What a study of eight leaders can and cannot tell us

The question

Behavior analysts often move into leadership because they are good clinicians. The skills that earned the promotion, careful assessment and precise programming, do not guarantee the skills the new role demands: understanding how a decision lands with a direct support professional on an overnight shift, anticipating how today’s fix will look in six months, and seeing how one program connects to the rest of the organization. I wanted to know whether those harder-to-name skills could be measured, and whether they had anything to do with how effective a leader is.

That question became my dissertation and, this year, a journal article (Witterschein, 2026). This piece is a plain-language account of what the study did and, just as important, what it did not do.

What the ABLA 2.0 measures

The ABA Leadership Assessment Tool, Version 2.0 (ABLA 2.0), is a 48-item instrument with two foundations. The first is the transformational and transactional leadership tradition, which describes what effective leaders do, from inspiring and developing people to setting clear expectations and following through (Bass & Riggio, 2006). The second is relational frame theory, a behavioral account of human language and cognition (Hayes et al., 2001).

From relational frame theory the instrument draws three constructs. Perspective-taking is the ability to relate events from the standpoint of I and you, here and there, now and then. Temporal reasoning is reasoning about before and after, and about consequences that unfold over time. Systems thinking is reasoning in hierarchies, relating a single shift to a home, a home to a program, and a program to an organization. My hypothesis was that leaders who are more flexible in these kinds of relating would also be more effective.

What the pilot found

Eight Board Certified Behavior Analysts in organizational leadership roles completed the instrument. Leaders who showed greater flexibility in taking others’ perspectives and in thinking across organizational levels scored higher on overall leadership effectiveness. Internal consistency looked encouraging, which suggests the items hang together well enough to keep refining them.

That is a promising first result. It is also exactly that: a first result.

What it cannot tell us

Eight people are not a sample to generalize from. Eight leaders answering 48 items produce 384 item responses, but that is still eight people, and any correlation or reliability estimate drawn from eight people can move a great deal when the ninth and tenth are added.

The sample was also a convenience sample. Leaders who volunteer for a study about leadership may differ in meaningful ways from those who do not.

The measures shared a single method. The same instrument captured both the relational skills and the effectiveness ratings, so some of the relationship between them may reflect how people respond to questionnaires rather than how they lead. This is the problem methodologists call common method variance, and it is one of the most familiar threats in organizational research (Podsakoff et al., 2003).

Nothing was checked against independent outcomes. The study did not include ratings from the people these leaders supervise, staff retention, or results for the people their programs serve.

Finally, the findings are correlational. Perspective-taking may help leaders become effective, effective leaders may get more practice taking perspective, or both may reflect something else, such as years of experience. A pilot study is not designed to sort that out. Its proper job is to test whether an instrument and a procedure are workable and to sharpen the questions for a larger study, not to estimate effects that others should rely on (Leon et al., 2011).

What it would take to know more

The next study needs more leaders, drawn from more organizations, so the sample looks more like the field. It needs effectiveness measured independently of the leaders themselves, for example through ratings by supervisees, retention data, and program outcomes. It needs test-retest data to show whether scores are stable, and enough participants to examine whether the items group into the constructs they are meant to measure. The strongest test would go further still: teach perspective-taking and systems thinking to a group of leaders, and see whether their behavior and their teams’ results change.

What a leader can take from it now

Because the evidence is preliminary, what follows is my practical suggestion, not a finding. The skills the ABLA 2.0 targets are cheap to practice. Before a significant decision, describe it out loud from the standpoint of the people it will reach: the direct support professional on the overnight shift, the family, the person served. Then describe it at three levels, for one person, for one home or classroom, and for the whole program. If the decision still makes sense from every one of those positions, it is probably a better decision. If it does not, you have learned where to look before anyone else has to tell you.

If you study leadership, organizational behavior management, or measurement in human services and want to help test these ideas properly, I would like to hear from you.