Primed to cheat: Stealing your co-worker's idea to stay in business

Table of Contents

1. Introduction

2. Theory and Hypotheses
2.1 The role of subliminal priming
2.2 Willingness to take credit without and with escalating severity of consequences
2.3 Perceived ethical costs
2.4 Perception of others’ mindset

3. Methodology

4. Results
4.1 Quantitative results
4.2 Qualitative results

5. Discussion

6. Limitations

7. Conclusion and implications

Appendix

REFERENCES

Acknowledgement

I would like to thank my wife Thea for her relentless emotional support and Dr Jonathan E. Booth for providing me with this great learning experience.

Abstract

This study aims to show that situational cues like semantic primes are able to influence a participant’s decision-making in the context of taking credit for someone else’s idea at the workplace. In a laboratory experiment either a competitive, cooperative or neutral environment was simulated by using subliminal priming techniques. Participants were then exposed to a hypothetical scenario in which they were urged – due to their heavy workload – to take credit for a colleague’s idea. In particular, the study examined four constructs: the participants’ willingness to take credit for this idea without and with escalating consequences for not choosing to do it, the perceived ethical costs related to that action and the perception of others mindsets. It was expected that participants in the competitive condition would be more willing to take credit for the colleague’s idea and perceive less ethical costs than in the neutral condition and vice versa for the participants in the cooperative condition. Additionally, the study expected the participants to perceive the mindset of a random other in the same situation to be similar to their own mindset. Furthermore, qualitative data was collected to explore the decision-making process in such a situation. The results showed that situational cues did not significantly influence the individual decision-making in this particular context. However, several unexpected findings about the relationships between the tested constructs deliver valuable implications for future research on the topic of taking credit for other people’s ideas. In addition, the findings from the qualitative analysis suggest that future studies have to differentiate between ethical and utilitarian considerations to gain further insights into the individual decision-making process.

1. Introduction

In his book “The rise of the knowledge worker” James Cortada (1998) describes the change in industrialized and increasingly computerized countries: More and more workers are required not to manufacture new goods anymore, but to provide knowledge and information to those who still do. In fact, with the manufacturing industries declining and the shift to service and knowledge sectors in developed countries, the importance of generating new ideas and innovation has risen over the last decades (Schettkat/Yocarini 2003).

With increasing importance of ideas, they more and more pose a competitive advantage for workers who have ideas over those who don’t. Following this thought, Wang and Noe (2010: 124) found in the literature about knowledge sharing that ideas can act as positive factors in one’s performance evaluation and thus in promotions, bonuses and other forms of personal gains. They also mention that employees in possession of valuable ideas will tend to guard them and thus keep their status as an expert in a particular area. As with any other resource, individuals who are in need of a valuable idea, but are not able to obtain it for whatever reason, might be tempted to illegally take credit for an idea of another person. A popular example for this poses the former German Minister of Defence, Karl-Theodor zu Guttenberg, who was recently stripped of doctorate due to plagiarism. He justified the substantial copying with his heavy workload as member of the German parliament while raising his two daughters (Pidd 2011).

The present study is primary concerned with the temptation to take credit for a colleague’s idea when unable to cope with one’s own job demands in competitive or in cooperative work environments. Thereby, it is important to clearly distinguish ideas from intellectual property. Intellectual property rights can only be claimed by the author of an idea if the respective idea has taken form in some way, e.g. as a drawing or a picture of an invention (patent, trade mark), a physical appearance (design) or a written document (copyright) that is then officially registered (Intellectual Property Office 2007). Thus, in contrast to intellectual property, it remains debatable if it is theft to take credit for an idea that was merely verbally mentioned by someone else, especially if that person is working within the same company.

So far, past research has mainly focused on the effects of competitive work environment on the willingness of workers to share information with each other (e.g. Steinel et al. 2010 ; Carpenter/Seki 2006 ; Burks et al. 2005 ; Drago/Garvey 1998). Actively taking credit for someone else’s ideas has rather been neglected. The only relevant study about taking credit for someone else’s idea at the workplace that could be identified in preparation of the present study was authored by OfficeTeam, a staffing specialist company (Krumrie 2010). This study included telephone interviews of 444 workers and found that 29 % had experienced a co-worker taking credit for one of their ideas. However, the survey didn’t ask if any of the interviewed persons ever took credit for someone else’s idea and about their motivations to do so.

