members (Brewer, 1979).
The above leads to a split in hypotheses. Firstly, both theories expect that H1:
Perceived future cooperation will lead to positive talent attribution of the in-group. But since there are multiple theories explaining in-group favoritism, we also get some extra secondary hypothesis. If interdependence causes in-group favoritism we expect that H2: Perceived future interdependence has a positive effect on talent attribution of the in-group. And H3: Perceived future interdependence has no effect on talent attribution of the out-group. However, if Social identity theory causes in-group favoritism we would expect that H4: Group based self-esteem has a positive effect on talent attribution of the in-group. And H5: Perceived future interdependence has a negative effect on talent attribution of the out-group.
Having looked at the existing literature, it can be expected that future cooperation will lead to in-group favoritism. Thus, for this research, it is expected that subjects will rate the artists of the painting with whom they are perceived to be working with in the future, as more talented than the other painters.
8 Methods
Participants
To determine the sample size we used an a priori power analysis based on a
previously conducted study about the effect of in-group favoritism on talent attribution by Millman (The Group Processes Class, 2015). The main effect size used was ηp2=.0235, this
translates as F=.1551 with a power of .80. The calculated sample size was n=69. Before the research started, we agreed to stop gathering data once the sample size was reached or if we didn’t manage to collect enough participants within a four weeks’ time frame.
In total 136 people filled out the questionnaire on Qualtrics, however 54 participants were discarded from the analysis for not filling out the questionnaire completely. The 82 participants who did complete the experiment had an age between 17 and 69 with an average age of 29.73 (SD=13.97). 19 of the participants were male, 63 were female.
Materials
All paintings used during the experiment were used in a previous study about talent attribution (Millman, 2014). Each painting had a small vignette with some additional information about the painter. The vignette either showed that the painter was from group “A”, group “B” or an “Unknown” group. All conditions were randomized by Qualtrics. The participants were asked to rate the paintings on the following questions: “This painter is talented.” and “This painting is beautiful.” on a 7-Points Likert Scale (1=Disagree Completely, 7=Agree Completely).
The next part of the experiment consisted of two small questionnaires that the participants had to fill in. The questionnaires measured future interdependence and group based self-esteem and could both be answered with a 7-Points Likert scale (1=Strongly Disagree, 2=Disagree, 3=Disagree Somewhat, 4=Neutral, 5=Agree Somewhat, 6=Agree,
9 7=Strongly Agree). The questionnaires were offered in a random order. The future
interdependence questions were a subset of six questions used in previous research by
Gerpott, Bailliet and de Vries (2016). Examples of questions used are: “How we behave now will have consequences for future outcomes.” and “Our future interactions are not affected by the outcomes of this situation.”. For measuring group-based esteem, we used a self-esteem scale by Luhtanen and Crocker (1992). This scale contains a subset of questions about social identity, but we used only the subset of membership self-esteem to measure group based self-esteem. “I am a worthy member of the social groups I belong to.” is an example of a question used for this scale.
The cooperation game we used was a modification of the golden balls game. Participants had to choose to either “Steal” or “Share” the tickets from a random group member they were linked to. If both linked participants chose to share, they both got 2 lottery tickets. If one of the participants chose to steal and the other to share, then the participant who chose to steal got 3 lottery tickets, and the participant who chose to share did not get any tickets. If both participants chose to steal, then they both got nothing. In the end, we
randomly picked two participants who both got 10 euros.
Procedures
For the gathering of participants we had no restrictions at all. Which means that everyone could participate in our experiment. The participants were approached via
Facebook. They were asked to participate in a small psychology experiment of ten minutes with a chance to win a prize of 20 euro’s.
When participants agreed to participate in our experiment, we asked them to fill out a questionnaire in Qualtrics. The questionnaire started with a form of consent. Once they signed the consent form the participants were given the instructions of the experiment. After
10 reading the instructions, participants were randomly assigned to either group “A” or group “B”. Participants were told that at the end of the experiment they would have to do a small cooperation game with someone from their own group. To assure that participants knew which group they had been assigned, we asked them to fill out their group before continuing with the experiment.
