Transactional and transformational leadership impacts on organizational learning
Table 1. Factor loadings and construct validity for the constructs used in the study Completely Unstandardized
Latent variable Transformation al leadership (Tf) Transactional leadership (Ts) Laissez-faire leadership (Lfl) Information acquisition (Infoacq) Information distribution (Distinfo) Information interpretation (Infoint) Behavioural and cognitive changes (Bcc)
Indicator factor loading standardized factor loading .84 t-values 31.97 Idealized influence .78 (attributed) II(a) Idealized influence .60 (behavioural) II(b) Inspirational motivation IM .85 Intellectual stimulation IS .75 Individual consideration IC .95 Contingent reward (CR) .88 Active management by .16 exception MBE(a) Passive management by -.28 exception MBE(p) LFL1 1.01 LFL2 .99 LFL3 .99 LFL4 .95 Internal information .54 acquisition (INTERNAL) External information .45 acquisition (EXTERNAL) Training as information .74 acquisition (TRAINING) Information distribution via .81 systems (SYSTEM) Information distribution via .78 organizational members (PEOPLE) Informal means of information interpretation .62 (INFORMAL) Formal means of information .63 interpretation (FORMAL) Behavioural changes (BC) .51 .74 .83 .83 .88 .86 23.02 30.12 28.86 36.41 27.25 5.16 .21 -.32 .74 .76 .74 .73 -8.39 25.27 25.59 24.90 22.67 19.69 .69 .57 15.58 20.38 .67 .84 29.00 .80 27.33 .76 20.09 21.28 21.03 .72 .76 .88 Cognitive changes (CC) .77 31.17 154 JEEMS 2/2009
Hugo Zagor?ek,Vlado Dimovski, Miha ?kerlavaj
Table 2. Internal consistency and reliability Construct Transformational leadership Contingent reward leadership Information acquisition Information distribution Information interpretation Behavioural and cognitive changes
Cronbach ? .91 .81 .67 .81 .71 .78 CRI .95 .88 .78 .88 .81 .88 AVE .68 .52 .42 .67 .55 .68 Figure 2. The impact of transformational leadership on organisational perfermance
* Statistically significant at p<0.001.
JEEMS 2/2009 155
Transactional and transformational leadership impacts on organizational learning
The relationship between leadership and organizational learning
Model 1: The impact of transformational leadership on organizational learning. Figure 2 presents a structural model of the relationship between transformational leadership and organizational learning along with the standardized values of path coefficients. Statistically significant structural coefficients are marked with an asterisk. Figure 2 also provides the overall coefficients of determination (R2) for each of the endogenous constructs. The model has a good fit to the data (?2 = 351.46 , df = 70, NNFI = .98, CFI = .98, and GFI = .93). In addition, the high values of the determination coefficients indicate that the model explains a large percentage of the variance in the endogenous latent variables.
Hypothesis 1 predicts that transformational leadership will have a strong and positive influence on all four constructs of organizational learning. The results show that only two of these four relationships (H1a and H1d) are statistically significant at p<.001. Transformational leadership demonstrates a strong direct impact on information acquisition (? = .72) as well as on behavioural and cognitive changes (? = .50). Transformational leadership demonstrates a strong indirect effect on information distribution via information acquisition (? = .71), rather than a direct one. Similarly, the indirect effect of transformational leadership (via information acquisition and information distribution) on information interpretation is statistically significant, positive, and strong (? = .61). The total effect of transformational leadership on behavioural and cognitive changes, which includes direct and indirect effects, amounts to .79.
Model 2: The impact of contingent reward leadership on organizational learning. Figure 3 presents the results of fitting the structural model of the impact of contingency reward leadership on the organizational learning process. The model shows a good fit (?2 = 265.32 , df = 58, NNFI = .98, CFI = .98, and GFI =.94) and has relatively large coefficients of determination. With regard to the second hypothesis, a very similar pattern of structural coefficients emerges as in the first model. Contingent reward leadership as a proxy for transactional leadership demonstrates a statistically significant, positive, and strong impact on both information acquisition (? =.72) and behavioural and cognitive changes (? = .55). Surprisingly, the direct impact of contingent reward leadership on behavioural and cognitive changes is even a little stronger than with transformational leadership. Again, only indirect effects of contingent reward leadership on information distribution (? = .65) and information interpretation (? = .61) are evident. For both constructs, these indirect impacts are strong and even slightly stronger than with transformational leadership.
156 JEEMS 2/2009
Hugo Zagor?ek,Vlado Dimovski, Miha ?kerlavaj
Figure 3. The impact of contingent reward leadership on orgnisational performance
Statistically significant at p<0.001.
Table 3 summarizes the direct and total effects of both types of leadership on the four organizational learning constructs. The total effects of transformational and transactional leadership are similar in their magnitude. Hence, we must reject Hypothesis 3.
JEEMS 2/2009
157
Transactional and transformational leadership impacts on organizational learning
Table 3. Standardized direct, indirect, and total effects of the two types of leadership on organizational learning Transformational leadership (H1) Contingent reward leadership (H2*)
Construct Direct effect Indirect effect Total effect Direct effect Indirect effect Total effect Information .72 --- .72 .72 --- .72 acquisition Information n.sn.s .70 .70 .65 .75 .00 . .10 . distribution Information n.s n.s. .61 .61 .61 .71 .00 . .10 interpretation Behavioural and .50 .29 .79 .55 .25 .80 cognitive changes All structural coefficients (except those marked) statistically significant at p>0.001; n.s. - not statistically significant; --- - not specified in the model.
Discussion and conclusions
Implications for theory
This study examines the influence of the transformational and transactional types of leadership on the organizational learning process. Two out of the three proposed hypotheses were confirmed in general, although not entirely. The results show that the same pattern of relationships emerges for both types of leadership. Leadership affects all four constructs of the organizational learning process, even though not directly. One explanation lies in the great correlation between the first three phases (the information processing part of organizational learning). The structural coefficients between information acquisition and information distribution, for example, range from .90 to .98 in the three models we tested. Similarly, the structural coefficients between the information- distribution and information-interpretation phases range from .88 to .94. Despite high correlations, they are distinct constructs as proved by both theoretical arguments (Huber 1991) and previous empirical research (?kerlavaj et al. 2007). The influence of leadership is greatest on behavioural and cognitive changes, which are the final and apparently the most important phase of the learning process in organizations. The total effect of transformational leadership on behavioural and cognitive changes amounts to .79, while the total effect of contingent reward leadership equals .80. Leadership influences behavioural and cognitive changes in two ways. First, it affects them through the previous information-processing phases of the organizational learning process. By facilitating or impeding information processing in an organization, leaders encourage or impede changes in the mentality or behaviour of organizational members in order to address changes in the internal or external business environment. However, leaders also influence changes in behaviour and
158 JEEMS 2/2009