In 1973, a young surgeon named Amy Edmondson was studying medication error rates in hospital nursing teams as part of her doctoral research at Harvard. She expected to find that the teams that reported the most errors were the worst teams — the ones making the most mistakes. Instead, she found the opposite. The teams that reported the highest error rates were the highest-performing teams by every measure of patient care quality. Her initial confusion resolved into a more unsettling finding: the high-performing teams reported more errors not because they made more errors but because they felt safe enough to acknowledge them. The lower-performing teams were making similar or greater numbers of mistakes, but the culture of fear and blame meant those errors were hidden rather than learned from. Edmondson named the critical variable she had identified — the degree to which team members believe they will not be punished or humiliated for speaking up — psychological safety. It would take another twenty years and a large-scale study by Google before the rest of the organizational world caught up with what her nursing team data had shown.

The history of leadership research is full of confident claims that have not survived sustained empirical scrutiny. In the early twentieth century, "great man" theories proposed that leadership was the expression of innate, heroic traits — strength, charisma, decisiveness — that could not be taught or cultivated, only recognized in those born with them. When psychologists in the 1940s and 1950s set about testing this idea systematically, comparing the traits of leaders and non-leaders across dozens of studies, they found weak and inconsistent relationships, leading many to conclude that there were no reliable leadership traits at all. That overcorrection gave way to situational and contingency theories — the idea that leadership effectiveness was entirely contextual. By the 1980s and 1990s, more sophisticated meta-analytic methods and better theoretical frameworks had produced a more nuanced picture: some traits do matter, context shapes their expression, and specific behavioral competencies can be developed. What followed has been half a century of progressively richer understanding of what actually separates transformative leaders from those who merely occupy leadership positions.

The findings are not always comfortable for the people they study. Research by Tomas Chamorro-Premuzic and others has documented a systematic and durable mismatch between the traits that get people promoted into leadership roles and the traits that make them effective once there. The modern workplace promotes for confidence and selects for charisma, and routinely fails to notice that neither confidence nor charisma correlates reliably with the ability to build high-performing teams, develop other people, make good decisions under uncertainty, or create the conditions in which talented people choose to do their best work.

"The most dangerous leadership myth is that leaders are born — that there is a genetic factor to leadership. This myth asserts that people simply either have certain charismatic qualities or not. That's nonsense; in fact, the opposite is true. Leaders are made rather than born." -- Warren Bennis

Key Definitions

Transformational leadership — A leadership style, theorized by James MacGregor Burns (1978) and operationalized by Bernard Bass and Bruce Avolio, characterized by inspiring followers to transcend self-interest for collective goals through idealized influence (role modeling), inspirational motivation (articulating a compelling vision), intellectual stimulation (encouraging creative thinking), and individualized consideration (attending to each follower's development needs). Consistently outperforms transactional leadership on measures of team performance, innovation, and follower wellbeing in cross-cultural meta-analyses.

Transactional leadership — Leadership that operates through contingent exchange: rewards for performance, correction for deviation from standards. Effective for managing predictable, well-defined tasks but limited in its ability to generate discretionary effort, creativity, or intrinsic commitment.

Psychological safety — Amy Edmondson's term for a shared belief held by members of a team that the team is safe for interpersonal risk-taking — that speaking up with ideas, questions, disagreements, or errors will not result in punishment or humiliation. The single most predictive factor in Google's Project Aristotle analysis of team effectiveness.

Servant leadership — Robert Greenleaf's concept (formalized in 1970) of leadership as fundamentally oriented toward serving the needs of followers and enabling their growth and effectiveness, rather than toward the leader's authority or goals. Associated with follower satisfaction, organizational commitment, and ethical organizational culture.

Level 5 Leadership — Jim Collins' designation for the highest level of leadership capability identified in his "Good to Great" research (2001): the combination of fierce professional will (determination to produce exceptional results) and personal humility (attribution of success to others and to luck, personal accountability for failures). Described as paradoxical, rare, and consistently associated with organizations that sustained exceptional performance over decades.

Emotional intelligence (EI) — As defined by Goleman (1995, 1998), drawing on the academic work of Salovey and Mayer: the cluster of capacities involving self-awareness, self-regulation, empathy, motivation, and social skill that enable effective emotional functioning in interpersonal contexts. Distinguished from cognitive intelligence (IQ) and from personality traits, though with some overlap.

