Control and randomization in leadership research
The inclusion of a control group and randomization of the subjects who are to be studied are critical steps in all research involving human subjects and this holds equally true for leadership research (Kaplan and Kaiser, 2003, p 24). Without the inclusion of an appropriate control group the interpretation of the data becomes very difficult if not impossible as the control group allows the comparison of the outcome measured to be assessed in the absence of the variable that is being studied (Kaplan and Kaiser, 2003, p 24). To examine the consequences of trying to analyze the effect of a new leadership technique without the inclusion of a control group consider the following fictional scenario. The introduction of a new leadership strategy, where staff were paid a bonus for months in which no sick leave is taken in a large for profit organization may have led to a significant decrease in staff absenteeism, the management staff are deciding on whether to implement permanently this apparently effective but costly strategy. By lucky coincidence for the management team it was discovered that at the same time the bonus strategy was being studied a staff implemented competition had been set as to which team would reach the highest monthly sales and this was seen by staff to be more important an incentive to stay at work than the bonus paid by the company. Control groups are critical for the accurate interpretation of any leadership study.
The incorporation of a control group into a study may prove valueless if the control group does not represent appropriately the study group (Parry and Meindl, 2002, p 129). The most commonly used approach to ensure the control group does represent the study group is by randomization (Cassell and Symon, 2004, p 195). Leadership research subjects are randomly allocated to either the control group or the study group, by following this method both groups should be representative of each other and be free from subject selection bias (Lord et al, 1986, p 376).
Internal validity in leadership research
In common with other fields of behavioral studies leadership research methods must have internal validity. Internal validity ensures that the study is capable of detecting a significant difference between the control and study groups if indeed there is one present beyond the possibility of chance alone (Judge et al, 2002, p 778). For example for a study to possess internal validity it must be able to demonstrate a causal relationship between the variable being tested and the outcome being measured. There are three factors that are important in establishing internal validity of a leadership research protocol. Firstly, the cause must precede the effect (Kaplin and Kaiser, 2003, p 23), in the case of leadership research; the variable being studied must precede the change in the outcome being measured. Secondly, the cause and effect must be related (Kaplin and Kaiser, 2003, p 23), in the case of leadership research the variable being studied must co vary with the outcome being measured. Thirdly, there must be no other plausible explanation possible to explain the co variation between the cause and effect (Kaplin and Kaiser, 2003, p 23) in the case of leadership research the co variation of the variable being studied and the outcome being measured must not be able to be explained by another unrelated variable.
External validity in leadership research
For a study to hold external validity the interpretation of the data must be consistent with other studies of the same variable and outcomes being measured in different experimental settings, with different procedures and with different subjects (Peterson et al, 2003, p 807). In the absence of external validity a study may fail to stand up to the rigors of peer review and academic evaluation. Another implication of failing to show external validity is that the study may be showing the result of unexpected biases that render the interpretation of the results vulnerable to errors. External viability of a study is one of the strongest supporting evidence that the theory developed from the interpretation of the data obtained is sound.
There are three factors that are important to consider when examining a leadership research protocol for external validity. Firstly, the sample size must be large enough to avoid a local bias (Lord et al, 1986, p 408). A local bias in this instance means a small subject size is particularly vulnerable to reflect the potentially specific nature of that study and control group. A larger subject number will lower the risk of a study failing to show external viability. Secondly, for the interpretation of the analyzed data to show external viability the same outcomes must be seen when the same variable is studied in a different experimental setting. Thirdly, a study to be considered to have external viability it must produce the same outcomes when studied again in different subjects.
Role for leadership research in both profit and non-profit realms
Leadership research plays a valuable role in all organizations regardless of whether they are for profit or are non-profit. Any organization that wishes to remain flexible in the face of the constantly changing business environment should have an ongoing policy for leadership research. For profit organizations may benefit from developing their own leadership research studies to determine which specific leadership styles provide the most productivity from their staff. Non-profit organizations have similar productivity goals as for profit organizations so the same values hold true in that realm however another value of research surfaces. Non-profit organizations may benefit form implementing leadership studies to determine which strategies may best influence the community they serve so as to create a more successful outreach of their services.
While we may expect that research is the business of academic institutions rather than for private organizations, the ability to accurately assess current as well as the possible implementation of new leadership initiatives directly in the specific setting of the organization is extremely valuable. The organizations who manage to conduct their own leadership research compared to those who rely on waiting for academic institutions to publish such work and then having to modify any innovative strategies to their organizations will be years ahead in terms of creative leadership strategies.
Cassell, C. and Symon, G. (2004). Essential guide to qualitative methods in organizational research. London, Sage Publications.
Judge, T.A., Bono, J.E., Ilies, R. and Gerhardt, M.W. (2002). Personality and leadership: a qualitative and quantitative review. Journal of Applied Psychology, 87, 765- 780.
Kaplan, R.E. and Kaiser, R.B. (2003), Developing versatile leadership. MIT Sloan Management Review, 44, 19-26.
Lord, R.G., DeVader, C.L. and Alliger, G. (1986). A meta-analysis of the relation between personality traits and leader perceptions. Journal of Applied Psychology, 71, 402-410.
Parry, K.W. and Meindl, J.R. (2002). Grounding Leadership Theory and Research: Issues, Perspectives and Methods. Leadership Horizon Series. Connecticut, Information Age Publishing.
Peterson, R.S., Smith, D.B., Martorana, P.V. and Owens, P.D. (2003). The impact of chief executive officer personality on top management team dynamics. Journal of Applied Psychology, 88, 795-808.