Research Articles

Home -> Research -> Articles

By: Scott G. Isaksen, Barbara J. Babij and Kenneth J. Lauer

Abstract:  Creative leadership incorporates the full spectrum of change and cognitive style, from innovative/exploratory to adaptive/developmental.  Leadership literature and measures tend to describe characteristics at the innovative/exploratory end of the spectrum, while management literature and measures portray characteristics at the adaptive/developmental end of the spectrum. The purpose of this study was to examine the relationship between cognitive style and leadership behaviors.  Two popular measures were used: the Kirton Adaption-Innovation Inventory and the Leadership Practices Inventory. 

What do we mean by creative leadership?  Creative leadership incorporates the full spectrum of change, as well as the full spectrum of cognitive style, from innovative/exploratory to adaptive/developmental.  At the innovative/exploratory end of the spectrum are the fundamentally different kinds of change, and people who prefer to do things in fundamentally different ways.  At the adaptive/developmental end of the spectrum are incremental types of change, and people who prefer to improve upon existing ways of doing things.  Literature on leadership tends to depict characteristics at the innovative/exploratory end of the spectrum, while literature on management describes characteristics at the adaptive/developmental end of the spectrum.  Innovative/exploratory and adaptive/developmental represent the two poles of cognitive style that manifest as leadership behavior.

That there is great interest in leadership is uncontested; as a field, it has grown in theory and research over the past fifty years.  Bass (1990) identified 7,500 studies of leadership.  By 1999, PsychLit and Erich listed 17,000 titles on leadership (Babij, 2001).  In 2001 alone, offered 902 titles on leadership.  Indeed, one of the more critical needs facing companies in the current century is for leadership (Bennis & Nanus, 1985; Greenleaf, 1991; Heifetz, 1998; Hock, 1999; Kouzes and Posner, 1997). 

The word “leadership” is commonly used with respect to people, position, or actions (Heifetz, 1998).  The historical trend in the literature progressed from trait theories to situational theories, to contingency models and to transactionalists (Bass, 1981; Heifetz, 1998).  Much of the literature pertains to a context of change and examines many different approaches. For some scholars it is a process, for some it is about personal characteristics; for some it is a question of position; and for some there is interplay between the environment and the person (Bass, 1981).   Still others base their paradigms of leadership on the “new science” and a systems view (Senge, 1990; Wheatley 1992, 1999).  Another popular perspective differentiates between management and leadership (Bennis, 1989; Kotter, 1996).  This perspective includes theories that distinguish leadership behavior as transformational and management behavior as transactional (Bass, 1985; Burke & Litwin, 1992; Burns, 1978; Kouzes & Posner, 1978).  And there are some like Gardner (1990) who realize that these two constructs may in fact, not be mutually exclusive.  Table 1 below lists some of the distinctions between leadership and management found in the literature. 

In addition to the literature, there are more than forty measures on the market today that represent some theory of leadership or management effectiveness (Pfeiffer, 1991; Shipper, 1995).  These instruments measure a vast range of traits, characteristics, behaviors and styles.  As with the literature, leadership inventories tend to focus on the innovative/exploratory end of the spectrum of cognitive style.  This tendency seems to create a bias toward innovation, i.e. those with a natural preference for doings things differently are considered better leaders  (Lauer & Babij, 2002).  This range in natural preference, from doing things differently to doing things better is known as cognitive style (Kirton, 1976, 1989, 2000; Talbot, 1997). 

The purpose of this study was to empirically examine the relationship between cognitive style and leadership behaviors, to either confirm or disprove the apparent bias in the literature and measures.  Two widely used measures were utilized: the Kirton Adaption-Innovation Inventory (KAI) and the Leadership Practices Inventory (LPI).  The KAI measures cognitive style and the LPI measures “best-practice” leadership behaviors.  The authors used the KAI to explore three hypothesis:
Hypothesis 1: There is a relationship between leadership behaviors and cognitive style as measured by the LPI and KAI.

Hypothesis 2: The LPI reflects an innovative bias.
Hypothesis 3: The LPI does not consider the full spectrum of cognitive style.

The methods used for this study, including a description of the sample, a brief overview of the theory underpinning each measure, and a brief description of the measures and sub-scales will be presented first.



A random sample of files was selected.  The total sample consisted of data collected from 179 participants who completed both the LPI (Kouzes & Posner, ) and the KAI (Kirton, 1977, 1985, 1991).  The data were collected from a diverse population across companies and groups: 86 worked for a multi-national direct mail company , 38 participated in public creative problem-solving courses and worked in a variety of professions, 24 worked for a manufacturer of household goods in the U.S., and 37 worked for an international accountancy firm.  The sample included 109 men and 67 women; three did not indicate gender.  Seven of these participants provided only sub-scale scores and a total score for the KAI, based on a previous score.  Ages of participants ranged from 20-63.


