Perceived Moderating Ability of Relational Interaction vs. Reciprocal Investments in Pursuing Exploitation vs. Exploration in RFID Supply Chains - Enterprise Information Systems Design, Implementation and Management: Organizational Applications - page 92

Table 1. Moderated regression for exploitation as the dependent variable with relational interaction as
moderator (N=87)

Independent Variables : Data consistency and cross-functional application integration (both are components of IT infrastructure integration);
financial flow integration, information flow integration, and physical flow integration (all are components of supply chain process integration)
Dependent Variable : Exploitation
Moderator Variable : Relational Interaction whose variable name is “RelaInteract3Cat1” (Nominal variable for the mean of the relational
interaction items)
Independent Vari-
able

Model 1: R 2 Without
ProductTerm

Model 2:R 2 With
Product Term

% Variance Ex-
plained by Mod-
erator with Product
Term

F Value of Model 2
(degrees of freedom)

Significance of F
Change

Financial flow inte-
gration

.614 .669 5.5% 56.023 (3, 83) p<.000

Cross-functional
application inte-
gration

.475 .527 5.2% 30.829 (3, 83) p<.003

Data consistency .557 .591 3.4% 40.026 (3, 83) p<.01

Information flow
integration

.631 .660 2.9% 53.631 (3, 83) p<.01

Physical flow inte-
gration

.659 .680 2.1% 58.852 (3, 83) p<.023

in this case, relational interaction and the specific
predictor variable. And so, for instance, in the case
of financial flow integration, the product term
would be the product of coordination informa-
tion exchanged and IT infrastructure integration
(i.e., FinancialFlow3XRelaInteract3Cat1). The
next column label shows “F Value of Model 2
(degrees of freedom), which means that the F
value of model 2 which includes the product term
is shown along with the degrees of freedom for
that regression model. The significance of the F
change from model 1 to model 2 is indicated by
the last column.
The relationships between the predictor vari-
ables and exploitation as moderated by relational
interaction should be interpreted accordingly.
Let's take the case of financial flow integration,
the predictor variable whose relationship with
exploitation is significantly moderated to the
greatest extent by relational interaction (i.e., largest
percentage increase in variance of 5.5 percent).
About 61.40 percent of the variance in exploita-
tion is explained by financial flow integration and

relational interaction as indicated by model 1 in
Table 1. Model 2 is, then, introduced by including
the product term (i.e., i.e., FinancialFlow3XRe-
laInteract3Cat1) which represents the interaction
between financial flow integration and relational
interaction. As shown on Table 1, the addition of
the product term resulted in an R 2 change of .669,
F(3,83) = 56.023, p<.000. This result supports the
presence of a moderating effect. In other words, the
moderating effect of relation interaction explains
5.5 percent in the increase in exploitation over and
above the variance explained by financial flow
integration and relational interaction as separate
independent variables.
The relationships between the remaining
predictor variables and operational exploitation
should be conducted accordingly as well: cross-
functional application integration, data consis-
tency, information flow integration, and physical
flow integration.