Content area
Abstract
The purpose of this study was to develop an instrument that operationalizes the Social Change Model of Leadership Development (A. Astin, et al., 1996). The production of this instrument helps to fill a void in available and affordable measures of leadership development with an application for college students.
R. F. DeVellis' (1991) steps for scale development served as the outline for the research process. The phenomenon of socially responsible leadership was comprised of eight constructs identified by the Model: consciousness of self, congruence, commitment, common purpose, collaboration, controversy with civility, citizenship, and change. Items for the instrument were generated around these eight constructs.
Three phases of data collection were utilized to identify the strongest items for each construct. The first, a rater exercise, focused on the content validity of the instrument. In this exercise, a group of experts and students sorted the original 291 items into each of the eight constructs they believed it measured. Frequency distributions were calculated, producing 202 items for which there was a high level of agreement.
The second phase of data collection included distributing the 202-item pilot instrument to 101 undergraduate students at the University of Maryland. The same instrument was distributed to the same clusters of students four weeks later to explore the test-retest reliability of the instrument. Both sets of data were also evaluated for internal-consistency reliability using Cronbach's alphas and construct validity using factor analysis. Evolving from these data were 104 items that supported the validity and reliability of the instrument.
In the final phase of data collection, 342 undergraduate students, from a random sample of 675, returned the 104-item instrument. The tests of internal-consistency reliability produced results indicating that seven of the eight constructs are likely to continually yield accurate outcomes. The Cronbach's alpha for the eighth construct was only minimally acceptable. Using factor analyses and correlational analyses, there was also strong support for the construct validity of the instrument. The results from this study support the value of this instrument for use in future research and assessment.
