The purpose of this paper is to examine the reliability and validity of the Multifactor Leadership Questionnaire (MLQ). While the MLQ is widely used to measure a variety of aspects of transformational and transaction leadership, it is also criticized for not being a valid and reliable measure of leadership traits and qualities. In this paper, an overview of the MLQ is provided, as well as a discussion of the reliability and validity issues of the MLQ that have been addressed in recent academic studies. In addition, a discussion is provided regarding the MLQ is effective and accurate as a quantitative research tool as compared to qualitative research methods in analyzing the impact that leadership styles can have on successfully implementing organization change. The information and analysis that is provided in this paper will demonstrate that the MLQ is not a reliable nor effective tool for analyzing leadership styles or their impact on organizational change.
OVERVIEW OF THE MULTIFACTOR LEADERSHIP QUESTIONNAIRE
The MLQ is a 45 item questionnaire that is designed to identify nine different leadership factors and three leadership outcomes. The nine leadership factors include five transformational leadership factors, three transactional leadership factors, and one non-leadership factor. The non-leadership component is laissez-faire leadership, which is viewed as not being a leadership style because little actual influence is exerted on followers. Based on the scoring of the 45 items, the questionnaire is supposed to provide a means to determine if individuals have a high, average, or low level of behavior and impact with regards to each of the nine leadership factors. This information can then be used by organizations to determine the best candidates for leadership positions, as well as the types of leadership outcomes that can be expected from existing leaders.
RELIABILITY AND VALIDITY
The validity of the MLQ rests with the idea that this questionnaire is able to provide information about nine distinct leadership factors. It is this quality of the measures of the MLQ that make it an important means to determine leadership behaviors and outcomes. However, the research that will be reviewed in this section suggests that the distinctions between the leadership factors included in the MLQ are very low, which means that the validity of the questionnaire is low. In addition, the low level of validity of the MLQ makes obtaining reliable results that can be used to make real-world business decisions problematic.
Muenjohn and Armstrong explain that several studies have found that the nine different leadership factors in the MLQ are actually highly correlated with each other. This high level of correlation exists not only across transformational and transactional leadership, but also within each of the individual leadership factors that are part of the MLQ. The problem with high levels of correlation between the major leadership traits and the specific leadership factors in the MLQ is that it is difficult to determine not only whether a person is more of a transformational or transaction leaders, but also whether a person will exhibit individual characteristics such as idealized influence, inspirational motivation or individualized consideration.
These findings mean that rather than the leadership factors that are measured in the MLQ being distinct from one another, they are instead related to each other. This makes being able to identify specific leadership traits of individuals and predictions of future leadership outcomes highly unreliable. For the purposes of attempting to make decisions within organizations about the people to place into leadership positions, the MLQ is not likely to provide reliable results.
Rowold and Heinitz conducted an examination of the MLQ to determine if the leadership qualities measured in the questionnaire were convergent or divergent with each other. The idea of the research being that if the leadership qualities as measured in the MLQ were convergent, then creating distinctions between them to make determinations about whether a person is a strong transformational or transactions leader is difficult. The researchers found that the leadership qualities that are part of the MLQ are convergent on each other. The factors related to transformational leadership were highly correlated with the transactional leadership factors. It was noted by the researchers that some of the individual leadership factors in the questionnaire did diverge from each other. However, the overall results of the research indicated that the MLQ's leadership factors were not highly distinctive from each other.
Heintz, Liepmann and Felfe conducted research to determine if the high level of correlation among the leadership factors could be reduced by removing some of the items could reduce the MLQ. The researchers conducted statistical analyses on results from the MLQ and found that three of the leadership factors could be removed without changing the results of the questionnaire. However, the researchers also found that even by removing three of the nine leadership factors that are part of the MLQ, the overall variance in findings is only reduced by 14%. The conclusion that the researchers put forth from these findings is that even with some modifications, the MLQ is not highly valid and is not useful for making reliable decisions about leadership personnel within organizations.
The conclusion that must be drawn from recent research and analysis of the MLQ is that it lacks validity with regards to providing information about distinct behaviors related to transformational and transactional leadership behaviors. The high degree of correlation between each of the components in the MLQ makes any analysis of questionnaire results difficult because trying to separate specific leadership factors and outcomes is not appropriate. Even more, the findings from the recent research suggest that using the MLQ to determine whether a person is likely to be a transformational or transactional leader regardless of individual leadership traits is also inappropriate. The high levels of correlation are exist both the individual components of the questionnaire, as well as between the transformational and transactional leadership qualities in general.
The overall lack of validity further suggests that being able to use to MLQ to achieve reliability across a large number of potential leadership candidates within an organization is not possible. Overall a large selection of potential leadership candidates, the findings of the MLQ may vary greatly in terms of suggesting that some candidates are both strong transformational and transactional leaders when that is not accurate. Based on all of these issues and the findings, the MLQ does not seem to be valid or reliable in making decisions within organizations.