Kirton developed the KAI as a measure of cognitive style.  He (1984, 1989, 2000) suggested the concept of a creativity continuum, distinguishing cognitive capacity (level) from cognitive style.  Kirton posited that individuals have a natural style for creativity, problem solving and decision-making (Holland, Bowskill, & Bailey, 1991; Kirton, 1976, 1989, 2000; Talbot, 1997).  When left alone, this preference will manifest in one’s behavior. 

The KAI is a 32-item paper and pencil self-report instrument.  The total scale ranges from 32-160; the theoretical mean is 96.  According to one’s total score on the measure, individuals can be placed on a continuum from highly adaptive to highly innovative (Kirton, 1984).   A more adaptive score is positioned to the left of the mean; a more innovative score is positioned to the right of the mean.  The stem question asks “how easy or difficult do you find it to present yourself, over time as…”  The scale for each question ranges from very easy to very hard.  The nature of the stem sets up the KAI as a measure of cognitive style rather than a measure of level.  The measure is value neutral as no judgment attaches to either end of the spectrum; i.e., neither is better than the other.

The more one is an adaptor, the more one prefers to do novel things in better ways.    More adaptive individuals tend to remain within the current paradigm to define the type of change needed as well as to implement change.  More innovative individuals prefer to be novel in fundamentally different ways, ignoring or even redefining the current paradigm (Kirton, 1993). That these differences are real is supported by a number of studies (CPS-B, 1997; Dewan, 1982; Gryskiewicz, 1987). 

The measure itself consists of three sub-scores: Sufficiency of Originality, Efficiency, and Rule/Group Conformity. 
Sufficiency of Originality pertains to the generation and the nature of the ideas produced, with regard to a particular problem (Kirton, 1999).  The more highly innovative tend to generate many ideas.  The relationship between quantity and quality almost ensures that one idea will represent an appropriate breakthrough.  In addition, these ideas tend to be more divorced from the current paradigm, thus requiring a longer implementation cycle.  Conversely, the more highly adaptive produce fewer ideas, more closely in accord with the prevailing paradigm.  These ideas are more developmental in nature and typically require a shorter implementation period. 

Efficiency is associated with the type of structure preferred (Kirton, 1999).  Innovator efficiency consists of redefining the task at hand, eschewing a detailed and orderly approach.  The more highly innovative tend to challenge the status quo.  Among its characteristics at the adaptive pole are precision, reliability, thoroughness, and attention to detail.   With regard to process and method, the more highly adaptive tend to be more developmental.  “The more adaptive will define more carefully, search more methodically for information, and arrange it in more orderly ways” (Kirton, 1999, p. 45).  This scale is reverse scored, the more efficient one is, the lower the score.  The less efficient one is, the higher the score.
Rule/Group Conformity pertains to the social cohesion of the group, and how much of the structure needs to be consensually agreed (Kirton, 1999).  The more innovative tend to develop their ideas at the expense of structural rules and group cohesion.  The more adaptive prefer conforming to the existing rules and achieving consensus within the group.  

Contrary to the approach adopted by Kirton, Kouzes and Posner place their inventory within the realm of behavior rather than cognitive style.  They also clearly differentiate their measure from one of IQ, personality or general management skills (Kouzes and Posner, 1997).  Kouzes and Posner developed the LPI based on an examination of “best practice”. Believing that leadership is learnable and accessible to everyone, Kouzes and Posner (1987) investigated how ordinary people accomplish extra-ordinary things in organizations.  They interviewed approximately 1,100 people asking them to relate their personal  “best leadership experience”.   From the data they collected, five main groups of actions or behaviors emerged from which they developed the Leadership Practices Inventory (LPI). 

The LPI consists of a 30-item paper & pencil questionnaire in both self and observer format and can be used for 360 degree feedback when the respondent asks others to complete it.  The stem question asks to what extent do you engage in the following actions and behaviors?  The scale for each question ranges from 1-5, from rarely or very seldom, to very frequently or almost always, with 5 representing most use of a particular leadership behavior.  The nature of the stem equates more or higher to better, making the LPI a measure of level rather than cognitive style. 
The LPI consists of five subscales each consisting of two strategies.  The practices (and strategies associated with the practice) are: challenging the process, (searching for opportunities, and experimenting and taking risks); inspiring a shared vision, (envisioning the future and enlisting others); enabling others to act, (fostering collaboration and strengthening others); modeling the way, (setting the example and planning small wins); and encouraging the heart (recognizing contributions and celebrating accomplishments). 

