**5. Results**

To verify whether teams with high CSE are more inclined to intuitive thinking rather than average and low CSE groups (H1), a one-way ANOVA was firstly implemented considering the di fferent CSE clusters (low, average, high) and their groups' results on the CRT test.

As shown in Table 1, there was a statistically significant di fference between groups as determined by the one-way ANOVA (F(2,27) = 80.510, *p* = 0.000). A Tukey post hoc test, shown in Table 2, revealed that high CSE groups were more inclined to the average CSE groups for intuitive thinking (−4.2 right answers compared with them; *p* = 0.00), but they were equally inclined to intuitive thinking with respect to low CSE groups (*p* = 0.964) (*p* = 0.989). So, H1 is *partly* verified.


**Table 1.** ANOVA Table—CSE on intuitive-reflective thinking.


**Table 2.** Post-hoc test—CSE on intuitive-reflective thinking.

\* The mean difference is significant at the 0.05 level.

In order to verify whether teams with high CSE are more inclined to be victims of the overconfidence bias compared to teams with low and average CSE (H2), a one-way ANOVA was implemented considering the di fferent CSE clusters (low, average, high) and their average teams' estimation of performance.

As shown in Table 3, there was a statistically significant di fference between groups as determined by the one-way ANOVA (F(2,27) = 295.962, *p* = 0.000). A Tukey post hoc test, shown in Table 4, revealed that high CSE groups were more inclined to the average CSE groups (+\$12,000 of net worth overestimation compared with them; *p* = 0.00) and low CSE groups (+\$30,000 of net worth overestimation compared with them; *p* = 0.00) to be victims of the overconfidence bias. So, high CSE groups were the ones that overestimated their performance more than other CSE groups, while low CSE groups were the ones that overestimated their performance least compared to other CSE groups. So, H2 is verified.

**Table 3.** ANOVA Table—CSE on overconfidence.




\* The mean difference is significant at the 0.05 level.

In order to verify whether teams with high CSE reach higher positive performance compared to teams with low and average CSE (H3), a one-way ANOVA was implemented (see Table 5) considering the di fferent CSE clusters (low, average, high) and their average teams' actual performance in terms of net worth.


**Table 5.** ANOVA Table—CSE on actual performance.

As shown in Table 5, there was a statistically significant difference between groups as determined by the one-way ANOVA (F(2,27) = 113.384, *p* = 0.000). A Tukey post hoc test, shown in Table 6, revealed that high CSE groups reached greater performance than low CSE groups (+\$50,800 of net worth overestimation compared with them; *p* = 0.00), but they reached lower performance than average CSE groups (who, on average, gained +\$53,000 of net worth). So, H3 is partly verified.

**Table 6.** Post-hoc test—CSE on actual performance.


\* The mean difference is significant at the 0.05 level.

### **6. Discussion and Implications**

This work offers a contribution to the debate about how CSE influences team decision-making processes. In order to do so, the research has been based on the analysis of four variables: CSE, intuitive/reflective thinking, overconfidence, and performance. Accordingly, three hypotheses have been formulated and tested—through one-way ANOVA and Tukey post hoc tests—on a sample population composed of 120 students while taking part in a simulation game in which they were asked to make decisions, in groups, acting in the role of the General Manager of a small-sized manufacturing firm.

The results only partially verified the first hypothesis. In fact, both teams with high and low CSE are equally inclined to intuitive thinking. This aspect deserves particular attention; indeed, if it is true that previous literature on the topic (e.g., Hiller and Hambrick 2005; Jordan et al. 2007; Claxton et al. 2015) has already recognized the existence of a link between high levels of CSE and intuitive thinking, it is surprising to see that also groups with a low level of CSE have the same predisposition to intuitive thinking. Therefore, the high self-consideration by individual/groups seems to lead to the same consequence of having low self-consideration: being inclined to intuitive thinking. Despite that, this result can be considered in line with the study of Rudolph et al. (2009) (see also the similar one by Cristofaro 2016) who found, through a computer-based simulation on data collected in clinical decision making, that the decision-making behavior that usually leads to wrong decision options is usually carried out by people that take too little, or too much, time to make a decision. These results complete these studies by providing the explanation at a personal trait level of why this different decision behavior occurs. The results totally confirm the second hypothesis. In fact, teams with a high CSE level were more predisposed to both average and low CSE groups of falling into the overconfidence trap. This is perfectly in line with previous literature on these topics (e.g., Baumeister et al. 1993; Kramer et al. 1993; Zacharakis and Shepherd 2001) and highlights how high CSE groups, overestimating their capabilities, tend to "destroy" their decision-making ability (Abatecola et al. 2018; Abatecola and Cristofaro 2019). Lastly, results only partially verified the third hypothesis. In fact, if it is true that high CSE groups reached higher positive performance compared to low CSE groups—as already demonstrated by previous researches (e.g., Hiller and Hambrick 2005; Cristofaro 2017b); it is also (surprisingly) true that average CSE groups reached more positive performance compared to high CSE groups. By linking the results of the test of the first and third hypothesis, it emerges that groups with a high level of intuitive thinking (corresponding to the ones having high or low CSE scores) are not the best performers in decision-making terms. This result contributes to the debate on the consequences of intuition in managemen<sup>t</sup> decision making. In particular, it supports a stream of prior results highlighting that intuitive thinking leads to poor quality of decisions (Elbanna et al. 2013), which consequently leads to poor firm performance (Goll and Rashe 1997). This happens, as suggested by Elbanna et al. (2013), because intuitive decision makers are impatient with routine or details—i.e., they have a poor systemic search for information—and are pushed, by their nature, to quickly reach conclusions and to ignore negative problems. However, despite reinforcing this stream of works, another important one has found a positive relationship between intuition and firm performance; such as in developing technologies, sizing new opportunities, and providing e ffective responses to crises (Bullini Orlandi and Pierce 2020). What is the determinant for the success of intuitive thinking seems to be, according to Bullini Orlandi and Pierce (2020), the dynamicity of the industry environment; indeed, in cases of highly dynamic and turbulent environments with the presence of real-time data, intuitive thinking is preferred rather than the reflective one. In sum, despite the confirming results of this work in substantiating a negative role of intuitive thinking in decision making, the rapid change of contextual and environmental variables can lead to positive e ffects of intuitive thinking—in line with the ecological rationality approach (Gigerenzer and Brighton 2009).

