Development of an Instrument to Measure Level 5 Leadership
Submitted to Regent University
School of Global Leadership and Entrepreneurship
In partial fulfillment of the requirements
for the degree of
Doctor of Philosophy in Organizational Leadership
Wilbur A. Reid III
December 2012
Instrument to Measure Level 5 Leadership ii
Instrument to Measure Level 5 Leadership
Abstract
In Good to Great, one of the best-selling business books in history, Jim Collins
(2001) evaluated companies whose stock market performance rose from good to
great and found they were all led by a CEO with a surprising blend of personal
humility and professional will. These leaders became known as Level 5 leaders.
Although the concept has gained great acceptance in the business community and
popular press, a validated instrument to measure Level 5 leadership has not yet
been developed. The objective of this research project is to develop a validated
scale utilizing the attributes that Collins (2001) used to describe these leaders. This
study began with 99 attributes from Collins (2001) which an expert review
committee refined to 74. An online survey was developed that included the 74
attributes, and 349 subjects evaluated their bosses on a 10-point semantic
differential scale for each attribute. Literature has suggested that Level 5 is the
same as servant leadership, and Collins (2006) proposed eight untested questions to
determine if an individual is Level 5. To test these items, the survey also included a
10-item validated servant leadership scale and Collins’ (2006) eight questions. The
results showed that there are two very distinct constructs within the 74 attributes
that match Collins’ (2001) proposed personal humility and professional will
constructs and explain 55.2% of the variance within the attributes. The final scale
contains five attributes of personal humility and five attributes of professional will.
Reliability is very good with Cronbach’s alpha of .833 and .845 respectively. The
analysis also showed that there is a statistically significant positive relationship
between the Level 5 attributes, servant leadership, and Collins’ (2006) eight
questions. The results of this research open up the doors for implementation in
organizations to identify Level 5 leaders and for a wealth of future research on this
important leadership construct.
Instrument to Measure Level 5 Leadership
Dedication
This dissertation is dedicated to the first two Wilbur Reids on whose
shoulders I stand. My dad and my grandfather have been the embodiment of Level
5 leadership: personal humility based on the servant leadership modeled by Christ
and an intense professional will to do whatever it took to advance His kingdom. My
grandfather once wrote to me and said: “You wear my name with honor. For this, I
am very pleased.” I pray that this dissertation and my life will bring honor to our
name and to His name.
Instrument to Measure Level 5 Leadership
Acknowledgements
The process of earning a Ph.D. requires support from a variety of sources,
and it is appropriate to acknowledge and thank four specific groups of people who
have provided valuable support throughout the past 4 years.
First, it has been a privilege and a pleasure to sit and learn at the feet of the
professors of the School of Business and Leadership at Regent University.
Specifically, Dr. Bruce Winston, Dr. Corne’ Bekker, Dr. Dail Fields, Dr. Paul Carr,
and Dr. Mihai Bocarnea exemplify the personal humility and professional will of
Level 5 leadership. Special thanks for the coaching and guidance of Dr. Winston as
the chair of this dissertation as well as Dr. West and Dr. Wood as committee
members who provided valuable insight.
The journey throughout this Ph.D. process has been immeasurably more
enjoyable due to my fellow members of the 2009 cohort. We studied hard together,
dialogued online, and truly enjoyed each other’s company. Though not practical to
list everyone, there are several, mostly from Group 3, who deserve special mention:
John Wilson, David Peltz, Andrea Ramirez, David Oginde, Laurel Emory, and
Heidi Frederick.
The support of friends, notably at Journey Christian Church, SP Richards,
and on Facebook, has been a great encouragement to continue forward.
Finally, my family has been awesome. When I told my beautiful wife Chris
that I wanted to go back to school to get a Ph.D., her first reaction was, “You are
nuts!” When she realized that I was serious, she was very supportive and
encouraging. My teenage daughters Chelsea, Kerri, and Kayleigh were
understanding and patient when I retreated to the mancave for hours at a time to
work on papers. Dad and mom, Wilbur and Linda Reid, Jr., have shown their pride
in what I am doing, which has also been an encouragement.
Instrument to Measure Level 5 Leadership
Table of Contents
Abstract.....................................................................................................................iii
Dedication.................................................................................................................iv
Acknowledgements....................................................................................................v
List of Tables and Figures......................................................................................viii
Chapter 1 – Introduction............................................................................................1
Theory of Level 5 Leadership.............................................................................2
Method and Analysis..........................................................................................4
Research Goals....................................................................................................6
Limitations of Study...........................................................................................6
Chapter 2 – Literature Review...................................................................................7
Personal Humility...............................................................................................8
Professional Will...............................................................................................10
Servant Leadership............................................................................................12
Charisma...........................................................................................................15
Secondary Literature.........................................................................................16
Summary of Characteristics..............................................................................18
Chapter 3 – Method..................................................................................................22
Determine Clearly What it is You Want to Measure........................................22
Generate an Item Pool.......................................................................................22
Determine the Format for Measurement...........................................................23
Expert Panel Review.........................................................................................23
Consider Inclusion of Validation Items............................................................28
Administer Items to a Development Sample....................................................29
Evaluate the Samples........................................................................................29
Reliability and Validity.....................................................................................30
Chapter 4 – Results..................................................................................................32
Demographic Variables....................................................................................32
Factor Analysis.................................................................................................34
Defining Variables............................................................................................42
Discriminant Validity Tests..............................................................................46
Instrument to Measure Level 5 Leadership
Final Instrument................................................................................................54
Summary...........................................................................................................56
Chapter 5 – Discussion.............................................................................................57
Evaluation of Findings......................................................................................57
Implications of Research...................................................................................58
Limitations of Study.........................................................................................59
Recommendations for Future Research............................................................59
Summary...........................................................................................................60
References................................................................................................................61
Appendix A – Survey Summary..............................................................................67
Appendix B – Human Subjects Review Board Application....................................79
Instrument to Measure Level 5 Leadership
List of Tables and Figures
Table 1: Professional Will and Personal Humility Results........................................5
Table 2: Positive and Negative Terms Describing Personal Humility and
Professional Will (Collins, 2001, 2002, 2005, 2006, 2009, 2011; Collins &
Hanson, 2001; Collins & Rose, 2009)..............................................................19
Table 3: Expert Panel for Instrument Verification...................................................24
Table 4: Average Scores of Relevance from Expert Panel......................................25
Table 5: Reliability as Measured by Cronbach’s Alpha..........................................31
Table 6: Gender and Age of Subjects......................................................................32
Table 7: Gender of Participant and Boss..................................................................33
Table 8: Age of Participant and Boss.......................................................................33
Table 9: Boss Position and Type of Organization...................................................34
Table 10: Religious Affiliation and Religious Commitment of Boss......................34
Figure 1: Scree plot for principal component analysis.............................................35
Table 11: Total Variance Explained........................................................................36
Table 12: Structure Matrix of Two Components.....................................................40
Table 13: Descriptive Statistics of Four Key Variables for Each Leader (N = 349)42
Table 14: Reliability of Scales.................................................................................43
Figure 2: Professional will and personal humility results (N =349)........................44
Table 15: Collins’ (2006) Eight Questions Total Variance Explained....................44
Table 16: Tests of Normality...................................................................................45
Table 17: Correlation Coefficients Using Spearman’s Rho.....................................46
Table 18: Leaders Identified as Level 5 Compared to Individual Constructs..........47
Table 19: Levene’s Test for Equality of Variances to Determine if Responses are
Normally Distributed........................................................................................47
Table 20: Independent Samples Test of L5 Within Key Variables.........................48
Table 21: Demographic Comparisons Between Level 5 and non-Level 5 Leaders.49
Table 22: Significance of Level 5 Leader by Demographic Variable.....................51
Table 23: Bonferroni Post Hoc Test for Religious Commitment of Level 5 Leaders
...........................................................................................................................52
Table 24: Bonferroni Post Hoc Test for Organization Type of Level 5 Leaders... .53
Instrument to Measure Level 5 Leadership
Table 25: Level 5 Leadership Scale Attributes........................................................54
Table 26: Final Level 5 Leadership Scale (L5LS) Attributes..................................55
Table A1: Level 5 Leadership..................................................................................68
Table A2: Servant leadership...................................................................................72
Table A3: Collins’ Eight Questions.........................................................................73
Table A4: Demographics.........................................................................................76
Instrument to Measure Level 5 Leadership
Chapter 1 – Introduction
Jim Collins is a business consultant, professor, and author who has studied
the rise and fall of organizations and sold more than 10 million books with
observations and conclusions regarding the drivers of success and failure (Collins,
2012; Woolridge, 2011). Collins’ first two books, Built to Last (Collins & Porras,
1994) and Good to Great (Collins, 2001), became seminal publications for this
generation of business leaders. They were both listed among the top 20 most
influential business books by Forbes Magazine (Ackman, 2002) and occupied
positions on the Businessweek best-seller list for over 6 years each, selling millions
of copies in dozens of languages (Bennett, 2011). The books have “made a
worldwide impact on management and leadership practice and research” (Caulkins,
2008, p. 217). The books’ peculiar terms and phrases have become part of the
lexicon of American business: getting the right people on the bus, facing the brutal
facts, big hairy audacious goals (BHAGs), first who. . . then what, the flywheel and
the doom loop, the hedgehog concept, clock building and time telling, and Level 5
leadership (Collins, 2001; Collins & Porras, 1994). Of all the ideas that Collins has
shared in his books, perhaps the most surprising and meaningful concept is Level 5
leadership. Collins found that companies that rose from good to great were all led
by humble CEOs who had “an absolute, obsessed, burning, compulsive ambition
for the organization” (Collins, 2009, 1:15). Collins identified this as Level 5
leadership. There are, however, no instruments to measure Level 5 leaders
(Liccardo, 2007). May (2006) summarized this problem in a book review:
Level 5 leadership is vague. The only trait people seem to agree on is that
level 5 leaders have humility. Humble leaders can be a good thing, but if
Jim Collins can’t even tell whether or not Jack Welch was a level 5, what
chance do the rest of us have? Isn’t Collins supposed to be the expert on
Level 5 leadership? Hasn’t more been written about Jack Welch than about
most other CEOs? And Collins can’t tell? He’s either being diplomatic and
refusing to say “no, Welch wasn’t” or Level 5 leadership is business jibber
jabber. (para. 8)
1
Instrument to Measure Level 5 Leadership
The purpose of this dissertation is to develop a scientifically validated
instrument to measure Collins’ (2001) Level 5 Leadership. Collins theorized that
there are many people who have the seed of Level 5 leadership but have not
achieved the higher titles of leadership such as CEO because the ambition and
egoistic needs that often drive people into leadership stand at odds with the
humility and servanthood that is required for Level 5 leadership. Collins (2001)
stated,
I believe—although I cannot prove—that potential Level 5 leaders are
prevalent in our society. The problem is not, in my estimation, a dearth of
potential Level 5 leaders. They exist all around us, if we just know what to
look for. (p. 37)
In a recent speech, Collins (2011) added that “our problem is not a lack of Level 5
leadership. Our problem is the lack of wisdom to put Level 5 at the top” (0:01).
Collins’ organization, ChimpWorks LLC, confirmed, however, that there is no
research data in this area (S. B. Toll, personal communication, February 28, 2012).
An instrument to identify the complementary presence of professional will and
personal humility in individuals provides an actionable tool to identify the seeds of
Level 5 leadership.
Theory of Level 5 Leadership
An overview of the literature from Collins provides the foundational
concepts to research and understand the levels of leadership. Collins and his
research team examined the stock performance of 1,435 corporations over a period
of 40 years, looking for companies that had a history of being good but then
became great. They identified 11 corporations as having a period of sustained
growth and success, far outpacing the market and industry.
The company's fifteen-year cumulative stock returns had to be at or below
the general stock market, punctuated by a transition point, and then
cumulative returns had to be at least three times the market value over the
next fifteen years. (Collins, 2001, p. 6)
2
Instrument to Measure Level 5 Leadership
The good-to-great company performance had to be independent of its industry.
These good-to-great companies were benchmarked to a comparable company from
the same industry that did not achieve greatness. One of the key characteristics that
separated these good-to-great companies from comparison companies in the same
industry that did not become great was the leadership style of the CEO. Collins
identified a hierarchy of five levels of people in an organization: “1) highly capable
individual, 2) contributing team member, 3) competent manager, 4) effective
leader, and 5) level 5 executive” (Collins, 2001, p. 20). Descriptions of Level 3 and
Level 4 leaders are brief because Collins (2001) indicated that they are “discussed
extensively by other authors” (p. 21). A Level 3 leader is a competent manager who
“organizes people and resources toward the effective and efficient pursuit of
predetermined objectives” (Collins, 2001, p. 20). A Level 4 leader is an effective
leader who “catalyzes commitment to and vigorous pursuit of a clear and
compelling vision, and stimulates the group to high performance standards”
(Collins, 2001, p. 20). Since it took 15 years of great performance to identify Level
5 leadership, it is difficult to recognize these valuable leaders quickly.
Level 5 leadership is unique. Collins (2001) defined a Level 5 leader as one
who “builds enduring greatness through a paradoxical blend of personal humility
and professional will” (p. 20). This is the highest and most effective level of
leadership. Level 4 leaders are often charismatic and egocentric and do not set up
their successors for success, whereas Level 5 leaders are more “plow horse than
show horse” (Collins, 2001, p. 39). They are self-effacing and understated and
make sure that those around them are set up for success. They defer praise and
share it with their team; however, this meekness and servant approach should not
be misinterpreted as weakness. Level 5 leaders have a fierce stoic resolve and are
ambitious first and foremost for the organization.
The study of Level 5 leaders must utilize two distinct constructs. “Level 5
leaders are a study in duality: modest and willful, humble and fearless” (Collins,
2001, p. 22). The first construct is one of a humble, servant leader. Collins used the
following words to describe this characteristic of Level 5 leaders: modest, servant,
shy, awkward, humble, quiet, reserved, placid persona, gracious, mild-mannered,
3
Instrument to Measure Level 5 Leadership
self-effacing, and understated. A Level 5 leader demonstrates a compelling
modesty, shunning public adulation and never boasting. A Level 5 leader acts with
quiet, calm determination and relies principally on inspired standards rather than
charisma to motivate. He or she channels ambition into the company, not the self,
and sets up successors for even greater success in the next generation. Collins
(2001) described how Level 5 leaders look out the window, not in the mirror, to
apportion credit for the success of the company, attributing the success to other
people, external factors, and good luck.
In trying to decide on a term to describe these Level 5 leaders, the research
team considered using the term servant leader (Collins, 2001). They decided
against it, however, because it was not a comprehensive description that accounted
for the idea of professional will, and the research team was afraid people would
perceive that these leaders were meek or weak. Despite this concern, servant
leadership may be a valid description of the first construct of Level 5 leadership.
The second construct of Level 5 leadership is professional will. Collins
(2001) also described this as ferocious and unwavering resolve, inner intensity,
stoic resolve, ambition first and foremost for the company, and “an absolute,
obsessed, burning, compulsive ambition for the organization” (Collins, 2009, 1:25).
