Leadership - A reflection on evidence-based practice

TL;DR: Leadership styles overlap, and focusing on behaviour might be a more effective way to understand what works. This article outlines key leadership behaviours, the importance of trust, and the role of values.

Starting 2025 in a formal leadership role, I wanted to review the literature on evidence-based leadership practices. As a Psychology graduate, I naturally turned to research. But here’s the thing—there are hundreds and thousands of papers on leadership. So, what exactly should an effective leader do? How should I behave in my new role?

The Challenge of Leadership Research

A quick heads-up: this field is messy. It’s filled with countless theories, models, and jargon that make it hard for non-scientists to apply findings. My goal here is to simplify (some of) the available knowledge and identify practical, evidence-based recommendations for new leaders like myself. Of course, this inevitably means that I’ll have to leave some things out (See this from 2017 if you are interested in a summary of leadership theory & research).

Defining Leadership

Every organisation wants high-performing leaders. But what does that even mean? Leadership performance is about achieving the goals of the role...duh. But defining these goals is tricky. Should a leader focus on individual and team performance, organisational commitment, job satisfaction, well-being, motivation, … maybe a combination of these or all of them taken together? This is the classic criterion problem in psychology—defining performance is context-dependent (See this for a good summary).

Here are a few definitions from reputable sources:

• “Leadership is a set of mindsets and behaviours that aligns people in a collective direction, enables them to work together and accomplish shared goals, and helps them adjust to changing environments.” (See McKinsey & Company, 2024).
• “Leadership is the capacity to translate vision into reality” (See Psychology Today)
• Leadership is “the processes involved in leading others, including organising, directing, coordinating, and motivating their efforts toward achieving certain group or organisational goals” (See the American Psychology Association dictionary)
• The ability of an individual to influence, motivate, and enable others to contribute towards the effectiveness and success of the organisation of which they are members’ (See here)
• “The process of influencing others to understand and agree about what needs to be done and how to do it, and the process of facilitating individual and collective efforts to accomplish shared objectives (Note: full disclosure, I didn't have access to the source material. I took Gary Yukl's quote from this reference)

These are all valid definitions of leadership. However, the focus on vision, team member goal setting, and attainment is missing two critical elements. In my view, an effective leader should also:

• Balance goal attainment with team members' job satisfaction & organisational commitment.
• Operate within legal, organisational, social and moral norms.

Note: Leadership may have a whole range of other correlates beyond job satisfaction & organisational commitment. For example, well-being, thriving, burnout, job stress, turnover intention, affective commitment, normative commitment, continuance commitment, affective trust, work engagement, innovate behaviours, organisational citizenship behaviour, organisational workplace deviance, trust, and team viability.

I also believe that understanding leadership as a two-way process between leader & follower(s) is more helpful than a single unidirectional and finite activity. Given that leader performance is to a large extent about team members goal attainment, understanding your organisation’s goals and translating them into meaningful objectives for your team is key.

Leadership Styles

One popular way of categorising leadership is through styles. Unfortunately, the field is fragmented and often falls short of its promise to offer comprehensive, evidence-based recommendations. The purpose of this article is obviously not to explore why and how this fragmentation happens, but it is important to note. To give you an idea of the problem, here are some of the dominant leadership styles from the past decades:

A recent large-scale study (see here) found that taken as a whole and compared to each other, the 12 leadership styles are not very different from each other (specific lingo: share significant variance). This is in direct opposition to what many papers are claiming when a “new” leadership style is shared with the wider world. The authors also suggests that it is behaviour that explains differences in leadership and not styles.

This doesn’t mean previous research on leadership styles is useless—it just suggests that treating them as distinct categories may not be the best approach. If you are interested, there are plenty of papers on specific styles (See this from 2019 on Servant Leadership, this from 2016 on Authentic and Transformational Leadership, this from 2011 on Transformational Leadership, or this from 2004 on Transformational and Transactional Leadership). Given this, I won’t spend too much time on styles. Instead, let’s focus on leadership behaviours.

