Are Ratings of Transformational Leadership Comparable? Measurement Invariance of the Leadership Practices Inventory Across Rating Sources

OBJECTIVE
The purpose of this study was to use confirmatory factor analysis techniques to examine the construct validity of one multisource measure of transformational leadership – the Leadership Practices Inventory for Individual Contributors.

METHODOLOGY
The self-rating sample consisted of graduate students pursing an MBA at a large western university (N=416). Supervisor (N=323), peer (N=687) and subordinate (N=156) raters were completed by individuals selected in each of these positions by the target manager. Internal reliabilities for this sample ranged from .73 to .86 for self-ratings, .84 to .90 for supervisor ratings, .81 to .89 for peer ratings, and .79 to .90 for subordinate ratings.

KEY FINDINGS
For all rater groups, the five-factor model fit best, followed by the five-factor model with a higher-order factor, and lastly, the one-factor model. Constraining the factor loadings to be invariant across rating groups “did not result in a significant loss of fit, suggesting that different raters (self, supervisor, peer, and subordinate) held similar conceptualizations of the transformational leadership constructs measured by the LPI-IC......Results from this study lend support to the assertion that conceptualizations of transformational leadership are similar for individuals at different levels in the organization” (p. 27).

“Compared to other leadership constructs, transformational leadership may be recognized by individuals at all levels of the organization because of its universality – the fact that regardless of an individual’s station or culture, conceptualizations of transformational leadership will be similar” (p. 27-28).

“This study is unique in that it establishes the factorial invariance of the LPI-IC across rater sources” (p. 29)....suggesting that “while dimensions of the LPI-IC may be highly correlated, they may each be measuring an important and distinct facet of transformational leadership” (p. 30).

The author concludes: “The finding of factorial invariance supports the use of the LPI-IC as a multisource feedback measure in business and educational settings and suggests that raters from difference sources do not necessarily view leadership performance differently. These findings should be good news to consultants, educators, and others that frequently use ratings from the LPI and LPI-IC from different organizational constituents in the service of leadership development, or even pay and promotion decisions (Conger, 1999; Kouzes & Posner, 1995, 2002; Posner & Kouzes, 1993). The finding of factorial invariance establishes a necessary prerequisite for comparing scores across raters on different LPI-IC dimensions” (p. 32).