The present research seeks to fill this gap by pursuing two goals. The first goal is to show that situational influences can activate competitively or cooperatively processing mindsets that are strong enough to alter an individual’s decision to take credit for a colleague’s idea regardless one’s personality or personal disposition. In particular, four aspects are examined: the willingness of an individual to take credit for someone else’s idea without or with escalating severity of anticipated consequences for not taking credit, the individual’s ethical considerations about this action and the individual’s perception of the mindsets of others. The situational influences in form of either a competitive or cooperative environment will be thereby artificially created by using subliminal priming techniques in an artificial laboratory experiment. The second goal is to develop a comprehensive basis for future research in the field of stealing ideas by asking questions about why it would be right or wrong to take credit for someone else’s ideas. This shall enhance our understanding of the underlying reasoning in the participants’ decision-making process.

2. Theory and Hypotheses

2.1 The role of subliminal priming

The method of subliminal priming was used in this study to artificially produce either a competitive or cooperative environment that affects the participants’ behaviour with situational cues. Subliminal in this context means – as opposed to supraliminal – that primes affect individuals below their threshold of cognitive awareness (Bargh/Pietromonaco 1982).

In the literature about social cognition primes represent subtle manipulations that activate certain mentally represented concepts via associative links (Bargh et al. 1996, Dijksterhuis et al. 1998). Research has shown that priming can alter a person’s perceptions of others (e.g. Herr 1986), which then influence behavioural responses to or processing of a certain piece of information. In Herr’s study (1986), participants were primed with the concept of hostility, which made them perceiving others more hostile towards them and therefore reduced the level of cooperation. Notably, these studies focused merely on the participants’ perceptions of other actors with a mediating effect on behavioural responses.

Other studies have shown that those activated concepts can directly trigger related behavioural responses, even if only weak cues were used to prime the participants in an experiment (e.g. Srull&Wyer 1979 ; Bargh et al. 2001 ; Maxwell et al. 1999 ; Good 1973). A study by Bargh et al. (1996) depicts the impact of such cues: merely exposing participants to an elderly stereotype induced them to walk down a hallway slower than those in the neutral control group did. In another study, Kay & Ross (2003) sought to show that situational constructs and norms have a mediating role on behaviour. They argued that such situational cues can greatly predict social behaviour. Their findings demonstrated that priming persons either exposed to competitive or cooperative cues had a significant effect on the persons’ subsequent behavioural intentions. The present study seeks to build on these findings by showing that competitive and cooperative cues can also influence behavioural perceptions and intentions towards stealing someone else’s ideas.

It has to be noted that such manipulations are expected to function best in ambiguous or novel situations. Extending Fiske and Taylor’s assumption that people try to minimize cognitive resources by established scripts about how to deal with familiar situations, Kay et al. (2004: 84) showed in a study about priming with physical objects that people’s decisions and perceptions in ambiguous or unfamiliar situations tend to rely on external (environmental) cues. Consequently, participants in the present study’s experiment were confronted with a hypothetical situation that was kept as ambiguous as possible.

2.2 Willingness to take credit without and with escalating severity of consequences

In the literature about the dark side of the organisation – involving all intentional behaviour by an entity in an organization with negative consequences for another entity (Griffin/O’Leary-Kelly 2004) – two perspectives are generally considered: the bad individual per se and the bad organisation with negative influence on the individual.

Evidence exists that there are certain personality traits, which can be responsible for dark side behaviour such as taking credit for someone else’s idea. Blickle and Schlegel (2006) found e.g. in their study among 150 managers and convicted 76 white-collar felons that certain traits are likely to be responsible for engaging in white-collar crime: higher levels of hedonism and narcissistic personality disorder and lower levels of behavioural self-control and conscientiousness indicated a higher likeliness to commit offences.

