Ask an Expert January 2018

Ask an Expert

Q: You present multivariate statistical analyses in the most recent research published in the 6th edition of The Leadership Challenge which are most appreciated and impressive. But I have a statistical question. Using regression analysis, with employee engagement as the dependent variable, the data shows that how a leader behaves is substantially significant in accounting for levels of employee engagement, while individual and organizational characteristics are totally insignificant in explaining why direct reports are or are not engaged. What else accounts for variation in employee engagement levels? Is there anything more important than leadership?

A: Participants in workshops and academic programs often ask a similar question: “It is impressive that leadership explains nearly 40% of the engagement levels of direct reports, but what about the remaining 60%? What’s that comprised of?”

Statisticians refer to the unexplained variability of the data as “noise.” Since social science is not perfect, it can’t ever explain perfectly, 100% of the time, why someone or something happens the way that it does. The unexplained portion is referred to as statistical noise or randomness, and it can be attributed to a host of factors. These could be measurement issues, incorrect data entry or understanding, or entirely random variables that are outside of the topic being investigated—e.g., being interrupted when completing the instrument, feeling ill, having just received a birthday greeting, and the like.

What we do know is that there isn’t any other variable other than leadership that explains such a significant amount of variance in studies involving employee engagement—not compensation, working conditions, technology, location, industry, function, nor any of a host of demographic variables. In addition, depending upon how engagement is measured, or even how leadership is conceptualized, can change the equation and results. For instance, studies by the Gallup organization have claimed that over 70 percent of the variance in how they measure engagement can be explained by how they measure leadership.

The key point is that the data clearly shows that how leaders behave impacts, in a significant and substantial way, how people feel about their workplace, and their levels of motivation, commitment, and productivity. We also know that leadership, or how one behaves as a leader, is a factor comprised of actionable behaviors that a person can directly do something about, if he/she chooses to do so.

Finally, while it is not a perfect world or precise science, leadership is the lever with the highest probability of changing the level of workplace engagement. Leadership matters. And our mission with The Leadership Challenge is to increase the quality and quantity of leaders for the world.


Barry Posner,
Ph.D., is the Accolti Endowed Professor of Leadership at the Leavey School of Business, Santa Clara University, where he served as Dean for 12 years. Together with Jim Kouzes, he is author of over 30 books and workbooks on leadership and leadership development, including the fully-revised and updated sixth edition of the international bestseller, The Leadership Challenge, and Learning Leadership, selected by Strategy+Business as one of the 2016 Best Business Books of Year.

RELATED RESOURCES

Articles & Stories