The leader with a strong personal will
creates superb results, a clear catalyst in the transition from good to
great . . . . Demonstrates an unwavering resolve to do whatever must be
done to produce the best long-term results, no matter how difficult. . . . Sets
the standard of building an enduring great company; will settle for nothing
less . . . . Looks in the mirror, not out the window, to apportion
responsibility for poor results, never blaming other people, external factors,
or bad luck. (Collins, 2001, p. 36)
Method and Analysis
An instrument to measure Level 5 leadership required an online survey with
a list of leadership characteristics compiled from a review of the literature,
interviews, and presentations from Collins regarding Level 5 leadership and all of
4
Instrument to Measure Level 5 Leadership
the relevant keywords and synonyms. Subjects were asked to think of their current
or most recent boss and to what extent their boss exhibits the characteristics of the
various Level 5 traits (e.g., “To what extent does your boss exhibit humility?”). The
subjects chose from a 10-point semantic differential scale, ranging from 1 (not at
all) to 10 (exactly). Factor analysis determined the number of latent variables that
exist in the original pool of terms that describe personal humility and professional
will. The items were condensed so that the variation was accounted for by the
smallest number of items that best describe the variation. Cronbach’s alpha of at
least .70 insures reliability (Hair, Black, Babin, Anderson, & Tatham, 2005). The
results were tested for content validity, criterion-related validity, and construct
validity. DeVellis (2012) developed a methodology for scale development that
functioned as a guide for developing a scale to measure Level 5 leadership.
The personal humility and professional will responses were summed for
each leader to determine two separate scores for humility and will. These scores
were then benchmarked against the aggregate score for each attribute to determine
if each leader was either above average or below average in each category. The
leaders were assigned to one of four categories: (a) weak leadership, below average
in both categories; (b) humble, above average humility but below average will; (c)
strong will, above average professional will but below average humility; and (d)
Level 5, above average humility and above average will (see Table 1).
Table 1: Professional Will and Personal Humility Results
Professional will weak Professional will strong
Humility weak Weak leadership Strong will
Humility strong Humble Level 5
In addition to this list of characteristics to determine humility and will, there
are two instruments that have been suggested as possible measures of Level 5
5
Instrument to Measure Level 5 Leadership
leadership. Patterson, Redmer and Stone (2003) and Drury (2004) suggested that
servant leadership may be the same as Level 5 leadership and recommended
empirical research to determine the relationship between the two. In addition,
Collins (2006) suggested that there are eight questions that can be asked to
determine if a person is a Level 5 leader. These eight questions and a validated
servant leadership scale were added to the extensive list of factored and validated
leader traits to determine if there is a relationship between the leadership
characteristics of servant leadership, Collins’ Level 5 questions, and the
characteristics of Level 5 leaders. The literature review in Chapter 2 articulates the
specific details of this theory. The survey was administered through Survey
Monkey™ to a wide variety of subjects who were reached using social media.
Research Goals
The core research problem is that there is no way for leaders to measure
Level 5 leadership within their organizations. This study will answer three research
questions: (a) Utilizing the attributes and characteristics from literature, can a
statistically valid instrument be developed to measure Level 5 leadership? (b) Is the
personal humility construct of Level 5 leadership the same as servant leadership?
(c) Do Collins’ (2006) eight questions to test Level 5 leadership correlate with the
attributes and characteristics that he used to describe Level 5 leadership? The goal
of this study is to provide an accurate and effective instrument to measure Level 5
leadership within individuals.
Limitations of Study
There are a number of applications of an instrument to measure Level 5
leadership that are not addressed in this study. For example, how does Level 5
leadership compare to John Maxwell’s (2011) five levels of leadership or Likert’s
System 5 leadership (Likert & Likert, 1976)? Common method variance and social
desirability response bias are minimized because (a) this is not a self-reporting
survey and (b) responses are evaluated relative to each other instead of a
comparison to an established benchmark.
6
Instrument to Measure Level 5 Leadership
Chapter 2 – Literature Review
Jim Collins (2001) provided the primary literature surrounding Level 5
leadership. In addition to the introduction of the term in Good to Great (Collins,
2001), Collins either authored or was extensively quoted in diverse journal and
popular press articles published in Harvard Business Review (Collins, 2005),
Strategy and Leadership (Finnie & Abraham, 2002), and Newsweek (McGinn &
Silver-Greenberg, 2005). In addition, he published three videos and 16 audio clips
regarding Level 5 leadership on his website (www.jimcollins.com) and provided
interviews and presentations that are available on his YouTube channel (Collins,
2009, 2011; Collins & Rose, 2009). Secondary literature sources come from a
variety of sources from a diverse pool of authors.
Collins embarked on the good-to-great project with a team of researchers
and a tremendous amount of data. Once the research team identified the 11 good-
to-great companies and the comparison companies, they began to pour through the
data to identify the characteristics that distinguished the great company from the
good company. Collins specifically instructed the team to ignore leadership as a
factor because he believed it was a general answer that was often used to explain
anything that researchers did not understand. To his surprise, he walked into the
research room one day and found that the research team had locked arms and stated
“Today is the day, Jim, that we have decided to tell you that you are wrong”
(Collins & Rose, 2009, 4:22). The research team had discovered that, although
there was strong leadership in both the great and comparison companies, the great
companies were led by people who were “cut from the same cloth” (Collins, 2001,
p. 22). They were different than that of the comparison leaders.
To illustrate the difference between these leaders who presided over the
change from good to great, Collins (2001) presented stories about three of the
leaders: Darwin Smith from Kimberly-Clark, Colman Mockler from Gillette, and
David Maxwell from Fannie Mae. Smith was Kimberly-Clark’s in-house attorney
who did not feel qualified to take the reins of the paper giant. He and his leadership
team decided that K-C’s core business of paper was doomed to mediocrity and
made the stunning decision to sell the paper mills and reinvest the proceeds in the
7
Instrument to Measure Level 5 Leadership
consumer business to develop brands like Kleenex, Huggies, Kotex, Depends, and
Cottonelle. Mockler was a quiet gentleman who displayed fierce resolve to stave
off a hostile takeover bid and retain control of the company that he guided to
greatness. Maxwell displayed a commitment first and foremost for the company
when he gave up $5.5 million in bonuses at retirement for the good of Fannie Mae
(Collins, 2001). The following sections provide an overview of the literature
regarding the two constructs of Level 5 leadership: personal humility and
professional will.
Personal Humility
The first construct of Level 5 leadership is personal humility. Collins (2001)
defined this idea of humility by first describing the characteristics of some of the 11
good-to-great leaders. Darwin Smith of Kimberly-Clark was described as a shy
man who had a lack of any pretense or air of self-importance. He felt unqualified to
accept the job of CEO. At his retirement 20 years later, he said that “he never
stopped trying to become qualified for the job” (Wicks, 1997, p. 10). Though Smith
was the CEO of a Fortune 500 company, he avoided the spotlight and was an
ordinary man who usually associated with people that society would not consider
important. Colman Mockler was described as a quiet, reserved, courteous, gracious,
gentleman with a placid persona. David Maxwell was highlighted because his
ambition was first and foremost for the company and not himself. Ken Iverson’s
lifestyle was described as simple, humble, and modest. The Level 5 leaders did not
talk about themselves; when others talked about them they said that “it wasn’t just
false modesty. Those who worked with or wrote about the good-to-great leaders
continually used words like quiet, humble, modest, reserved, shy, gracious, mild
mannered, self-effacing, understated, did not believe his own clippings; and so
forth” (Collins, 2001, p. 27). In addition to these three good-to-great CEOs, Collins
(2001) also featured Abraham Lincoln as a Level 5 leader and cited his personal
modesty, shy nature, and awkward manner. Level 5 leaders are selfless and servant
leaders. In the research interview, Alan Wurtzel of Circuit City was asked to
describe the difference between himself and his Level 4 counterpart at the
8
Instrument to Measure Level 5 Leadership
comparison company. He stated that it was the difference between “the show horse
and the plow horse—he was more of a show horse, whereas I was more of a plow
horse” (Collins, 2001, p. 33).
Although people in the organization and outside observers credited the
Level 5 leaders as the key to elevating the company from good to great, these
leaders do not accept the credit and often credit luck.
Level 5 leaders look out the window to apportion credit to factors outside
themselves when things go well (and if they cannot find a specific person or
event to give credit to, they credit luck). At the same time, they look in the
mirror to apportion responsibility, never blaming bad luck when things go
poorly. (Collins, 2001, p. 35)
In contrast, the Level 4 leaders of the comparison company did just the opposite,
taking credit for success and blaming bad luck for failures. The Level 5 leaders of
the good-to-great companies were contrasted with the Level 4 leaders in the
comparison companies like Lee Iacocca of Chrysler and Al Dunlap of Scott Paper.
The Level 4 CEOs were described as charismatic, egocentric, tyrannical,
celebrated, personally ambitious, and larger-than-life celebrities. They had
gargantuan personal egos and were seekers of fame, fortune, adulation, and power.
These leaders were boastful and concerned with their own personal greatness,
loudly beating their own chest and bragging about their accomplishments. Level 4
leaders do not set up successors for success. Iacocca was described as treating
successor candidates “the way Henry the VIII treated wives” (Taylor, 1992, p. 1).
In contrast to the Level 4 leader, the Level 5 leader can be summarized as
one who demonstrates a compelling modesty, shunning public adulation and never
boastfing. He or she acts with quiet, calm determination, relying principally on
inspired standards, not inspiring charisma, to motivate. He or she channels
ambition into the company, not the self, and sets up successors for even greater
success in the next generation. The Level 5 leader looks out the window, not in the
mirror, to apportion credit for the success of the company, crediting other people,
external factors, and good luck (Collins, 2001). Collins (as cited in Serfontein &
Hough, 2011) concluded that humility is key to successful leadership: “we cannot
9
Instrument to Measure Level 5 Leadership
see something from the perspective of another if we do not have deep humility,
because without it we impose our own perspective or analyze things from our own
perspective only; we will not see the other person’s viewpoint” (p. 396). Leaders,
therefore, must reassess their role regarding practice and power within the
organization, and the organization must consider whether their leaders recognize
and appreciate the implications of their power (Goleman, 2000).
Professional Will
The second construct of Level 5 leadership is professional will. Ten years
after Good to Great was published, Collins acknowledged that his description of
Level 5 leadership “focused heavily on the humility aspect of Level 5 leaders”
(Collins & Hanson, 2011, p. 32) but said the most important trait of Level 5 leaders
is that they are “incredibly ambitious, but their ambition is first and foremost for
the cause, for the company, for the work, not themselves” (Collins & Hanson, 2011,
p. 32). Although Collins and secondary writers seem to have dwelled more on the
concept of personal humility in leaders because it seems to be a novel concept in
the corporate world, Level 5 leadership is equal parts humility and “ferocious
resolve, an almost stoic determination to do whatever needs to be done to make the
company great” (Collins, 2001, p. 30). Collins (2009) described this professional
will in Level 5 leaders as “an absolute, obsessed, burning, compulsive ambition that
was not about them” (1:15). After describing Darwin Smith’s personal humility
previously referenced, Collins (2001) stated that “if you were to think of Darwin
Smith as somehow meek or soft, you would be terribly mistaken. His awkward
shyness and lack of pretense was coupled with a fierce, even stoic, resolve toward
life” (p. 18). This intense, ferocious resolve was coupled with an incredible work
ethic (Collins, 2001). Colman Mockler was described as a strong and tireless
fighter with an inner intensity to make whatever he touched the best that it could
be. David Maxwell was ambitious for the company, not himself. “Level 5 leaders
are fanatically driven, infected with an incurable need to produce results” (Collins,
2001, p. 30). Level 5 leaders are fearless and not afraid to “draw a line in the sand”
(Collins, 2005, p. 8). They have a workmanlike diligence, are a clear catalyst in the
10
Instrument to Measure Level 5 Leadership
transitions from good to great, and set the standard of greatness. They will settle for
nothing less.
Although the term professional will has not been utilized in academic
research, there are other terms that Collins (2001) used to describe professional will
that are supported by academic research. Three of the more prominent terms that
have been utilized in research are intrinsic motivation, self-determination, and work
ethic. Like the professional will of a leader who is focused intently on the
organization, motivation is about energy and direction of behavior (Deci & Ryan,
1985). Maslow (1946) stated that “man is a perpetually wanting animal” (p. 370)
and defined a hierarchy of needs that man seeks to fulfill (Maslow, 1954).
However, there are varying levels of motivation within each person based on innate
and social influences (Markus & Kitayama, 1991; Ryan & Deci, 2000).
Motivation concerns energy, direction, persistence, and equifinality—all
aspects of activation and intention. Motivation has been a central and
perennial issue in the field of psychology, for it is at the core of biological,
cognitive, and social recognition. Perhaps more important, in the real world,
motivation is highly valued because of its consequences: Motivation
produces. It is therefore of preeminent concern to those in roles such as
manager, teacher, religious leader, coach, health care provider, and parent
that involve mobilizing others to act. (Ryan & Deci, 2000, p. 69)
Intrinsic motivation does not rely on exterior sources of influence but is a person’s
“natural inclination toward assimilation, mastery, spontaneous interest, and
exploration” (Ryan & Deci, 2000, p. 70).
Since it is often difficult to differentiate between extrinsic and intrinsic
motivation, self-determination theory has gained widespread attention as a theory
of work motivation (Gagne & Deci, 2005). Self-determination theory is concerned
with the intrinsic motivation behind the choices that people make and identifies
three innate needs that must be satisfied for optimum growth: competence,
relatedness, and autonomy (Deci & Ryan, 1985). Individuals with high levels of
intrinsic motivation and self-determination are more likely to succeed in their work
(Deci & Ryan, 1985). A leader who exhibits intrinsic motivation and self-
11
Instrument to Measure Level 5 Leadership
determination to perform at a high level on the job will likely meet the definition of
professional will that is found in Collins literature (Collins, 2001, 2002, 2005,
2009).
Within the construct of professional will, Collins (2001) identified Level 5
leaders as exhibiting a strong work ethic. Work ethic has been described as “the
complete and relentless devotion to one’s economic role on earth” (Lim, Woehr,
You, & Gorman, 2007, p. 319). Although the term work ethic has been used for
centuries, it was popularized and connected to religion in the 20th century by
Weber (1958) in his seminal publication titled The Protestant Ethic and the Spirit
of Capitalism (Miller, Woehr, & Hudspeth, 2001).
The religious valuation of restless, continuous, systematic work in a worldly
calling, as the highest means to asceticism, and at the same time the surest
and most evident proof of rebirth and genuine faith, must have been the
most powerful conceivable lever for the expansion of that attitude toward
life which we have here called the spirit of capitalism. (Weber, 1958, p.
172)
Protestants in early America adopted scriptural commands such as the following:
“Whatever you do in word or deed, do all in the name of the Lord Jesus” (Col 3:17,
New International Version), “Let us not become weary in doing good, for at the
proper time we will reap a harvest if we do not give up” (Gal 6:9), and “The one
who is unwilling to work shall not eat” (2 Thess 3:10). The concept of professional
will, therefore, may have some relationship to religious commitment.
Servant Leadership
A common perception among business leaders is that Level 5 leadership is
just another name for servant leadership (Lichtenwalner, 2010). Additionally, in
academia, Patterson et al. (2003) and Drury (2004) suggested that Level 5
leadership may be the same as servant leadership. Wong and Davey (2007)
concluded that servant leaders are more likely to be Level 5 leaders, van
Dierendonck (2011) stated that there is a clear overlap between Level 5 and servant
leadership, and Morris, Brotheridge, and Urbanski (2005) found that “there are
12
Instrument to Measure Level 5 Leadership
marked similarities between the behavior of those termed Level 5 leaders and the
servant or humble leader” (p. 1323). Greenleaf (1970, 1977) introduced the concept
of servant leadership in the modern era. The focus of servant leadership is on the
development and performance of the follower (Winston & Fields, in press).
Greenleaf (1977) described the motivation behind the desire to lead:
The servant-leader is servant first. . . . It begins with the natural feeling that
one wants to serve first. Then conscious choice brings one to aspire to lead.
