Page 59 - IJET_July2021_final

average variance extracted (AVE). to be the second subset of
The factors loading value of latent construct validity with an objective
to observed variable and AVE values of measuring the distinctiveness of
should be above 0.05 to confirm each construct. This can be done by
convergent validity (Hair et.al.2010). comparing Square roots of AVE with
It was identified that the values of the correlation squared. To obtain
both AVE and standardized factor discriminant validity, the square
loading for all latent and observed root of Average Variances Extracted
values were above the acceptable (AVE) of the individual factors must
limit, thereby making our construct be higher than the variances of
confirm convergent validity (Table- constructs. The same was achieved
3 Summary of Reliability & Validity in this research after doing analysis.
for Measured Model). (diagonal element in the correlation
matrix) (Table-4 Discriminant
• Discriminant Validity: This happens
Validity).
Table-4: Discriminant Validity

PU PCI ATU BIU

Perceived Usefulness 0.372 0.017956 0.0719 0.0651
(PU)
Perceived Complexity in 0.365 0.0225 0.025921
using the Internet (PCI)

Attitude towards Usage 0.394 0.08208
(ATU)
Behavioural Intention 0.394
to Use Online learning
(BIU)
**. Correlation is significant at the 0.01 level (2-tailed).
Source: AMOS 21.00, Excel calculations

The Structure Model of vital indices are performed to

confirm the research model after which
The measured model above helped in
understanding that the latent variables the results of hypothesis testing are
used for study had good reliability and revealed.
validity. Now the structure model of SEM Model Fit Analysis
will help in measuring the relationships
between the latent and observed The verification of the model has been
variables and testing the hypothesis. done in this research by using 03 types
Structured model is performed in two of fit measures. (Table-5).
parts; in the first section calculations Table-5: Goodness-of-fit indices for
Indian Journal of Educational Technology 49
Volume 3, Issue 2, July 2021