EFFECTIVENESS AND ACCURACY OF THE MULTIFACTOR LEADERSHIP QUESTIONNAIRE
The issue of whether the MLQ should be used to make actual business decisions has been touched upon in relation to the concerns about its lack of validity and reliability. However, a larger issue that deserves attention is how effective using the MLQ and its quantitative methodology as compared to using a qualitative approach to examine leadership styles and the impact that they have on successful organizational change. The research and analysis that exist on this subject suggest that attempting to use quantitative methods such as the MLQ can be effective compared to qualitative means to examine leadership styles, but the specific tools that are used must take into account the variables that impact leadership behaviors and outcomes.
One of the arguments that has been put forth about the lack of effectiveness and accuracy in using the MLQ is that the specific leadership traits and characteristics contained within the questionnaire do not accurately reflect leadership behaviors. This argument is not entirely based on the larger issue of whether a qualitative method of examining leadership styles is better or worse than a quantitative method. Instead, it is simply stated that the specific factors that are part of the MLQ may not accurately reflect elements and dimensions of leadership behaviors. In order to make the MLQ more accurate, a reassessment of the factors that are identified as being part of transformational and transactional leadership would need to be performed. This would also require a change in the composition of the MLQ.
This concern that connections between leadership factors and actual leadership behaviors and performance are inaccurate has been expressed in a variety of studies. The arguments that have been made is that simply trying to link specific traits to a strictly transformational or transactional leadership behavior and outcome is not accurate, particularly in real-world settings in which a variety of other variables can intervene. First, in order for the MLQ to be identified as highly accurate, it would need to be used and evaluated across a large number of organizations in a variety of different industries and disciplines. This type of testing is the only way in which it would be possible to determine whether the nine specific leadership factors that are part of the MLQ are strong enough regardless of industry or company to allow for other situational variables to be ignored or viewed as insignificant.
In addition, cultural variables in a specific country have to be considered with regards to actual leadership behaviors and outcomes. What is viewed as defining transformational or transactional leadership in one country may not be the same in another country. Cultural differences may impact these definitions and the way in which these leadership styles are perceived by followers. Even Bass notes that transformational and transactional leadership styles have an important connection to ethical and moral actions and behaviors. The issue would seem to be how ethics and morals are defined, and how people view certain behaviors with regards to ethics and morality. Talking about ethics and morals is much different than being able to define them, and then explain how they impact leadership factors and behaviors, particularly in the larger context of attempting to achieve successful organizational change.
While these issues suggest that the MLQ is indeed not an effective means of analyzing the impact of leadership styles on successfully implementing organizational change, they do not fully suggest whether quantitative or qualitative evaluation methods are better or more appropriate. With regards to whether a statistical means to leadership evaluation and prediction of outcomes is appropriate as opposed to a qualitative method, research indicates that quantitative methods can be effective. While the MLQ has been shown to lack a high degree of validity and reliability does not mean that other quantitative measures of leadership behaviors and outcomes are ineffective. Instead, any quantitative measure of leadership behaviors and outcomes that has a high level of accuracy and effectiveness has to take into account not just individual attitudes toward leadership, but also the impact that intervening variables such as job satisfaction, interaction with co-workers, and even the type of work that is performed.
Research has also shown that the level of engagement at work, such as engagement with the specific duties that must be performed and the people with whom a person works, impacts leadership behaviors. In addition, job experience is another factor that has been found to significantly impact leadership styles and behaviors. Once again, what is demonstrated is that while a person may believe that certain actions with followers is beneficial or necessary, which is essentially what is measured by the MLQ, these beliefs are likely to change because of the impact of other variables related to the specific situation or organization in question. By ignoring these variables and not asking questions about how such characteristics are likely to impact the leadership styles and behaviors of a person, then any leadership measurement is ineffective and lacks accuracy in determining leadership outcomes such as achieving successful organizational change.
In fact, regional variations of the MLQ have been created and shown to have a high level of accuracy and effectiveness in predicting the actions and behaviors of leaders, as well as the outcomes that might be achieved from people with specific types of transformational and transactional leadership characteristics. What is unique about such regional variations of the MLQ is that they include questions that take into account regional and local considerations for such things as ethical and moral behaviors, as well as the impact that specific worker characteristics may have on the decisions that leaders make and the outward actions and behaviors that are manifested.
The answer to the question of whether quantitative means of leadership evaluation, such as that of the MLQ, is effective and accurate for making leadership decisions and determining the impact that leadership styles can have on making successfully organizational changes is that that quantitative means can be effective and accurate. The issue with the MLQ is less about its use of quantitative means to determine leadership qualities and characteristics and more about the specific leadership characteristics and factors that it seeks to examine. The research that has been reviewed has shown that the MLQ is too narrowly focused on nine specific traits. Even more, the research has shown that the lack of distinction between these traits as they are measured by the MLQ makes it impossible to use this questionnaire with any degree of validity or reliability.