In the next part of the questionnaire, participants had to rate six different paintings. They were asked to rate the paintings on how talented they thought the painter was, as well as how beautiful they thought the paintings were. The next part of the experiment consisted of filling out two small questionnaires about group-based self-esteem and future
interdependence and a small questionnaire about the participant’s demographics. After doing so, the participants entered the cooperation game. At the end of the experiment, participants were asked to leave their email address if they wanted to have a chance of winning some money. This was voluntary not mandatory.
Statistical Analyses
Data Preparation
Outlier analysis. For this study, participants were excluded if they were perceived to be under the influence of alcohol or any other drugs. Participants were also excluded when we noticed suspicious answer patterns on the Likert scale.
Scale reliability. Internal consistency of all multiple item scales will be determined using Alpha and must be higher than .70.
11 Sphericity test. Normality will be assessed using the Mauchly's Test of Sphericity for equal covariances. If the test for Sphericity is significant, we continue with Greenhouse-Geisser.
Assumption tests. Normality will be assessed using Q-Q plots. Unless there is severe non-normality, we assume the sample size is sufficient.
Omnibus test. Repeated measures ANOVA will be used with perceived future cooperation as an independent variable and talent attribution as dependent variable. If the repeated measures is significant I will do three post-hoc tests comparing all levels of perceived future cooperation using paired sample t-test using Bonferronie. I will analyze mediation by perceived future interdependence and group based Self-Esteem by PROCESS.
Results
Checks and data screening
After analyzing the data we found that there were no outliers. The variable of talent rating of the painters was normally distributed as seen in the Q-Q plots. This means that the sample size is sufficient and that the assumption is met.
Main effect
To test the Main hypothesis we conducted a repeated measures ANOVA in SPSS. This repeated measures ANOVA had talent rating as dependent variable and group (In-group/Out-group/Unknown-group) as independent variable. Mauchly’s test of Sphericity was not significant (p=.978) so the assumption is met.
The repeated measures ANOVA revealed a significant difference between the groups
F(2, 80)=3.60, p=.030, n2=.053. Further analysis using contrasts revealed that there was a
12
p=.009, n2=.080. However, a non-significant difference was found between the Out-group
and the Unknown-group F(1, 81)=.00, p=1, n2=.000. Looking at the means the In-group has a
significant higher score on talent attribution (M=10.68, SD=1.98) then the Out-group (M=10.22, SD=2.37) and the Unknown-group (M=10.22, SD=2.35). This means that Hypothesis 1 is supported. Since there is no significant difference between the Out-group and the Unknown-group we can assume that Hypothesis 3 is supported and at the same time we can also assume that Hypothesis 5 is not supported.
Secondary analysis
To test the effect of the Moderators (Future Interdependence/Group Based Self-Esteem) on talent attribution we did a new repeated measures ANOVA. Mauchly’s test of Sphericity was not significant (p=.965) which means that the assumptions is met.
The test of Between-Subject Effects was significant for both FIF(1, 81)=7=0.50, p=.483, n2=.006 and GBSE F(1, 81)=0.02, p=.880, n2=.000. However the test of
Within-Subjects Effects was not significant for FI F(2, 80)=0.12, p=.889, n2=.001, meaning that
Hypothesis 2 is not supported. Furthermore, the analysis also revealed non-significance for GBSE F(2, 80)=0.569, p=.567, n2=.007, meaning that Hypothesis 4 is also not supported.
Discussion
The research question of this experiment was: “What is the effect of perceived future cooperation on talent attribution?” Our results show that perceiving future work leads to a higher attribution of talent. Our results also show that there is no difference in how you perceive the talent of someone of the out-group compared with the unknown group. These results suggest that interdependence based theories are the more likely explanation of in-group favoritism then Social identity based theories. However, both the future
13 attribution, which might suggest that there is further research needed to draw a conclusion on which theories might explain in-group favoritism.
Our findings match the theories of favoritism by Turner, Brown and Tajfel, (1979) and Rabbie and Wilkens (1971) on which our main hypothesis that perceived future
cooperation will lead to in-group favoritisms was based. However our findings do not match the theory of Gerpott, Bailliet and de Vries (2016) on which our second hypothesis was based. Due to this theory, we would expect a significant effect of the moderator’s future interdependence on talent attribution. To explain this difference we should check our methods. During our experiment, we made a mistake by only measuring the future interdependence of the in-group. Instead, we should have also measured the future
interdependence of the out-group, which would have enabled us to compare these two group and led to more reliable results.