Leadership Behaviours

The best framework I've found is Gary Yukl's leadership behavioural categories (See the original paper from 2002 and the updated version from 2012). This model organises leadership into four main behavioural groups:

1. Task-Oriented Behaviours (planning, clarifying, monitoring, problem-solving)
2. Relations-Oriented Behaviours (supporting, recognising, developing, empowering)
3. Change-Oriented Behaviours (advocating change, envisioning change, encouraging innovation, facilitating learning)
4. External Behaviours (networking, external monitoring, representing)

Before deep-diving into the specific behaviours within each category, it is important to note that not all categories are created equal. Their relationship differs across multiple variables - in simple terms certain behaviours seem more effective than others. The below table summarises the effects.

The two meta-analysis I used for the above table did not include the 4th category of behaviour (external behaviours). Therefore, these are marked as "missing". The two relevant papers are Integrating leadership research: a meta-analytical test of Yukl’s meta-categories of leadership and What Type of Leader Behaviors Are Functional in Teams? A Meta Analysis.

Here is a quick summary of the findings:

• Job satisfaction: Relations- and change-oriented behaviours seem most effective. Importantly, task-oriented behaviours seem to have a negative relationship with job satisfaction.
• Commitment: Relations-oriented behaviours are key.
• Performance: Relations- and task-oriented behaviours show the strongest link.
• Team effectiveness & productivity: Both task- and relationship-oriented behaviours matter.
• Team learning: Strongly linked to relationship-oriented behaviours.

Unfortunately, I could not find any meta-analysis looking at External behaviours, so the relationship between those behaviours and outcomes are more speculative than for the other three behaviour categories.

Couple of points here are important to mention:

• Even "strong" effects are weaker than you'd expect. In technical terms, difference in leadership behaviours explain approx 30% (r = 0.57) variance in outcome variable. And these are the "strongest" effects.
• The usual mantra of "correlation does not equal causation" also applies here. This is just to say be careful how and what you read into the table above.
• Leadership behaviours rarely have isolated effects. Not much is formally known about the complex interaction of behaviours and their effects.
• The context has some effect on the effectiveness of leadership behaviour: in other words, some contexts are more conducive to certain leadership behaviours than others. This is (in my opinion) where leadership experience might play a central role. Experienced leaders might be better at adjusting their behaviour to the context.

So now onto the specific leadership behaviours within each category. Acknowledgement: The following is from the Appendix of Gary Yukl's 2012 paper: Effective Leadership Behavior: What We Know and What Questions Need More Attention.

Task-oriented behaviours

Planning:

Develops short-term plans for the work; determines how to schedule and coordinate activities to use people and resources efficiently; determines the action steps and resources needed to accomplish a project or activity.

Clarifying:

Clearly explains task assignments and member responsibilities; sets specific goals and deadlines for important aspects of the work; explains priorities for different objectives; explains rules, policies, and standard procedures.

Monitoring:

Checks on the progress and quality of the work; examines relevant sources of information to determine how well important tasks are being performed; evaluates the performance of members in a systematic way.

Problem Solving:

Identifies work-related problems that can disrupt operations, makes a systematic but rapid diagnosis, and takes action to resolve the problems in a decisive and confident way.

Relationship-oriented behaviours

Supporting:

Shows concern for the needs and feelings of individual members; provides support and encouragement when there is a difficult or stressful task, and expresses confidence members can successfully complete it.

Recognising:

Praises effective performance by members; provides recognition for member achievements and contributions to the organisation; recommends appropriate rewards for members with high performance.

Developing:

Provides helpful feedback and coaching for members who need it; provides helpful career advice; encourages members to take advantage of opportunities for skill development.

Empowering:

Involves members in making important work-related decisions and considers their suggestions and concerns; delegates responsibility and authority to members for important tasks and allows them to resolve work-related problems without prior approval.

Change-oriented behaviours

Advocating Change:

Explains an emerging threat or opportunity; explains why a policy or procedure is no longer appropriate and should be changed; proposes desirable changes; takes personal risks to push for approval of essential but difficult changes.

