Higher Education Managers/Executives/Administrators
To determine if the leadership practices of presidents of higher educational
institutions in the economically challenged northeast were different from those of leaders
in business and industry.
All college presidents in New England (N=248) were asked to complete
the LPI-Self (N=142) and distribute five LPI-Observers (N=512). Effectiveness, as per
Kouzes and Posner, was measured by eight items (alpha = .93). Internal reliability for the
LPI-Self ranged between .71 to .84, for the LPI-Observer from .85 to .93, and combined Self
and Observer reliabilities ranged between .84 to .92. Principal factoring extraction
technique combined with the equamax rotation achieved comparable factor structure with
Posner and Kouzes (1993).
LPI-Self scores for academic leaders were significantly higher than
business leaders; as were the LPI-Observer scores for Challenging, Inspiring, and
Modeling. Enabling was ranked highest by college presidents (as was true for business
leaders), followed by Inspiring (5th), Challenging (2nd), Modeling (3rd), and Encouraging
(4th). The rank order on LPI-Observer scores for college presidents was the same. The
rank ordering was identical between the LPI-Self and LPI-Observer when examining each
sample separately. The most substantive difference was on Inspiring a Shared Vision.
Using LPI-Observer scores, regression analysis accounted for 55 percent of the
explained variance on effectiveness scores. Enabling accounted for the most significance
(also true for business leaders); Modeling accounting for more variance with academic
leaders than with business leaders. For institutions "undergoing significant change" LPIObserver
scores accounted for 59 percent of the variance in effectiveness, and for "no
change" institutions R2 = .53. For both the change and no-change institutions Modeling
and Enabling significantly contributed to predictions of leadership effectiveness.
Institutional ownership (public vs. private) did not affect the leadership practices profiles
of academic leaders (Self or Observers). Neither the mean or rank order of LPI-Self or LPIObserver
scores for female college presidents (23.4%) were significantly different from their
male counterparts. The effectiveness measure did not vary by respondent gender,
institutional type, or degree of environmental change.
The author discusses the possibility of two meta-practices: Envisioning
(Challenging, Inspiring, and Modeling) and Implementation (Enabling and Encouraging).