I'm sorry but I cannot be at the LMX agreement session at AOM, but I would like to suggest a number of reasons that some researchers have not found strong agreement between leadership motivated xcellence (LMX) ratings of the parties to the agreement. This agreement is to support and defend each other on projects and the big three of mutual respect for competence, trust in motives, and commitment to the welfare of both the corporation and the parties. These characteristics can be assessed reliably and validly using the proper protocol:
1. Business associates involved most approve the use of this information and personally vouch for its accuracy. (Best accomplished in individual interviews in a longitudinal study.)
2. Properly worded questions focusing on the partner's side of the agreement must be employed.
3. Only reputational reports can be used and not identity reports (Hogan, 2007).
When the two parties disagree significantly, the disagreement must be resolved before acceptance.
5. The agreement must be tested by both parties (Graen, Hui, & Taylor, 2006).
6. The assessors must make clear that the concept under investigation is an agreement that will be checked with the other party.
This protocol when properly followed produces acceptable agreement (Graen, Hui, & Taylor, 2006). However, when a requirement is not met, this can contribute to lack of agreement. When no personal approval is obtained, socially desirable responses can be expected. When different questions are asked of each party, little agreement should be expected. When identity questions are asked of a manager and reputational questions are asked of a direct report, a mismatch occurs. A common mistake is to mix identity and reputation. Reputation is a description by an observer of another person's behavior. Identity is a self-report of a person's self-concept. Reputation is used to predict future behavior and identity is used to explain one's own behavior. In addition, when parties disagree and the discrepancy is not resolved, error is added. Until a firm agreement has been tested by both parties, the strength of the alliance is unknown. Finally, the concept to be assessed is: Do the two parties maintain an LMX agreement or not? When they do, agreement is expected, but when they do not disagreement is expected. This is what is found in our research. This issue is discussed in more detail in my chapter in Mike Rumsey (Ed.) Oxford Handbook of Leadership, (2011).
George Graen
/jag