That person is sharply different from one who is leader first, perhaps
because of the need to assuage an unusual power drive or to acquire
material possessions. For such it will be a later choice to serve—after
leadership is established. The leader-first and the servant-first are two
extreme types. Between them are shadings and blends that are part of the
infinite variety of human nature. . . . The difference manifests itself in the
care taken by the servant-first to make sure that other people’s highest
priority needs are being served. (p. 13)
As Collins’ research team was searching for a term to describe this new type of
leadership seen in the good-to-great companies, there was some discussion
regarding calling it servant leadership. According to Collins (2001), however,
Members of the team violently objected to these characterizations. Those
labels don’t ring true. . . . It makes them sound weak or meek, but that is not
at all the way that I think of Darwin Smith or Colman Mockler. They would
do almost anything to make the company great. (p. 30)
The modern idea of servant leadership is particularly popular in Christian
cultures as a result of the teachings of Jesus nearly 2000 years ago. Although Jesus
did not use terms that are translated as leader or leadership, he described the
positions, characteristics, and actions of people in leadership such as teacher, lord,
and great. An example of the teachings of Jesus on servant leadership followed a
mother’s request to exalt her sons:
Jesus called them together and said, “You know that the rulers of the
Gentiles lord it over them, and their high officials exercise authority over
them. Not so with you. Instead, whoever wants to become great among you
13
Instrument to Measure Level 5 Leadership
must be your servant, and whoever wants to be first must be your slave—
just as the Son of Man did not come to be served, but to serve. (Matt 20:25-
27)
Jesus put his teachings into actions in the upper room before the Passover shortly
before his death. There was no servant to wash the feet of the group. The custom
was that the lowest ranked non-Jewish slave would wash the feet of the dinner
guests upon arrival; however, the Messiah, son of God, put a towel around his waist
and proceeded to wash the feet of each individual disciple.
When he had finished washing their feet, he put on his clothes and returned
to his place. “Do you understand what I have done for you?” he asked them.
“You call me ‘Teacher’ and ‘Lord,’ and rightly so, for that is what I am.
Now that I, your Lord and Teacher, have washed your feet, you also should
wash one another’s feet. I have set you an example that you should do as I
have done for you. Very truly I tell you, no servant is greater than his
master, nor is a messenger greater than the one who sent him. Now that you
know these things, you will be blessed if you do them. (John 13:12-17)
In his letter to the Philippians, Paul captured the significance of the humility of
Jesus and exhorted his readers to follow His example.
Do nothing out of selfish ambition or vain conceit. Rather, in humility value
others above yourselves, not looking to your own interests but each of you
to the interests of the others. In your relationships with one another, have
the same mindset as Christ Jesus: Who, being in very nature God, did not
consider equality with God something to be used to his own advantage;
rather, he made himself nothing by taking the very nature of a servant,
being made in human likeness. And being found in appearance as a man, he
humbled himself by becoming obedient to death—even death on a cross!
Therefore God exalted him to the highest place and gave him the name that
is above every name, that at the name of Jesus every knee should bow, in
heaven and on earth and under the earth, and every tongue acknowledge
that Jesus Christ is Lord, to the glory of God the Father. (Phil 2:3-11)
14
Instrument to Measure Level 5 Leadership
Following the introduction of servant leadership into modern leadership
research by Greenleaf (1970, 1977), studies have sought to define measures to
quantify the attributes of a servant leader (Dennis & Bocarnea, 2005; Patterson,
2003; Winston, 2003; Winston & Fields, in press). Winston and Fields identified 10
essential servant leader behaviors:
1. Practices what he or she preaches;
2. Serves people without regard to their nationality, gender, or race;
3. Sees serving as a mission of responsibility to others;
4. Genuinely is interested in employees as people;
5. Understands that serving others is most important;
6. Is willing to make sacrifices to help others;
7. Seeks to instill trust rather than fear or insecurity;
8. Is always honest;
9. Is driven by a sense of higher calling; and
10. Promotes values that transcend self-interest and material success.
Winston and Fields’ behaviors help determine the relationship between personal
humility and Level 5 leadership. Based on a review of the key attributes of servant
leadership, it seems that servant leadership does not account for the professional
will element of Level 5 leadership, but it may be the same as the construct part of
personal humility within Level 5 leadership.
Charisma
Charismatic leaders have been notably successful in political, religious, and
societal contexts and are now being studied more in the context of business,
military, and educational contexts (Bass, 2008). Because “charismatics exude
confidence, dominance, a sense of purpose, and the ability to articulate the goals
and ideas for which followers are already prepared psychologically” (Bass, 2008, p.
576), it was quite a surprise in Good to Great (Collins, 2001) when the comparison
companies were the ones that were led by charismatic leaders and the good-to-great
companies were not. The fact that the findings “fly in the face of our modern
business culture and will, quite frankly, upset some people” (Collins, 2002, p. 1).
15
Instrument to Measure Level 5 Leadership
The idea of charisma and charismatic leadership is scarcely mentioned in
Good to Great (Collins, 2001), but the reader response to this surprising conclusion
of Level 5 leadership led Collins to address the topic in subsequent speeches,
interviews, and publications. In a 2009 speech, Collins said that some of the great
Level 5 leaders seemed to have a charisma bypass and that “we should never
confuse charisma for leadership” (0:21) . In an interview, Collins stated that
charisma is negatively correlated to leading a company from good to great, and a
leader must overcome the handicap of charisma to be successful (Collins & Rose,
2009). Charisma is considered to be a handicap because, if you are a charismatic
leader, you can “convince everyone you are right to the power of your personality”
(Collins, 2009, 3:08); however, leaders without charisma must lead by the merit of
their ideas and strong team support. Sam Walton is recognized as a great leader
who overcame his charisma to build a strong team and become a Level 5 leader
(Collins & Rose, 2009).
Collins and Porras (1994) described the difference between a time teller and
a clock maker. A time teller is a great visionary leader who can look out on the
horizon, determine what time it is, and then communicate the discovery to the rest
of the organization. When the time teller is no longer there, however, nobody else
in the organization can determine the time. Time tellers are not Level 5 leaders and
may be referred to as a “genius with a thousand helpers” (Collins & Rose, 2009,
1:46). Lee Iacocca is an example of a charismatic time teller who led Chrysler to
greatness for a few years; but, he was not able to sustain the greatness by himself,
and the company slipped back to mediocrity (Collins & Rose, 2009). In contrast to
the time teller leader, clock makers build a clock so that others in the organization
can determine the time long after the clock maker is gone. Level 5 leaders are clock
makers who surround themselves with quality individuals and strive to develop
them into leaders.
Secondary Literature
Since the publication of Good to Great (Collins, 2001), interest in the
concept of Level 5 leadership has remained popular in press and literature around
16
Instrument to Measure Level 5 Leadership
the world. The majority of the secondary references simply pass along descriptions
of Level 5 leadership as Collins (2001) described it, without adding value to the
discussion other than educating readers to the concept. Although professional will
seems to be accepted as a given for effective leadership, and humility in leadership
is not a new concept in American leadership, the idea of effective leadership being
driven by personal humility seems to resonate with leadership experts around the
world (Baale, 2011; Ling, 2009; “Taking Russia From Good to Great,” 2012;
Pimolsaengsuriya, 2012; Smith, 2005). Many leadership books now include
references to Level 5 leadership (Fullan, 2004; Ogden & Meyer, 2007; Tokunaga,
2003; Williams, 2005). The Schumpeter column in The Economist bemoaned the
lack of flamboyant and colorful leaders in business across the globe and blamed
Collins as one of the reasons:
Facelessness—or at least humility—is also the height of fashion among
management consultants and business gurus. Corporate headhunters are
helping firms find “humble” bosses. Jim Collins, one of America's most
popular gurus, argues that the best chief executives are not flamboyant
visionaries but “humble, self-effacing, diligent and resolute souls”. Business
journalists have taken to producing glowing profiles of self-effacing and
self-denying bosses such as Haruka Nishimatsu, the boss of Japan Airlines,
who travels to work on the bus and pays himself less than his pilots, and
Mike Eskew, the former boss of UPS, who flew coach and shares an
administrative assistant with three other people. (Woolridge, 2009, p. 1)
Despite the desire of some for a star CEO, Malmendier and Tate (2008) found that
CEOs who have won prestigious business awards subsequently underperform both
their own previous performance and the performance of CEOs who do not win
awards. “Our results suggest that the ex-post consequences of media-induced
superstar status for shareholders are negative” (Malmendier & Tate, 2008, p. 1593).
A common question regarding Level 5 leadership is: “Can you learn to
become Level 5?” (Collins, 2001, p. 35). Collins’ understanding of the answer to
this question has evolved with him now seeing Level 5 leadership as more
17
Instrument to Measure Level 5 Leadership
learnable than he used to believe (Bisoux, 2007). Yet, he cautioned that “the truth is
that Level 5 leadership is painful—not everyone is up to it” (Bisoux, 2007, p. 18).
There are two primary criticisms of the research that led to Level 5
leadership: lack of disconfirming research (May, 2006) and evidence of leaders
who are successful and clearly do not exhibit Level 5 characteristics. The lack of
disconfirming evidence suggests that there may be numerous Level 5 leaders who
are not successful.
Collins’ team looked at the companies that went from good to great and said
“What do all these have in common?” They never went back and said “are
there any companies that have these traits that did not make the leap from
good to great?” And I understand why they didn’t. Because these principles
are vague and it would be hard to debate whether or not an unsuccessful
company was doing these good-to-great things. (May, 2006, p. 1)
Some critics have pointed to great leaders like Lee Iacocca, Steve Jobs, Jack
Welch, and Donald Trump who have tremendous egos and, yet, are considered
great leaders. Collins readily admitted that there are egocentric leaders who
demonstrate great results for a period of time, but Level 5 leaders build a strong
team around them so the organization remains great when they are no longer there.
Summary of Characteristics
A simple 2x2 grid summarizes the characteristics the literature review
captures. There were originally a total of 99 items: (a) 55 describing personal
humility and (b) 44 describing professional will. Personal humility and professional
will can be described with both positive and negative terms. For example, positive
attributes such as humble describe the Level 5 leader, while negative attributes such
as egocentric describe a leader who is not Level 5. Table 2 summarizes the terms
that Collins (2001, 2002, 2005, 2006, 2009, 2011; Collins & Hanson, 2001; Collins
& Rose, 2009) has used to describe Level 5 leaders.
18
Instrument to Measure Level 5 Leadership
Table 2: Positive and Negative Terms Describing Personal Humility and
Professional Will (Collins, 2001, 2002, 2005, 2006, 2009, 2011; Collins & Hanson,
2001; Collins & Rose, 2009)
Positive Negative
Personal humility
Humble Egocentric
Set up others for success Arrogant
Genuine Overbearing
Lack of pretense Condescending
Gives others credit for success Patronizing
Accepts responsibility when things
don’t go well
Pretentious
Modest Talks about themselves a lot
Authentic Does not set up successors for success
Team player Large personal ego
Sets up successors for success Seeks fame, fortune, adulation, and
power
Self-effacing Boastful
Gracious Self-serving (at the expense of others)
Unpretentious Acts superior
Courteous Large ego
Understated Air of self-importance
Servant attitude Rude
Doesn't seek spotlight Snooty
Reserved personality Seeks personal greatness
Placid persona (pleasantly calm or
peaceful)
Charismatic leadership
Selfless (puts the needs of others first) Brash
Unassuming Confident
Low-key Personal ambition
19
Instrument to Measure Level 5 Leadership
Positive Negative
Reserved Larger-than-life
Quiet Gutsy
Associates with unimportant people Big personality
Shy Socially awkward
Simple
Mild-mannered
May feel unqualified for the job
Professional will
Resolve Unmotivated
Inner intensity Lazy
Gets results Undisciplined
Self-motivated Weak leader
Ferocious resolve Afraid to take a chance
Intense resolve Surrounded by “yes-men”
Drive Cowardly
Self-control Motivated by financial gain
Strength of character
Courageous
Bold
Builds strong team
Strong work ethic
Clear catalyst in achieving results
Ambitious for organization
Fearless
Results oriented
Self-determination (determination not
influenced by outsiders)
Enthusiastic desire to produce results
Unafraid to take risk
20
Instrument to Measure Level 5 Leadership
Positive Negative
Intense personality
Will settle for nothing less than the best
Pride in the organization
Dedication to the organization
Willpower (able to control one's
impulses and actions)
Backbone
Daring: willing to take a chance
Fierce resolve to life that is not deterred
based on emotional ups and downs
Fanatically driven to achieve results
Stoic determination
Desire for organization
Will
Gutsy
Workmanlike diligence
Obsession for organization
Fanatically driven
21
Instrument to Measure Level 5 Leadership
Chapter 3 – Method
With the lack of an instrument to measure this important concept of Level 5
leadership, the characteristics of Level 5 leadership were explored to develop a
parsimonious scale to identify the Level 5 traits of leaders. Three research
questions were asked: (a) Utilizing the attributes and characteristics from literature,
can a statistically valid instrument be developed to measure Level 5 leadership? (b)
Is the personal humility construct of Level 5 leadership the same as servant
leadership? and (c) Do Collins’ (2006) eight questions to test Level 5 leadership
correlate with the attributes and characteristics that he used to describe Level 5
leadership? This study provides an accurate and effective instrument to measure
Level 5 leadership within individuals.
This chapter outlines the steps utilized to develop the Level 5 leadership
instrument. The literature review in the previous chapter identified 99 attributes in
the item pool that Collins (2001, 2002, 2005, 2006, 2009, 2011) highlighted to
describe Level 5 leaders. DeVellis (2012) defined eight steps for scale
development: (a) determine clearly what it is you want to measure, (b) generate an
item pool, (c) determine the format for measurement, (d) have initial item pool
reviewed by experts, (e) consider inclusion of validation items, (f) administer items
to a development sample, (g) evaluate the samples, and (h) optimize scale length.
Determine Clearly What it is You Want to Measure
The research task is to develop an instrument to measure Level 5 leadership,
as described by Collins (2001). Since there are no empirically tested tools currently
to identify Level 5 leaders within organizations, the constructed tool will be
necessarily new and unique. Servant leadership was also measured to determine if
there is a relationship between servant leadership and Level 5 leadership, as has
been suggested by Patterson (2003) and Drury (2004).
Generate an Item Pool
The researcher identified 99 unique attributes and characteristics that
Collins (2001, 2002, 2005, 2006, 2009, 2011) utilized in literature as well as video
22
Instrument to Measure Level 5 Leadership
interviews and speeches. There is some redundancy in the list, but DeVellis (2012)
pointed out that redundancy enhances reliability and is not bad: “At this stage of the
scale development process, it is better to be more inclusive” (p. 78). A key
objective in this task is to ensure that items will be clear and unambiguous to the
person responding to the survey. The items selected have a blend of positive and
negative characteristics.
Determine the Format for Measurement
The semantic differential scaling method was utilized to collect the
responses of subjects to the attributes and characteristics describing Level 5 leaders.
This method is highly compatible with theoretical models like the researcher is
addressing in this study (DeVellis, 2012). Subjects were asked the following
question:
On a scale of 1 to 10, to what extent do the following characteristics
describe your boss? 1 indicates that this characteristic does not describe
your boss at all, while a 10 indicates that it describes him/her exactly.
Not At All 1 2 3 4 5 6 7 8 9 10 Exactly
Expert Panel Review
After identifying the attributes and characteristics that Collins (2001, 2002,
2005, 2006, 2009, 2011) used to describe Level 5 leaders, it is necessary to
determine which of these characteristics is the most relevant for further
consideration. This was effectively accomplished by engaging a panel of experts.
Four authors who have published on the topic of Level 5 leadership or scale
development of servant leadership agreed to participate in the study (see Table 3).
Collins’ organization, ChimpWorks LLC, declined to participate in the expert
panel.
23
Instrument to Measure Level 5 Leadership
Table 3: Expert Panel for Instrument Verification
Expert Relevant publication titles
Dr. Kathleen Patterson,
Regent University, Virginia
Beach, VA
Transformational Leaders to Servant Leaders
versus Level 4 Leaders to Level 5 Leaders—The
Move from Good to Great (2003)
Dr. Douglas Caulkins,
Grinnell University, Grinell,
Iowa
Re-theorizing Jim Collins’s Culture of Discipline
in Good to Great (2008)
Dr. Dail Fields, Regent
University, Virginia Beach,
VA
Seeking the Essence of Servant Leadership
(Winston & Fields, in press)
Dr. Sabrina Liccardo,
University of
Witwatersrand,
Johannesburg, South Africa
Level 5 Leaders and the Romance of Leadership
Construct (2007)
An online survey was created using Survey Monkey™ that listed the 99
attributes Collins (2001, 2002, 2005, 2006, 2009, 2011) has used to describe Level
5 leaders. The experts were asked to rate each of the attributes as either highly
relevant, somewhat relevant, or not at all relevant. In addition, a comment box was
provided by each attribute to allow the expert to expand on the rating by indicating
when an attribute was ambiguous, confusing, or required context. The results were
quantified by assigning a score of 3 to highly relevant attributes, 2 to somewhat
relevant, and 1 to not at all relevant. An average score of at least 2 indicates that the
experts believe the attribute has relevance for measuring Level 5 leadership.
Twenty-five attributes received a score of less than 2 and were removed from the
final survey. Some of the items removed included items that were ambiguous such
as low-key or big personality, and others were removed because they were not clear
without additional context, including gutsy and simple. Table 4 shows the 99
24
Instrument to Measure Level 5 Leadership
attributes reviewed by the expert panel, including the average score that the panel
rated the relevance of the item (1.00 to 3.00) with an indication that a score of less
than 2.00 was to be removed from the list. The outcome of the feedback from the
expert team was a list of 74 attributes to be tested.
Table 4: Average Scores of Relevance from Expert Panel
Attributes Average score Action
Positive attributes of humility
Humble 3.00
Set up others for success 3.00
Genuine 3.00
Lack of pretense 3.00
Gives others credit for success 3.00
Accepts responsibility when things don’t go well 3.00
Modest 2.75
Authentic 2.75
Team player 2.75
Sets up successors for success 2.75
Self-effacing 2.50
Gracious 2.50
Unpretentious 2.50
Courteous 2.50
Understated 2.25
Servant attitude 2.25
Doesn't seek spotlight 2.25
Reserved personality 2.00
Placid persona (pleasantly calm or peaceful) 2.00
Selfless (puts the needs of others first) 2.00
Unassuming 1.75 REMOVE
25
Instrument to Measure Level 5 Leadership
Attributes Average score Action
Low-key 1.67 REMOVE
Reserved 1.67 REMOVE
Quiet 1.50 REMOVE
Associates with unimportant people 1.50 REMOVE
Shy 1.25 REMOVE
Simple 1.25 REMOVE
Mild-mannered 1.25 REMOVE
May feel unqualified for the job 1.25 REMOVE
Socially awkward 1.00 REMOVE
Negative attributes of humility
Egocentric 3.00
Arrogant 3.00
Overbearing 3.00
Condescending 3.00
Patronizing 3.00
Pretentious 3.00
Talks about themselves a lot 3.00
Does not set up successors for success 3.00
Large personal ego 3.00
Seeks fame, fortune, adulation, and power 3.00
Boastful 2.67
Self-serving (at the expense of others) 2.33
Acts superior 2.33
Large ego 2.33
Air of self-importance 2.33
Rude 2.00
Snooty 2.00
Seeks personal greatness 2.00
Charismatic leadership 2.00
26
Instrument to Measure Level 5 Leadership
Attributes Average score Action
Brash 1.67 REMOVE
Confident 1.67 REMOVE
Personal ambition 1.67 REMOVE
Larger-than-life 1.33 REMOVE
Gutsy 1.00 REMOVE
Big personality 1.00 REMOVE
Positive attributes of will
Resolve 3.00
Inner intensity 3.00
Gets results 3.00
Self-motivated 3.00
Ferocious resolve 3.00
Intense resolve 3.00
Drive 2.75
Self-control 2.75
Strength of character 2.75
Courageous 2.75
Bold 2.75
Strong work ethic 2.75
Clear catalyst in achieving results 2.75
Ambitious for organization 2.67
Fearless 2.50
Results oriented 2.50
Self-determination (determination not influenced
by outsiders)
2.50
Enthusiastic desire to produce results 2.50
Unafraid to take risk 2.33
Intense personality 2.33
Will settle for nothing less than the best 2.33
27
Instrument to Measure Level 5 Leadership
Attributes Average score Action
Pride in the organization 2.25
Dedication to the organization 2.25
Willpower (able to control one's impulses and
actions)
2.00
Backbone 2.00
Daring: willing to take a chance 2.00
Fierce resolve to life that is not deterred based on
emotional ups and downs
2.00
Fanatically driven to achieve results 2.00
Stoic determination 1.75 REMOVE
Desire for organization 1.75 REMOVE
Will 1.67 REMOVE
Gutsy 1.67 REMOVE
Workmanlike diligence 2.67 REMOVE
Obsession for organization 1.50 REMOVE
Fanatically driven 1.33 REMOVE
Negative attributes of professional will
Unmotivated 3.00
Lazy 3.00
Undisciplined 3.00
Weak leader 3.00
Afraid to take a chance 3.00
Surrounded by “yes-men” 3.00
Cowardly 2.67
Motivated by financial gain 1.67 REMOVE
28
Instrument to Measure Level 5 Leadership
Consider Inclusion of Validation Items
DeVellis (2012) recommended utilizing validation items for scales,
particularly when they are self-assessments, because some individuals are “strongly
motivated to present herself or himself in a way that society regards as positive” (p.
101). A scale measuring social desirability can help eliminate this error. However,
since this instrument is not based on self-assessment, validation items would not be
meaningful.
Administer Items to a Development Sample
The online survey created for the experts on Survey Monkey™ contained
four sections: (a) the 74 attributes of Level 5 leaders taken from Collins’ literature,
speeches, and interviews; (b) the 10 items from the servant leadership scale
(Winston & Fields, in press); (c) the eight questions that Collins (2006) claimed
can identify Level 5 leaders; and (d) demographic information. The survey was
disseminated through the snowballing technique utilizing email and the social
media platforms Facebook and LinkedIn. Nunnally (1978) recommended 300
subjects for scale development, and DeVellis (2012) confirmed that 300 “should be
sufficiently large to eliminate subject variance as a significant concern” (p. 102).
Evaluate the Samples
The researcher constructed three different components of the survey: Level
5 attributes, servant leadership, and Collins’ (2006) eight questions for Level 5
leadership. The objective of developing this scale was to identify items highly
correlated to a true score of Level 5 leadership. However, since the true score is not
known, the next best option is to identify the scale items that are highly
intercorrelated (DeVellis, 2012). The first step of analysis is to determine the
number of latent variables that underlie this set of 74 items. This was accomplished
utilizing principal component analysis for factor analysis in SPSS. In the factor
analysis, rotation increased “interpretability by identifying clusters of variables that
can be characterized predominantly in terms of a single latent variable” (DeVellis,
2012, p. 133). Direct Oblimin rotation is more effective when the factors are
correlated (DeVellis, 2012). By condensing the information so that the variation
29
Instrument to Measure Level 5 Leadership
can be accounted for by the fewest number of variables, the instrument becomes
more desirable and usable for respondents. Since the number of factors is not
known prior to the research, the researcher relied on existing theory and literature
that indicated that one or more factors will comprise personal humility and one or
more factors will comprise professional will. To determine the items that load on
each factor, a .30 level is considered to be an acceptable minimum factor loading
because it shows that about 10% of the variance for a corresponding variable has
been explained by a factor (Tinsley & Tinsley, 1987). The length of the scale was
optimized to balance the desired reliability of lengthy surveys and the desirability
of shorter surveys that reduce the burden of the respondents.
The responses from the validated Winston and Fields (in press) servant
leadership instrument were summed into one servant leadership score per
participant, so that a higher score indicates a higher level of servant leadership.
Likewise, responses from Collins’ (2006) eight questions were summed into a
single score per participant, so that a higher score indicates a higher level of Level
5 leadership. The servant leadership scores were then compared to the personal
humility attribute scores using paired-samples t tests to determine if servant
leadership is the same as the humility construct of Level 5 leadership. The results
from Collins’ (2006) eight questions were then compared to the list of Level 5
leaders who were determined using the attributes of personal humility and
professional will. The leader designations from the Level 5 attributes were
dichotomous; the leader is either Level 5 or is not Level 5. The scores from
Collins’ (2006) eight questions are continuous. Therefore, a biserial correlation was
utilized. A biserial correlation is simply a special case of a Pearson product moment
correlation and is used when one variable is artificially dichotomous and the other
is truly continuous (IBM, 2012).
Reliability and Validity
The reliability of a scale is the measure of how accurately the scale
represents the true score of the latent variable in a consistent and predictable
manner (DeVellis, 2012). “If the items of a scale have a strong relationship to their
30
Instrument to Measure Level 5 Leadership
latent variable, they will have a strong relationship to each other” (DeVellis, 2012,
p. 34). The internal consistency reliability of the Level 5 leadership scale was
measured using Cronbach’s alpha. The guidelines provided by DeVellis were used
to determine the acceptability of Cronbach’s alpha (see Table 5).
Table 5: Reliability as Measured by Cronbach’s Alpha
Cronbach’s alpha Acceptability
Below 0.60 Unacceptable
.60 to .65 Undesirable
.65 to .70 Minimally acceptable
.70 to .80 Respectable
.80 to .90 Very good
Above .90 Consider shortening scale
Note. From Scale Development. Theory and Applications (p. 109), by R.F. DeVellis, 2012, Los
Angeles, CA: Sage. Copyright 2012 by DeVellis. Adapted with permission.
Once the reliability of the scale has been determined to be consistently and
reliably describing the latent variable, one must determine if that variable being
described is actually Level 5 leadership. There were three different types of validity
considered: content validity, construct validity, and criterion-related validity
(DeVellis, 2012). Content validity reflects the accuracy that the items in the scale
reflect the latent variable. Since the original item set contained all of the
descriptions that Collins (2001) used for Level 5 leaders, the scale achieved content
validity. Construct validity reflects the relationship between the theoretical and
actual behavior of the construct.
31
Instrument to Measure Level 5 Leadership
Chapter 4 – Results
To establish a scientifically validated instrument to measure Level 5
leadership, 349 subjects evaluated their bosses in an online survey. Four key
variables were analyzed: (a) personal humility, attributes derived from Collins
(2001); (b) professional will, attributes derived from Collins (2001); (c) Collins’
(2006) eight questions to determine Level 5 leadership; and (d) servant leadership.
The scores from the personal humility and professional will results were utilized to
define a new Level 5 variable. The new Level 5 variable was then compared to
Collins’ (2006) eight questions and servant leadership to determine if all of these
constructs are actually the same, as literature has suggested.
Demographic Variables
The sample size of 349 exceeds the 300 recommended by Nunnally (1978)
and DeVellis (2012). Participation was solicited through email and social media
platforms, and responses were collected between September 26 and October 6,
2012. Demographic questions were included to better understand the subjects and
the bosses being evaluated. The demographic questions identified the gender and
age of the participant, the gender and age of the boss, the position of the boss in the
organization, the type of organization, and the religious affiliation and commitment
of the boss (See Tables 6-10).
Table 6: Gender and Age of Subjects
Age
Total24 or
younge
r
25-34 35-44 45-54 55-6465 or
older
Gende
r
Male 2 16 63 81 44 16 222
Femal
e
2 19 28 46 19 11 125
32
Instrument to Measure Level 5 Leadership
Total 4 35 91 127 63 27 347
Table 7: Gender of Participant and Boss
Boss gender
TotalMale Female
Subject
gender
Male 201 21 222
Female 88 37 125
Total 289 58 347
Table 8: Age of Participant and Boss
Boss age
Total25-34 35-44 45-54 55-64
65 or
older
Subject
age
24 or
younger
0 1 2 0 1 4
25-34 5 8 10 10 1 34
35-44 2 28 41 20 0 91
45-54 3 16 69 35 5 128
55-64 1 14 27 17 3 62
65 or
older
0 2 7 7 9 25
Total 11 69 156 89 19 344
33
Instrument to Measure Level 5 Leadership
Table 9: Boss Position and Type of Organization
Organization type
Total
Fortune
500 Public Private
Nonprofi
t
Boss
position
Executive 17 9 36 37 99
Officer 31 19 16 38 104
Director 36 14 16 23 89
Manager 30 9 4 4 47
Supervisor 2 1 1 4 8
Total 116 52 73 106 347
Table 10: Religious Affiliation and Religious Commitment of Boss
Religious commitment
Total
Leade
r Committed Somewhat
Not
committed
Religious
affiliation
Catholic 4 17 8 9 38
Evangelical 55 37 12 6 110
Mainline 8 18 9 1 36
Christian -
other
2 6 4 5 17
Other
religion
0 3 1 8 12
Nonreligious 1 0 1 14 16
Total 70 81 35 43 229
Factor Analysis
The first step in analyzing the data was to determine if the 74 attributes of
Level 5 leaders that were derived from Collins’ (2001) literature can be factored
34
Instrument to Measure Level 5 Leadership
into meaningful scales. Negative attributes such as arrogant were recoded so that
the scores correspond with the positive attributes such as humble. Therefore, a boss
who was rated a 10 in arrogance was recoded as a 1 so that the negative attributes
could be scored with the positive attributes. Principal component analysis in SPSS
showed that the significant majority of variability in the data could be explained in
two components (see Figure 1). As detailed in Table 11, 55.24% of the total
variance is explained in the first two components. Since the factors are correlated
with one another, direct oblimin rotation was used.
Figure 1: Scree plot for principal component analysis.
35
Table 11: Total Variance Explained
Component
Initial eigenvalues Extraction sums of squared loadingsRotation sums of
squared loadingsa
Total % of Variance Cumulative % Total% of
VarianceCumulative % Total
1 19.21 39.21 39.21 19.21 39.21 39.21 18.06
2 7.85 16.03 55.24 7.85 16.03 55.24 11.12
3 1.64 3.34 58.58
4 1.29 2.63 61.20
5 1.24 2.53 63.73
6 .99 2.03 65.76
7 .98 1.99 67.75
8 .89 1.82 69.57
9 .87 1.77 71.34
10 .74 1.52 72.89
11 .72 1.48 74.33
12 .68 1.39 75.72
13 .63 1.29 77.00
14 .58 1.19 78.19
Component
Initial eigenvalues Extraction sums of squared loadingsRotation sums of
squared loadingsa
Total % of Variance Cumulative % Total% of
VarianceCumulative % Total
15 .58 1.18 79.37
16 .56 1.13 80.50
17 .51 1.04 81.54
18 .50 1.02 82.56
19 .47 .97 83.53
20 .46 .94 84.46
21 .44 .90 85.36
22 .43 .87 86.23
23 .42 .85 87.08
24 .41 .84 87.91
25 .38 .78 88.70
26 .37 .74 89.44
27 .34 .69 90.14
28 .33 .68 90.82
29 .33 .66 91.48
Component
Initial eigenvalues Extraction sums of squared loadingsRotation sums of
squared loadingsa
Total % of Variance Cumulative % Total% of
VarianceCumulative % Total
30 .32 .65 92.12
31 .29 .60 92.72
32 .28 .57 93.30
33 .27 .55 93.85
34 .26 .54 94.39
35 .26 .52 94.91
36 .24 .50 95.41
37 .24 .49 95.90
38 .22 .45 96.35
39 .21 .44 96.79
40 .21 .43 97.21
41 .20 .40 97.61
42 .19 .39 98.01
43 .18 .36 98.37
44 .16 .32 98.69
Component
Initial eigenvalues Extraction sums of squared loadingsRotation sums of
squared loadingsa
Total % of Variance Cumulative % Total% of
VarianceCumulative % Total
45 .15 .31 98.10
46 .14 .28 99.27
47 .13 .27 99.54
48 .13 .26 99.80
49 .10 .20 100.00
a When components are correlated, sums of squared loadings cannot be added to obtain a total variance.
Instrument to Measure Level 5 Leadership
As was expected from literature and theory, the two components of the
attributes match the two constructs of Level 5 leadership: personal humility and
professional will. Values below .50 and crossloaded items were removed from the
analysis. Component 1 consists of 30 of the attributes that describe the humility of
the leader. Component 2 consists of 19 of the attributes that describe the
professional will of the leader (see Table 12).
Table 12: Structure Matrix of Two Components
AttributesComponent
1 2
Arrogant .88
Acts superior .84
Egocentric .84
Large personal ego .82
Air of self-importance .84
Genuine .81
Pretentious .80
Condescending .80
Humble .80
Talks about themselves .78
Overbearing .78
Self-serving .78
Modest .77
Boastful .77
Seeks fame .77
Snooty .77
Gracious .75
Rude .74
Selfless .73
40
Instrument to Measure Level 5 Leadership
AttributesComponent
1 2
Team player .73
Patronizing .72
Unpretentious .72
Courteous .70
Does not set up successors .64
Placid .64
Will power .58
Self-control .58
Servant attitude .55
Doesn’t seek spotlight .54
Surrounded by yes men .53
Drive .81
Intense resolve .76
Courageous .78
Catalyst .78
Gets results .77
Will Not Settle .76
Backbone .76
Resolve .75
Bold .72
Fearless .71
Results oriented .69
Strong work ethic .68
Self-motivated .68
Inner intensity .65
Charismatic -.61
Dedication to the organization .59
Fierce resolve .56
41
Instrument to Measure Level 5 Leadership
AttributesComponent
1 2
Fanatically driven .53
Intense personality .50
Since Collins believed that charisma is negatively correlated to Level 5
leadership (Collins & Rose, 2009), charisma scores were recoded to reflect this
belief. However, the recoded charisma attribute is now negatively correlated to the
other Level 5 leadership attributes, meaning that charisma is, in fact, positively
correlated to Level 5 leadership attributes.
Defining Variables
In order to simplify the scales for analysis and further measure the Level 5
attributes, a new score for each leader was created by summing the scores of the top
10 factor-loaded items for personal humility and the top 10 for professional will.
The four key scales are now personal humility, professional will, servant
leadership, and Collins’ (2006) eight questions. Table 13 presents descriptive
statistics for the four scales. Cronbach’s alpha is very high with at least .92 on each
(see Table 14).
Table 13: Descriptive Statistics of Four Key Variables for Each Leader (N = 349)
Scale M SD
Humility 69.69 24.38
Will 73.12 17.54
Servant leadership 37.34 9.05
Collins’ (2006) eight questions 30.17 7.49
42
Instrument to Measure Level 5 Leadership
Table 14: Reliability of Scales
Scale Cronbach’s alpha
Servant .95
Collins .93
Humility .95
Will .92
To derive a score for Level 5 leadership, leaders were divided into four
categories based on their humility and will score. Each leader was identified as
having either an above average or below average score for humility, and above or
below average score for professional will. Based on those identifications, the
leaders were then assigned to one of four leadership identifiers: (a) weak
leadership, below average in both categories; (b) humble, above average humility
but below average will; (c) strong will, above average professional will but below
average humility; and (d) Level 5, above average humility and above average will.
The results showed that 31% of the leaders were categorized as Level 5 (see Figure
2).
The scores from the 10-item, validated servant leadership scale from
Winston and Fields (in press) and the eight questions from Collins (2006) were also
summed. A component factor analysis of Collins’ (2006) eight questions showed
that the questions factored into one component that explained 68.73% of the
variance (see Table 15).
43
Instrument to Measure Level 5 Leadership
Professional will
weak
Professional will
strong
Humility
weak
Bad leaders
n =102
Strong will
n =71n =173
Humility
strong
Humble
n =67
Level 5
n =109n =176
n =169 n =180
Figure 2: Professional will and personal humility results (N =349).
Table 15: Collins’ (2006) Eight Questions Total Variance Explained
Component
Initial eigenvalues
Extraction sums of squared
loadings
Total
% of
Variance
Cumulative
% Total
% of
Variance
Cumulative
%
1 5.50 68.73 68.73 5.50 68.73 68.73
2 .55 6.83 75.56
3 .42 5.22 80.78
4 .41 5.09 85.87
5 .34 4.23 90.10
6 .32 3.98 94.08
7 .26 3.20 97.28
8 .22 2.72 100.00
44
Instrument to Measure Level 5 Leadership
The Kolmogorov-Smirnov and Shaprio-Wilk tests measure the normality of
a dataset. Shaprio-Wilk is the most effective for datasets of less than 2,000 items.
Both Shapiro-Wilk and Kolmogorov-Smirnov show a significance less than .05, so
the data are not normally distributed (see Table 16). This is consistent with the
observation that most subjects identified their bosses in a positive manner and the
median on the personal humility and professional will scores were 7.5, indicating
that the distribution of responses was negatively skewed. Since the data of all four
scales are not normally distributed, nonparametric analysis is used when applicable.
Table 16: Tests of Normality
ScaleKolmogorov-Smirnova Shapiro-Wilk
Statistic df p Statistic df p
Servant .09 349 .000 .95 349 .000
Collins .11 349 .000 .93 349 .000
Humility .12 349 .000 .92 349 .000
Will .11 349 .000 .94 349 .000
a Lilliefors significance correction.
Spearman’s Rho was used to show the convergent validity between the four
variables (see Table 17). The strongest correlation is between Collins and servant
leadership, but all correlations are statistically significant at the .01 level.
45
Instrument to Measure Level 5 Leadership
Table 17: Correlation Coefficients Using Spearman’s Rho
Servant Collins Humility Will
Servant –
Collins .83** –
Humility .75** .68** –
Will .52** .65** .34** –
**p < .01.
Discriminant Validity Tests
In his literature, Collins (2001) did not usually refer to leaders along a
continuum to Level 5 or as weak, just humble, or just strong will. He referred to
them as either Level 5 or not Level 5. Therefore, the leaders in this study were
separated into the group of 109 Level 5 leaders and 240 non-Level 5 leaders.
Discriminant validity tests were run to determine if these designations were
consistent with the new Level 5 scale, servant leadership, and Collins’ (2006) eight
questions (see Table 18). The various scores of Level 5 leaders (L5) were
consistently higher than the scores of leaders who were not Level 5 (Not L5).
Levene’s test for equality of variances was used on the variables together and
shows that equal variances between the means cannot be assumed (see Table 19).
An independent samples t test shows that the difference between means is
statistically significant (see Table 20).
46
Instrument to Measure Level 5 Leadership
Table 18: Leaders Identified as Level 5 Compared to Individual Constructs
Variables N M SD SE
ServantNot L5 240 34.03 8.74 .56
L5 109 44.62 4.14 .40
CollinsNot L5 240 27.40 7.18 .46
L5 109 36.28 3.44 .33
HumilityNot L5 240 60.09 23.39 1.51
L5 109 90.83 6.94 .67
WillNot L5 240 66.77 17.23 1.11
L5 109 87.12 6.86 .66
Table 19: Levene’s Test for Equality of Variances to Determine if Responses are
Normally Distributed
F p
Servant 39.38 .000
Collins 45.03 .000
Humility 145.93 .000
Will 55.32 .000
Level 5 380.05 .000
47
Table 20: Independent Samples Test of L5 Within Key Variables
Scale t df p ΔM ΔSE
95% Confidence interval of
the difference
Lower Upper
Servant -15.36 346.41 .000 -10.59 .69 -11.95 -9.23
Collins -15.61 346.02 .000 -8.88 .57 -10.00 -7.76
Humility -18.63 314.47 .000 -30.74 1.65 -33.99 -27.49
Will -15.75 342.62 .000 -20.35 1.29 -22.89 -17.81
Note. Equal variances not assumed.
Instrument to Measure Level 5 Leadership
Overall, 31% of bosses were considered to be Level 5 leaders. Table 21
shows the percent of Level 5 bosses by demographic category. ANOVA was
performed to determine if there was a statistically significant difference between
the Level 5 leaders and non-Level 5 leaders within a demographic category.
Table 21: Demographic Comparisons Between Level 5 and non-Level 5 Leaders
non-Level 5 Level 5
Subject gender
Male 68% 32%
Female 70% 30%
Boss gender
Male 69% 31%
Female 67% 33%
Participant age
24 or younger 50% 50%
25-34 77% 23%
35-44 67% 33%
45-54 75% 25%
55-64 62% 38%
65 or older 52% 48%
Boss age
24 or younger 0% 0%
25-34 45% 55%
35-44 70% 30%
45-54 72% 28%
55-64 66% 34%
65 or older 53% 47%
Boss position
Executive 64% 36%
49
Instrument to Measure Level 5 Leadership
non-Level 5 Level 5
Officer 67% 33%
Director 74% 26%
Manager 70% 30%
Supervisor 75% 25%
Boss religious affiliation
Catholic 74% 26%
Evangelical 59% 41%
Mainline 74% 26%
Christian - Other 86% 14%
Other Religion 54% 46%
Nonreligious 84% 16%
Boss religious commitment
Leader 48% 52%
Committed 63% 37%
Somewhat 86% 14%
Not Committed 88% 13%
Organization type
Fortune 500 78% 22%
Public 81% 19%
Private 70% 30%
Nonprofit 52% 48%
Total 69% 31%
Religious commitment and organization type are statistically significant
variables in identifying Level 5 leaders. Other demographic variables do not have a
statistically significant impact on the data (see Table 22).
50
Instrument to Measure Level 5 Leadership
Table 22: Significance of Level 5 Leader by Demographic Variable
SS df MS F p
Gender
Between groups .02 1 .02 .09 .761
Within groups 79.95 345 .23
Total 79.97 346
Boss gender
Between groups .01 1 .01 .07 .797
Within groups 48.32 346 .14
Total 48.33 347
Age
Between groups 3.23 1 3.23 2.62 .106
Within groups 425.76 346 1.23
Total 428.99 347
Boss age
Between groups .17 1 .17 .22 .642
Within groups 274.06 342 .80
Total 274.23 343
Boss position
Between groups 2.09 1 2.09 1.75 .187
Within groups 414.01 346 1.20
Total 416.10 347
Religious affiliation
Between groups 3.90 1 3.90 1.96 .162
Within groups 498.77 251 1.99
Total 502.68 252
Religious commitment
Between groups 30.38 1 30.38 28.28 .000
Within groups 252.43 235 1.07
Total 282.81 236
51
Instrument to Measure Level 5 Leadership
SS df MS F p
Organization type
Between groups 26.98 1 26.98 18.48 .000
Within groups 503.71 345 1.46
Total 530.69 346
Bonferroni post hoc tests show that Level 5 leaders are more likely to be
leaders in their religious community or at least committed to their religion and that
they are more likely to work for a nonprofit organization than any other type of
organization (see Table 23).
Table 23: Bonferroni Post Hoc Test for Religious Commitment of Level 5 Leaders
ΔM (I-J) SE p95% Confidence interval
Lower bound Upper bound
Leader
Committed .59 .2
9
.255 -.18 1.36
Somewhat 1.51* .3
7
.000 .53 2.50
Not committed 1.56* .3
4
.000 .69 2.47
Committed
Leader -.59 .2
9
.255 -1.36 .18
Somewhat .92 .3
6
.067 -.04 1.88
Not committed .99* .3 .015 .13 1.86
52
Instrument to Measure Level 5 Leadership
ΔM (I-J) SE p95% Confidence interval
Lower bound Upper bound
3
Somewhat
Leader -1.51* .3
7
.000 -2.50 -.53
Committed -.92 .3
6
.067 -1.88 .04
Not committed .07 .4
0
1.00
0
-.99 1.13
Not committed
Leader -1.56* .3
4
.000 -2.47 -.69
Committed -.99* .3
3
.015 -1.86 -.13
Somewhat -.07 .4
0
1.00
0
-1.13 .99
*p < .05.
Table 24: Bonferroni Post Hoc Test for Organization Type of Level 5 Leaders
ΔM (I-J) SE p95% Confidence interval
Lower bound Upper bound
Fortune 500
Public .13 .30 1.000 -.67 .93
Private -.31 .27 1.000 -1.03 .41
Nonprofit -1.03* .24 .000 -1.67 -.38
Public
Fortune 500 -.13 .30 1.000 -.93 .67
53
Instrument to Measure Level 5 Leadership
ΔM (I-J) SE p95% Confidence interval
Lower bound Upper bound
Private -.44 .33 1.000 -1.31 .43
Nonprofit -1.16* .31 .001 -1.97 -.34
Private
Fortune 500 .31 .27 1.000 -.41 1.03
Public .44 .33 1.000 -.43 1.31
Nonprofit -.72 .28 .056 -1.45 .01
Nonprofit
Fortune 500 1.03* .24 .000 .38 1.67
Public 1.16* .31 .001 .34 1.97
Private .72 .28 .056 -.01 1.45
*p < .05.
Final Instrument
The final step in developing an instrument to measure Level 5 leadership is
to select the attributes that will provide a parsimonious scale with sufficient
reliability. Principal component analysis identified 30 attributes of personal
humility and 19 attributes of professional will (see Table 12) with a Cronbach’s
alpha of .951 and .923, respectively. For a Cronbach’s alpha greater than .900,
DeVellis (2012) recommended shortening the scale (see Table 5).
To shorten the scale, negative attributes that would require recoding were
removed. This left 13 attributes for personal humility and 18 attributes for
professional will. From the personal humility list, placid was removed because
many subjects may not know the meaning of the word, and will power and self-
control were removed because Collins (2001) used them to describe professional
will, not personal humility. The remaining 18 professional will attributes were
narrowed to 10 by removing duplicate concepts with lower factor loadings. For
example, intense resolve was retained, and resolve and fierce resolve were
removed. Likewise, clear catalyst for results was retained, and gets results and
54
Instrument to Measure Level 5 Leadership
results oriented were removed. The remaining top 10 factor-loaded attributes for
personal humility and professional will were retained to create a 20-item scale (see
Table 25).
Table 25: Level 5 Leadership Scale Attributes
Personal humility Profession will
1 Humble Drive
2 Genuine Intense resolve
3 Modest Courageous
4 Selfless Clear catalyst for change
5 Gracious Will not settle
6 Courteous Bold
7 Unpretentious Strong work ethic
8 Team player Self-motivated
9 Placid Dedication to the organization
10 Self-control Gets results
Cronbach’s alpha is .929 for personal humility and .918 for professional
will. DeVellis (2012) recommended reducing the number of items if Cronbach’s
alpha is over .900. Again, duplicate concepts with lower factor loadings were
removed. For example, humble was retained, and modest was removed. In addition,
concepts that Collins did not use often to discuss Level 5 leaders, such as courteous
and courageous, were also removed. The final 10-item instrument to measure Level
5 leadership contains 5 items that capture personal humility and 5 items that
capture professional will (see Table 26). Cronbach’s alpha is .833 for personal
humility and .826 for professional will. This exceeds the threshold of .800 that
DeVellis established to be considered very good reliability. This scale will be
referred to as the Level 5 Leadership Scale (L5LS). In this research study, a score
55
Instrument to Measure Level 5 Leadership
of more than 37 for BOTH personal humility and professional will were considered
to identify a Level 5 leader.
Table 26: Final Level 5 Leadership Scale (L5LS) Attributes
Personal humility Profession will
1 Genuine Intense resolve
2 Humble Dedication to the organization
3 A team player A clear catalyst in achieving results
4 Servant attitude Strong work ethic
5 Doesn’t seek spotlight Self-motivated
Summary
In order to address the questions regarding Level 5 leadership, this chapter
presented findings from the responses of 349 subjects who described their bosses.
Three research questions were proposed in Chapter 1: (a) Utilizing the attributes
and characteristics from literature, can a statistically valid instrument be developed
to measure Level 5 leadership? (b) Is the personal humility construct of Level 5
leadership the same as servant leadership? (c) Do Collins’ (2006) eight questions to
test Level 5 leadership correlate with the attributes and characteristics that he used
to describe Level 5 leadership? To answer the first question, principal component
analysis confirmed Collins’ (2001) claim that there are two primary components of
the Level 5 construct: personal humility and professional will. Therefore, a
statistically validated instrument can be used to measure Level 5 leadership. The
second question was answered by the statistically significant correlation between
the personal humility score and the servant leadership score. Finally, the third
question was also answered positively with the statistically significant correlation
between Collins’ (2006) eight questions and the Level 5 leadership scores derived
from the personal humility and professional will scores.
56
Instrument to Measure Level 5 Leadership 57
Instrument to Measure Level 5 Leadership
Chapter 5 – Discussion
This study explored Collins’ (2001) concept of Level 5 leadership to
determine if, in fact, the attributes described in the literature could be developed
into a statistically validated instrument. This research shows that Level 5 is a
construct of leadership that can be measured with a valid instrument to identify
Level 5 leadership. This chapter discusses the findings of the research, the
implications of the research, the limitations of the study, and recommendations for
future research.
Evaluation of Findings
The research findings were consistent with expectations established from
existing literature. Although Collins’ research team wanted to stay away from the
term servant leadership to describe this new breed of leader that had been
identified as taking organizations from good to great (Collins, 2001), Patterson,
Redmer, and Stone (2003) and Drury (2004) were correct in suggesting that they
are actually the same leaders. Although the data show that servant leaders and
Level 5 leaders are the same people, the qualities of servant leadership align most
closely with the personal humility construct of Level 5 leadership and do not seem
to account for the idea of professional will. Therefore, a scale that measures equal
parts personal humility and professional will provide the most robust measure of
Level 5 leadership.
Since the eight questions from Collins (2006) were shown to identify the
same Level 5 leaders, it might seem reasonable to utilize those questions instead of
a new scale. However, there is an advantage to using the new Level 5 Leadership
Scale (L5LS). By separating the two constructs of Level 5 leadership, a more
accurate and comprehensive assessment of Level 5 leadership can be ascertained.
The eight questions from Collins are focused more on the personal humility
attributes. When a principal components analysis encompasses all of the attributes
and the eight questions together, all eight questions factor with the personal
humility attributes. None of the questions factor with the professional will
attributes. By balancing the personal humility and professional will attributes
58
Instrument to Measure Level 5 Leadership
evenly, both factors are given equal weight so that the scale does not favor humble
leaders who may not have professional will to outperform their peers.
Implications of Research
Collins (2001) identified at least two practical implications of this research:
(a) identifying CEOs to lead organizations from good to great and (b) finding the
seeds of Level 5 leadership within organizations and developing them. There is
now a validated two-factor instrument so that organizations can identify Level 5
leaders. The premise of Good to Great (Collins, 2001) is that organizations that
grew from good to great were led by CEOs who were Level 5 leaders. Therefore, a
board of directors searching for their next CEO should utilize the L5LS to
determine which candidates are Level 5 leaders. This process may also be utilized
in the process of hiring leaders who are below the CEO level. The greatest
challenge may be finding objective subjects who will provide an unbiased
assessment of a candidate’s attributes.
Within organizations, Level 5 leaders must be identified and developed.
Collins (2001) stated:
I believe—although I cannot prove—that potential Level 5 leaders are
prevalent in our society. The problem is not, in my estimation, a dearth of
potential Level 5 leaders. They exist all around us, if we just know what to
look for. (p. 37)
The use of 360-degree performance ratings is growing to more organizations and to
more people within the organizations (DeVito, 2012). This simple 10-item scale
can be included in a 360-degree evaluation so that individuals, peers, subordinates,
and superiors can identify Level 5 leaders within their organization.
Once the Level 5 leaders are identified within the organization or in the
interview process, they should be developed through mentoring. In American
corporations, mentoring is correlated with “increased job satisfaction, higher salary,
faster promotion, firmer career plans, and the increased probability that the protégé
will also become a mentor” (Wright & Wright, 1987, p. 204). One of the
characteristics of Level 5 leaders is that they surround themselves with strong
59
Instrument to Measure Level 5 Leadership
leaders and develop them into stronger Level 5 leaders so that the organization
remains healthy and vibrant, regardless of the individual at the helm (Collins,
2001).
Limitations of Study
This research effort forwards the body of literature on Level 5 leadership by
providing the first instrument to measure Level 5 leadership. However, there are
limitations. By utilizing a snowball methodology to gain participation in the survey,
the subjects fit a demographic profile similar to the researcher: middle-aged,
evangelical Christian men. Though the questions were not asked on the survey,
anecdotal feedback would indicate that most of the subjects are also Caucasians
who live in Georgia and Tennessee.
Recommendations for Future Research
The next step in the utilization of the L5LS to measure Level 5 leadership
could be to compare the actual results and accomplishments of leaders and potential
leaders to their Level 5 scores. Theory would indicate that Level 5 leaders should
be more effective in their outcomes. This research would utilize success criteria
within the organization such as sales growth or a 360-degree performance review
compared with L5LS results to determine if, in fact, Level 5 leaders are more
effective.
It is possible that good leadership is being identified with the L5LS, but
there may not be much difference between Level 5 leadership and a variety of other
types of good leadership such as authentic leadership (Avolio & Gardner, 2005) as
measured by the Authentic Leadership Questionnaire (Walumbwa, Avolio,
Gardner, Wernsing, & Peterson, 2008) or transformational leadership as measured
by the Multifactor Leadership Questionnaire (Bass, 1990, 2008). This study has
already shown the similarity between Level 5 leadership and servant leadership;
therefore, future research should evaluate the differences between these various
leadership constructs to confirm that they are not all measuring the same concepts.
Collins (2009) described professional will in Level 5 leaders as “an
absolute, obsessed, burning, compulsive ambition that was not about them” (1:15).
60
Instrument to Measure Level 5 Leadership
The results of this research show a correlation between personal humility and
dedication to the organization, a relationship that should be explored further. There
also seems to be a correlation between religious commitment and subjugating the
desires of self in leadership to a greater good, as evidenced in personal humility and
dedication to the organization.
Future research in the area of Level 5 leadership is also recommended
across a variety of cultural settings. As noted in the section with regards to
limitations, this research was performed on a fairly homogenous population.
Additional research in a variety of cultures and nationalities would provide
assurance that these concepts are universally valid.
Summary
This study explored the concept of Level 5 leadership from an academic
perspective to develop an instrument to accurately measure the concept of Level 5
leadership. Research has confirmed that the constructs of personal humility and
professional will that Collins (2001) proposed in Good to Great are valid and
separate constructs. Research also has confirmed that Level 5 attributes, Collins’
(2006) eight questions, and servant leadership are statistically the same. A simple
10-item scale (the L5LS) was developed that now can be utilized to identify these
leaders.
61
Instrument to Measure Level 5 Leadership
References
Ackman, D. (2002, September). The 20 most influential business books. Retrieved
from http://www.forbes.com/2002/09/30/0930booksintro_2.html
Avolio, B., & Gardner, W. (2005). Authentic leadership development: Getting to
the root of positive forms of leadership. The Leadership Quarterly, 16, 315-
338.
Baale, L. (2011, October 22). Nigeria: Humility as a great attribute of true
leadership. Africa News. Retrieved from http://allafrica.com/stories/
201110220126.html
Bass, B. (1990). From transactional to transformational leadership: Learning to
share the vision. Organizational Dynamics, 18(3), 19-31.
Bass, B. (2008). The Bass handbook of leadership: Theory, research, and
managerial application (4th ed.). New York, NY: Free Press.
Bennett, D. (2011, August 11). Business bestseller. Businessweek. Retrieved from
http://www.businessweek.com/magazine/tom-rath-08112011.html
Bisoux, T. (2007, January-February). Thinking big. BizEd, 6(1), 16-21. Retrieved
from http://www.e-digitaleditions.com/i/58065/17
Caulkins, D. (2008). Re-theorizing Jim Collins’ culture of discipline in Good to
Great. Innovation: The European Journal of Social Science Research,
21(3), 217-232.
Collins, J. (2001). Good to great: Why some companies make the leap . . . and
others don’t. New York, NY: Harper Collins.
Collins, J. (2002). Reading guide: Good to great. New York, NY: Harper Collins.
Collins, J. (2005). Level 5 leadership: The triumph of humility and fierce resolve.
Harvard Business Review, 79(1), 66-76.
Collins, J. (2006). Where are you on your journey from good to great? Good to
Great™ diagnostic tool. Retrieved from
http://www.jimcollins.com/tools/diagnostic-tool.pdf
Collins, J. (2009, September 21). From good to great: What defines a Level V
leader? [Video]. Retrieved from http://www.youtube.com/watch?v=q-
KyQ90XByY
62
Instrument to Measure Level 5 Leadership
Collins, J. (2011, March 25). Our problem is not a lack of Level 5 . . . [Video].
Retrieved from http://www.youtube.com/watch?v=at9-u8xv4vE
Collins, J. (2012). About Jim. Retrieved from http://www.jimcollins.com/about-
jim.html
Collins, J., & Hanson, M. (2011). Great by choice: Uncertainty, chaos, and luck—
Why some thrive despite them all. New York, NY: Harper Collins.
Collins, J., & Porras, J. (1994). Built to last: Successful habits of visionary
companies. New York, NY: Harper Collins.
Collins, J., & Rose, C. (2009, March 4). Level 5 Good to Great [Video]. Retrieved
from http://www.youtube.com/watch?NR=1&feature
=endscreen&v=wfaZ4pw99hc
Deci, E., & Ryan, R. (1985). Intrinsic motivation and self-determination in human
bebavior. New York, NY: Plenum Press.
Dennis, R., & Bocarnea, M. (2005). Development of the servant leadership
assessment instrument. Leadership & Organization Development Journal,
25(4), 600-615.
DeVellis, R. (2012). Scale development. Thousand Oaks, CA: Sage.
DeVito, A. (2012, October 25). Performance management specialists echoSpan
discuss 360-degree reviews getting longer, more widely used. Retrieved
from http://www.digitaljournal.com/pr/903908
Drury, S. (2004). Employee perceptions of servant leadership: Comparisons by
level and with job satisfaction and organizational commitment. Virginia
Beach, VA: Regent University.
Finnie, W., & Abraham, S. (2002). Getting from good to great: A conversation with
Jim Collins. Strategy and Leadership, 30(5), 10-14.
Fullan, M. (2004). Leadership & sustainability: System thinkers in action.
Thousand Oaks, CA: Corwin Press.
Gagne, M., & Deci, E. (2005). Self-determination theory and work motivation.
Journal of Organizational Behavior, 26(4), 331-362.
Goleman, D. (2000, March-April). Leadership that gets results. Harvard Business
Review , 79-90.
63
Instrument to Measure Level 5 Leadership
Greenleaf, R. (1970). The servant as leader. Retrieved from
http://www.greenleaf.org/whatissl/
Greenleaf, R. (1977). Servant leadership. A journey into the nature of legitimate
power and greatness. New York, NY: Paulist Press.
Hair, J., Black, B., Babin, B., Anderson, R., & Tatham, R. (2005). Multivariate
data analysis. Upper Saddle River, NJ: Prentice Hall.
IBM. (2012, September). Point-biserial correlations in SPSS. Retrieved from
http://www-01.ibm.com/support/docview.wss?uid=swg21476004
Liccardo, S. (2007). Level 5 leaders and the romance of leadership. Johannesburg,
South Africa: University of Witwatersrand.
Lichtenwalner, B. (2010). Servant leadership lessons: Jim Collins at Chick-Fil-A
leadercast. Retrieved from http://modernservantleader.com/servant-
leadership/servant-leadership-lesson-jim-collins-at-chick-fil-a-leadercast/
Likert, R., & Likert, J. (1976). New ways of managing conflict. New York, NY:
McGraw Hill Book.
Lim, D., Woehr, D., You, Y., & Gorman, C. (2007). The transalation and
development of a short form of the Korean language version of the
multidimensional work ethic profile. Human Resource Development
International, 10(3), 319-331.
Ling, T. (2009, December 12). Bland CEOs versus raging egomaniacs: Are the
world’s best leaders flamboyant visionaries or self-effacing individuals. The
Business Times Singapore. Retrieved from
http://a1preview.asia1.com.sg:90/vgn-ext-templating/v/ index.jsp?
vgnextoid=4fc4c0d825b85210VgnVCM100000430a0a0aRCRD&vgnextch
annel=73873060b6822110VgnVCM100000bd0a0a0aRCRD&vgnextfmt=pr
int
Malmendier, U., & Tate, G. (2008). Superstar CEOs. The Quarterly Journal of
Economics, 124(4), 1593-1638.
Markus, H., & Kitayama, S. (1991). Culture and the self: Impications for cognition,
emotion, and motivation. Psychological Review, 98(2), 224-253.
64
Instrument to Measure Level 5 Leadership
Maslow, A. (1946). A theory of human motivation. Psychological Review, 50(4),
370-396.
Maslow, A. (1954). Motivation and personailty. New York, NY: Harper and Row.
Maxwell, J. (2011). The 5 levels of leadership. New York, NY: Hachette Book
Group.
May, R. (2006, January 31). Why “Good to Great” isn’t very good. Retrieved from
http://www.businesspundit.com/why-good-to-great-isnt-very-good/
McGinn, D., & Silver-Greenberg, J. (2005, December 19). A “Great” second act:
Biz guru Jim Collins has advice for do-gooders. Newsweek, 146(25).
Retrieved from http://www.thedailybeast.com/newsweek/2005/12/11/a-
good-to-great-second-act.html
Miller, M., Woehr, D., & Hudspeth, N. (2001). The meaning and measurement of
work ethic: Construction and initial validation of a multidimensional
inventory. Journal of Vocational Behavior, 59, 1-39.
Morris, J., Brotheridge, C., & Urbanski, J. (2005). Bringing humility to leadership:
Antecedents and consequences of leader humility. Human Relations,
58(10), 1323-1350.
Nunnally, J. (1978). Psychometric theory. New York, NY: McGraw Hill.
Ogden, G., & Meyer, D. (2007). Leadership essentials: Shaping vision, multiplying
influence, defining character. Downers Grove, IL: IVP Connect.
Patterson, K. (2003). Servant leadership: A theoretical model. Virginia Beach, VA:
Regent University.
Patterson, K., Redmer, T., & Stone, A. (2003). Transformational leaders to servant
leaders versus Level 4 leaders to Level 5 leaders—The move from good to
great. CBFA Annual Conference. Virginia Beach, VA: Regent University.
Pimolsaengsuriya, A. (2012, May 13). Taking leadership to the next level. The
Nation (Thailand). Retrieved from
http://www.nationmultimedia.com/business/Coach-Talk-Column-Taking-
leadership-to-the-next-le-30181862.html
Ryan, R., & Deci, E. (2000). Self-determination theory and the facilitaion of
intrinsic motivation, social development, and well-being. American
65
Instrument to Measure Level 5 Leadership
Psychologist, 55(1), 68-78.
Serfontein, K., & Hough, J. (2011). Nature of the relationship between strategic
leadership, operational strategy, and organizational performance. SAJEMS
NS, 14(4), 393-406.
Smith, F. (2005, July 19). Great companies know a humble truth. Australian
Financial News, 53.
Taking Russia from good to great. (2012, June 3). Moscow Times. Retreived from
http://www.investmentclimate.ru/en/invest-climate-in-media/1421/
Taylor, A. (1992, April 20). Iacocca’s last stand at Chrysler: His anointed successor
must launch new models, shore up a shaky balance sheet, and find the
controls of the company fast. But is Chairman Lee ready to give them up?
Fortune. Retrieved from http://money.cnn.com/magazines/fortune/
fortune_archive/1992/04/20/76315/index.htm.
Tinsley, H., & Tinsley, D. (1987). Uses of factor analysis in counseling psychology
research. Journal of Counseling Psychology, 34(4), 414-424.
Tokunaga, P. (2003). Invitation to lead: Guidance for emerging Asian American
leaders. Downers Grove, IL: InterVarsity Press.
van Dierendonck, D. (2011). Servant leadership: A review and synthesis. Journal
of Management , 37(4), 1228-1261.
Walumbwa, F., Avolio, B., Gardner, W., Wernsing, T., & Peterson, S. (2008).
Authentic leadership: Development and validation of a theory-based
measure. Journal of Management, 34(1), 89-126.
Weber, M. (1958). The Protestant ethic and the spirit of capitalism. New York,
NY: Scribners.
Wicks, W. (1997). Shared values: A history of Kimberly-Clark. Greenwich, CT:
Greenwich.
Williams, M. (2005). Leadership for leaders. Oxfordshire, UK: Thorogood.
Winston, B. (2003). Extending the servant leadership model: Coming full circle.
Paper presented at Regent University’s Servant Leadership Roundtable,
Virginia Beach, VA.
66
Instrument to Measure Level 5 Leadership
Winston, B., & Fields, D. (in press). Seeking the essense of servant leadership:
Identifying core servant leader behaviors.
Wong, P., & Davey, D. (2007). Best practices in servant leadership. Paper
presented at Regent University’s Servant Leadership Roundtable, Virginia
Beach, VA.
Woolridge, A. (2009, November 12). The cult of the faceless boss. The Economist.
Retrieved from http://www.economist.com/node/14844995
Woolridge, A. (2011, November 26). Built to last: Jim Collins has stayed at the top
by practicing what he preaches. The Economist. Retreived from
http://www.economist.com/node/21540219
Wright, C., & Wright, C. (1987). The role of mentors in the career development of
young professionls. Family Relations, 36(2), 204-208.
67
Instrument to Measure Level 5 Leadership
Appendix A
Survey Summary
Informed Consent Statement: I would appreciate your assistance with this research project on developing an instrument to measure Level 5 leadership, as defined by Jim Collins in Good to Great. The project is being conducted by Wilbur Reid for a dissertation for Regent University.
All you need to do is complete this survey, which should take approximately 15 minutes. Your participation is completely voluntary, so you have the option to skip questions or to stop participating at any time. If you do not wish to participate, simply exit from the web site.
Responses will be completely anonymous; your name will not appear anywhere on the survey.
By completing and returning the questionnaire you are acknowledging that you are 18 years of age or older and are consenting to participate in this study.
If you have any questions regarding the research, contact Wilbur Reid at [email protected] or 404-202-1924.
Thank you again for your help.
68
Table A1: Level 5 Leadership
On a scale of 1 to 10, to what extent do the following characteristics describe your boss? A 1 indicates that this characteristic does not describe your boss at all,
while a 10 indicates that it describes him/her exactly.
Answer options1
Not At All2 3 4 5 6 7 8 9
10Exactly
N/ARating average
Response count
Humble 26 26 34 27 36 22 33 43 55 40 6 6.05 348Set up others for success
24 18 21 13 31 21 36 60 61 56 5 6.74 346
Genuine 12 11 13 17 20 21 30 51 77 88 8 7.52 348Lack of pretense 27 26 30 27 47 21 23 49 42 37 13 5.89 342
Gives others credit for success
7 13 16 13 20 19 27 56 85 83 8 7.61 347
Accepts responsibility when things don’t go well
16 9 22 16 24 20 39 62 61 61 16 7.07 346
Modest 22 27 23 42 29 30 37 50 41 43 4 6.09 348Authentic 9 14 8 18 31 29 21 55 71 88 4 7.47 348Team player 8 11 15 16 26 24 33 64 63 79 8 7.41 347
Sets up successors for success
24 22 15 28 30 24 34 55 46 39 30 6.31 347
Self-effacing 40 28 44 24 46 35 34 39 27 16 12 5.16 345Gracious 8 21 20 17 34 35 31 61 53 59 8 6.87 347Unpretentious 22 21 25 26 31 22 26 42 58 63 7 6.55 343Courteous 6 9 19 18 31 23 29 66 55 83 8 7.38 347Understated 31 34 42 34 40 27 31 38 30 22 8 5.27 337Servant attitude 52 35 25 33 28 21 28 33 38 48 7 5.50 348Doesn't seek spotlight 29 32 37 35 28 24 27 52 37 45 3 5.83 349Reserved personality 52 56 43 33 27 20 21 34 27 31 3 4.80 347
Placid persona 23 31 41 21 39 27 24 53 45 41 3 5.96 348
Answer options1
Not At All2 3 4 5 6 7 8 9
10Exactly
N/ARating average
Response count
(pleasantly calm or peaceful)
Selfless (puts the needs of others first)
24 19 39 21 34 30 44 53 38 42 4 6.13 348
Egocentric 89 76 41 15 25 18 24 20 17 20 2 3.88 347Arrogant 109 68 33 15 19 24 22 24 15 16 4 3.72 349Overbearing 79 63 52 21 25 24 26 23 13 17 5 3.99 348Condescending 95 85 32 23 21 18 21 19 17 12 3 3.61 346Patronizing 91 70 44 21 29 21 22 19 14 13 3 3.72 347Pretentious 95 74 46 19 23 26 18 11 15 15 4 3.59 346
Talks about themselves a lot
82 66 45 18 38 17 22 17 17 20 7 3.96 349
Does not set up successors for success
96 53 36 22 23 19 13 22 18 20 23 3.90 345
Large personal ego 83 62 39 23 25 16 24 25 20 26 4 4.21 347
Seeks fame, fortune, adulation, and power
90 59 42 23 33 18 22 16 20 22 2 4.00 347
Boastful 89 68 50 23 23 28 20 16 16 12 3 3.72 348
Self-serving (at the expense of others)
114 76 41 16 21 13 16 19 16 14 2 3.42 348
Acts superior 75 59 49 24 23 21 24 25 23 22 4 4.27 349Large ego 79 59 41 24 18 22 17 33 23 28 4 4.38 348Air of self-importance 75 59 41 12 27 25 25 29 26 24 4 4.46 347Rude 149 67 38 17 23 16 13 8 11 5 1 2.82 348Snooty 148 60 41 24 16 11 21 5 9 9 4 2.88 348Seeks personal greatness
44 51 35 18 30 28 31 42 35 27 6 5.22 347
Charismatic leadership
21 23 42 29 27 33 38 54 39 38 2 6.02 346
Answer options1
Not At All2 3 4 5 6 7 8 9
10Exactly
N/ARating average
Response count
Resolve 8 7 10 7 28 24 46 69 80 60 5 7.55 344Inner intensity 7 8 11 19 29 30 49 62 70 56 5 7.29 346Gets results 5 2 8 11 24 25 41 71 93 61 4 7.77 345Self-motivated 4 5 3 3 18 10 31 53 102 109 10 8.38 348Ferocious resolve 20 18 20 17 38 42 47 50 53 34 5 6.41 344Intense resolve 10 8 19 16 29 36 43 62 76 42 4 7.06 345Drive 1 6 9 3 22 26 43 59 88 84 7 7.99 348Self-control 9 4 13 16 33 27 41 62 73 65 3 7.40 346Strength of character 7 7 9 13 22 18 43 47 80 90 10 7.78 346Courageous 5 10 12 20 24 36 59 68 58 47 7 7.16 346Bold 7 9 18 17 28 40 49 74 61 43 2 7.05 348Builds strong team 15 20 13 15 30 25 41 62 55 66 3 6.99 345Strong work ethic 9 4 7 5 15 10 31 48 83 122 13 8.27 347
Clear catalyst in achieving results
8 13 13 18 26 29 36 77 63 54 7 7.19 344
Ambitious for organization
7 9 12 9 16 18 34 70 88 77 8 7.78 348
Fearless 10 13 19 19 40 31 52 77 53 28 6 6.70 348Results oriented 2 4 9 10 16 22 28 54 102 96 5 8.15 348Self-determination (determination not influenced by outsiders)
9 7 19 17 20 21 41 60 88 59 7 7.40 348
Enthusiastic desire to produce results
3 3 8 11 21 17 34 67 79 101 2 8.08 346
Unafraid to take risk 20 11 24 10 32 27 48 69 66 38 2 6.78 347Intense personality 21 29 32 24 35 30 45 38 42 46 5 6.09 347
Will settle for nothing less than the best
8 12 10 13 39 27 44 76 71 43 5 7.16 348
Answer options1
Not At All2 3 4 5 6 7 8 9
10Exactly
N/ARating average
Response count
Pride in the organization
4 2 2 6 19 14 28 59 78 124 9 8.41 345
Dedication to the organization
4 2 5 7 10 9 30 55 80 136 9 8.54 347
Willpower (able to control one's impulses and actions)
5 11 19 19 30 26 44 55 73 59 7 7.22 348
Backbone 8 11 10 18 22 32 37 65 80 62 5 7.40 350
Daring: willing to take a chance
13 14 19 17 29 30 51 71 59 44 1 6.88 348
Fierce resolve to life that is not deterred based on emotional ups and downs
16 12 15 19 42 28 45 49 60 49 12 6.79 347
Fanatically driven to achieve results
29 23 26 19 31 45 44 53 42 31 4 6.05 347
Unmotivated 173 92 46 7 8 2 3 6 7 3 2 2.12 349Lazy 204 60 36 7 11 7 4 5 5 6 3 2.12 348Undisciplined 136 75 57 24 14 10 5 8 9 6 2 2.68 346Weak leader 107 67 40 19 18 21 26 14 16 16 3 3.64 347Afraid to take a chance
79 80 59 31 25 20 15 14 12 8 6 3.47 349
Surrounded by “yes-men”
82 59 52 32 27 19 19 20 9 21 6 3.87 346
Cowardly 143 81 47 27 12 11 3 7 7 3 5 2.49 346
Instrument to Measure Level 5 Leadership
Table A2: Servant leadership
To what extent does each of these statements describe the behavior of your boss?
Answer optionsDefinitel
y noNo Neutral Yes
Definitely yes
Rating average
Response count
Practices what he/she preaches
11 33 45 156 105 3.89 350
Serves people without regard to their nationality, gender, or race
6 17 28 127 172 4.26 350
Sees serving as a mission of responsibility to others
12 52 72 123 91 3.65 350
Genuinely interested in employees as people
19 26 60 126 119 3.86 350
Understands that serving others is most important
18 56 92 118 65 3.45 349
Willing to make sacrifices to help others
13 41 86 132 78 3.63 350
Seeks to instill trust rather than fear or insecurity
23 44 49 122 112 3.73 350
Is always honest 16 44 49 128 112 3.79 349Is driven by a sense of higher calling
17 45 104 92 91 3.56 349
Promotes values that transcend self-interest and material success
16 51 81 125 76 3.56 349
73
Table A3: Collins’ Eight Questions
How does your boss exemplify the following characteristics?
Answer options
A = Exemplifies this trait
exceptionally well—there is
limited room for improvement.
B = Often exemplifies this
trait, but has room for
improvement.
C = Some evidence of this trait, but record
is spotty.
D = Little evidence that this
trait is exemplified, and there are obvious contradictions.
F = Operates almost entirely contrary to this
trait.
Not applicable
Response count
Is ambitious first and foremost for the cause, the organization, the work—not themselves—and they have an iron will to do whatever it takes to make good on that ambition.
111 133 56 30 17 1 348
Displays an ever-improving track record of making decisions that prove best for the long-term greatness of the company and its work.
97 139 66 22 16 6 346
Practice the window and the mirror. They point out the window to people and factors other than themselves to give credit for success. When confronted with failures, they look in the mirror and say, “I am responsible.”
84 126 65 39 26 6 346
Although he or she might be charismatic, this is not the primary source of their effectiveness. They inspire
113 109 62 36 24 4 348
Answer options
A = Exemplifies this trait
exceptionally well—there is
limited room for improvement.
B = Often exemplifies this
trait, but has room for
improvement.
C = Some evidence of this trait, but record
is spotty.
D = Little evidence that this
trait is exemplified, and there are obvious contradictions.
F = Operates almost entirely contrary to this
trait.
Not applicable
Response count
others primarily via inspired standards—excellence, hard work, sacrifice, and integrity—not with an inspiring public persona.Values substance over style, integrity over personality, and results over intentions. 131 136 39 31 11 1 349
Dialogues and debates in search of the best answer (not for the sake of looking smart or winning a point) up until the point of decision.
120 113 61 31 23 0 348
Unifies behind a decision to ensure success—even if disagreed with the decision. 84 142 69 33 17 3 348
Cultivate leaders who are highly capable individuals, strong contributing team members, competent managers, and effective leaders.
91 137 62 34 22 2 348
Instrument to Measure Level 5 Leadership
Table A4: Demographics
Answer options % n
What is your gender?
Male 64.1% 223
Female 35.9% 125
What is the gender of your boss?
Male 83.4% 291
Female 16.6% 58
What is your age?
24 or younger 1.1% 4
25-34 10.0% 35
35-44 26.1% 91
45-54 36.7% 128
55-64 18.3% 64
65 or older 7.7% 27
What is the approximate age of your boss?
24 or younger .0% 0
25-34 3.2% 11
35-44 19.8% 69
45-54 45.0% 157
55-64 25.5% 89
65 or older 5.4% 19
Which level best describes your boss’ position?
Executive, C-level (i.e.,
CEO, COO, CIO,
president, senior pastor,
etc.)
28.9% 101
Officer of the company
(i.e., vice president, dean,
principal, etc.)
29.8% 104
77
Instrument to Measure Level 5 Leadership
Answer options % n
Director/senior manager 25.5% 89
Manager 13.5% 47
Supervisor/team leader 2.3% 8
Which of the following best describes your boss’ religious affiliation?
Christianity - Catholic 13.2% 46
Christianity - Evangelical
(Christian church,
Church of Christ,
Baptist, etc.)
