4. Study design
This study used two statistical methods – structural equation modeling (SEM) and
simple group difference analysis. Ordinary least squares (OLS) or linear least squares is
a method for estimating the unknown parameters in a linear regression model.
This method minimizes the sum of squared vertical distances between the observed
responses in the dataset, as well as the responses predicted by the linear approximation.
However, OLS does not provide structural relations among the variables. This limitation
often gives way to a SEM, which is a very powerful multivariate analysis method.
A major benefit of using SEM analysis is that measurement error is estimated while
parameter estimates are adjusted accordingly, as opposed to conventional regression,
which assumes that variables are measured without error (Bollen, 1989). Studies using
this approach provide valuable insights into cross-country comparative models
involved in technology acceptance (Shin, 2009; Walsh et al., 2010).
SEM was carried out to assess structural relationships among the factors and
investigate their moderating effects, and it was conducted using AMOS, a maximum
likelihood-based SEM software. Simple group difference analysis was used to
categorize groups and to determine the statistical significance of the differences
between them.