[an error occurred while processing this directive]

About BRINT | News About BRINT | Help & FAQs | Users Guide | Advertise Here |
Welcome to the World's No. 1 Resource for Business Technology Management and Knowledge Management

Services: Knowledge Portals · Knowledge Map · Knowledge Network · Book of Knowledge · NEWS· INFORMATION Channels: General Business · Business Technology · E-Business · Knowledge Management Community: Join the Network! · Global Network · Events Calendar · Executive Jobs	SEARCH [HELP]

3. Two Stage Least Squares Method for Determining the Parameters of the Structural Equations

Two stage least squares method was used to determine the parameters of the structural equations. Since, ISOUT is the only endogenous explanatory variable in the model and is itself independent of any other endogenous variable, the parameters of the reduced equation for ISOUT would also be valid in case of its structural equation. Nevertheless, the parameters of the other endogenous (non-explanatory) variable, specifically IPI, would need to be determined using the estimated values of ISOUT. First, an ordinary stepwise regression was done on the equations of the reduced model. Then the parameters from this regression were used to derive the estimated values of the endogenous explanatory variable (ISOUT index of the OP variable). The estimated values of the endogenous explanatory variable were used for the next stepwise regression to determine the parameters of the structural equation for the other endogenous (non-explanatory) variable.

3 A. Determinants of Information Systems Outsourcing Policy

The objective of this section is to analyze the relative influence of financial considerations and business strategy on IS outsourcing policies, that is, to analyze equation 1. For this purpose, IS outsourcing policy will be represented by the indicator of percentage of IS budget spent outside IS department (ISOUT). The explanatory variables are divided into two groups. First, financial considerations are represented by the four factors -- performance index, long-term index, business cost index, short-term index -- obtained from factor analysis A. Second, business strategy is represented by the two factors -- position index and competitive index -- obtained from factor analysis B.

The standard regression methods were chosen to do the analyses because the variables were essentially indices and hence continuous in nature. Stepwise regression was used to select the indicators of financial considerations and business strategy that were to be included in the equations. Only the indicators that maximize the explained variance of the information systems outsourcing policy variable (ISOUT) and have at least a .05 significance level appear in the results in Table 3. As mentioned before, the parameters obtained for this reduced equation of ISOUT will also be valid in case of the structural equation of ISOUT

                Table 3: Determinants of IS Outsourcing (ISOUT)
____________________________________________________________________________________
                                 Explained Variable
                        _______________________________________
Explanatory Variables a,b,c   % IS Budget Spent Outside IS Department
____________________________________________________________________________________
        R-Square                        .1421
        F valuec                        [2,72]
                                        5.96
        Sig F                           .004
        Constant                         138.93
        A. Financial Considerations
                Performance             91.23
                                        (1)
                Business cost            -77.50
                                        (2)
____________________________________________________________________________________

a  Only the explanatory variables selected by stepwise method up to the
   0.05 level of significance are included in this table
b  For each variable, data are values of the regression coefficients;
   numbers in parentheses indicate step when entered into the
   corresponding regression.
c  Values in brackets are corresponding degrees of freedom for F
   coefficient of the equation. 

The overall relationship between the explained variable and the explanatory variables is significant. The R-square value of .1421 indicates that over fourteen percent variance is explained by the explanatory variables. The F-value indicates that the regression is significant at .01 significance level. The regression suggests the relevance of the two indices of performance and business cost structure. Interestingly, neither the business strategy indicators nor the short-term and long-term financial indicators appear in the regression equation. In the regression equation with ISOUT as the explained variable, return on technology and business cost structure, in that order, show significant explanatory powers. The respective signs of the regression coefficients indicate that a higher return on technology has a positive effect on IS outsourcing while higher business costs have a negative effect on IS outsourcing.

In conclusion, only the financial considerations influence the IS outsourcing policies.

3 B. Determinants of Information Systems Productivity

Since IPI (the index for information systems productivity) is explained by another endogenous variable ISOUT, we need to find the estimated values of ISOUT for the reduced model and then use these values (instead of original ISOUT data) for the stepwise regression to explain the variance in IPI. Based on the parameters of the reduced equation for ISOUT, estimated values for ISOUT were computed and labelled EISOUT. Using the structural equation (3b) for ISP along with the original indices (factors) for variable F (performance index, long-term index, business cost index and short-term index) and the EISOUT values, the second-stage stepwise regression was done. The results of this analyses are presented in Table 4.

                Table 4: Determinants of IS Productivity (IPI)
____________________________________________________________________________________

                                                Explained Variable
                                        
_______________________________________
Explanatory Variables a b               Information Productivity Index
____________________________________________________________________________________
        R-Square                                        .1210
        F valuec                                        [2,81]
                                                        5.574
        Sig F                                           .005
        Constant                                        .2775
        A. Financial Considerations
                        Business cost                   -.068086
                                                           (1)
                        Short-term                      -.056296
                                                           (2)
____________________________________________________________________________________
a b c   Same as for Table 3

The results of these stepwise regression analyses suggest that financial considerations are the more important determinants of IS productivity. Specifically, higher business cost structure has a negative influence on the information productivity index. Similarly, short-term effects of IT investments on financial performance may adversely effect the IPI. A surprising result is that the indicators of business strategy do not enter into the regression equation for IS productivity. This suggests that [within the assumptions of the proposed framework] the strategic position of IS within the organization, or its competitive resource investment do not have significant influence on the information productivity. According to the results listed in table 4, the answer to the question addressed in this study is that financial considerations are the primary determinant of the IS productivity.