Leadership Frameworks Compared

Framework Core Claim Key Research Distinguishing Concept
Transformational leadership Leaders who inspire transcend transactional exchange Bass & Avolio; 87-study meta-analysis (Judge & Piccolo, 2003) Idealized influence, inspirational motivation, intellectual stimulation, individualized consideration
Level 5 Leadership Sustained great performance requires humility + fierce will Jim Collins, Good to Great (2001) "Window and mirror" — credit others, accept blame
Psychological safety Team learning requires safety from interpersonal risk Amy Edmondson (1999); Google Project Aristotle (2012) Teams report more errors not because they make more, but because they feel safe to say so
Servant leadership Leaders serve followers' needs first Robert Greenleaf (1970) Power flows from serving, not commanding
Authentic leadership Genuine self-awareness and values-alignment enables trust Bruce Avolio; various Leader self-knowledge and consistency between values and actions

Transformational Leadership: What the Evidence Shows

The most extensively researched leadership construct in modern organizational science is transformational leadership. Bernard Bass, building on Burns' earlier theoretical work, developed the Multifactor Leadership Questionnaire (MLQ) in the 1980s as a tool for measuring leadership style, and the research that accumulated around it represents hundreds of studies across dozens of countries and organizational contexts.

The meta-analytic picture is consistent. A landmark 2003 meta-analysis by Judge and Piccolo, published in the Journal of Applied Psychology, analyzed 87 independent studies and found that transformational leadership correlated with follower job satisfaction at r = 0.58, with leader effectiveness ratings at r = 0.44, and with organizational performance measures at r = 0.26. The relationships held across cultures, organizational types, and measurement methods, making transformational leadership one of the more robust constructs in applied psychology.

What makes transformational leadership effective? Bass and Avolio's research identified the mechanisms as operating at multiple levels. Individually, transformational leaders develop followers' self-efficacy — their belief in their own capability — by consistently communicating confidence in their potential, providing growth-oriented challenges, and recognizing achievement. At the group level, transformational leaders build shared identity and purpose, creating the motivational conditions in which discretionary effort (effort beyond what role requirements specify) becomes the norm rather than the exception. At the organizational level, they articulate and embody values that persist in culture beyond their own tenure.

Importantly, Bass's research found that transformational and transactional leadership are not opposites on a single dimension but distinct patterns that can coexist. The most effective leaders in Bass and Avolio's studies were those who demonstrated both transformational behaviors — for inspiration, development, and innovation — and effective transactional management — for clear expectations, accountability, and contingent recognition. Pure transactional management without transformational elements produces compliant performance without discretionary engagement. Pure transformational inspiration without transactional structure produces enthusiasm without direction or accountability.

Jim Collins and the Level 5 Paradox

Jim Collins began the Good to Great research project in 1996 with a team of researchers tasked with identifying what distinguished companies that had made the leap from good to great sustained performance from those that had not. They expected to find visionary strategies, innovative products, lucky market timing, or aggressive acquisitions. They found something they had not been looking for and initially resisted reporting: the single most important variable was leadership.

Specifically, every company in the "great" comparison group had been led, at the pivotal transition moment, by what Collins came to call a Level 5 leader — someone who combined "an extreme personal humility with an intense professional will." These were not the charismatic visionaries that business media celebrated. They were, in many cases, almost invisible externally — quiet, self-effacing, prone to crediting their teams and circumstances for success and taking personal responsibility for failures, but relentlessly, almost ferociously determined to produce results.

Collins noted a specific behavioral pattern he called "the window and the mirror." When things went well, Level 5 leaders looked out the window — attributing success to their team, to timing, to favorable circumstances. When things went wrong, they looked in the mirror — taking personal accountability without deflection. Comparison leaders (those who led companies that did not make the sustained transition to great performance) showed the reverse: personal credit for successes, external blame for failures.

The research also found that Level 5 leaders consistently made what Collins called "first who, then what" decisions: getting the right people into the organization and into key positions before determining strategic direction. This reflected a recognition that great strategies are insufficient compensation for having the wrong people implementing them, and that the right people with genuine alignment and capability will find or create the right strategies.

The Level 5 construct has attracted both significant subsequent empirical attention and some methodological criticism. Collins' methodology was retrospective and qualitative — identifying the leaders of companies that had already been defined as great and characterizing their leadership — which raises questions about selection bias and post-hoc attribution. Nonetheless, the specific combination of humility and will that Collins described has found support in subsequent research on authentic leadership and in the psychological safety literature, where leader humility is consistently identified as a prerequisite for team learning and risk-taking.

