ERIC Identifier: ED409605
Publication Date: 1997-06-00
Author: Lashway, Larry
Source: ERIC Clearinghouse on
Educational Management Eugene OR.
Measuring Leadership Potential. ERIC Digest, Number 115.
In an age when business executives take leadership advice from Chinese sages
and Gothic warriors, it's clear that our notions of leadership are diverse, if
not confused. Everyone agrees that schools require effective leaders, but
identifying the necessary traits and skills is not easy.
People often believe they can recognize leadership when they see it, but most
schools can tell at least one rueful story about an administrator who turned out
to be less than the paragon everyone expected. Thus, interest is always high in
finding more reliable ways of identifying and selecting effective leaders.
Today, there are dozens of instruments that claim to measure leadership
capacities. Do they work? Is it possible to administer a test that will identify
The answer appears to be a qualified "yes," though every instrument has
limitations and must be used with care. This Digest provides a brief survey of
basic issues in the measurement of leadership potential.
WHY USE FORMAL ASSESSMENT?
Selection of a school leader is
a high-stakes decision that is often hindered by too many applicants and too
little information. After screening applicants through resumes and cover
letters, most districts rely on the personal interview to make their final
choice. Unfortunately, interviews are highly subjective and easily influenced by
appearance, mannerisms, and conversational skills (Mark Anderson 1991).
Frederick Wendel and colleagues (1992) point out that a poor choice has
economic and organizational consequences. "The cost of paying an ineffective
administrator thousands of dollars a year is readily apparent but putting a
price on lost opportunities is nearly impossible." Viewed in that light, using
objective information to supplement the usual interview process is simply good
sense. Kenneth and Miriam Clark (1996) cite evidence that using a carefully
chosen battery of tests along with other information provides better prediction
than using only professional judgment.
Formal assessments provide several advantages. First, they are objective;
scoring is either done by outsiders or is so straightforward there is no room
for personal biases. Second, they are designed to systematically explore key
leadership qualities. For example, a test can ask half a dozen well-structured
questions that probe the candidate's attitude toward teamwork, whereas an
interviewer may settle for one open-ended question. Third, feedback from the
assessment allows districts to measure candidates against a much larger pool,
comparing them to a field of thousands rather than the handful typically invited
WHAT DO LEADERSHIP TESTS MEASURE?
To be useful, any test
requires a clear conception of what is being measured. Unfortunately, after
decades of debate, researchers are no closer than ever to a universal definition
of leadership (Larry Lashway and colleagues 1997). Over the years, leadership
has been correlated to social dominance, vision, intelligence, interpersonal
competence, energy, technical skills, charisma, and many other qualities. (One
researcher claims to have found over 800 definitions.)
Every test thus reflects a particular set of assumptions about leadership.
For example, theorists have endlessly debated the difference between
"management" (routine administration) and "leadership" (inspiration and
A test such as "BENCHMARKS" zeroes in on the managerial dimensions
(delegating, mastering technical knowledge, solving problems). Conversely, the
Leadership Practices Inventory seeks to identify "extraordinary leadership" in
categories such as Challenging the Process, Inspiring a Shared Vision, and
Encouraging the Heart (Ellen Van Velsor and Jean Leslie 1991). Obviously, the
same candidate might perform very differently on these two instruments.
Whatever the definition, formal tests seldom provide a direct measure of
leadership performance; instead, they probe a candidate's PERCEPTIONS of his or
her performance. Because self-perceptions are highly subjective, test-makers
must validate results by correlating scores with actual performance. If those
scoring high on the test are shown to perform well on the job, the test has
value in predicting success.
HOW IS LEADERSHIP POTENTIAL MEASURED?
The most common
measures are paper-and-pencil instruments. These typically ask subjects to agree
or disagree with statements about their behaviors or beliefs, often using a
Likert-type scale (1 means "always" and 5 means "never"). In some cases, a
similar instrument is given to superiors or subordinates, asking them to rate
the leader on the same criteria.
Scoring of tests is done in several ways. In some cases, the test must be
sent to the publisher for scoring; sometimes it can be scored locally; and
sometimes it can be scored by anyone who has received special training or
certification. The meaning of the scores is determined by comparing them to
others who have taken the test or to a group of proved leaders. In most cases,
certain answers are presumed to be better (more predictive of future success); a
high score on "problem-solving" is considered preferable to a low score.