Envisioning Change:

Communicates a clear, appealing vision of what could be accomplished; links the vision to member values and ideals; describes a proposed change or new initiative with enthusiasm and optimism.

Encouraging Innovation:

Talks about the importance of innovation and flexibility; encourages innovative thinking and new approaches for solving problems; encourages and supports efforts to develop innovative new products, services, or processes.

Facilitating Collective Learning:

Uses systematic procedures for learning how to improve work unit performance; helps members understand causes of work unit performance; encourages members to share new knowledge with each other.

External behaviours

Networking:

Attends meetings or events; joins professional associations or social clubs; uses social networks to build and maintain favourable relationships with peers, superiors, and outsiders who can provide useful information or assistance.

External Monitoring:

Analyses information about events, trends, and changes in the external environment to identify threats, opportunities, and other implications for the work unit.

Representing:

Lobbies for essential funding or resources; promotes and defends the reputation of the work unit or organisation; negotiates agreements and coordinates related activities with other parts of the organisation or with outsiders.

Leader values and integrity

So we got the behaviour part of leadership covered, and as complex as it is, the behaviour aspect is probably easier to describe than the following two sections, especially the one on values. This is because the domain of values are extremely hard to talk about from a scientific perspective… Here we go.

Trust

So let’s cover the less difficult one of the duo first: trust. Trust is the belief that someone or something can be relied on to do what they say they will (see here), and it is an essential component for productive and mutually beneficial relationships. There is a lot of research on trust in general, which I am not going to cover in-depth in this article. If you are really interested, see How and why humans trust: A meta-analysis and elaborated model or Models of Interpersonal Trust Development: Theoretical Approaches, Empirical Evidence, and Future Directions.

However, there is surprisingly little research on trust in leadership. My main takeaway from the available studies (See these meta-analyses from 2020 and 2000) is that trust seem to mediate / moderate the relationship between leader behaviour and individual-, team-, and organisational-level outcomes (e.g performance, organisational citizenship behaviour, commitment, job satisfaction, etc.).

Nothing new under the sun, trust is important. But how do we develop trust as a leader? Now, this is where the number of (published and available) studies drop even more dramatically. Here are a couple of models that make sense on a conceptual level:

I believe the best we can say at this point, is that open cooperation, reciprocity, communication, and expectations play key roles in trust development / decline. So as a leader, do your best to identify expectations from the team towards you and be open about your performance on those expectations.

As some of the guiding principles, consider the following as a leader:

• Integrity: Be honest and truthful.
• Competence: Have or develop the necessary knowledge and skills (very context dependent what these are!)
• Consistency: Be reliable and fair.
• Loyalty: Support your team, don’t act opportunistically.
• Openness: Be transparent about decisions and actions.

Although the meta-analytical evidence is sparse, I have no issue recommending the above. It seems like the right thing to do as a leader, and even if it doesn’t lead to better performance directly, the moral basis of these actions is sufficient justification for me. Which leads us to values.

Values

Values are important in leadership. Recall that leadership is to a large extent about goal attainment of the leader, team and members within specific constraints, two of them being social and moral norms. This is just to say if the organisation the leader and team are part of has malicious goals, a leader should not necessarily try to reach that goal. But this is where the conversation gets really hard, because values, social and moral norms are about rights and wrongs that are inherently subjective and context/ culture-dependent. What's acceptable today might not be in 50 years. That said, a good leader should:

1. Be clear about their own values – Know what you stand for.
2. Understand the values of their team – This helps build cohesion and trust.
3. Recognise social and moral norms – Know what is considered normal in your context.

Final Thoughts

So, what should an effective leader do? Based on this review, I plan to:

• Focus on leadership behaviours rather than getting caught up in styles.
• Be aware of how my team perceives my behaviour and adjust accordingly.
• Build trust through integrity, competence, consistency, loyalty, and openness.
• Be transparent about my values and respect those of my team.

Is this a magic bullet for leadership? Probably not. But it’s a solid, evidence-based foundation to build on.