Google's Project Aristotle and Psychological Safety

In 2012, Google launched Project Aristotle, a systematic two-year investigation into what made some Google teams dramatically more effective than others. The research team analyzed data from 180 teams, examining dozens of variables: individual talent levels, educational backgrounds, personality types, demographic composition, geographic co-location, social density, and many others. They expected to find that the best teams had the best individuals, or the best skill mix, or the most experienced members. They did not find this.

The strongest predictor of team effectiveness was not who was on the team but how the team interacted. Five norms consistently characterized the highest-performing teams, with psychological safety — Edmondson's construct from 1999 — at the top. The finding validated and dramatically amplified Edmondson's work: teams in which members believed they would not be punished for taking risks were significantly more innovative, more productive, and better at learning from mistakes than teams where members felt they needed to manage impressions and avoid visible failure.

The implications for leaders are specific and behavioral. Edmondson's subsequent research identified the concrete behaviors that create or destroy psychological safety. Leaders who create it: frame work as a learning problem requiring multiple perspectives rather than a performance problem requiring individual expertise; acknowledge their own uncertainty and fallibility; model intellectual curiosity rather than pretending to possess all the answers; respond to failures and errors with inquiry rather than blame; and explicitly invite dissent and challenge to their own views. Leaders who destroy it: react to mistakes with punishment or public humiliation; create hierarchical distance that signals their own views are not to be questioned; prioritize appearing certain and decisive over being accurate; and use asymmetric information access as a control mechanism.

The insight that psychological safety is a leadership-created condition, not an inherent team characteristic, has been the most practically influential finding in organizational research in the past two decades. It reframes leadership as primarily about creating conditions rather than directing activity.

Emotional Intelligence: The Interpersonal Foundations

Daniel Goleman's 1995 book Emotional Intelligence popularized a concept that had been introduced in academic psychology by Peter Salovey and John Mayer in 1990. Goleman's application of the concept to leadership, developed in his 1998 Harvard Business Review article and elaborated in Primal Leadership (2002), argued that emotional intelligence competencies account for up to 85 percent of the variance in leadership performance at senior levels, once baseline competencies are controlled.

This claim has been contested. The scientific standing of emotional intelligence as a distinct construct separate from personality and general intelligence remains debated, and some of Goleman's quantitative claims lack robust empirical support. What the more rigorous subsequent research does support is a more modest but still important conclusion: specific EI competencies — particularly self-awareness (accurately perceiving one's own emotional states and their effects on others), empathy (accurately perceiving others' emotional states and needs), and self-regulation (managing emotional reactions in ways that serve rather than undermine interpersonal goals) — predict leadership effectiveness above and beyond general intelligence and the Big Five personality traits, particularly in complex interpersonal leadership contexts.

A 2015 meta-analysis by Harms and Crede found that EI explained incremental variance in leadership effectiveness over personality and general mental ability in most studies, though the incremental effect was smaller than Goleman's popular writing suggested. The practical import is that EI competencies respond to targeted development. Unlike general intelligence, which has very limited trainability in adults, emotional self-awareness and empathic accuracy can be improved through structured feedback, coaching, and mindfulness practices — making them meaningful targets for leadership development.

Who Actually Gets Promoted and Why It's Often Wrong

Tomas Chamorro-Premuzic, a business psychologist at University College London, has spent years documenting what he calls "the leadership crisis" — the systematic failure of organizations to select the most capable leaders and the systematic promotion of those with surface-level charisma and confidence over those with genuine leadership competence.

His 2019 book, Why Do So Many Incompetent Men Become Leaders?, presents evidence that narcissism — characterized by grandiosity, self-promotion, entitlement, and charisma — is positively associated with leader emergence (getting into leadership roles) and negatively associated with leader effectiveness (performance once in those roles). The problem is that the traits that make someone appear capable in selection processes — confidence, self-promotion, decisiveness — are systematically unrelated or inversely related to the traits that predict leader effectiveness: intellectual humility, openness to feedback, capacity for genuine collaboration, and willingness to develop others.

The WEIRD bias Chamorro-Premuzic identifies (Western, Educated, Industrial, Rich, Democratic) in leadership research points to a related problem: much leadership research has been conducted in contexts that reflect specific cultural values around individualism, assertiveness, and visible achievement that are not universal. Research by Mansour Javidan and the GLOBE project across 62 societies found substantial cross-cultural variation in which leadership behaviors are valued and effective — attributes universally desirable across cultures include integrity, being inspirational, visionary, and decisive, but many other attributes show strong cultural variation.

Growth Mindset and the Learning Leader

Carol Dweck's research on fixed versus growth mindsets — developed primarily in educational contexts but extensively applied to leadership — identifies a specific cognitive orientation toward effort and failure that has strong implications for leadership behavior. Leaders with growth mindsets (who believe abilities can be developed through effort and learning) respond to organizational failures and personal limitations with inquiry and strategy adjustment. Leaders with fixed mindsets (who believe abilities are innate and stable) respond to failures with defensiveness, external attribution, and suppression of information that challenges their self-concept.

The leadership implications are behavioral: growth-mindset leaders create psychologically safe environments because failure is, for them, informative rather than threatening. They seek critical feedback because feedback serves their learning orientation. They develop other people because others' growth does not threaten their own relative standing. They are more likely to hire people better than themselves in specific domains — a behavior frequently cited as a hallmark of exceptional leaders and rarely practiced by those primarily motivated by status maintenance.

Practical Takeaways

Create psychological safety before demanding performance. The Google Project Aristotle research and Edmondson's two decades of study are unambiguous: the conditions for exceptional team performance are interpersonal, not technical. Model fallibility, respond to errors with inquiry rather than blame, and explicitly invite challenge to your own views.

Practice the window and mirror discipline. Collins' Level 5 finding is a behavioral habit, not a personality trait: attribute successes to your team and circumstances; take personal accountability for failures without deflecting.

Develop your emotional self-awareness. The most consistent finding in EI research is that leaders who accurately perceive their own emotional states make better interpersonal decisions. Regular reflection, 360 feedback, and coaching are evidence-based tools for developing this.

Hire for growth mindset as deliberately as for technical skill. A team of learners who develop rapidly outperforms a team of high-aptitude fixed-mindset performers over any time horizon longer than six months.

Recognize the difference between what makes you look like a leader and what makes you effective as one. Confidence is not competence; charisma is not vision; decisiveness is not accuracy. The traits the selection process rewards are often orthogonal to the traits that produce team and organizational outcomes.

References

  1. Bass, B. M., & Avolio, B. J. (1994). Improving Organizational Effectiveness through Transformational Leadership. Sage.
  2. Collins, J. (2001). Good to Great: Why Some Companies Make the Leap and Others Don't. HarperCollins.
  3. Edmondson, A. C. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350-383.
  4. Google Re:Work. (2016). Guide: Understand team effectiveness. re.work/guides.
  5. Goleman, D. (1998). What makes a leader? Harvard Business Review, 76(6), 93-102.
  6. Judge, T. A., & Piccolo, R. F. (2004). Transformational and transactional leadership: A meta-analytic test of their relative validity. Journal of Applied Psychology, 89(5), 755-768.
  7. Chamorro-Premuzic, T. (2019). Why Do So Many Incompetent Men Become Leaders? (And How to Fix It). Harvard Business Review Press.
  8. Dweck, C. S. (2006). Mindset: The New Psychology of Success. Random House.
  9. Brown, B. (2012). Daring Greatly: How the Courage to Be Vulnerable Transforms the Way We Live, Love, Parent, and Lead. Gotham Books.
  10. Harms, P. D., & Crede, M. (2010). Emotional intelligence and transformational and transactional leadership. Journal of Leadership and Organizational Studies, 17(1), 5-17.
  11. House, R. J., Hanges, P. J., Javidan, M., Dorfman, P. W., & Gupta, V. (2004). Culture, Leadership, and Organizations: The GLOBE Study of 62 Societies. Sage.
  12. Avolio, B. J., & Gardner, W. L. (2005). Authentic leadership development: Getting to the root of positive forms of leadership. Leadership Quarterly, 16(3), 315-338.

Tags

leadership management psychology organizational behavior emotional intelligence research work skills

Frequently Asked Questions

What traits do the best leaders consistently have?

Meta-analytic research on leader effectiveness consistently identifies a cluster of traits: high general cognitive ability, conscientiousness, openness to experience, emotional stability, extraversion (though with meaningful caveats), and relatively high but not extreme self-monitoring (social adaptability). More specifically, emotional intelligence competencies — self-awareness, empathy, self-regulation, social skill — show reliable relationships with leadership effectiveness across cultures and contexts. Amy Edmondson's research adds psychological safety creation as a critical behavioral skill. Jim Collins' Level 5 analysis found the combination of fierce professional will and personal humility as the distinguishing characteristic of leaders who produced sustained exceptional organizational performance.

Is leadership ability born or developed?

Twin studies estimate heritability of leadership at around 30 to 40 percent — meaningfully but not dominantly genetic. The balance reflects developmental experience, deliberate practice, mentoring, and environmental conditions. Research by Bruce Avolio and colleagues on 'authentic leadership development' provides a framework for how leadership grows through experience, particularly through crucible experiences: significant challenges that force self-examination and adaptation. There is no evidence for an immutable 'natural leader' type that development cannot produce, and considerable evidence that specific leadership competencies — including emotional intelligence and psychological safety creation — respond to targeted training and feedback.

What does Google's research say makes great team leaders?

Google's Project Aristotle (2016), which analyzed 180 teams over two years, found that team composition and individual talent mattered far less than team dynamics. The single most predictive factor of team performance was psychological safety — team members' confidence that they would not be punished or humiliated for speaking up with ideas, questions, or mistakes. This concept was developed by Amy Edmondson at Harvard in her 1999 study of hospital nursing teams, which found that higher-performing teams reported more errors, not fewer, because they felt safe enough to acknowledge mistakes and learn from them. Google's analysis found that leaders who created psychological safety, set clear goals, and demonstrated care for team members' personal wellbeing produced teams that were consistently more innovative, more productive, and more successful.

How important is emotional intelligence for leadership?

Daniel Goleman's 1998 Harvard Business Review article, 'What Makes a Leader,' analyzed data from nearly 200 large global companies and found that while technical skills and IQ were threshold competencies (necessary but not differentiating), emotional intelligence competencies — self-awareness, self-regulation, motivation, empathy, and social skill — distinguished top performers from average performers, particularly at senior leadership levels. A 2011 meta-analysis by Joseph and Newman found that emotional intelligence predicted leadership effectiveness with a corrected correlation of approximately r = 0.24 to 0.39 depending on the emotional intelligence measure used. The relationship is stronger for transformational leadership (inspiring and motivating others) than for transactional leadership (managing through contingent reward and correction).

What leadership styles are most effective according to research?

Transformational leadership — inspiring change through vision, intellectual stimulation, individualized consideration, and role modeling — consistently outperforms transactional leadership (managing through contingent reward and exception management) on outcomes including team performance, innovation, follower motivation, and organizational commitment. This finding, based on Bass and Avolio's research from the 1990s, has been replicated in hundreds of studies across cultures. However, research also supports situational leadership theory (Hersey and Blanchard): different contexts genuinely call for different approaches. In crisis situations requiring immediate direction, directive styles outperform collaborative ones. In stable, expert-heavy environments requiring innovation, participative and coaching styles dominate. Adaptive leaders — those who can shift their style based on situational demands — outperform those locked into single-style repertoires.

Why do bad leaders get promoted?

Tomas Chamorro-Premuzic's research, synthesized in his 2019 book 'Why Do So Many Incompetent Men Become Leaders?', identifies a systematic mismatch between the traits that facilitate leader selection and the traits that produce leader effectiveness. Confidence, charisma, and self-promotion — traits that predict career advancement — show weak or negative relationships with actual leadership effectiveness. Narcissism, which is associated with confidence and social boldness, is positively associated with leader emergence (getting chosen for leadership roles) but negatively associated with leader effectiveness once in those roles. The result is systematic over-promotion of charismatic, confident, self-promoting individuals at the expense of those with genuine leadership competencies. Research by Chamorro-Premuzic and colleagues finds this bias is particularly strong in selection processes based on unstructured interviews and first impressions.

How do great leaders handle failure and uncertainty?

Research on resilient leadership and psychological capital (PsyCap) by Luthans, Youssef, and Avolio identifies four components of psychological resources that distinguish effective leaders in adversity: hope, efficacy, resilience, and optimism (HERO). Brene Brown's qualitative research on vulnerable leadership adds that leaders who acknowledge uncertainty and their own fallibility — who model intellectual humility rather than projecting false certainty — generate significantly greater trust, psychological safety, and authentic followership than those who project invulnerability. Carol Dweck's research on growth mindset in leaders finds that leaders with incremental (growth) theories of ability respond to failure with increased effort and strategy adjustment, modeling for their teams that failure is informative rather than defining.

Share this article