Most leadership research asks followers to describe their leader. This asked something else: what is that leadership doing to the person answering the question? Here is the study, the number, the hypothesis that failed, and what none of it proves.
Davis, L. (2018)
The Relationship Between Servant Leadership
and Psychological Capital:
A Follower’s Perspective
Ed.D. dissertation, George Washington University
Leadership is almost always measured from the leader's end. Rate your manager. Describe their style. Score them against a model. The result is a description of a person, and the person being described is rarely the one whose behaviour you were hoping to change.
This study turned the instrument around. It asked employees to report on their own psychological capital — the hope, efficacy, resilience and optimism they had available for their work — and specifically on how their relationship with their leader affected it.
That framing matters commercially as well as academically. An executive can disagree with a description of their style forever. It is much harder to argue with a measured change in what their organization has left to spend on a difficult year.
r² = .442 · F(1,400) = 317.06 · p < .001
By the conventional benchmark for the social sciences, anything above .25 is a large effect. This is well above it.
With all seven dimensions of servant leadership entered together, the model accounted for 46.6% — F(7,394) = 49.134, p < .001. No individual dimension outperformed servant leadership taken as a whole, which is itself a useful result: it supports treating the construct as a single thing rather than shopping among its parts.
The second hypothesis was that helping followers grow and succeed — the mentoring-and-development dimension, the one every leadership programme is built around — would relate to psychological capital more strongly than any other dimension.
It did not. Taken on its own it placed third. Entered alongside the other six, it accounted for roughly one percent of the variance. The hypothesis was not supported.
What did hold up was empowerment, and it was the only dimension that did. Empowerment was the single dimension that predicted psychological capital over and above all six others — 5.66% of the variance with the rest held constant, 3.2% with the rest already in the model.
The dissertation records the reaction in four words: “This finding was unexpected.”
A failed hypothesis is the most informative thing a practitioner can show you, because it is the one result that could not have been arranged. Everything else on this site could, in principle, have been selected to flatter it. This could not.
| Dimension, measured alone | r² |
|---|---|
| Putting followers first | .343 |
| Empowerment | .334 |
| Helping followers grow and succeed | .290 |
| Behaving ethically | .255 |
| Creating value for the community | .225 |
| Conceptual skills | .212 |
| Emotional healing | .193 |
All seven significant at p < .001, N = 402. Measured alone, putting followers first edges empowerment. Measured against each other, empowerment is the only one that survives — which is the finding that matters, and the one worth stating carefully.
Reproduced from the limitations chapter of the dissertation itself, not written for this page.
If what a leader does to their people is measurable from the follower's side, and accounts for something on the order of forty percent of the psychological resources those people bring to a difficult change, then surveying the executive team about itself is the wrong instrument pointed in the wrong direction.
It is also why psychological capital is measured before a change begins rather than after. And it is why the empowerment result — decision rights, latitude, the authority to act — gets more attention in an engagement than the mentoring conversation that most coaching programmes are built around.
With the caveat, stated again because it belongs next to the claim every time: one study, one data collection, a single-item measure of the dimension in question.
Davis, L. (2018). The Relationship Between Servant Leadership and Psychological Capital: A Follower’s Perspective. Doctoral dissertation, George Washington University. Graduate School of Education and Human Development, Executive Leadership in Human and Organizational Learning.
Read the full dissertation on GW ScholarSpace →
Open access. Every figure on this page can be checked against it.
Instruments referenced: the SL-7 (Liden, Wayne, Meuser, Hu, Wu & Liao, 2015) and the Psychological Capital Questionnaire (Luthans, Avolio, Avey & Norman, 2007). Both are copyrighted and licensed by their authors; neither is reproduced here.
The wider evidence base for the practice, including the parts that do not support it, is set out in The 6% and on the diagnostics page.
If you want to know whether any of this applies to the change you are actually carrying, that is a conversation rather than a report.