However, there is reason to assume that an individual’s surroundings significantly affect his or her intentions and behaviour regardless the individual’s personal disposition. There is e.g. empirical evidence that an individual’s level of dark side behaviour is in fact influenced by the overall level of this behaviour in his or her work group. In a study by Robinson and O’Leary-Kelly (1998) perceived antisocial group behaviour (such as stealing or saying something bad about a colleague) was positively related to each group member’s antisocial behaviour.

The mere working conditions also present a viable influence: Employees who perceive that their job demands exceed their coping resources (one’s own effort to succeed in one’s job) feel overwhelmed (Demerouti et al. 2001). Agnew’s general strain theory (1992, in Langton/Piquero 2007: 2) assumes that the failure to achieve a goal or to meet job demands is likely to result in mental strain. Thus, employees will be urged to look for ways to regain control over the work environment. The decision or rather the justification which way is the right to choose is thereby also likely to be influenced by the employees’ environment itself. Social information processing theory (Salancik/Pfeffer 1978: 224) suggests that an individual develops attitudes and socially acceptable rationalisations based on the available social information. Next to the consequences of own past choices this information is provided by other employees and the organisation itself, i.e. the social environment of the individual. The following paragraphs consider two kind of environments and their expected effect on an individual under pressure, which is relevant for this study: a competitive and a cooperative environment.

Labour economists have addressed in the past how employees under pressure will behave in a competitive work environment (Lazear 1989, Lazear/Gibbs 2009, Chen 2003, Green/Stokey 1983). In a so-called “tournament” (Lazear/Gibbs 2009: 298) a firm rewards or promotes its employees for their performance. The competitive factor derives from a relative performance measurement rather than from an absolute standard: it does not primarily matter if an employee reaches a certain performance, but it does matter if he or she is doing better than her colleagues (Lazear/Gibbs 2009). This implies two principles. First, competition motivates employees to increase their effort in order to win (Lazear/Rosen 1981). Second, it can decrease cooperation among them and push them to engage in sabotaging their colleagues (Lazear 1989). Such “predatory behaviour” (Lazear 1989: 562) can include stealing information from them to make up for one’s – insufficient – effort. This paragraph explained how workers react in a environment with high pressure from competition. The next paragraph will go further into detail about why they react that way.

This does not necessarily imply the intention to harm the co-worker who is sabotaged, i.e. is ripped of his or her idea, but primary to secure one’s survival at the workplace (Rosenfeld et al. 1995). As found in a study by Harrell-Cook et al. (1999), this kind of behaviour serves as a coping mechanism for one’s own incompetence. Such a form of impression management is described by Bratton and Kacmar and involves next to blaming, discrediting, intimidation and negative projection also “taking credit for someone else’s work” (2004: 296). They further explain that this harmful form of impression management is triggered by contextual variables such as uncertainty that partially arises when one’s own competence doesn’t seem enough to succeed (Bratton/Kacmar 2004).

In contrast to a competitive environment, a cooperative environment, in which individual performance evaluation is based on group performance or on absolute standards, has a rather opposite effect. A study, conducted by Carpenter and Seki (2006) in a Japanese fishing community shows that the nature of a respective work environment has distinct outcomes on an individual’s attitude. Their findings are that the fishermen who are not organized in collaborative groups faced a highly competitive environment. They engaged in less cooperation than the fish traders and the administrative staff of the community, who faced a less competitive work environment. This study demonstrated how the work environment and job demands lead to different outcomes on individuals who even shared a common background after all: all workplaces were situated in the same community (Carpenter/Seki 2006).

Notably, the level of collaboration is positively related to the level of cohesiveness in a group of employees (Bakker et al. 2006; Sawng et al 2006). The level of group cohesiveness indicates the level of mutual trust and assurance that none of them will take advantage of the other members and also to what extent they comply to shared norms and goals (Lauring/Selmer 2010).

If faced by a heavy workload or similar threats, individuals in a competitive environment might therefore be less inclined to collaborate and instead engage in taking credit for a colleague’s idea. In contrast, individuals in a cooperative environment might be eager to uphold the level of collaboration and resist the urge to take credit. Based on these thoughts, it can be predicted that:

Hypothesis 1: When unable to cope with their job demands, individuals in the competitive priming condition are more willing and individuals in the cooperative priming condition are less willing to take credit for someone else’s idea than individuals in the neutral condition are.

Following the same rationale, it can also be assumed that the willingness to take credit for someone else’s idea is likely to increase if the anticipated consequences for not doing it cause increasing costs to the individual, like e.g. getting a worse performance appraisal or being dismissed. In a study about an individual’s commitment to engaging in software piracy, i.e. stealing intellectual property, Glass and Wood (1996) found that the willingness to commit piracy is positively correlated with the likelihood of obtaining favourable outcomes in return. Despite the initially mentioned distinction between ideas and intellectual property, it seems reasonable to assume that the motivation to engage in such behaviour is similar in the case of stealing ideas: The negative consequences pose an additional source of strain to the employee that he or she will try to minimize in order to obtain a more favourable outcome:

Hypothesis 2: When unable to cope with their job demands, individuals in the competitive priming condition are more willing and individuals in the cooperative priming condition are less willing to take credit for someone else’s idea with escalating severity of anticipated consequences for not taking credit for the idea than individuals in the neutral condition.

2.3 Perceived ethical costs

The issue with claiming intellectual property rights for an idea was initially mentioned in the introduction. One could assume that taking credit for someone else’s idea is stealing and therefore violates legal or societal norms, i.e. causes ethical costs to the individual and society as a whole. Still, stealing always entails by definition a violation of another person’s rights of ownership in a fundamental way (Green 2007). With ownership and ideas being not a clear-cut relationship, taking credit for someone else’s idea allows a lot of subjectivity in perceiving ethical costs associated with that action, especially if one takes into account the culturally and historically conditioned mental model of ethical behaviour of an individual (Chen/Choi 2005). Individuals might simply fail to recognize taking credit for someone else’s idea as a moral issue (Glass/Wood 1996). Indeed, studies have shown that ethical decision-making is a highly complex process that is subject to a variety of personal dispositions. Various studies show that ethical reasoning is influenced e.g. by age, work experience, gender and cultural heritage (Chiu/Spindel 2009).

However, other authors mention that the decision-making process is influenced by an individual’s “contextual rationality” (Guy 1990: 34), i.e. the generation and selection of feasible alternatives is embedded in the context. In fact, Bekkers (2004) expects in a study about the stability of individual social value orientations, which also entail personal competitive and cooperative dispositions, that situational cues like primes have been responsible for the low stability of social value orientations in the study’s experiment.

Guy (1990) describes the ethical decision-making process in four stages. In the first stage, the problem and the desired outcomes in all relevant dimensions are determined. Stage two involves an evaluation which values might be affected in the process. In step three, an individual identifies all feasible alternatives considering all dimensions of the problem. Additionally, it is evaluated which values would be maximised or minimised by choosing a respective alternative. In the final step, the individual chooses the alternative, which maximises the value that is the most important to the individual and still solves the problem. The whole process is highly subjective and the weighting to what extent which values are affected by an alternative depends largely on the individual.

Assuming that situational cues influence the individual’s ethical decision-making, those who are exposed to competition primes might consider pro-social values (e.g. mutual trust in one’s team) as less important than pro-self values (e.g. success in one’s career) and vice versa for participants in a cooperative priming condition. For competitively primed individuals the decision to take credit for someone else’s idea would incur a minor, if any, norm violation, i.e. negligible ethical costs. As Sims (2003) notes that people who face competition may focus only on winning and accept behaving unethically as necessary mean toward one’s own end. Considering that individuals in a cooperative environment are likely to try to maintain a high level of collaboration – as mentioned above – taking credit for someone else’s idea would pose a severe norm violation. This would cause the individual to perceive higher ethical costs despite the inability to cope with one’s job demands.

Hypothesis 3: When unable to cope with their job demands, individuals in the competitive priming condition perceive less ethical costs and individuals in the cooperative priming condition perceive more ethical costs in taking credit for someone else’s idea than individuals in the neutral condition.

Due to the complexity of the ethical decision-making process, the present study also explores why an individual would think that taking credit for someone else’s idea is the right or not the right thing to do and also what that individual thinks what someone else in the same situation would consider as the right or wrong thing to do. Therefore, participants in the present study were asked four open-ended questions. These will be explained in more detail in the methodology section of this study.

