Core Self-Evaluations, Self-Leadership, and the Self-Serving Bias in Managerial Decision Making: A Laboratory Experiment

Round 1

Reviewer 1 Report

Comments for the Authors

This study explores the relationship between Core Self-Evaluations (CSE) and self-leadership, as well as the relationship between self-leadership and the Self-Serving Bias (SSB).  This study is among the first to examine the relationship between self-leadership and CSE.  Exploring CSE as a potential antecedent of self-leadership makes an important contribution to the self-leadership literature.  Beyond a few studies that have explored empowering leadership (e.g., Amundsen & Martinsen, 2015) and some broad personality traits (e.g., Bailey et al., 2016), very little research has explored the antecedents of self-leadership.  As you describe in the paper, the outcomes of self-leadership have been explored extensively, but research examining self-leadership predictors is more limited.  Your study has potential for contributing to our understanding of self-leadership by probing an additional possible antecedent of self-leadership.  This study is also among the first to explore possible deleterious effects of excessive use of self-leadership strategies.  Just as with a vitamin that may be beneficial at certain levels but toxic at higher levels, this study makes a potentially important contribution by examining possible the SSB that could arise from excessive self-leadership.

While there is much to like about this paper, a number of issues prevent the paper from being published in the Administrative Science in its current form.  I provide a detailed discussion of these concerns below and provide some suggestions for revising the paper.  

1. Specific vs. general self-efficacy distinction.  On p. 2 you introduce the concept of self-efficacy.  Self-efficacy has been conceptualized in two ways in the literature.  As originally introduced by Bandura (e.g., 1986), self-efficacy is task specific belief in one's capabilities to perform.  Later, others proposed a more generalized (or trait) self-efficacy that refers to one's general disposition regarding one's overall capabilities to perform across many achievement situations (e.g., Chen et al., 2001).  This distinction is very important in the current study.  As you note, self-efficacy has been proposed as a primary outcome of self-leadership (Neck & Houghton, 2006).  However, generalized self-efficacy is likely to be an antecedent of self-leadership.  In the discussion on p. 12, you do a very nice job describing about how trait self-efficacy as part of CSE can increase self-leadership, which in turn can increase task specific self-efficacy in a co-evolutionary fashion.  It is important that you set up this discussion by clearly distinguishing between the two types of self-efficacy earlier in the manuscript.
2. CSE as an antecedent of self-leadership contribution.  One of the major contributions of this paper is providing evidence in support of CSE as an antecedent of self-leadership.  You could highlight this contribution more clearly, both in the introduction and in the discussion.
3. Clarification of terminology. Throughout the paper, you use several terms somewhat interchangeably, including self-serving bias, self-attribution bias, and overconfidence.  For example, in H2 you use "self-serving bias," in your measurement section you use "self-attribution bias," and in Table 4 you use "overconfidence."  I suggest that you standardize your terminology more across the manuscript.
4. Typos. On p. 6, the section 2.3 heading is incorrect.  Self-leadership was section 2.2.  This section should be "Self-Serving Bias" (or whichever term you decide to use; see point 3 above).  On p. 9, the second part of H2 is incorrect.  It should be "low and average self-leadership." On p. 9, I'm not sure from where the quote in the middle of the page comes.  It seems to come from Neck and Houghton (2006), but it is not from that paper and that paper does not have a p. 1160.  Please check this citation.  On top of p. 10, I think there is a "CSE" in the second line that should not be there.
5. Supporting arguments for H2.  I don't think the case was clearly made in support of H2.  Consider some theoretical support and/or stronger logical rationale.
6. Clearer explanation of results.  I am confused by the results shown in Table 2.  All of the mean differences are shown as significant, yet the CIs for four of the six pairs contain zero.  Please explain these results more clearly.  Similarly, I am unsure how to interpret Table 4.  The mean differences shown are in the thousands.  Shouldn't the means fall between 4-36, the range of the scale?  Please explain the mean differences shown.
7. Comparison of findings. In your discussion on p. 13, you might wish to compare your findings with recent evidence suggesting that too much task specific self-efficacy may be harmful to performance (e.g., Vancouver et al., 2002).
8. Implications of self-leadership to self-serving bias relationship. Throughout the manuscript you seem to imply that a high level of self-serving bias, in which a person attributes failure to external circumstances and success to internal circumstances, is bad thing.  But is this always the case?  Why is this the case?  For example, research on explanatory styles (e.g., Gillham et al., 2001) suggests that people with an optimistic explanatory style that includes external attributions (it's not all my own fault) may perform better than people with a pessimistic explanatory style who make internal attributes (it's all my own fault).  Consider expanding your discussion of the implications of H2 at the top of p. 13.  You do a nice job at the bottom of p. 13 discussion managerial implications, but these are predicated upon the notion that high SSB is a bad thing.  You need to lay the groundwork earlier.

I really enjoyed reading this paper.  Good luck moving forward with it!

References:

Amundsen, S., & Martinsen, Ø. L. (2015). Linking empowering leadership to job satisfaction, work effort, and creativity: The role of self-leadership and psychological empowerment. Journal of Leadership & Organizational Studies, 22(3), 304-323.

Bailey, S. F., Barber, L. K., & Justice, L. M. (2018). Is self-leadership just self-regulation? Exploring construct validity with HEXACO and self-regulatory traits. Current Psychology, 37(1), 149-161.

Bandura, A. (1986). The explanatory and predictive scope of self-efficacy theory. Journal of social and clinical psychology, 4(3), 359-373.

Chen, G., Gully, S. M., & Eden, D. (2001). Validation of a new general self-efficacy scale. Organizational research methods, 4(1), 62-83.

Gillham, J. E., Shatté, A. J., Reivich, K. J., & Seligman, M. E. P. (2001). Optimism, pessimism, and explanatory style. In E. C. Chang (Ed.), Optimism & pessimism: Implications for theory, research, and practice (p. 53–75). American Psychological Association.

Vancouver, J. B., Thompson, C. M., Tischner, E. C., & Putka, D. J. (2002). Two studies examining the negative effect of self-efficacy on performance. Journal of applied psychology, 87(3), 506.       

Author Response

Ms. Ref. No.: admsci-917765

Core self-evaluations, self-leadership, and the self-serving bias in managerial decision making: A laboratory experiment

Administrative Sciences

Replies to reviewer 1

Dear Reviewer and Editor,

We would like to express our gratitude for your comments in this round of review. We thank the editors and reviewers for the guidance to refine our work.

Please find attached the revised version of our manuscript entitled “Core self-evaluations, self-leadership, and the self-serving bias in managerial decision making: A laboratory experiment”.

Thanks for reading and taking into consideration our manuscript. Your comments were well-received and the relative suggestions have been incorporated into this improved version. Your work truly enables us to significantly improve our own.

In the following section, we have reported the comments and our replies. In the text, all of the changes are highlighted in yellow.

***

Reviewer: 1

Recommendation: Minor Revision

General comment

This study explores the relationship between Core Self-Evaluations (CSE) and self-leadership, as well as the relationship between self-leadership and the Self-Serving Bias (SSB). This study is among the first to examine the relationship between self-leadership and CSE. Exploring CSE as a potential antecedent of self-leadership makes an important contribution to the self-leadership literature. Beyond a few studies that have explored empowering leadership (e.g., Amundsen & Martinsen, 2015) and some broad personality traits (e.g., Bailey et al., 2016), very little research has explored the antecedents of self-leadership. As you describe in the paper, the outcomes of self-leadership have been explored extensively, but research examining self-leadership predictors is more limited. Your study has potential for contributing to our understanding of self-leadership by probing an additional possible antecedent of self-leadership. This study is also among the first to explore possible deleterious effects of excessive use of self-leadership strategies. Just as with a vitamin that may be beneficial at certain levels but toxic at higher levels, this study makes a potentially important contribution by examining possible the SSB that could arise from excessive self-leadership.

While there is much to like about this paper, a number of issues prevent the paper from being published in the Administrative Science in its current form. I provide a detailed discussion of these concerns below and provide some suggestions for revising the paper.

Reply to general comment: Thank you for having appreciated the contribution, really! We have worked hard on the issues you raised and we really think we have fixed them. Please, see the answers to the following comments.

Additional Questions

Comment 1: Specific vs. general self-efficacy distinction. On p. 2 you introduce the concept of self-efficacy. Self-efficacy has been conceptualized in two ways in the literature. As originally introduced by Bandura (e.g., 1986), self-efficacy is task specific belief in one's capabilities to perform. Later, others proposed a more generalized (or trait) self-efficacy that refers to one's general disposition regarding one's overall capabilities to perform across many achievement situations (e.g., Chen et al., 2001). This distinction is very important in the current study. As you note, self-efficacy has been proposed as a primary outcome of self-leadership (Neck & Houghton, 2006). However, generalized self-efficacy is likely to be an antecedent of self-leadership. In the discussion on p. 12, you do a very nice job describing about how trait self-efficacy as part of CSE can increase self-leadership, which in turn can increase task specific self-efficacy in a co-evolutionary fashion. It is important that you set up this discussion by clearly distinguishing between the two types of self-efficacy earlier in the manuscript.

Reply to Comment 1: Thank you for this comment. You are right in saying that we skipped the distinction between generalized self-efficacy, conceived as a trait, and specific self-efficacy, as originally conceived by Bandura. In order to fill this gap, we have added the following part within the new theoretical sub-section 2.1 of the manuscript – built in order not to overload section 2.2 about CSE and offering, at the same time, the necessary theoretical premises about self-efficacy and self-esteem (the latter has been asked by reviewer 2). In particular:

“Albert Bandura (1986) defined self-efficacy as directly related with individuals’ beliefs in their own personal abilities, both at physical and mental levels, such as how well they will be capable of managing their skills and cognitive abilities, and how much effort they will be able to carry in order to achieve a predetermined goal. According to Bandura et al. (1999), individuals that have a high level of self-efficacy are more expected to perceive challenging duties as something to be mastered instead of perceiving them as something to be bypassed. Consequently, individuals with a high level of self-efficacy are more inclined to work hard in order to accomplish their duties, even if they have to keep up their endeavors in the long term. However, according to Bandura (1977), the perception of self-efficacy mainly depends on the feedback received from other individuals – as previously stated by Weinberg et al. (1979) – thus: if people receive negative feedback about their physical and mental efficiency, they will perform inadequately, while, if they receive positive feedback, they will be likely to show outstanding performances. Yet, this feedback, which determines the perceived self-efficacy, varies according to the situational circumstances of the task that the individual is facing; indeed, Bandura stated that: “The extent to which people will alter their perceived self-efficacy on the basis of performance feedback will depend on such factors as the difficulty of the task, the amount of effort they expend, the amount of external aid they receive, the situational circumstances under which they perform, and their mood and physical state at the time” (Bandura 1986; p. 363). In contrast to this task-related perception of self-efficacy, other scholars have subsequently conceived self-efficacy in a more general dimension that has been called General Self-Efficacy (GSE) (Gardner and Pierce 1998; Judge et al. 1998), which, within this work, is the meaning adopted when speaking about self-efficacy. GSE can be described as the attitude of an individual’s belief in one’s global capacity to face challenges across a broad range of circumstances (Judge et al. 1998; Scherbaum et al. 2006). This stream of studies assumes that GSE is able to catch distinctive variation among people in their propensity to see themselves as able to manage and achieve different goals in different contexts; thus, GSE is a ‘situation-independent context belief’ rather than a fluctuant perception of their own abilities. However, GSE and specific self-efficacy are linked by a high inter-relationship (Judge et al. 1998; 1999). Indeed, various studies demonstrated that GSE and specific self-efficacy are positively correlated as well as the former having been found to be a determinant of the latter (e.g., Chen et al. 2001; Scherbaum et al. 2006)”.

Comment 2: CSE as an antecedent of self-leadership contribution. One of the major contributions of this paper is providing evidence in support of CSE as an antecedent of self-leadership. You could highlight this contribution more clearly, both in the introduction and in the discussion.

Reply to Comment 2: Thank you for this comment. Within the introduction has been reported that: “Hence, it is posited – in accordance with the established literature that assumed and proved the antecedent role of personality traits with regard to behavior (Ajzen 2005) – that CSEs is an antecedent of self-leadership. In this regard, and due to the fact that no other contributions, to the best of the authors’ knowledge, have been produced on the relationship between CSEs and self-leadership, the following research question emerges as to understand how CSEs work as an antecedent of self-leadership: “How is self-leadership related with CSE?”. Then, about the value added of the paper reported within the introduction has been added: “In particular, it is advanced that CSEs is an antecedent of self-leadership; thus, the degree of the former personality trait influences the process through which individuals control their own behavior, influencing and leading themselves through.”

Within discussion and conclusions I added the following:

“From that, it emerges that CSE is a personality trait that works as an antecedent of self-leadership, reinforcing – by studying a personality trait that has not been ever considered in self-leadership literature – the assumption that: “self-leadership represents a unique constellation of strategies that are founded upon, related to, and yet distinct from these various theories as well as from various personality traits” (Neck and Houghton 2006; p. 275)”.

Comment 3: Clarification of terminology. Throughout the paper, you use several terms somewhat interchangeably, including self-serving bias, self-attribution bias, and overconfidence. For example, in H2 you use "self-serving bias," in your measurement section you use "self-attribution bias," and in Table 4 you use "overconfidence." I suggest that you standardize your terminology more across the manuscript.

Reply to Comment 3: Thank you for these comments. Due to the fact that there is a strong and extensive stream of literature speaking about the Self-Serving Bias by using the word ‘Self-attribution bias’ we cannot use just one terminology instead of the other. Due to that, we decided to say the following within the introduction: “In particular, due to the already discovered relationship between self-efficacy and Self-Serving Bias (SSB) – also called Self-attribution bias (used interchangeably within the text) – thus the internal attribution of successes and external attribution of failures (see Watt and Martin 1994; Kalish and Luria 2016), a second research questions emerges as follows: “How does self-leadership influence the attribution of successes/failures?”.

At line 162 and 276 we eliminated the word ‘overconfidence’ because not necessary, while we replaced the word ‘overconfidence’ with SSB within the heading of Table 4. It was a typo.

Comment 4: Typos. On p. 6, the section 2.3 heading is incorrect. Self-leadership was section 2.2. This section should be "Self-Serving Bias" (or whichever term you decide to use; see point 3 above). On p. 9, the second part of H2 is incorrect. It should be "low and average self-leadership." On p. 9, I'm not sure from where the quote in the middle of the page comes. It seems to come from Neck and Houghton (2006), but it is not from that paper and that paper does not have a p. 1160. Please check this citation. On top of p. 10, I think there is a "CSE" in the second line that should not be there.

Reply to Comment 4: Thank you for these comments.

We substituted the heading of section 2.3 with the correct heading ‘Self-Serving     Bias’.

We substituted the second part of H2 with the correct wording: “low and average self-leadership”.

The quote, as you correctly said, did not come from Neck and Houghton but from p. 1260 of Edmondson and McManus (2007)--> Edmondson, Amy C., and Stacy E. McManus. 2007. Methodological fit in management field research. Academy of Management Review 32: 1246–1264. We have fixed it.

You are right, at line 457 there was a CSE that should not have been there. We eliminated it; thank you.

Comment 5: Supporting arguments for H2. I don't think the case was clearly made in support of H2. Consider some theoretical support and/or stronger logical rationale.

Reply to Comment 5: Thank you for this comment. You are right in saying that H2 was not so incisively supported. So, we have tried making some stronger connections through the pieces of literature about self-efficacy and self-enhancement (related with SSB) that have been already reported with the concept of self-leadership. The section has been partly rewritten and now it reads as follows:

“Silver et al. (1995) investigated the link between job performance and outcome attribution – the basis of SSB – through the mitigating effect of self-efficacy. In doing so, they delivered practical problems, similar to those happening during the Graduate Management Admission Test (GMAT) exams, to a sample of 100 undergraduate students. The results of this investigation revealed the importance for reaching high levels of job performance and of how people understand the origin of their performance. In fact, it appeared that different self-efficacy expectations emerge in divergent acknowledgments of performance, directly affecting the guess: outstanding performances are associated with subjective aspects that boost the self-beliefs of efficacy; negative performance, instead, leads to a decrease in self-belief (on this point, see also Watt and Martin 1994). Thus, from the study carried out by Silver et al. (1995), we can hypothesize that when people truly believe in their capabilities – thus demonstrating a high level of self-efficacy (at the basis of the CSE) – and they achieve successful goals both in their private and professional lives, they are automatically more likely to experience the SSB bias rather than people with an average-medium level of CSE. Yet, Campbell and Sedikides (1999), who collected data from over 6,900 participants, studied the effects of the SSB in the workplace, also investigating a wide range of moderators, such as self-esteem, locus of control, gender, and job position. In particular, this study showed that people, according to their level of self-esteem, react in a different manner when they have to face negative outcomes: those with a high level of self-esteem are likely to behave more defensively (Baumeister et al. 1993, 1996; Blaine and Crocker 1993), embodying this defensiveness via an emphasized level of SSB. In contrast, people with a small degree of self-esteem did not fall into the SSB. Another important moderator element investigated by Campbell and Sedikides (1999) concerns success ambition. In fact, these scholars demonstrated that people with a large degree of success ambition – a personal orientation at the basis of self-leadership – are affected by high levels of self-threat, leading them to be more likely to fail in SSB. Implicitly, stemming from the fact that self-esteem is one of the four dispositions at the basis of the CSE trait, we can assume that people with a high level of CSE are inclined to allocate the responsibility of their failures to external factors.

An important element to bear in mind when scholars want to assess the inclination of individuals to be victim of SSB concern self-enhancement – which represents a whole set of mental activity actions aimed to establish, preserve, or augment a positive self-image (Kwan et al. 2004) – at the basis of the self-leadership concept. Hence, some scholars investigated the impact of self-enhancement on the private and professional lives of people (Dunning 1999; Taylor and Brown 1988). For example, Sedikides (1993) showed that people pursue self-knowledge, a pillar of self-enhancement, in order to achieve laudatory outcomes, leading to internally attributing successes and externally attributing failures. Hence, from the work of Sedikides (1993), it can be again derived that people with high self-enhancement, thus strongly oriented to self-leadership, tend to internally attribute success and externally attribute failures.

Finally, it is worth noticing that the SSB is a viable trait to manifest both self-enhancement and self-protection (Campbell and Sedikides 1999; Mezulis et al. 2004). In fact, Sedikides and Alicke (2012) argued that this aspect, on the one hand, requires taking more consideration than one merits for positive results (i.e., self-enhancement) and, on the other hand, fails to take sufficient liability for the negative results (i.e., self-protection). Thus, the attempt to boost the benefits of positive results of their own actions is directly connected with self-enhancement, with the consequential increase in the perception that individuals have of their own abilities. On the contrary, attempting to deflect responsibility for negative results is correlated with self-protection, which can be perceived as the aim of not reducing their self-esteem and, consequently, the evaluation that people make about them. From what has just been reported and stemming from the fact that: “A major objective of all self-leadership strategies [...] is the enhancement of self-efficacy perceptions in advance of higher performance levels” (Neck and Houghton 2006; p. 279), it can be derived that:

H2: Individuals with great self-leadership are more inclined to be victims of the self-serving bias compared to individuals with low and average self-leadership”.

Comment 6: Clearer explanation of results. I am confused by the results shown in Table 2. All of the mean differences are shown as significant, yet the CIs for four of the six pairs contain zero. Please explain these results more clearly. Similarly, I am unsure how to interpret Table 4. The mean differences shown are in the thousands. Shouldn't the means fall between 4-36, the range of the scale? Please explain the mean differences shown.

Reply to Comment 6: Thank you for this comment. All the relationships among CSE, divided in high-average-low, and self-leadership are significant and this has been added as explanation within results through adding the following text: “Moreover, apart from the relationships just highlighted, it can be derived from Table 2 that: i) managers with average CSE have greater self-leadership compared to individuals with a low rating (+2.6; p = 0.00), but lower than self-leadership managers with high CSE (-4.9; p = 0.00); and ii) managers with low CSE have lower self-leadership compared to both individuals with an average rating (+2.6; p = 0.00) and high CSE (-6.1; p = 0.00)”. The interest of the paper is in the significant difference between high CSE and low and leverage CSE in terms of self-leadership, however, also these relationships have been found as significant and have been pointed out. About the confidence intervals, you are right. It was a problem of us in reporting results from the statistical software. Now, confidence intervals are correct. Thank you a lot for having spotted it.

About table 4, we had a problem from reporting results from the Italian version of the statistical software – in Italy comas have the meaning of full stops in the English format. We have fixed it by reporting the correct results. About the range of values within table 4, they can exceed the value present in the scale (-4, 36) due to the fact that the Tukey test aims at identifying the existing differences among values and not at spotting the values of the groups under investigation. In other words, recurring to a practical example, if you look at the last row: High SL compared to Low SL, the value pointed out in the column Mean Difference (I-J) is 17.000*. This means that high SL individuals have a value of SSB that is greater than the one of Low SL people by 17 points; thus SL people can have 22 as a SSB value and Low SL individuals have 5; there exists a difference of 17 points from one value to another.

Comment 7: Comparison of findings. In your discussion on p. 13, you might wish to compare your findings with recent evidence suggesting that too much task specific self-efficacy may be harmful to performance (e.g., Vancouver et al., 2002).

Reply to Comment 7: Thank you for this comment. We have followed your suggestion and we included within discussions a part in which we debate our results in the light of Vancouver et al. (2002) and other similar works (Powers, 1973; 1991). In particular:

“This result reinforces the stream of studies that advanced self-efficacy, one of the four pillars of CSE, has a negative facet due to conducting cognitive errors (see Powers 1973; 1991) – in contrast with the works of Bandura (1986; 1997; Bandura et al. 1999) that assumed a positive influence of self-efficacy on task performance. Indeed, Vancouver et al. (2002), through two laboratory experiments with 87 and 104 undergraduates, respectively, found that self-efficacy has a negative influence on performance, explained by the fact that self-efficacy leads to the cognitive error of overconfidence, which increases the likelihood of committing logic errors during tasks. However, in the proposed contribution, the link between CSE, self-leadership, SSB and task performance has not been considered; so, self-efficacy is supposed as a ‘negative’ influence of high self-leadership because it leads to a cognitive error substantiated by the SSB – without due consideration of task performance. However, other works suggested that people with an optimistic explanatory style, oriented to the external attribution of failures, can perform better than people with a pessimistic explanatory style, who make internal attributes of failures (Gillham et al. 2001). From that, future research can expand the results of this work to try to measure also the task performance of individuals present in the sample”.

Comment 8: Implications of self-leadership to self-serving bias relationship. Throughout the manuscript you seem to imply that a high level of self-serving bias, in which a person attributes failure to external circumstances and success to internal circumstances, is bad thing. But is this always the case? Why is this the case? For example, research on explanatory styles (e.g., Gillham et al., 2001) suggests that people with an optimistic explanatory style that includes external attributions (it's not all my own fault) may perform better than people with a pessimistic explanatory style who make internal attributes (it's all my own fault). Consider expanding your discussion of the implications of H2 at the top of p. 13. You do a nice job at the bottom of p. 13 discussion managerial implications, but these are predicated upon the notion that high SSB is a bad thing. You need to lay the groundwork earlier.

Reply to Comment 8: Thank you for this comment. You are right in your affirmation; indeed, there is a huge stream of literature about the ecological rationality that postulates that the occurrence of biases is not always detrimental for decision making processes within organizations. From that, we have enriched our discussion taking into account this point as follows:

“However, in the proposed contribution it has not been considered the link between CSE, self-leadership, SSB and task performance; so, it is supposed a ‘negative’ influence of high self-leadership because leading to a cognitive error substantiated by the SSB – without due consideration of task performance. However, other works suggested that people with an optimistic explanatory style, oriented to the external attribution of failures, can perform better than people with a pessimistic explanatory style, who make internal attributes of failures (Gillham et al., 2001). From that, future research can expand the results of this work trying to measure also the task performance of individuals present in the sample”.

[...]

“However, it is also true that, under some circumstances, biases can lead to results and performances that are greater than the ones reached through the use of other ‘more rational’ decision-making tools (e.g., inferential statistics) (Gigerenzer et al. 1999; Luan et al. 2019). Hence, if biases occur within decision-making processes due to the personal characteristics of decision makers, they should be individually assessed in relation to the performed task to identify if their occurrence was totally negative or not”.

References to add:

Amundsen, S., & Martinsen, Ø. L. (2015). Linking empowering leadership to job satisfaction, work effort, and creativity: The role of self-leadership and psychological empowerment. Journal of Leadership & Organizational Studies, 22(3), 304-323.

Bailey, S. F., Barber, L. K., & Justice, L. M. (2018). Is self-leadership just self-regulation? Exploring construct validity with HEXACO and self-regulatory traits. Current Psychology, 37(1), 149-161.

Bandura, A. (1986). The explanatory and predictive scope of self-efficacy theory. Journal of social and clinical psychology, 4(3), 359-373.

Chen, G., Gully, S. M., & Eden, D. (2001). Validation of a new general self-efficacy scale. Organizational research methods, 4(1), 62-83.

Gillham, J. E., Shatté, A. J., Reivich, K. J., & Seligman, M. E. P. (2001). Optimism, pessimism, and explanatory style. In E. C. Chang (Ed.), Optimism & pessimism: Implications for theory, research, and practice (p. 53–75). American Psychological Association.

Vancouver, J. B., Thompson, C. M., Tischner, E. C., & Putka, D. J. (2002). Two studies examining the negative effect of self-efficacy on performance. Journal of applied psychology, 87(3), 506.    

Thank you for these references, that we added within the list of references as well as cited and discussed in different parts of the text mainly to support the difference between general and specific self-efficacy.

***

Finally, we again thank the Editors and Reviewers for their time, which has resulted in highly thoughtful reviews and invaluable guidance for the authors. The paper has evolved – structurally and in terms of content – and we anticipate that the improvements will meet and exceed expectations.

We hope that the revisions in the manuscript, and our accompanying responses, will be sufficient to make our manuscript suitable for publication in Administrative Sciences.

We shall look forward to hearing from you at your earliest convenience.

Yours sincerely,

The Authors

Reviewer 2 Report

High levels of self efficacy do NOT cause people to blame failures on external causes – this is confusing in this statement. - This work reinforces the stream of (few) studies that considers a high level of CSE and 81 self-leadership as not always desirable for managerial decision-making processes and consequent 82 job performance (e.g., Barker 1993; Langfred 2004; Hiller and Hambrick 2005).

Is self-leadership autonomy?  What is the relationship between them???

Line 92 – “good ness” – better “ accuracy”

Line 149 – replace “liaison” with Link

 Line 222 -  “inborn benefits”  - define or find alternative word

“Another important source of external antecedents comes from the national mindset, 231 which leads people to have considerable discrepancies in the adoption of self-leadership, and it also 232 touches the intrinsic connotation of self-leadership inside that specific national scenario (Stewart et 233 al. 2019).” This statement needs further clarification and some examples to illustrate this point.

Section 2.3 self- leadership – very good explanation of SSb and interesting theoretical material included here.

Line 284 – replace “enunciated” with “articulated”

Line 336 “liaison” is problematic – “relationship” or” link” better serves this statement.  Liaison - communication or cooperation which facilitates a close working relationship between people or organizations.

******There needs to be discussion of the construct of self -efficacy and why it is used in this study. Also, there is a need to distinguish between self-efficacy and self-esteem and make clear those differences.   This is the largest aspect of the literature review that has been omitted and should be included earlier in the document ( with regard to self-efficacy theory)  - the distinction between self-efficacy and self-esteem can be drawn later before section 4 ( lines 375-400)

Need more information about Amazon Mechanical Turk  - this may be familiar to the authors of this paper but not to a broader audience of scholars.

Line 419 – more information on the participants – what positions did they hold ( management is a very broad term) this needs more thorough explanation. What were the characteristics in the organizations where they worked.

Line 529 – use the term you had previously used (self-evaluation) – the “self-perception” term causes additional confusion because you have so many like terms – Keep the same terms throughout your paper

Either in Section 5 “results” or section 6 “Discussion” – I would report the results by each research question – possibly making the research question a sub-heading and putting the data/results/explanation under each – this would help clarify how each of these analyses help to answer the specific research questions. In the present format, the research questions feel secondary.

Line 544 – “Thanks to the result emerging from the test of 545 the first hypothesis, the self-leadership literature is completed??? and enhanced with regard to the 546 relationship with self-efficacy. Indeed, prior literature found that self-leadership, thanks to the 547 adopted strategies, influences self-efficacy – as demonstrated by Latham and Frayne (1989), Prussia 548 et al. (1998), and Saks and Ashforth (1996).”  Can you really make this statement?   Rewording needed – maybe this adds to the literature base – but “completes” it is far from accurate.  Here again is a bold claim made in reference to self -efficacy with no mention or citation of Bandura’s work.  Albert Bandura was the father of “self-efficacy” theory – yet no reference even exists in this paper.  This makes me wonder if the authors thoroughly understand the construct of self-efficacy.   If self-efficacy is indeed a pillar of the CSE construct there needs to be much more work done on helping the audience understand the significance of each “pillar” and the theoretical genesis of each.   Try this resource – Albert Bandura – 1997  Self -Efficacy: The Exercise of Control   Freeman and CO – New York.

Author Response

Ms. Ref. No.: admsci-917765

Core self-evaluations, self-leadership, and the self-serving bias in managerial decision making: A laboratory experiment

Administrative Sciences

Replies to reviewer 2

Dear Reviewer and Editor,

We would like to express our gratitude for your comments in this round of review. We thank the editors and reviewers for the guidance to refine our work.

Please find attached the revised version of our manuscript entitled “Core self-evaluations, self-leadership, and the self-serving bias in managerial decision making: A laboratory experiment”.

Thanks for reading and taking into consideration our manuscript. Your comments were well-received and the relative suggestions have been incorporated into this improved version. Your work truly enables us to significantly improve our own.

In the following section, we have reported the comments and our replies. In the text, all of the changes are highlighted in yellow.

***

Reviewer: 2

Recommendation: Minor Revision

Comment 1: High levels of self efficacy do NOT cause people to blame failures on external causes – this is confusing in this statement. - This work reinforces the stream of (few) studies that considers a high level of CSE and 81 self-leadership as not always desirable for managerial decision-making processes and consequent 82 job performance (e.g., Barker 1993; Langfred 2004; Hiller and Hambrick 2005).

Reply to Comment 1: Thank you for this comment. We deleted parts in which it was directly assumed that “High levels of self efficacy do NOT cause people to blame failures on external causes – this is confusing in this statement”.

Comment 2: Is self-leadership autonomy? What is the relationship between them???

Reply to Comment 2: Thank you for this comment. As to distinguish self-leadership from autonomy we reported the following within the theoretical section dedicated to self-leadership:

“It is worth stating from the beginning that self-leadership is distinct from the concept of autonomy; indeed, the latter is a need, together with freedom, which forms the basis for self-leadership (Stewart et al. 2019). These two, therefore, are variables that are inter-related with each other, but do not underline the same phenomena; on the contrary, they should be well combined in order to avoid dysfunctional results (Langfred 2005). On this relationship between self-leadership and autonomy, through the study of 71 self-managed teams composed of MBA students, Langfred (2004) found that members of self-leading teams are generally reluctant to monitor each other given their effort to maintain high individual autonomy, which leads to poor performance when individual autonomy is high and to excellent performance when individual autonomy is low. In another study of longitudinal data from 35 self-managing teams, Langfred (2007) confirmed that very high self-leadership undermines individual autonomy; thus further advancing a negative relationship between self-leadership and autonomy”.

Comment 3:

Line 92 – “good ness” – better “ accuracy”

Line 149 – replace “liaison” with Link

Line 222 - “inborn benefits” - define or find alternative word

“Another important source of external antecedents comes from the national mindset, 231 which leads people to have considerable discrepancies in the adoption of self-leadership, and it also 232 touches the intrinsic connotation of self-leadership inside that specific national scenario (Stewart et 233 al. 2019).” This statement needs further clarification and some examples to illustrate this point.

Line 284 – replace “enunciated” with “articulated”

Line 336 “liaison” is problematic – “relationship” or” link” better serves this statement. Liaison - communication or cooperation which facilitates a close working relationship between people or organizations.

Line 529 – use the term you had previously used (self-evaluation) – the “self-perception” term causes additional confusion because you have so many like terms – Keep the same terms throughout your paper

Reply to Comment 3: Thank you for this comment.

At line 92 we substituted goodness with accuracy.

At lines 149 and 336 we substituted liaison with link.

At line 222 we substituted goodness with internal or natural rewards as advanced by Stewart et al. 2011 (“With regard to internal antecedents, at the individual level, studies have demonstrated that they are crucial for individual self-leadership, such as internal or natural rewards and ‘thought self-leadership’ (a mean for individuals to manage their own thinking tendencies). In particular, people, who choose their work position with internal or natural rewards [...]”).

Lines from 231 to 233 were better explained by substituting the word national mindset with ‘national culture’ as well as by adding this example: “For example, Neubert and Wu (2006) found that certain self-leadership strategies, such as self-goal setting, were more adopted in the Chinese context rather than in Western countries, while other self-leadership strategies, such as self-observation, were more used in Western countries rather than in China”.

At line 284 we replaced “enunciated” with “articulated”.

At lines 441 and 529 we used the word self-evaluation as to be consistent.

Comment 4: ******There needs to be discussion of the construct of self-efficacy and why it is used in this study. Also, there is a need to distinguish between self-efficacy and self-esteem and make clear those differences. This is the largest aspect of the literature review that has been omitted and should be included earlier in the document (with regard to self-efficacy theory) - the distinction between self-efficacy and self-esteem can be drawn later before section 4 (lines 375-400).

Reply to Comment 4: Thank you for this comment. You are right in saying that a good discussion of the concept of self-efficacy and self-esteem is necessary as well as to identify why, at a certain point, we introduce them in our hypotheses development. As to make a great discussion of these points, we have decided to build a new sub-section entitled: “. Self-efficacy and self-esteem: A theoretical premise” which opens to the next one about Core Self-Evaluations (2.3). This follows the rationale that self-efficacy and self-esteem are two out of four pillars of CSE and because of that they should be treated before speaking about CSE – it would have not been useful to report it around lines 375-400 because of the fact that these are due theoretical premises of CSE. Moreover, we have decided to build this new sub-section as not to overload section 2.3 about CSE. This new sub-section is reported as follows:

“Albert Bandura (1986) defined self-efficacy as directly related with individuals’ beliefs in their own personal abilities, both at physical and mental levels, such as how well they will be capable of managing their skills and cognitive abilities, and how much effort they will be able to carry in order to achieve a predetermined goal. According to Bandura et al. (1999), individuals that have a high level of self-efficacy are more expected to perceive challenging duties as something to be mastered instead of perceiving them as something to be bypassed. Consequently, individuals with a high level of self-efficacy are more inclined to work hard in order to accomplish their duties, even if they have to keep up their endeavors in the long term. However, according to Bandura (1977), the perception of self-efficacy mainly depends on the feedback received from other individuals – as previously stated by Weinberg et al. (1979) – thus: if people receive negative feedback about their physical and mental efficiency, they will perform inadequately, while, if they receive positive feedback, they will be likely to show outstanding performances. Yet, this feedback, which determines the perceived self-efficacy, varies according to the situational circumstances of the task that the individual is facing; indeed, Bandura stated that: “The extent to which people will alter their perceived self-efficacy on the basis of performance feedback will depend on such factors as the difficulty of the task, the amount of effort they expend, the amount of external aid they receive, the situational circumstances under which they perform, and their mood and physical state at the time” (Bandura 1986; p. 363). In contrast to this task-related perception of self-efficacy, other scholars have subsequently conceived self-efficacy in a more general dimension that has been called General Self-Efficacy (GSE) (Gardner and Pierce 1998; Judge et al. 1998), which, within this work, is the meaning adopted when speaking about self-efficacy. GSE can be described as the attitude of an individual’s belief in one’s global capacity to face challenges across a broad range of circumstances (Judge et al. 1998; Scherbaum et al. 2006). This stream of studies assumes that GSE is able to catch distinctive variation among people in their propensity to see themselves as able to manage and achieve different goals in different contexts; thus, GSE is a ‘situation-independent context belief’ rather than a fluctuant perception of their own abilities. However, GSE and specific self-efficacy are linked by a high inter-relationship (Judge et al. 1998; 1999). Indeed, various studies demonstrated that GSE and specific self-efficacy are positively correlated as well as the former having been found to be a determinant of the latter (e.g., Chen et al. 2001; Scherbaum et al. 2006).

In relation to self-efficacy, other scholars defined and advanced the concept of self-esteem, conceived as the whole set of personal introspective beliefs that people have about themselves as well as their perception about the emotional states affecting both their professional and personal lives (Smith et al. 2007). In particular, some scholars (e.g., Leary and Baumeister 2000) identified that self-esteem can be used for testing the perception that individuals have about a specific personal quality, i.e. "I believe I am a good student and I feel satisfied about that – or about a global evaluation of themselves – i.e. "I believe I am not a good person, and I feel bad about myself in general”. Due to the obvious link between the evaluation that a person can make about her/himself in general or about a specific quality (self-esteem) and the evaluation that the same person can make about her/his abilities (self-efficacy), it has been derived and proved that self-esteem is often directly affected by self-efficacy (Judge and Bono 2001; Lane et al. 2004; Asakereh and Yousofi 2018). Indeed, the perception that people make about their own abilities is influenced by the feedback received from others around them about their reached performance, affecting the perception of their own worth.

Thanks to the huge number of existing studies, self-efficacy and self-esteem are two of the four main personality traits investigated within the domain of industrial and organizational psychology (Judge et al. 1998), which is then aggregated under the same umbrella term (trait), called Core Evaluations”.

Comment 5: Need more information about Amazon Mechanical Turk - this may be familiar to the authors of this paper but not to a broader audience of scholars.

Reply to Comment 5: Thank you for this comment. We added the following text as to explain Amazon Mechanical Turk: “In particular, Amazon Mechanical Turk is a “crowdsourcing marketplace that makes it easier for individuals and businesses to outsource their processes and jobs to a distributed workforce who can perform these tasks virtually. This could include anything from conducting simple data validation and research to more subjective tasks like survey participation, content moderation, and more” (Mturk 2020). Due to its features, this platform has been extensively used by scholars for collecting reliable responses for experiments on executives (Paolacci et al. 2010)”.

Comment 6: Line 419 – more information on the participants – what positions did they hold (management is a very broad term) this needs more thorough explanation. What were the characteristics in the organizations where they worked.

Reply to Comment 6: Thank you for this comment. Unfortunately, despite being reliable and extensively used for management scholar research, Amazon Mechanical Turk does not allow knowing the type of companies in which managers work as well as it is not possible to know which kind of management functions they operate. Amazon Mechanical Turk only allows to select ‘Job function – Management’. So, we know that this is not the best for a scientific research; however, as already said citing Paolacci et al. (2010) – but there are a lot of other studies – Amazon Mechanical Turk is valid and reliable; it is just a matter of not perfectly knowing the sample. We know that this is a limit and because of that we inserted this point as a limitation of our study stating the following: “Another limitation of this study, stemming from the terms and conditions of the use of Amazon Mechanical Turk in selecting the sample of respondents, lies in not having known in which sector and organizations the sample managers work, as well as not knowing their specific positions. Future research should use less virtually mediated sampling methods so that these sample features are collected”.