However, quantitative questionnaires have been created that include the additional variables that have been discussed, such as interactions with followers, cultural attitudes toward morals and ethics, and attitudes toward the job and organization. The conclusion that can be drawn is that quantitative measures do not lack effectiveness or accuracy as a whole. Instead, each quantitative measure of leadership behaviors and outcomes must be evaluated on its own merits to determine whether it provides reliable and valid information. Simply choosing an existing quantitative questionnaire, regardless of how much attention it has received within the professional or academic community, is not enough to receive effective and accurate information about the outcomes that can be expected in terms of implementing successful organizational change.
Finally, all of this also suggests that choosing a quantitative questionnaire over a qualitative evaluation method is not an indication by itself of effectiveness or accuracy. What is important in any leadership evaluation method that is used is the specific measurements that are used and the validity and reliability of the factors that are included in the evaluation method. The MLQ has been shown to not take into account the range of variables that impact a leader in a real-world setting for it to be effective or accurate for decision-making purposes.
The purpose of this paper has been to examine issues of validity and reliability related to the Multifactor Leadership Questionnaire (MLQ). The research that has been reviewed and the accompanying analysis has shown that the MLQ is not a validity and reliable measure of leadership behaviors with followers. The questionnaire does not provide sufficient distinctions between the leadership factors that are measured. Because of the high correlation between the individual leadership factors, a lack of reliability is present in making determinations about the transformational or transactional leadership styles that are likely to be used by an individual.
The lack of validity and reliability of the MLQ should not be taken to mean that all quantitative measures of leadership styles, or their use to determine the impact of leadership styles on successful organizational change. Instead, whether a qualitative or quantitative measure is used, the important issue is including variables that accurately reflect the conditions in which a leader is to operate. Rather than focusing solely on the beliefs and attitudes of the leaders, variables about the attitudes and actions of followers and the larger internal environment of an organization should be included. Cultural differences in attitudes and opinions about moral and ethical behaviors must also be considered. By taking into account a more complete range of variables that are likely to impact the decisions that leaders make and the behaviors they are likely to use to influence followers, a valid and reliable measure of leadership styles can be created, and in fact, such measures have been created. The MLQ is simple not an valid and reliable measure and its use should be avoided in favor of measures that have been shown to have higher levels of validity.
Arnold, K. A., Turner, N., Barling, J., Kelloway, E. K. & McKee, M. C. Transformational leadership and psychological well-being: The mediating role of meaningful work. Journal of Occupational Health Psychology, 12(3), 193-203.
Babcock-Roberson, M. E. & Strickland, O. J.. The relationship between charismatic leadership, work engagement, and organizational citizenship behaviors. The Journal of Psychology, 144(3), 313-326.
Bass, B. M.. Leadership and Performance Beyond Expectations. New York: Free Press.
Bass, B. M.. From transactional to transformational leadership: Learning to share the vision. Organizational Dynamics (Winter): 19-31.
Bass, B. M. & Steidlmeier, P.. Ethics, character and authentic transformational leadership, Retrieved on September 10, 2009 from .
Brown, F. W. & Reilly, M. D.. The Myers-Briggs type indicator and transformational leadership. Journal of Management Development, 28(10), 916-932.
EssayForum. Psychology Research Method Course. Online. essayforum.com/research/method-section-psychology-methods-29079/
Giri, V. N. & Santra, T.. Effects of job experience, career stage, and hierarchy on leadership style. Singapore Management Review, 32(1), 85-93.
Heinitz, K., Liepmann, D. & Felfe, J.. Examining the factor structure of the MLQ. European Journal of Psychological Assessment, 21(3), 182-190.
Hinking, T. R. & Schriesheim, C. A.. A theoretical and empirical examination of the transactional and non-leadership dimensions of the multifactor leadership questionnaire (MLQ). The Leadership Quarterly, 19(5), 501-513.
Muenjohn, N. & Armstrong, A.. Evaluating the structural validity of the multifactor leadership questionnaire (MLQ), capturing the leadership factors of transformational-transactional leadership. Contemporary Management Research, 4(1), 3-14.
Rowold, J. & Heinitz, K.. Transformational and charismatic leadership: Assessing the convergent, divergent and criterion validity of the MLQ and CKS. The Leadership Quality, 18(2), 121-133.
Schriesheim, C. A., Wu, J. B. & Scandura, T. A.. A meso measure? Examining of the levels of analysis of the multifactor leadership questionnaire (MLQ). The Leadership Quarterly, 21(1), 604-616.
Sparrowe, R. T.. Authentic leadership and the narrative self. The Leadership Quarterly, 16, 419-439.
Sugreen, G. & Schepers, J. M.. The association of selected speex-battery indices with the constructs of the multifactor leadership questionnaire. South African Journal of Industrial Psychology, 32(1), 42-51.
Turner, J. R. & Muller, R.. The project manager's leadership style as a success factor on projects: A literature review. Project Management Journal (June): 49-61.
Author InfoQuestionnaire Researcher
More about Author
Research Studies on Multifactor Leadership Questionnaire Methods.