By Challenging the Process (Kouzes & Posner, 1987) leaders aim to change the status quo in radical or revolutionary ways.  They search for opportunities for ways to create something new and to “go where no one has gone before”.  Leaders experiment with new approaches to old problems, taking risks and using failures and mistakes as tools for learning, development and growth of both themselves and others.  Leaders compulsively toy with ideas. 

A vision is a unique and ideal image of a desired future state (Kouzes and Posner, 1987).  Leaders articulate this image, inviting others to participate in the possibilities.  The values (deep seated pervasive standards) and dreams of others are engaged when leaders are successful in inspiring a shared vision.  Enlisting others entails finding common ground among group members in order to move forward.
Building on Challenging the Process and Inspiring a Shared Vision, Enabling Others to Act (Kouzes & Posner, 1987) involves ensuring that people have the skills and knowledge they need to feel empowered enough to take initiative.  The orientation is towards giving power away in order to strengthen others.  Allowing others discretion and autonomy in performing their jobs is also intrinsic to strengthening others.  Fostering collaboration involves the social support – the quality of interpersonal relationships.  Collaboration and understanding group dynamics lead to deepening trust and greater commitment to the organization. 

Leaders Model the Way (Kouzes & Posner, 1987) by behaving in ways that are consistent with their stated values and visions.  They set the example by practicing the values they espouse.  They position incremental change to fit into accepted belief.  They plan small wins that build commitment by achieving consistent progress.   Small wins break down the big challenges and obtain recurring affirmation numerous times, not just once. 

And finally, Encouraging the Heart (Kouzes & Posner, 1987) focuses on key values, the personal involvement of the leader, and visible and public recognition.  Praise and coaching are significant forms of recognition, both enhancing intrinsic motivation.  Public ceremonies and rituals crystallize personal commitment and help bond people together.  Thus, it is tied to values, risk-taking, vision, and fostering collaboration. 
By integrating these five practices, leaders are able to create a climate in which organizational members willingly accept the challenge of change, and the nature of the leader/follower dynamic is strengthened.  As a measure of level of behavior (“to what extent do you engage in the following...”), one would expect the LPI to be orthogonal to the KAI, a pure measure of style.  Prior research found this to be true  (Kirton, 1999; Wunderley,1996).  Wunderley looked at the relationship between the KAI and LPI and concluded that “with regard to the relationship between the five leadership factors and the KAI innovative factors, there are no significant correlations.  In particular, a positive correlation between Challenging the Process and both Rules and Originality might have been expected, but was not observed” (1996: 29).  A preliminary exploration of the language used to describe Challenging the Process and Inspiring a Shared Vision compared to the language describing the innovative end of the KAI spectrum led the authors to question these findings.
 Let us now turn our attention to the study itself.  


A random sample of files for individuals who had responded to both measures was selected. 
KAI and LPI item and scale information were collected and entered into an Excel spreadsheet. 
Data was checked for accuracy and 6 individuals with incomplete information were deleted (note: 7 individuals without KAI item data were included).  The data set was then analyzed using SPSS. 
Inter-item reliability and multivariate correlation calculations (???) were performed for each measure. 


Table 2 presents the descriptive statistics and alpha coefficients for both measures.  The KAI is reliable for the total score and across all sub-scales, with alphas higher than .7.  Published norms for the LPI (Kouzes and Posner, 1997, 1988) confirm the reliability of the LPI.  The authors’ data, however, contrast significantly to the norms published by Kouzes and Posner.  According to Table 2, Inspiring a Shared Vision and Encouraging the Heart have high internal reliabilities while Challenging the Process has an acceptable reliability.  Enabling Others to Act and Modeling the Way with alphas at .65 and .62 respectively, do not.  These relatively low alphas raise some concerns as to what these two dimensions are really measuring.

Table 3 presents the correlations of these two measures.  Both Challenging the Process and Inspiring a Shared Vision correlate positively to the KAI across all three subscales as well as the total KAI score.  Intuitively, this makes sense and is supported by looking at the language used to describe these two dimensions.  Radical, new, revolutionary change from the status quo, risk-taking, change is an opportunity to innovate, experiment with new approaches, are words associated with Challenging the Process (Kouzes and Posner, 1997).  Creating the future together, and reaching consensus about a common goal pertain to Inspiring a Shared Vision.  Enabling Others to Act, Modeling the Way, and Encouraging the Heart are orthogonal to the KAI and clearly measure something other than cognitive style.


This study examined the relationship between cognitive style and leadership behaviors as measured by the KAI and LPI.  It produced fundamentally different results from previous studies. The results provide preliminary evidence of the LPI’s internal structure and raise the concern that the LPI measures practices that are confounded with cognitive style.    Three dimensions, Challenging the Process, Inspiring a Shared Vision, and Encouraging the Heart accurately measure these leadership behaviors.  However, the positive correlation between the LPI dimensions, Challenging the Process and Inspiring a Shared Vision, and the KAI, suggests that cognitive style is related to the leadership behaviors described by these dimensions.  In addition, this positive correlation provides some evidence to support the notion that those who are more adaptive by nature would not score well on these two dimensions.
Inspiring a Shared Vision and Challenging the Process do correlate to a “pure” measure of cognitive style, suggesting a bias toward a more innovative cognitive style.  This imbalance reinforces the perception that those who prefer more adaptive novelty cannot make as great a contribution as those who prefer more innovative novelty.  This is problematic if we really believe that creative leadership requires the full spectrum of change and that leadership really is everyone’s business. 
            Some implications may be drawn from this research that should be considered when using the LPI for its stated purpose:

  1. Consider the full spectrum of cognitive style when interpreting the LPI.
  2. Use strengths as a starting point when interpreting results.
  1. People who do things differently are likely to be more highly regarded than those who do them better; avoid imposing value judgments on scores.
  2. LPI is a measure only of how often certain behaviors are practiced or observed.  It is not a measure of overall success.


The authors used the KAI as a “pure” measure of style, unrelated to capacity.  Therefore, these results require further inquiry and research.  The authors used the old version of the LPI (cite) because the data was more accessible.  They used only one measure of style; as this study is not broad-based, its conclusions are tentative.  Some results contradicted published LPI information.  The sample did not include some professions, which might have an impact on the outcome.  And finally, the KAI sample mean used is slightly biased on the innovator (those who do things differently) side.    Future studies should include a broader base of professions and additional measures of style.   Future studies should also utilize the 1997 version of the LPI (Kouzes & Posner, 1997), to obtain greater significance in the findings.  Despite these limitations, the authors believe that creative leadership requires consideration of the full spectrum of creativity and a mastery of both leadership and management skills. 

Table 1: Common Distinctions Between Leaders and Managers

From the Leadership Literature






Bennis, 1994

Do the right thing.

Do things right.

Kotter, 1996

Define what the future should look like; align people with that vision and inspire them to make it happen despite obstacles.

Identify how to achieve the future by planning, budgeting, organizing, staffing, controlling, and problem solving.

Kouzes & Posner, 1995

Take initiative; confront the established order.

Maintain order, organization and control.

Palus & Horth, 2002

Tell stories that begin with one’s identity and grow outward into stories of group identity and vision.

Tell stories that are passive and look backward.

Buckingham & Coffman, 1999

Great leaders look outward.

Great managers look inward.

Connors, 2000

Set corporate context.

Plan & execute.

Egan, 1988

Establish climate and communicates to others in a way that results in excitement about visions.

Provide resources and support, monitors progress, and rewards performance.

Katzenbach, 1995

Focus on keeping people and performance in balance.

See structure as the alignment method for channeling behavior and skills.

Shipper, 1995


Controls and manages details.

Wilson & O’Hare, 1991

Adapt strategies to fit organizational thrusts.

Organize and plans well.

Bardwick, 1996

Create a winning strategy, frame mission and values.

Gently tweak the existing system in order to improve.

Bethel, 1995

Inspire risk-taking.

Avoid risks.


Table 2: Means, Standard Deviations, and Reliability Indices for the Kirton Adaption-Innovation Inventory and the Leadership Practices Inventory





# of







KAI Total






Sufficiency of Originality












Rule/Group Conformity






Challenging the Process






Inspiring a Shared Vision






Enabling Others to Act






Modeling the Way






Encouraging the Heart







Table 3: Correlation of Leadership Practices Inventory with Kirton Adaption-Innovation Inventory








KAI Total







Sufficiency of Originality














Rule/Group Conformity







* Correlation is significant at the .001 level.



Babij, B.J. (2001).  Through the looking glass: Creativity and leadership juxtaposed.  Unpublished masters thesis.  State University at Buffalo, Center for Studies in Creativity, at Buffalo State College.

Bass, B. M. (1981).  Stodgill’s handbook of leadership: A survey of theory and research (revised and expanded edition).  NY: Free Press.