2.4 Perception of others’ mindset

As outlined above in the section about subliminal priming, there is reason to assume that priming effects alter an individual’s perception about how other people think. Therefore, this study also seeks to confirm that individuals in a competitive or cooperative priming condition would perceive another person in the same situation – without any further information about this person – to have a similar competitive or cooperative mindset. Theoretically and empirically, this is widely supported by authors about social cognition (e.g. Tversky/Kahneman 1974). The basic thought is that once a certain concept becomes activated, i.e. more accessible, it is more likely that an individual uses this concept in the assessment of others (Srull/Wyer 1979). Tversky and Kahneman named this a “judgemental heuristic” (1974: 1127) or shortcut that allows an individual to reduce information overload or – in the case of an ambiguous person or situation – information scarcity before rendering a judgment about another person’s mindset or attitude. The individual uses the recently activated concept because it comes more easily to his or her mind.

Multiple studies have examined this phenomenon. In an experiment by Higgins et al. (1997) subjects were asked to describe another person in a hypothetical situation subsequently to priming activities. The researchers found that the subjects in the priming condition used words to describe the person’s personality traits that were similar to the words that have been used to prime the subjects. In another study, Hertel and Kerr (2001) found that priming the concept loyalty made their participants more likely to believe that the other participants in the same group would expect loyalty as well. Notably, the target which individuals perceive to have a similar mindset do not have to be the person or group the individuals are competing against or collaborating with (Stapel/Koomen 2005).

Based on these thoughts, the present study expects:

Hypothesis 4: Individuals in the competitive priming condition perceive others’ mindsets to be more competitive and individuals in the cooperative priming condition perceive others’ mindsets to be more cooperative than individuals in the neutral condition do.

3. Methodology

Participants and design. The participants were 64 postgraduate and PhD students recruited at the London School of Economics and Political Science. 22 of them were native English speakers, 34 of them were female. The design of the experiment was between-groups with the three priming conditions as independent variables (competition vs neutral vs cooperation). The dependent variables were the participant’s Willingness to take credit, Willingness to take credit with escalating severity of consequences for not choosing to take credit for the idea, the Perceived ethical costs in connection with the decision to take credit and the Perception of others’ mindset.

Procedure and materials. Participants came to a laboratory at the university in varying numbers and were welcomed by a male experimenter. They were told that they would participate individually in three unrelated activities that would deal with word processing skills and decision-making in a specific work situation. After the introduction, they were brought into the main room with 20 individual booths and were told to choose a booth at their own convenience. Every booth had been outfitted with an envelope containing the three activities prior to the participants entering the room. That way each participant was blindly assigned to one condition (competition, cooperation or neutral). While the participants were working on the activities, they were not able to see each other. The three activities included (1) two word processing tasks representing the priming procedure, (2) the scenario and related questions (unless noted otherwise, all items were on a 7-point Likert response format) and (3) a manipulation check item, items for the Core Self Evaluation (CSE) score (Judge et al. 2003) and the Altruism score (on a 5-point Likert response format) (Penner et al. 2005) and demographic information (control variables) and funnel debriefing questions. After participants completed all activities, they individually submitted the envelope at the experimenter’s desk, were handed their reward and were thanked for their participation. On average, it took the participants 40 minutes to complete all three activities. Each participant was rewarded with a chocolate bar plus the chance to win a 50 £ voucher.

Priming procedure. In the first activity, participants were confronted with two word-processing tasks, which were used as a subliminal priming technique. The first task was a word search puzzle as used in a study by Kleiman and Hassin (2001) and required the participant to find and mark words (provided in a list below) in a square of letters. In the competition and cooperation group, 6 of the 12 words in this list were prime words. In the competition group participants were confronted with words like battle, compete and rivalry, whereas in the cooperation group participants were exposed to words like commune, cooperation and group. In the neutral group words for objects were used (e.g. carpet, chair, picture) to avoid any undesired priming effects.

In the second priming task, participants were asked to rearrange 24 scrambled sentences that were used in studies by Kay and Ross (2003) and Smeesters et al. (2003). 16 of those sentences contained prime words similar to the words from the first task. All tasks used in activity 1 were identical to tasks that have been used in prior studies and proven to have a significant priming effect. Those prime words were included to produce a semantic and subliminal priming effect (Marslen-Wilson et al. 1994).

In activity 3, participants were asked to indicate on a 7-point Likert response format whether they feel rather competitive or cooperative. This item was included to provide a manipulation check to control for the effectiveness of the prime words.

Measuring the constructs. Activity 2 included a fictional scenario in which the participant takes on the role of an employee that is urged by his/her boss to have developed an idea by the end of the week for an important marketing campaign. Every employee on this team has to develop an individual idea for the meeting. Due to his/her heavy workload, this employee is not able to produce an idea. Coincidentally, he/she overhears a colleague called Sam talking on the phone about a very good idea for the marketing campaign. On the next team meeting, the employee has the chance to take credit for his/her colleague’s idea. Based on this scenario, the participants then answered a questionnaire containing 26 items related to the four different constructs.

To test Hypothesis 1, five questions about the participants’ basic willingness to take credit for the colleague’s idea, whether they would hesitate or have doubts to do so and how often they would think about it prior to the team meeting. For comparative reasons, another item, which asked participants about the percentaged chance with that he or she would take credit for the colleague’s idea, was included using a continuous scale from 1 to a 100 percent.

For Hypothesis 3, participants were then asked whether they agree with statements arguing that e.g. taking credit for the colleague’s idea would be the right decision and therefore would not cause any harm to anyone. The items were created by drawing on literature about ethical decision-making (Guy 1990). In addition, two open-ended items gave the participants the possibility to explain:

Can you think of any other arguments why it would be right to take credit for Sam’s idea?

Can you think of any other arguments why it would NOT be right to take credit for Sam’s idea?

Subsequently, participants were confronted with four items that asked once again about their willingness to take credit for this idea, but this time in anticipation of exacerbating consequences (including receiving a bad performance appraisal, not being promoted, being demoted and being fired) (Hypothesis 2), if the participant decided not to take credit and didn’t have an idea during the team meeting.

To measure the perception of other’s mindset (Hypothesis 4), the participants were asked to imagine that not they, but another random teammate would overhear the colleague’s phone call and could take credit for this idea during the next team meeting. No additional information about this other teammate was provided to keep it as ambiguous as possible. The five items for this scale included aspects of the previous three scales with the difference that the participants had to indicate the expected behaviour of that teammate in the same context.

Furthermore, the participants were asked to further explain in two open-ended questions:

Why do you believe that your teammate would think that taking credit for Sam’s idea is the right thing to do?

Why do you believe that your teammate would think that taking credit for Sam’s idea is NOT the right thing to do?

For a more reliable and consistent measurement of the four constructs, scales for each of the constructs were developed based on inter-item correlations. This was necessary because no other study in which similar measures had already been used could be found prior to this study. In the first step of the data analysis, items were therefore grouped in varying combinations in order to reach a preferably high internal consistency. As measure, Cronbach’s  (on standardized items) was used. This measure reflects how well each included item correlates with the sum of the other items in the scale. To reach an acceptable level of consistency, it is advised to accept only scales for which Cronbach’s  is at least  ≥ 0,7, but not much higher than 0,9. Not accepting very high  -values should minimize the threat of item redundancy in which case the included items basically all ask the same question in a different wording (Streiner/Norman 1989). For each construct a factor analysis with orthogonal varimax rotation was run to support the results from the reliability analysis with Cronbach’s . The factor analysis is usually used to discover patterns in the relationships between items (Darlington et al. 1973).

Control variables. Activity 3 included the 12-item CSE scale to measure the participant’s general job satisfaction, job performance, life satisfaction and locus of control (internal/external), and a 5-item scale to measure the participant’s level of Altruism. Furthermore, funnel-debriefing questions (Stapel et al. 2002) were included to ask about the participant’s awareness of the priming manipulation and purposes of the study. Two participants reported that they were aware of the manipulation and had to be discarded prior to the data analysis. Additionally, participants were asked to provide background information about gender, age, whether English was their native language, work experience, country of origin and the degree and programme they were currently studying. The control variables were included to provide additional explanations for the effects on the dependent variables.

Coding strategy for qualitative data. The qualitative data in the four above-mentioned open-ended items^[1] was evaluated by roughly following the steps of an inductive analysis (Patton 2002, Lee 1999). In the first step, data was repeatedly reviewed while more data was collected in further trials of the experiment. This served to assign first-order codes on emerging topics in a way that all obtained qualitative data could be coded into a category by the end of the data collection. If new codes emerged, data was re-coded where necessary. For the second step, a minimum frequency for all coding categories (i.e. the minimal aggregated number of times that codes in the same category were mentioned in the written answers) of 8 was determined as suggested in Anderson (2007). Any category below this threshold was excluded from the further analysis. This procedure allowed condensing the extensive collected data. In a third step, underlying themes for the categories were developed for the further discussion. The goal of this analysis was to explore major themes in the participants’ argumentation strategies why a decision to take credit for someone else’s idea would be right or wrong.

4. Results

4.1 Quantitative results

Developing scales. For each construct, ^[2] the scale with the highest a was chosen after comparison with the results from the factor analysis. For Willingness to take credit (Scale 1), three items were included (a = 0.895), asking about the likelihood the subject would think and consider taking credit for the colleague’s idea and how frequently he/she would think about it. For Willingness to take credit with escalating severity of consequences (Scale 2), three items were included (a= 0.926^[3] ), asking the subject stepwise if he/she would consider it under anticipation that not taking credit would result in not being promoted, a demotion or being fired. For Perceived ethical costs (Scale 3), four items were selected (a = 0.886), asking if taking credit is the right thing to do, if it is fair and ethical and if one should feel ashamed about it. For Perception of others’ mindset (Scale 4) , three items were included (a = 0.812), asking whether another teammate in the same situation would consider taking credit for the colleague’s idea, if he/she would do so under anticipation of a demotion or a dismissal (as a consequence of not taking credit) and how frequently this teammate would think about it. The factor analyses showed similar results (in the rotated factor matrix) and supported the scale development based on Cronbach’s a

Manipulation check. The item asking participants whether they feel rather competitive or cooperative (on a scale from 1 to 7, “1” being rather competitive, “7” being rather cooperative) was included to check for priming effects. Ideally, participants in the competitive condition should on average yield a lower mean than in the neutral condition and participants in the cooperative condition a higher mean than the control group. However, participants in the competitive condition (M = 4.22 ; SD = 1.31) did not feel significantly more competitive than in the neutral condition (M = 4.39 ; SD = 1.24), t (34) = - 0.392, p = 0.698, d = 0.13. Also, participants in the cooperative condition (M = 4.47 ; SD = 1.39) did not feel significantly more cooperative than they did in the neutral condition (M = 4.39 ; SD = 1.24), t(35) = - 0.195 , p = 0.846 , d = 0.06.

Hypotheses. Independent-samples t-tests were used to test the hypotheses. Unexpectedly, none of the four hypotheses was supported. Participants in the competitive condition (M = 2.43 ; SD = 0.94) were not more willing to take credit for someone else’s idea than in the neutral condition (M = 2.65 ; SD = 1.38), t (38) = - 0.581, p = 0.564, d = 0.19. Also, participants in the cooperative condition (M = 2.67 ; SD = 1.45) were not less willing than in the neutral condition, t (40) = - 0.038 , p = 0.970 , d = 0.01 (Hypothesis 1)^[4]. Participants in the competitive condition (M = 3.27 ; SD = 1.48) were also not more willing to take credit for someone else’s idea with escalating severity of consequences than the participants in the neutral condition (M = 3.25 ; SD = 1.79), t (38) = 0.032, p = 0.975, d = 0.01. Participants in the cooperative condition (M = 3.39 ; SD = 1.58) were also not less willing under escalating severity of consequences than they were in the neutral condition, t (40) = - 0.277 , p = 0.783 , d = 0.08 (Hypothesis 2). Similarly, participants in the competitive condition (M = 1.31 ; SD = 0.49) did not perceive less ethical costs than the participants in the neutral condition (M = 1.39 ; SD = 0.60), t (38) = - 0.433 , p = 0.668 , d = 0.15 . Participants in the cooperative condition (M = 1.42 ; SD = 0.91) did also not perceive more ethical costs than in the neutral condition, t (40) = - 0.136 , p = 0.892, d = 0.04 (Hypothesis 3). Finally, participants in the competitive condition (M = 4.12 ; SD = 1.16) did not perceive others’ mindsets to be more competitive than in the neutral condition (M = 4.12 ; SD = 1.07), t (38) = .000 , p = 1.000 , d = 0. Likewise, participants in the cooperative condition (M = 4.53 ; SD = 1.10) did not perceive others to have a more cooperative mindset than in the neutral condition, t (40) = -1.231 , p = 0.226 , d = 0.38 (Hypothesis 4).