33.0% 115
Christianity - Mainline
Protestant (Lutheran,
Presbyterian, Methodist,
etc.)
11.2% 39
Christianity - Other 6.3% 22
Other religion (Judaism,
Islam, Hindu, Buddhist,
etc.)
3.7% 13
Nonreligious (atheist,
agnostic, secular)5.5% 19
Don’t know 27.0% 94
To the best of your knowledge, which description best describes the religious
commitment of your boss?
A leader: involved in
leadership and is very
committed
20.7% 72
Committed (attends
services regularly)23.9% 83
Somewhat committed
(attends services
10.1% 35
78
Instrument to Measure Level 5 Leadership
Answer options % n
occasionally)
Not committed 13.8% 48
Don’t know 31.6% 110
What type of organization do you work for?
Large, Fortune 500
corporation33.3% 116
Small to medium-sized
corporation (publicly
traded on a stock
exchange)
14.9% 52
Privately owned
company (not publicly
traded on a stock
exchange)
21.3% 74
Nonprofit organization 30.5% 106
79
Instrument to Measure Level 5 Leadership
Appendix B
Human Subjects Review Board Application
Please submit one electronic copy of this form and any supporting documents to your
dissertation chair or to the SBL IRB representative, Dr. Emilyn Cabanda at:
1. PROJECT REVIEW
New Project (The HSRB will assign an ID#) ___________________________
Revised Project (Enter ID#) ___________________________
Renewal (Enter ID#) ___________________________
2. PRINCIPAL INVESTIGATOR _____Wilbur Reid______________________
Address_2936 Summitop Road, Marietta, GA 30066_ Phone __404-202-1924____
E-Mail [email protected]_________________ Date __Nov. 10, 2012____
List of all project personnel (including faculty, staff, outside individuals or
agencies) Wilbur Reid, Dr. Bruce Winston (chair)_______________________
__________________________________________________________
If you are a student, please provide the following additional information:
This research is for Dissertation Thesis Independent Study
Other ___________________________________________
Faculty Advisor’s Name: ____Dr. Bruce Winston_______________________________
3. TRAINING: The National Institutes of Health Office of Extramural Research
offers free self-paced online training at phrp.nihtraining.com.
I have completed human subjects research training. Training Date: _9/30/12__
4. PROJECT TITLE Development of an instrument to measure Level 5 leadership
5. IS THIS RESEARCH BEING SUBMITTED AS PART OF A FUNDED
RESEARCH PROPOSAL? Yes No
80
Instrument to Measure Level 5 Leadership
If yes, please identify the funding source: ________________________________
6. ANTICIPATED LENGTH OF HUMAN SUBJECTS CONTACT:
Beginning Date Sept 2012___________ Ending Date __Nov. 2012_________
7. DESCRIPTION OF PARTICIPANTS:
Number __349_____ Age Range ___18-90________
Briefly describe subject population Participants are a diverse group of adults
working at a variety of levels in a variety of organizations solicited through
snowballing in social media. The majority of participants were male (70%),
between the ages of 35 and 64 (81%), Christian (87%) and/or work for a for-profit
organization (70%). Though the survey did not include questions regarding
geography, anecdotal evidence and feedback indicates that the majority of
participants are likely from the states of Georgia and Tennessee.
8. INDICATE THE REVIEW CATEGORY FOR WHICH YOU ARE
APPLYING.
Further information about each review category can be found at
http://www.regent.edu/academics/academic_affairs/IRB/guidelines.cfm
I am applying for an exempt review, based on one or more of the following categories (check all that apply):Note: Exempt review cannot be claimed for any research involving prisoners and most research involving children.
Research conducted in established or commonly accepted educational settings and involving normal educational practices such as (i) research on regular and special education instructional strategies, or (ii) research on the effectiveness of or the comparison among instructional techniques, curricula, or classroom management methods
Research involving the use of survey procedures, educational tests (cognitive, diagnostic, aptitude, achievement), interview procedures or observation of public behavior, if information from these sources is
81
Instrument to Measure Level 5 Leadership
recorded in such a manner that participants cannot be identified, directly or through identifiers linked to the subjects; and (ii) any disclosure of the human subjects' responses outside the research could not reasonably place the subjects at risk of criminal or civil liability or be damaging to the subjects' financial standing, employability, or reputation Note: This category cannot be used for research involving children
Research involving the use of survey procedures, educational tests (cognitive, diagnostic, aptitude, achievement), interview procedures, or observation of public behavior, if (i) the human subjects are elected or appointed public officials or candidates for public office; or (ii) federal statute(s) require(s) without exception that the confidentiality of the personally identifiable information will be maintained throughout the research and thereafter
Research involving the collection or study of existing data, documents, records, pathological specimens, or diagnostic specimens, if these sources are publicly available or if the information is recorded by the investigator in such a manner that subjects cannot be identified, directly or through identifiers linked to the subjects
Research and demonstration projects which are conducted by or subject to the approval of federal department or agency heads, and which are designed to study, evaluate, or otherwise examine (i) Public benefit or service programs; (ii) procedures for obtaining benefits or services under those programs; (iii) possible changes in or alternatives to those programs or procedures; or (iv) possible changes in methods or levels of payment for benefits or services under those programs
I am applying for an expedited review, based on meeting all of the following conditions (check all that apply):Note: Expedited review cannot be claimed for research involving prisoners.
Research poses no more than minimal risk to subjects (defined as "the probability and magnitude of harm or discomfort anticipated in the research are not greater in and of themselves than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests.")
Research limited to one or more of the following data collection procedures:
Collection of data through noninvasive procedures routinely employed in clinical practice
Research involving materials (data, documents, records, or specimens) that have been collected, or will be collected solely for nonresearch purposes
Collection of data from voice, video, digital, or image recordings made for research purposes
Research on individual or group characteristics or behavior (including, but not limited to, research on perception, cognition, motivation, identity, language, communication,
82
Instrument to Measure Level 5 Leadership
cultural beliefs or practices, and social behavior) or research employing survey, interview, oral history, focus group, program evaluation, human factors evaluation, or quality assurance methodologiesNote: Some research in this category may be classified as
exempt; this listing refers only to research that is not exempt.
Continuing review of research previously approved by the convened HSRB as follows: (a) where (i) the research is permanently closed to the enrollment of new subjects; (ii) all subjects have completed all research-related interventions; and (iii) the research remains active only for long-term follow-up of subjects; or (b) where no subjects have been enrolled and no additional risks have been identified; or (c) where the remaining research activities are limited to data analysis.
I am applying for full board review.
9. PROJECT DESCRIPTION
Briefly describe (or attach) the methodology and objectives of your research
(including hypotheses and/or research questions), the data collection procedures,
and any features of the research design that involve procedures or special
conditions for participants, including the frequency, duration, and location of their
participation. The description should be no longer than 3 pages single space.
Attach addendums for materials and detailed descriptions of the research if more
space is needed. Please note that complete chapters of thesis/dissertation
proposals will not be accepted.
There is currently no instrument to measure Level 5 leadership, as defined
by Jim Collins in Good to Great (2001).Therefore, an exploration of the
characteristics of Level 5 leadership was conducted to develop a
parsimonious scale to identify the Level 5 traits of leaders. The three
research questions that were answered were: Utilizing the attributes and
characteristics from literature, can a statistically valid instrument be
developed to measure Level 5 leadership? Is the personal humility construct
of Level 5 leadership the same as servant leadership? Do Collins’ eight
questions to test Level 5 leadership correlate with the attributes and
characteristics that he uses to describe Level 5 leadership? The goal of this
83
Instrument to Measure Level 5 Leadership
study is to provide an accurate and effective instrument to measure Level 5
leadership within individuals.
To accomplish this objective, a 92 item survey was administered to 349
adults representing a diverse set of organizations and work experiences.
Responses were collected via Survey Monkey ™ online and are
anonymous. Each participant completed one survey, which took about 15
minutes. Reliability was very high, with Cronbach’s alpha of at least .92 on
all components.
HSRB Project Description Checklist
a) Is your data completely anonymous, where there are no possible identifications of the participants.
No Yes
b) Will you be using existing data or records? If yes, describe in project description (#9 above)
No Yes
c) Will you be using surveys, questionnaires, interviews or focus groups with subjects? If yes, describe in #9 and include copies of all in application.
No Yes
d) Will you be using videotape, audiotape, film? If yes, describe in #9 No Yes
e) Do you plan to use any of the following populations? Regent students, Regent employees, Non-English speaking, cognitively impaired, patients/clients, prisoners, pregnant women? If yes, describe which ones in #9
No Yes
f) Do you plan to use minors (under 18)? If yes, describe in #9 and give age ranges
No Yes
g) Are sites outside of Regent engaged in the research? If yes, describe in #9 and give consent letter or their IRB information
No Yes
h) Are you collecting sensitive information such as sexual behavior, HIV status, recreational drug use, illegal behaviors, child/elder/physical abuse, immigrations status, etc? If yes, describe in #9.
No Yes
i) Are you using machines, software, internet devices? If so describe in #9 No Yes
j) Are you collecting any biological specimens? If yes, describe in #9 No Yes
k) Will any of the following identifying information be collected: names, telephone numbers, social security number, fax numbers, email addresses, medical records numbers, certificate/license numbers, Web
No Yes
84
Instrument to Measure Level 5 Leadership
universal resource locators (URLs), Internet protocol (IP) address numbers, fingerprint, voice recording, face photographic image, or any other unique identifying number, code or characteristic other than “dummy” identifiers? If yes, describe in #9
l) Will there be data sharing with any entity outside your research team? If so, describe who in #9
No Yes
m)Does any member of the research team or their family members have a personal financial interest in the project (for commercialization of product, process or technology, or stand to gain personal financial income from the project)? If yes, describe in #9.
No Yes
n) As applicable, do you plan to provide a debriefing to your participants? If written, include in application as addendum
No Yes
o) Will there be any inducement to participate, either monetary or nonmonetary? If there is inducement please describe how the amount is not coercive in #9.
No Yes
p) Will there be any costs that subjects will bear (travel expenses, parking fees, professional fees, etc. If no costs other than their time to participate, please indicate)? If yes describe in #9
No Yes
q) Will subjects be studied on Regent University campus? If yes, please describe where the study will be done in #9
No Yes
r) Will subjects be obtained by internet only? If yes, please describe what internet forums or venues will be used to obtain participants in #9
No Yes
s) Are you using the Regent University consent form template? Whether using the template or requesting an alternate form, you must include a copy in your submission.
No Yes
10. PARTICIPANT RECRUITMENT
Describe the sources of potential participants, how they will be selected and
recruited, and how and where you will contact them. Describe all relevant
characteristics of the participants with regard to age, ethnic background, sex,
institutional status (e.g., patients or prisoners), and their general state of mental and
physical health.
Participants are a diverse group of adults working at a variety of levels in a variety
of organizations solicited through snowballing in social media. The majority of
participants were male (70%), between the ages of 35 and 64 (81%), Christian
(87%) and/or work for a for-profit organization (70%). Though the survey did not
85
Instrument to Measure Level 5 Leadership
include questions regarding geography, anecdotal evidence and feedback indicates
that the majority of participants are likely from the states of Georgia and
Tennessee. Each of the participants is employed and the state of physical and
mental health of the participants is believed to be sound.
11. INFORMED CONSENT
Describe how you will inform participants of the nature of the study. Attach a
copy of your cover letter, script, informed consent form and other information
provided to potential participants.
Informed Consent Statement: I would appreciate your assistance with this research project on developing an instrument to measure Level 5 leadership, as defined by Jim Collins in Good to Great . The project is being conducted by Wilbur Reid for a dissertation for Regent University.
All you need to do is complete this survey, which should take approximately 15 minutes. Your participation is completely voluntary, so you have the option to skip questions or to stop participating at any time. If you do not wish to participate, simply exit from the web site.
Responses will be completely anonymous; your name will not appear anywhere on the survey.
By completing and returning the questionnaire you are acknowledging that you are 18 years of age or older and are consenting to participate in this study.
If you have any questions regarding the research, contact Wilbur Reid at [email protected] or 404-202-1924.
Thank you again for your help.
** EXEMPT APPLICATIONS SKIP TO QUESTION 17: ATTACHMENTS **
12. WRITTEN CONSENT
I am requesting permission to waive written consent, based on one or
more of the following categories (check all that apply):
86
Instrument to Measure Level 5 Leadership
The only record linking the subject and the research would be the
consent document, and the principal risk would be potential harm
resulting from a breach of confidentiality.
The research presents no more than minimal risk of harm to subjects
and involves no procedures for which written consent is normally
required outside of the research context.
I will be using a written consent form. Attach a copy of the written
consent form with this application.
13. CONFIDENTIALITY OF DATA
What procedures will be used to safeguard identifiable records of individuals and
protect the confidentiality of participants?
__________________________________________________________________
_________________________________________________________________
_________________________________________________________________
_________________________________________________________________
** EXPEDITED APPLICATIONS SKIP TO QUESTION 17: ATTACHMENTS **
14. RISKS AND BENEFITS
Describe in detail the immediate or long-range risks, if any, to participants that
may arise from the procedures used in this study. Indicate any precautions that will
be taken to minimize these risks. Also describe the anticipated benefits to
participants and to society from the knowledge that may be reasonably expected to
result from this study.
_________________________________________________________________
_________________________________________________________________
__________________________________________________________________
__________________________________________________________________
15. DEBRIEFING STATEMENT
The two major goals of debriefing are dehoaxing and desensitizing. Participants
should be debriefed about any deception that was used in the study. Participants
87
Instrument to Measure Level 5 Leadership
also should be debriefed about their behavioral response(s) to the study. Please
describe your debriefing plans and include any statements that you will be
providing to the participants.
_________________________________________________________________
_________________________________________________________________
__________________________________________________________________
__________________________________________________________________
16. DISSEMINATION & STORAGE OF RESULTS
a) How and where do you plan on disseminating the results of your study?b) For electronic data stored on a computer, how will it be stored and
secured (password, encryption, other comparable safeguard)?c) For hardcopy data, how will it be stored (locked office or suite, locked
cabinet, data coded by team with master list secured separately, other)?d) What are your plans for disposing of data once the study is ended (give
method and time)?__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
__________________________________________________________________
17. ATTACHMENTS:
Attach copies of all relevant project materials and documents, including (check all
that apply):
A copy of your training certificate (required for principal investigator)
Surveys, questionnaires, and/or interview instruments
Informed consent forms or statements
Letters of approval from cooperative agencies, schools, or education boards
Debriefing statements or explanation sheet
18. AFFIRMATION OF COMPLIANCE:
By submitting this application, I attest that I am aware of the applicable principles,
policies, regulations, and laws governing the protection of human subjects in
research and that I will be guided by them in the conduct of this research. I agree
88
Instrument to Measure Level 5 Leadership
to follow the university policy as outlined in the Faculty & Academic Policy
Handbook (available online at
http://www.regent.edu/academics/academic_affairs/handbook.cfm) to ensure that
the rights and welfare of human participants in my project are properly protected. I
understand that the study will not commence until I have received approval of
these procedures from the Human Subjects Review Board. I further understand
that if data collection continues for more than one year from the approval date, a
renewal application must be submitted.
I understand that failure to comply with Federal Regulations (45 CFR 46, available
online at http://www.hhs.gov/ohrp/humansubjects/guidance/45cfr46.html ) can
result in confiscation and possible destruction of data, suspension of all current and
future research involving human subjects, or other institutional sanctions, until
compliance is assured.
89
Instrument to Measure Level 5 Leadership 90