Measurement and Empirical Validity
20. Often exhibits very unique behavior that surprises other members of the organization
Strategic Vision and Articulation (SVA)
originally classified under the factor "environmental sensitivity." Three items each were hypothesized to constitute the factors sensitivity to member needs (SMN; Stage 1 of the model), personal risk (PR; Stage 3 of the model), and unconventional behavior (UB; Stage 3 of the model). These classifications are in line with the original classification of these items under the original factors sensitivity to member needs, personal risk, and unconventional behavior, respectively.
The five-factor model proposed above was then tested through a confirmatory factor analysis using LISREL 7. Convergent and dispriminant validities of the subscales were assessed by testing a structural model using LISREL 7, with the subscales as exogenous latent constructs and the four other leadership scales (task-oriented, participative, people-oriented, and Bass charisma) as endogenous latent variables.
Results of the Refinement
Confirmatory factor analysis of the 20-item five-factor model yielded a χ2(160) = 458.06, ρ = .000. Comparison to an absolute null model with χ2(190) = 4118.54, ρ = .000, yielded a Normed Fit Index (NFI) of .89 and a Non-Normed Fit Index (NNFI) of .91 (see Medsker, Williams, & Holahan [1994] for the method of calculation of these indexes). This was considered a very good fit considering the large sample size of more than 400. The Cronbach's alpha reliabilities of the individual subscales were .87 (SVA), .77 (SE), .84 (SMN), .85 (PR), and .74 (UB).
Convergent and Discriminant Validity
Because the five subscales of the Conger-Kanungo scale are hypothesized to be tapping different behavioral dimensions, they can be expected to have differing relationships with the four other leadership scales included in Study 2.
For example, given the nature of the two constructs, the subscale "strategic vision and articulation" (SVA) should be positively related to the Bass charisma scale (BASS). By the same token, given the context and scope of leader behaviors associated with formulation and articulation of a strategic vision, SVA should not be strongly related to leadership scales such as "task-oriented,"
"people-oriented," and "participative," which are more concerned with day-to- day, routine organizational matters. Similarly, the assessment of the internal environment of the organization is a component of successful task leadership;
therefore, the subscale "sensitivity to the environment" should be strongly and
TABLE 3.12 LISREL (Gamma) Estimates of Relationships Between the Refined Conger-Kanungo Subscales and Other Leadership Scales
Leadership
Scales SVA SE SMN PR UB BASS .381***
TASK .044 PPL .043 PART .139
.209** .248***
.580*** -.204***
.070 .682***
.209* .485***
.214*** _.099***
.133 -.113**
.143** .033 .035 -.065 SOURCE: Adapted from "Measuring Charisma: Dimensionality and Validity of the Conger-Kanungo Scale of Charismatic Leadership" (p. 295), Conger, Kanungo, Menon, and Mathur. © 1997 by the Canadian Journal of Administrative Sciences. Reprinted with permission.
NOTE: See Table 3.11 for definitions of subscales.
*p < .05. **p < .01. ***p < .001.
positively related to the task-oriented leadership scale (TASK). On the other hand, the subscale "sensitivity to member needs" (SMN) should be strongly and positively related to the people-oriented leadership scale (PPL) and the partici- pative leadership scale (PART). Given the oft-suggested dichotomy between task and relationship orientations, one can then expect a negative relationship between SMN and TASK leadership. There is no a priori reason to expect relationships between the subscales "personal risk" (PR) or "unconventional behavior" (UB) and each of the other leadership scales.
The above hypotheses were tested by a structural latent variable model, with the five subscales treated as independent exogenous variables and the four other leadership scales (BASS, TASK, PPL, and PART) treated as dependent, endo- genous variables. The alpha reliabilities for these latter scales are .87 (BASS), .72 (TASK), .81 (PPL), and .81 (PART). Table 3.12 shows the LISREL (gamma) estimates.
As hypothesized, the subscale "strategic vision and articulation" (SVA) is positively and significantly related to the Bass charisma scale (BASS). Further- more, it is not significantly related to any of the other leadership scales. As expected, the subscale "sensitivity to the environment" (SE) is strongly related to the task leadership scale (TASK). The hypotheses involving the subscale
"sensitivity to member needs" (SMN) also are substantiated. As hypothesized, SMN is strongly related to "people-oriented" (PPL) and "participative" (PART) leadership scales. It also is negatively related to the TASK leadership scale. The subscales "personal risk" (PR) and "unconventional behavior" (UB) were not expected to be strongly related to any of the leadership scales. Table 3.12 reveals that, in general, this expectation is supported. PR is significantly related to the BASS scale, and UB does not have strong positive relationships with any of the
leadership scales. These results provide evidence of convergent and discriminant validity for the Conger-Kanungo subscales.
This refinement of the Conger-Kanungo scale resulted in a revised 20-item measure of charismatic leadership with five behavioral dimensions measured by five distinct subscales. The alpha reliability of the total scale is .88, justifying its use as an overall measure of charismatic leadership as proposed in the Conger-Kanungo model.
Although the data in Table 3.12 provide evidence of the convergent and discriminant validity of the subscales, there is some evidence of the convergent and discriminant validity of the overall scale. The overall Conger-Kanungo scale has a correlation of .69 with the Bass charisma scale. Table 3.12 reveals that this degree of convergence is attributable to the positive relationship between four of the Conger-Kanungo subscales and the BASS scale. On the other hand, the dimension of "unconventional behavior" differentiates the Conger-Kanungo scale from the Bass scale, as is revealed by the slight but significant negative relationship between the subscale UB and the BASS scale.
Although the confirmatory factor analysis (CFA) supported a five-factor solution with adequate fit, it must be borne in mind that the CFA was performed on the same data collected in Study 2 as was the original principal components analysis. It is necessary, therefore, to replicate this result using other independent samples. This was done in Study 3.