As the availability of learning and business metrics increases, organizational requests for impact analysis continue to rise. Most follow a simple causal chain – from learning metrics to behavioral changes to strategic metrics. Each leading metric is expected to move in a specific direction and to influence the next downstream link. When applied to courses with clear, measurable learning objectives, this process is manageable. But few areas confound impact analysis like leadership development initiatives.

It seems illogical that leadership development – an estimated $24 billion industry – would expand its reach annually with little evidence of impact. It is obvious that employers find leadership development important and effective, but they are guided more by an anecdotal sense of progress, because assigning rigor to leadership impact analysis is confusing at best.

That confusion is rooted in the nebulous definition of leadership. How can organizations measure leadership development when leadership looks different from leader to leader and is applied differently from team to team? A more universal, measurable metric is leadership self-efficacy.

What Is Leadership Self-Efficacy?

Self-efficacy, a concept developed by psychologist Albert Bandura in the early 1960s, is probably best summed up in the Henry Ford quote, “Whether you believe you can or you can’t, you’re right.” Leadership self-efficacy identifies how the self-perception of a leader affects the people who follow him or her. Research in the last two decades has uncovered natural cognitive biases that keep people from accurate self-assessment, but studies in leadership self-efficacy find that even a poor self-assessment about leadership is irrelevant. People will change their behavior if their leader shows confidence in his or her own leadership skills.

In 2008, researchers set out to create a taxonomy that would define the factors of leadership self-efficacy. Their intent was to find the measurable factors of leadership, thereby making impact analysis possible. At the end of the research, the taxonomy consisted of 88 attributes across five categories. Further meta-analysis found that as multiple attributes were combined, the correlation with leadership effectiveness rose. Unfortunately, there was no consistent set of attributes that led to effectiveness.

What is interesting is that multiple studies have tested the correlation between a leader’s self-reported belief in their self-efficacy and the effectiveness of organizational change as a result of that leadership. The results indicate that “leadership self-efficacy” is a generalized definition consisting of a unique set of the discrete attributes. Self-belief in that generalized definition carries weight with people who look to leaders for direction. To paraphrase the Henry Ford quote, “Whether you believe you can or can’t, people will follow accordingly.”

Surveying for Self-Efficacy

The optimal survey for leadership self-efficacy is a standard Likert scale survey, with questions framed in first-person narrative (“I believe as a leader I can…”). The questions should focus on the five categories of the leadership self-efficacy taxonomy:

• Leadership attributes (how well you create a vision for the future)
• Management attributes (how well you oversee daily operations of your team)
• Problem-solving attributes (how well you resolve issues or empower your team to resolve issues)
• Social/communication attributes (how well you communicate your vision, team goals and responsibilities, etc.)
• General work attributes (how you approach work from an emotional perspective, with factors like humility or persistence)

The benefit of this type of metrics-gathering is that the rollout of leadership development is rarely enterprise-wide. Most leadership initiatives address specific needs, and that inconsistent delivery creates a natural control group to isolate the effects of the development training.

Causality and the Leadership Self-Efficacy Survey

Using self-reported leadership self-efficacy scores is a concise way to map the impact of leadership development from the learning opportunity to strategic impact. Instead of focusing on the many attributes of leadership, you can compare leaders’ belief in their self-efficacy before and after a development opportunity. You can also compare leaders’ changes to a control group (who did not receive the same development) to identify if the development opportunity correlates with an improvement in self-efficacy. If it does not, the effectiveness of the development opportunity is in question.

However, if there is an improvement in leadership self-efficacy post-training, you can connect the rise in self-efficacy to strategic impact. It is a direct causal link, without confusion and difficult-to-define nuance, and finally allows you to understand the impact of your leadership development programs.