Control variables and inter-scale correlations. Interestingly, Table 1 shows significant correlations between the four scales and certain control variables. As the priming conditions were found to have no significant effect on the outcome variables, some additional multiple linear regression analyses were conducted to further explore the correlations between the scales. As suggested by Allen and Bennett (2008), the necessary assumptions (multicollinearity, multivariate outliers, normality of residuals, linearity and homoscedasticity) for the regressions with the respective variables were tested. The original independent variables, the priming or neutral conditions, were not considered in these regressions. Control variables were included as predictors if a correlation with the respective scale existed and if deemed to be supported by theory. All predictor variables were entered directly into the regressions.

For Scale 1, a regression with the three other scales and Altruism (one-way ANOVA: F (4 , 57) = 17.6, p = 0.000) showed that the predictors explained 55.3 % of the variance. It yielded significant effects for Scale 2, t (57) = 3.32, b 0.367, p = 0.002, and for Scale 4, t (57) = 4.05 b = 0.422, p = 0.000. So, participants with higher scores on willingness to take credit with escalating severity and on perception of others’ mindsets were also more willing to take credit for someone else’s idea in the first place. Altruism and Scale 3 were not shown to have a significant effect on Scale 1.

For Scale 2, the predictors Scale 1, Scale 3, Scale 4, Altruism and Origin Anglo-Saxon and Western Europe countries explained 49,7 % of the variance (one-way ANOVA: F (5, 56) = 11.05, p = 0.000). Scale 1 ( t (56) = 3.36 ; b = 0.436; p = 0.001) and marginally also Origin Anglo Saxon and Western European countries ( t (56) = 1.98 ; b = 0.330 ; p = 0.053) had an effect. Therefore, participants who were more willing to take credit for someone else’s idea and originated from an Anglo-Saxon or Western European country were also more willing to take credit for this idea if they had to anticipate negative repercussion in case they would not take credit for the idea. The other scales and Altruism did not have a significant effect.

The regression for Scale 3 entailed the three other scales, Altruism, Origin Anglo- Saxon and Western Europe and Origin Asia as predictors. The two latter variables were included to see if there is a cultural difference regarding taking credit for someone else’s ideas and norm violations as suggested by Chin and Spindel (2009). Although the one-way ANOVA was significant for p ≤ 0.05 (F (6,55) = 2.36)), the predictor variables did not show any significant effects on Scale 3.

Finally, Scale 1 and Scale 2 were regressed against Scale 4. The ANOVA (F (2, 59) = 23.89 ; p = 0.000) showed that 44.7 % of the variance of Scale 1 could be explained by the predictors. Only Scale 1 yielded a significant effect, t (59) = 3.997, b = 0.504 ; p = 0.000. This showed that participants with a high willingness to take credit for someone else’s idea also perceive others to have a similar mindset as they do.

[...]

^[1] Multiple answers were possible.

^[2] Table 1 (Correlations) and Table 2 (Means and Standard Deviations) can be found in the appendix

^[3] Items were used for this scale despite the high a, because an item redundancy as consequence of alternating wording is not expected: The items address all negative, but different consequences.

^[4] The item which asked participants about the percentaged chance with which he or she would take credit for the colleague’s idea positively correlated with Scale 1 (b = 0.642, p = 0.000). The item was included to check if the participants’ answers for items with the Likert response format deviate from the answer on the same topic on a continuous scale.