Bass, B.M. (1985).  Leadership and performance beyond expectations.  New York: Free Press.

Bass, B.M. (1990).  Bass and Stogdill’s handbook of leadership (3rd ed.).  New York: Free Press.

Bennis, W. & Nanus, B. (1985).  Leaders: The strategies for taking charge.  New York: Harper & Row, Publishers, Inc.

Bennis, W. (1989).  On becoming a leader.  Reading, MA: Perseus Books.

Burke, W. W. & Litwin, G. H. (1992). A causal model for Organizational performance and change.  Journal of Management, 18, 523-545.

Burns, J. M. (1978). Leadership.  New York: Harper & Row.

CPS-B (1997). From slide 12 of presentation.

Dewan, (1982).  From slide 12 of presentation.

Gardner, J.W. (1990).  On leadership.  NY: The Free Press.

Greenleaf, R. K. (1991).  Servant leadership: A journey into the nature of legitmate power and greatness.  New York: Paulist Press.

Gryskiewicz. S. (1987).  From slide 12 of presentation.

Heifetz, R. A. (1998).  Leadership without easy answers (9th printing) Cambridge, MA: Harvard University Press.

Hock, D. (1999).  Birth of the chaordic age.  San Francisco, CA: Berrett-Koehler Publishers, Inc.

Holland, P.A., Bowskill, I., & Bailey, A, (1991).  Adaptors and innovators: Selection versus induction. Psychological Reports, 68, 1283-1290.

Kirton, M. J. (1976). Adaptors and innovators: A description and measure. Journal of Applied Psychology, (6) 15, pp.622-629.

Kirton, M. J. (1977, 1985, 1991). KAI Response Sheet, Form ref. EK9I R. Berkhamstead, Herts, United Kingdom: Occupational Research Centre.

Kirton, M. J. (1984). Adaptors and innovators – Why new initiatives get blocked.  Long Range Planning, 17 (2), 137-143.

Kirton, M. J. (1989). Adaptors and innovators: Styles of creativity and problem solving.  London: Routledge. 

Kirton, M. J. (1993).  KAI pre-course reading: Description summary.  M.J. Kirton.

Kirton, M. J. (1999).  KAI: Kirton Adaption-Innovation Inventory manual, 3rd edition. Berkhamstead, Herts, United Kingdom: Occupational Research Centre.

Kirton, M. J. (2000). Adaptors and innovators: Styles of creativity and problem solving (3rd edition), London: Routledge. 
Kotter, J. P. (1996).  Leading change.  Boston, MA: Harvard Business School Press.

Kouzes, J. M. & Posner, B. Z. (xxxx).  Leadership Practices Inventory – Self.  San Francisco: Jossey-Bass Pfeiffer.

Kouzes, J. M. & Posner, B. Z. (1987).  The leadership challenge: How to get extraordinary things done in organizations.  San Francisco: Jossey-Bass.

Kouzes, J. M. & Posner, B. Z. (1997).  The leadership challenge: How to keep getting extraordinary things done in organizations.  San Francisco, CA: Jossey-Bass Publishers.

Lauer, K.J. & Babij, B.J. (2002). Style and leadership: Emerging research. Paper presented at the Creative Problem Solving Institutes, Buffalo, NY.

Pfeiffer, J. W. (Ed.) (1991).  Theories and models in applied behavioral science: Management/Leadership.  Vol. 3. San Diego: Pfeiffer & Company.

Posner, B. Z.  &Kouzes, J. M. (1988). Development  and validation of the Leadership Practices Inventory.  Educational and Psychological Measurement, 48, (2), 483-496.

Senge, P. M. (1990).  The fifth discipline: The art and practice of the learning organization.  New York: Currency Doubleday.

Shipper, F. (1995).  A study of the psychometric properties of the managerial skill scales of the survey of management practices.  Educational and Psychological Measurement, 55(3), 468-479.  Newbury Park, CA: Sage Publications, Inc.

Talbot, R. J. (1997).  Taking style on board: How to get used to the idea of creative adaptors and uncreative innovators.  Creativity and Innovation Management, 6, (3), 177-184.

Wheatley, M. (1992).  Leadership and the new science: Learning about organization from an orderly universe.  San Francisco: Berrett-Koehler.

Wheatley, M. (1999).  Leadership and the new science: Discovering order in a chaotic world. (2nd ed.).  San Francisco: Berrett-Koehler.

Wunderley, L.J. (1996). The relationships among optimism, pessimism, vision, and effective leadership practices. Unpublished master’s thesis, University of Cincinnati, Cincinnati, Ohio