chose PROCESS for the following reasons: First, it integrates many of the
most functional procedures, such as
SOBEL’s test, RSQUARE, MODPROBE,
MBESS, and MODMED (Hayes, 2013).
Second, PROCESS allows researchers to
build up models that combine examinations of both direct and indirect mediation effects. In this work, PROCESS was
implemented to evaluate the mediation
role of team reflexivity.
Evaluation of the Structural Model
This study examined the hypotheses
using SmartPLS 3 (Ringle, Wende, &
Becker, 2014). It is important to note that
the PLS model is measured based on
the goodness-of-fit (GoF) of the model
(Vinzi, Trinchera, & Amato, 2010) and
R2, which evaluates its predictive power
(Chin, 1998). As shown in Figure 2, this
study achieved a GoF value of 0.549,
which surpasses the cutoff value of
0.36 for large effect sizes of R2 (Cohen,
1988). This shows that the research
model had a better estimation power
compared with the baseline values (the
GoF criteria). Additionally, the research
findings supported the predictive power
of the model, because the R2 values
all exceeded 0.3 (Hermann, Gassmann,
& Eisert, 2007). Moreover, this study
also computed the Stone-Gaisser test
criteria (Q2) to examine the model’s
predictive power (DeLeon & Chatterjee,
2015) by using the blindfolding proce-
dure. The results demonstrated that for
each endogenous construct, the Stone-
Gaisser criteria surpassed the critical
value of 0, ranging from 0.407 to 0.546.
Finally, this study computed the stan-
dardized root mean square residual
(SRMR), which is an important model-
fit criterion provided for PLS path mod-
eling (Henseler et al., 2015). The results
proved the appropriate fit of the model,
with an SRMR value of 0.029.
As specified by the path loadings,
the direct effect of existing knowledge
on team reflexivity shows a significant
and positive effect (b 5 0.263, p ,0.001).
Figure 2: Results of path analysis.
TR2 TR3 TR1
R2 = 66.2%
R2 = 65.28%
R2 = 49.4%
0.893 0.794 0.898
Fit Statistics: GoF = 0.549; SRMR = 0.029; R2>0.3; Q2>0
Note: p<0.001; p<0.01; p<0.05
Path coefficients are standardized.