Organizational Climate, Leadership, Organization Size, and Aspiration for Innovation in Government Agencies

ABSTRACT:

Scholars of organizational innovation emphasize the importance of employees’ perceptions and point to dissatisfaction with the status quo as a force facilitating innovation adoption and innovative behavior. This study applies the efficiency-focused perspective on innovative behavior to explore how perceptions of organizational climate and leadership explain employees’ aspiration for innovation in public organizations. It notes that the association of the predictors varies according to organization size (number of full-time employees). Using samples of full-time employees in South Korean government agencies, regression analyses suggest that employees’ perceptions of the hierarchical climate, as opposed to the innovative climate, relate positively to their aspirations for organizational innovation. One unexpected result is the positive association of the current facilitative leadership with the criterion variable. Organization size moderates the influence of the predictors, except for the current facilitative leadership. The academic and practical implications of these findings are presented in the discussion and conclusion section.

KEYWORDS::

• aspiration
• government agencies
• leadership
• organizational climate
• organization size

Notes

However, we make no claim that employees’ dissatisfaction with the status quo always leads to innovation adoption. In fact, Hirschman’s (1970) theory of exit, voice, and loyalty stands in contrast to our hypothesis, suggesting that organization members can leave their organization if their dissatisfaction with the status quo of current management conditions or culture is not resolved.

As an anonymous reviewer pointed out, this hypothesis, more exactly, seems to propose that public employees working under more hierarchical leadership might be discouraged by expecting that changes could not happen while perhaps also thinking that change is needed. This point could be the dilemma of this article in that the survey only allowed us to know whether employees thought their organization should be more dynamic, and not whether it is likely to become such.

These items were originally developed to measure the desired level of innovative culture or developmental culture under the competing value framework in Quinn and McGrath (1985). Here we used them as a proxy variable of aspiration for organizational innovation.

The correlation results in the Appendix and Table 3 suggest that aspiration for innovation (indicating what the organization should do) and current innovative climate (indicating what the organization is doing) are distinct concepts. In the Appendix, the items for each variable are highly correlated for what they are supposed to measure, but are not highly correlated with those for the other variable. Furthermore, in Table 3, these two variables share around 25% of the variance. Concerning the relationship between these two variables, we assumed that people’s judgments about aspiration for innovation or desired state should be based on their assessments of the current state. Moreover, we checked individual respondents’ current and aspired states of innovation and found that in many cases the current state was equal to or higher than the aspired state. That is, when employees perceived their organization as sufficiently or too innovative, they had lower desired states than current states. Thus, the concern that the aspiration for innovation can represent only the absence of the current state was not the case in the data.

We found a small heteroscedasticity problem in the residuals of the models. The robust standard errors take heteroscedasticity and lack of normality into consideration. Therefore, they can prevent the need to explore nonlinear modeling techniques (Dehart-Davis & Pandey, 2005; Neter, Kutner, Nachscheim, & Wasserman, 1996). Additionally, we checked no multicollinearity problem between the predictor variables by a variance inflation factor (VIF) test. This was because all the VIF scores were lower than 1.5. The results indicate that multicollinearity is not likely to distort our findings (Bowerman & O’Connell, 1990).

Owing to the nested data structures (i.e., employees nested in government agencies), we also conducted hierarchical linear modeling (HLM) (Hox, 2002; Raudenbush & Bryk, 2002). The HLM results are similar to those of the current analysis overall, with respect to the direction and statistical significance of the main predictor variables and interaction terms. The results are available from the authors on request.

In order to check the effect size of the main predictors, we calculated η² (eta-square): 0.22 for current climate for innovation, 0.017 for current hierarchical climate, and 0.0103 for current facilitative leadership. These η² levels indicate large, small, and small effect sizes, respectively. (The regression effect size option in STATA does not work with robust standard errors. It was only in calculating these effect sizes that robust standard error was not used.)

For the substantive effects of the interactions, we provided the changes of R² for each interaction. The first two interaction terms (between current innovative climate and organizational size, and between current hierarchical climate and organizational size) had a small effect on the criterion variable, showing a 0.011 and 0.012 R² change, respectively.

This interpretation suggests the need to investigate contingent situations that can influence the leadership effect on employees’ attitude toward organizational innovation.

In the analysis of variance, it was statistically significant (F = 16.29, p < 0.001), but the effect size (e.g., Cohen’s f = 0.1275) is relatively small. This result implies that the agency effect is not large.

Additional information

Notes on contributors

Chan Su Jung

Chan Su Jung is an Associate Professor in the Department of Public Policy at the City University of Hong Kong.

Geon Lee

Geon Lee is an Assistant Professor in the Department of Public Administration at Kyonggi University in Suwon, Korea.