Kruskal-Wallis tests for main constructs - Influence of department context

The goal of this study is to test hypotheses about the relationships between multiple independent variables and one dependent variable. As most of my latent constructs are measured on interval scales and I expect linear relationships between the variables, multiple linear regression analysis with ordinary least squares estimation was used (Cohen 2003 Tabachnick and Fidell 1989). The study had two thematically separate parts the first part is focused on the antecedents of lead usemess of employees in firms (n=83, hypotheses 1-3, dependent variable lead usemess). in the second part, h3 otheses about lead usemess of employees and behavioral outcomes are tested (n=149, hypotheses 4-8, dependent variables innovative work behavior, internal boundaiy spanning behavior, external boundary spanning behavior, organizational 

In order to conduct ordinaiy least square regression, some assumptions have to be met, which address linearity, normally of data distribution, constant variance of the error terms, independence of the error terms, and normality of the error term distribution (Cohen 2003 Hair et al. 1998], Whereas the former two can be assessed before performing the actual regression analysis, the latter three can only be evaluated ex post I will thus anticipate some of the regression results to check, if the assumptions with respect to the regression residuals are met 

B B O u s B Tenure Age Gender Use Expertise Technical Expertise Product Involvement Lead Usemess LeadUserness = bo + biTenure + b2Age + b Gender + b U seExpertise + b TechnicalExpertise + b Productlnvolvement + e 83 

Intent variable LeadUsemess mnovatlveWoric Behavior l enialltouKlaiySpamiii Boiavior 

---

Table 27 Kruskal-Wallis tests for main constructs - Influence of department context 3.5 Analysis...

---