Furthermore, an explanation for not finding a significant effect for the group based self-esteem moderator, might be the fact that our out-group was not salient enough. The literature states that Social identity theory will only have an effect on in-group favoritism when the out-group is salient enough (Brewer, 1979). During my experiment, I did not give that much information about the group, which may have resulted in a non-salient out-group. Furthermore, we also did not add a manipulation check to see if the out-group was perceived as salient by the participants. For future research it might be better to add such a manipulation check.
Despite this critique of our research, it is still important to interpret the current findings and see how they could be implicated in real life. Given that future cooperation has an influence on talent attribution, it is an important aspect of the hiring process. Since this is
14 now known to us, it is wise to think of methods that take this effect into account during hiring procedures.
Even though our research already found a direct link between perceived future cooperation and talent attribution, we would still advise to do additional research on this subject. Currently, there is still little research into the effects of in-group favoritism on talent attribution. By doing more research on this subject with different forms of in-group
favoritism, we will get a more complete image of the subject.
When we started our research we wanted to fill the gap in the literature when it came to talent attribution. We also wanted to check what kind of theories best explained in-group favoritism. With the results of our experiment, we found that perceived future cooperation has a positive effect on talent attribution. We also found that this effect of in-group favoritism is likely due to interdependence based theories.
15 Bibliography
Balliet, D., Wu, J. and De Dreu, C. (2014). Ingroup favoritism in cooperation: A meta-analysis. Psychological Bulletin, 140(6), pp.1556-1581.
Brewer, M. (1979). In-group bias in the minimal intergroup situation: A cognitive-motivational analysis. Psychological Bulletin, 86(2), pp.307-324.
Ellemers, N., De Gilder, D. and Haslam, S. (2004). Motivating Individuals and Groups at Work: A Social Identity Perspective on Leadership and Group Performance. The
Academy of Management Review, 29(3), p.459.
Fu, F., Tarnita, C., Christakis, N., Wang, L., Rand, D. and Nowak, M. (2012). Evolution of in-group favoritism. Sci. Rep., 2.
Gerpott, F., Balliet, D., & de Vries, R. E. (2016). How Do People Think About Interdependence? Testing a Multidimensional Model of Subjective Outcome Interdependence. Unpublished manuscript.
Gilovich, T., Keltner, D., & Nisbett, R. (2006). Social psychology (2nd ed.). New York: W.W. Norton & Co.
Gleitman, H., Gross, J., Reisberg, D. (2011). Psychology (8th ed.). New York: W.W. Norton & Company.
Lodewijkx, H., Rabbie, J. and Visser, L. (2006). “Better to be safe than to be sorry”:
Extinguishing the individual – group discontinuity effect in competition by cautious reciprocation. European Review of Social Psychology, 17(1), pp.185-232.
Luhtanen, R. and Crocker, J. (1992). A Collective Self-Esteem Scale: Self-Evaluation of One's Social Identity. Personality and Social Psychology Bulletin, 18(3), pp.302-318. Millman, N. (2014). In-group Favouritism in Talent Attribution. Unpublished pilot.
Rabbie, J. and Wilkens, G. (1971). Intergroup competition and its effect on intragroup and intergroup relations. European Journal of Social Psychology, 1(2), pp.215-234.
16 The Group Processes Class, PPLE. (2015). In-group Favouritism in Talent Attribution.
Unpublished manuscript.
Tajfel, H., Billig, M., Bundy, R. and Flament, C. (1971). Social categorization and intergroup behaviour. European Journal of Social Psychology, 1(2), pp.149-178.
Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 33– 47). Monterey, CA: Brooks/Cole
Turner, J., Brown, R. and Tajfel, H. (1979). Social comparison and group interest in ingroup favouritism. European Journal of Social Psychology, 9(2), pp.187-204.
Yamagishi, T., Jin, N., & Kiyonari, T. (1999). Bounded generalized reci-procity: Ingroup boasting and ingroup favoritism. In E. J. Lawler (Series Ed.) & S. R. Thye, E. J. Lawler, M. W. Macy, & H. A. Walker (Vol. Eds.), Advances in group processes (pp. 161–197). Bingley, England: Emerald