Thanks to this work, the results provided extend those of cited and discussed contributions, offering a more solid base for the highlighted assumptions by providing an empirical assessment of an established personal trait variable, CSE. Indeed, cited studies only assumed this relationship looking at one of the four pillars of the CSE, such as self-esteem (e.g., Kramer et al. 1993; Baumeister et al. 1993; Jordan et al. 2007), or by providing a theoretical explanation (Hiller and Hambrick 2005; Abatecola et al. 2018; Abatecola and Cristofaro 2019). Moreover, this is the first study that investigates the outlined relationship at a group level, practically overcoming the limits of the others that considered only the individual level of analysis.

Based on the exposed results, some important managerial implications can be derived for practitioners, especially the younger ones (Millennials) with a similar age to the sampled students. Firstly, as CSE is a personal trait, it is not possible to suppress it in an absolute sense; or, at least, it is very di fficult in a short or medium range timescale. However, practitioners can reduce its value to an average by composing an ad hoc team. They can, in practice, bring together people with di fferent CSE levels so that the CSE average will result as "acceptable"—namely, the score of their CSE needs to be between −8 to +8 points—there will be a balance between intuitive and reflective thinking in the team and, thus, the possibility of avoiding the overconfidence trap and the opportunity of achieving satisfactory performance. Thus, it is fundamental that human resource managers track the CSE level of each individual within organizations. In this way, they also have the possibility of appropriately suggesting—to department or unit heads—the "best team composition" to achieve better decisional and organizational performance. At the same time, if an organization needs to quickly respond to internal or external pressures, such as identifying a commercial strategy to counteract a sudden, huge price cut of a competitor, composing teams of individuals with exclusively high or low CSE levels can be beneficial for the production of intuitive (and quick) responses.

Despite the rigor with which the experiment was conducted, this study has some limitations, which also represent fruitful starting points for future researches on these topics. Firstly, the sample population is composed by students, which means that most have little or no work experience. A second and connected limit is determined by the fact that the experiment was conducted during course lectures; therefore, participants applied their strategies in an environment in which they were comfortable. In particular, they acted without being subjected to external pressure and this could have biased, for example, their risk orientation—with obvious consequences on the potential performance that would have been attained with another 'purer behavior'. A third and no less important limitation of this research arises from the fact that some variables have not been controlled, for example, work experience and risk orientation, although in real life they obviously a ffect the behavior and attitude of managers in managing situations and acting upon them. Finally, a fourth and last research limit results from the structure of the platform provided to the students. Indeed, like all simulation games, there are limitations about the representation of all the variables that are played within real world choices. In sum, despite the fact that respondents' decisions have been implemented (and tracked by the researchers) within a very well simulated environment (i.e., the simulation game), which is better than paper-based cases, sampling managers in real life situations would be necessary for extending the generalizability of these results. Future research can surely solidify the results of this work by avoiding the outlined limits; moreover, following Abatecola et al. (2018), future research avenues could investigate if there are other cognitive distortions (e.g., self-serving, emotion and cognition collision) in managerial decision making—in addition to overconfidence—that are linked to and influenced by CSE. Yet, the results of this work should be highly considered by scholars that want to deepen the antecedents of intuition and its outcomes in strategic decision-making (see Elbanna et al. 2013). In particular, future research can test whether trait variables—CSE above all—have more weight than contextual and environmental ones in determining the thinking style of decision makers. Last but not least, the link between CSE and performance could be deepened also by looking at the emotional answers that high, average, and low CSE groups have when facing some decisional situations. These can reinforce the debate and operationalization of intuitive thinking which, nowadays, still does not take the role of emotions in substantiating intuitive answers into very high consideration.

The originality of this work is threefold. Firstly, we are not aware of any study that has investigated the influence exercised on decision-making processes by CSE in relation to reflective/intuitive thinking, overconfidence, and performance. Secondly, these relationships have always been investigated at the "individual level". However, most of the tasks in an organization are performed at team level and decisions are rarely made individually; thus, this study adopts a "team/group level" to evaluate the importance and the influence of these aspects on organizations. Thirdly and lastly—despite the limitations previously exposed—this paper gives evidence of what really happens within organizations without resorting to case studies but, instead, is shown through the direct participation of individuals—i.e., students—acting as General Managers.

**Author Contributions:** Conceptualization, M.C.; methodology, M.C. and P.L.G.; formal analysis, M.C.; data curation, M.C.; writing—original draft preparation, M.C. and P.L.G.; writing—review and editing, L.L.; visualization, M.C.; supervision, M.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **Appendix A Core-Self Evaluation Scale**


### **Appendix B Seven-Item Cognitive Reflection Test**