An alternative approach to measurement is used by assessment centers such as
the ones sponsored by NASSP. Assessment centers provide a series of simulated
leadership tasks and use trained assessors to rate performance on a variety of
dimensions, including leadership, judgment, decisiveness, and stress tolerance.
Unlike written tests, assessment centers measure performance on tasks similar to
those encountered in real settings. (Anderson).
WHAT ARE THE LIMITS OF FORMAL ASSESSMENT?
instruments are relatively simple, inexpensive, and understandable. The major
disadvantage is that a written answer on a test is at least one step removed
from real behavior; the world is filled with people who test well but fail life.
Wendel and colleagues note that paper-and-pencil tests provide SIGNS rather than
SAMPLES of leadership capability (just as passing a written driver's test does
not guarantee the ability to handle a car effectively).
Because self-perceptions are so subjective (especially in the hiring process,
where job-seekers have every reason to paint a rosy picture), feedback from
others is a useful supplement. Clark and Clark note that "ratings of superiors
by subordinates are the best predictors of good performance as leaders."
Unfortunately, not all instruments offer this kind of "360-degree feedback," and
even those that do may be irrelevant for entry-level candidates.
Even assessment centers, while more performance oriented, do not directly
measure on-the-job behavior. Although "inbasket" exercises and similar problems
can provide helpful insights into a candidate's skills, they lack the rich
context of a real-life situation, which may confront a leader with dozens of
Finally, few leadership tests have been developed specifically to select
school leaders. Most have been developed in business settings, and their
appropriateness for school settings is always a matter for careful deliberation.
In addition, relatively few leadership tests have been studied for PREDICTIVE
VALIDITY; that is, there is no firm evidence that performance on the test today
will predict successful leadership over a period of years (Van Velsor and
HOW SHOULD TESTS BE USED IN SELECTING LEADERS?
tests in the selection process requires careful attention to several issues.
First, make sure the test matches closely the demands of the position. For
example, if district personnel believe the position will require exceptional
motivational skills, they should use a test that provides adequate feedback on
that dimension. (A detailed description of sixteen well-designed instruments can
be found in the review by Van Velsor and Leslie.)
Second, determine whether the test satisfies basic statistical criteria such
as validity and reliability. Van Velsor and Leslie warn that using tests for
selection requires special attention to predictive validity; for legal and
ethical reasons, schools should avoid putting too much weight on a test that has
not been shown to measure future (as opposed to current) leadership performance.
Finally, carefully interpret test results rather than reduce them to a simple
numerical comparison. What areas of strength and weakness does the candidate
show? How well does this profile mesh with the requirements of the position? To
what extent are test results consistent with the candidate's resumes and
Like all tests, measures of leadership are designed to provide reliable data,
not make decisions. A well-designed instrument will add depth and richness to
the selection process, but it will never eliminate the need for careful
Anderson, Mark. "Principals: How to Train,
Recruit, Select, Induct, and Evaluate Leaders for America's Schools." Eugene,
Oregon: ERIC Clearinghouse on Educational Management, 1991. ED 337 843.
Clark, Kenneth E., and Miriam Clark. "Choosing to Lead." Greensboro, North
Carolina: Center for Creative Leadership, 1996.
Lashway, Larry; JoAnn Mazzarella; and Thomas Grundy. "Portrait of a Leader."
In "School Leadership: Handbook for Excellence," third edition. Edited by Stuart
C. Smith and Philip K. Piele. Eugene, Oregon: ERIC Clearinghouse on Educational
Van Velsor, Ellen, and Jean Britton Leslie. "Feedback to Managers. Volume II:
A Review and Comparison of Sixteen Multi-rater Feedback Instruments."
Greensboro, North Carolina: Center for Creative Leadership, 1991. 301 pages. ED
Wendel, Frederick C.; Allan H. Schmidt; and James Loch. "Measurements of
Personality and Leadership: Some Relationships." Lincoln, Nebraska, University
of Nebraska, 1992. 121 pages. ED 350 694.