Comment 7: Either in Section 5 “results” or section 6 “Discussion” – I would report the results by each research question – possibly making the research question a sub-heading and putting the data/results/explanation under each – this would help clarify how each of these analyses help to answer the specific research questions. In the present format, the research questions feel secondary.

Reply to Comment 7: Thank you for this comment. We have opted for dividing the results section in two sub-section – i.e. 5.1 and 5.2 – with the following sub-headings “Results for research question 1” and “Results for research question 2”. We did not like, in terms of communication style, the option of entitling them with the research question themselves. However, we strongly recognize that a specific reference to them is necessary within the results section. Due to that, we have reported RQ1 and RQ2 within the respective results’ sub-section 5.1 and 5.2. The text reads as follows:

“To identify whether managers with high CSE have greater self-leadership compared to individuals with low and average CSE (H1), a one-way ANOVA was implemented considering the different CSE clusters (low, average, high), and their results on the RSLQ. This test helps answering research question 1: “How is self-leadership related with CSE?”.

“A second one-way ANOVA was conducted in order to verify whether individuals with high self-leadership are more inclined to be victims of the SSB compared to executives with low and average self-leadership (H2). This test helps answering research question 2: “How does self-leadership influence the attribution of successes/failures?”.

We decided to make this split here and not within the Discussion section as to maintain the latter as a unique platform for discussing the theoretical advancements made from this work.

Comment 8: Line 544 – “Thanks to the result emerging from the test of 545 the first hypothesis, the self-leadership literature is completed??? and enhanced with regard to the 546 relationship with self-efficacy. Indeed, prior literature found that self-leadership, thanks to the 547 adopted strategies, influences self-efficacy – as demonstrated by Latham and Frayne (1989), Prussia 548 et al. (1998), and Saks and Ashforth (1996).”  Can you really make this statement?   Rewording needed – maybe this adds to the literature base – but “completes” it is far from accurate.  Here again is a bold claim made in reference to self -efficacy with no mention or citation of Bandura’s work.  Albert Bandura was the father of “self-efficacy” theory – yet no reference even exists in this paper.  This makes me wonder if the authors thoroughly understand the construct of self-efficacy.   If self-efficacy is indeed a pillar of the CSE construct there needs to be much more work done on helping the audience understand the significance of each “pillar” and the theoretical genesis of each.   Try this resource – Albert Bandura – 1997  Self -Efficacy: The Exercise of Control   Freeman and CO – New York.

Reply to Comment 8: Thank you for these comments. First, we removed the word ‘completed’ and introduced the word ‘advanced’; this better fits with the scientific communication style.

About the reference to the works of Bandura and to the self-efficacy concept, you are more than right in saying that they should have a greater and more detailed space within the manuscript. In order to fix this flaw, we have built a new section called “2.1. Self-efficacy and self-esteem: A theoretical premise” in which we describe the concepts of self-efficacy and self-esteem, that are at the basis of CSE, as well as their inter-relationship. We decided not to extend this theoretical premise also to the other two pillars of the CSE – i.e., locus of control, and emotional stability – due to the fact that works centred on them have not been used for the hypotheses development. Moreover, the manuscript, right now, has a length around 14.500 words; so, it is useful not wasting time of the reader in reading theoretical premises on locus of control and emotional stability that will not be used for the scope of the manuscript. In particular, section 2.1 reads as follows:

“In relation to self-efficacy, other scholars defined and advanced the concept of self-esteem, conceived as the whole set of personal introspective beliefs that people have about themselves as well as their perception about the emotional states affecting both their professional and personal lives (Smith et al. 2007). In particular, some scholars (e.g., Leary and Baumeister 2000) identified that self-esteem can be used for testing the perception that individuals have about a specific personal quality, i.e. "I believe I am a good student and I feel satisfied about that – or about a global evaluation of themselves – i.e. "I believe I am not a good person, and I feel bad about myself in general”. Due to the obvious link between the evaluation that a person can make about her/himself in general or about a specific quality (self-esteem) and the evaluation that the same person can make about her/his abilities (self-efficacy), it has been derived and proved that self-esteem is often directly affected by self-efficacy (Judge and Bono 2001; Lane et al. 2004; Asakereh and Yousofi 2018). Indeed, the perception that people make about their own abilities is influenced by the feedback received from others around them about their reached performance, affecting the perception of their own worth.

Thanks to the huge number of existing studies, self-efficacy and self-esteem are two of the four main personality traits investigated within the domain of industrial and organizational psychology (Judge et al. 1998), which is then aggregated under the same umbrella term (trait), called Core Evaluations”.

Moreover, a passage in which the works of Bandura are read in the light of our results of H1 and H2 is made within discussion. In particular it has been written:

“This result reinforces the stream of studies that advanced that self-efficacy, one of the four pillars of CSE, has a negative effects on performance and that can conduct to cognitive errors (see Powers 1973; 1991) – in contrast with the works of Bandura (1986; 1997; Bandura et al., 1999) that assumed a positive influence. Indeed, Vancouver et al. (2002), through two laboratory experiments with 87 and 104 undergraduates respectively, found that self-efficacy has a negative influence on performance, explained by the fact that self-efficacy leads to the cognitive error of overconfidence and hence increased the likelihood of committing logic errors during tasks”.

***

Finally, we again thank the Editors and Reviewers for their time, which has resulted in highly thoughtful reviews and invaluable guidance for the authors. The paper has evolved – structurally and in terms of content – and we anticipate that the improvements will meet and exceed expectations.

We hope that the revisions in the manuscript, and our accompanying responses, will be sufficient to make our manuscript suitable for publication in Administrative Sciences.

We shall look forward to hearing from you at your earliest convenience.

Yours sincerely,

The Authors