25515
Convergent validity is a set of indicators representing one latent variable. This also
underlies the latent variable. This representation can be demonstrated through
unidimensionality, in which the average variance extracted (AVE) can be described. The
minimum AVE value is 0.5. Reliability can use Cronbach's Alpha where the value describes
the reliability of all indicators in the model. The minimum value is 0.7.
Table 2. Convergent Validity and
Reliability Test Results
Variab
le
Cronbach's
Alpha
Factor
Loading
Composite
Reliability
Average
Variance
Extracted
(AVE)
AT
0,6013889
0,610417
0,6375
0,547917
GM
0,5881944
0,629861
0,6270833
0,525
GP
0,5888889
0,614583
0,6291667
0,529167
GT
0,6284722
0,656944
0,6520833
0,581944
GW
0,6125
0,620139
0,6430556
0,561111
In Table 2 each indicator has met the results of the convergent validity test because the
AVE value is > 0.5 and the factor loading value is above 0.5 so that all items can be declared
valid. And the factor loading value > 0.5 so that all items are declared valid. And for all
indicators, it has composite reliability > 0.7 and Cronbach's alpha > 0.6 so it can be said that
all items are reliable.
In table X, the discriminant validity
test can be tested using AVE,
namely whether the correlation value is greater than the value of the latent construct. Each
item is greater than the correlation of the latent construct, so it can be concluded that all items
are valid.
Chia sẻ với bạn bè của bạn: