Convergent & Discriminant Validity using Correlation Coefficient

https://www.de.sarupub.org/wp-content/uploads/Convergent-and-Discriminant-Validity.jpg

Convergent and Discriminant Validity

Meaning of Convergent and Discriminant Validity

Campbell and Fiske (1959) proposed two aspects to asses the construct validity of a test:

1. Convergent validity: It is the degree of confidence we have that a trait is well measured by its indicators. ie. The statements measures the construct well.


2. Discriminant validity: It is the degree to which measures of different traits are unrelated. ie. Each construct measures a different attribute of the problem.

In structural equation modelling, Confirmatory Factor Analysis has been usually used to asses construct validity (Jöreskog, 1969).

Convergent & Discriminant Validity using Correlation Coefficient

To estimate the degree to which any two measures are related to each other we typically use the correlation coefficient. That is, we look at the patterns of intercorrelations among our measures. Correlations between theoretically similar measures should be “high” while correlations between theoretically dissimilar measures should be “low”. Use factor analysis to make correlation matrix in SPSS OR use Analyse»correlate»Bi-variate.

Convergent Validity

To confirm convergent validity, the results should show that there exist positive correlation between the statements that measures a construct.

Example of Convergent validity

We have four items/statements of “Self esteem” as “Respect, Care, Sincerity and Emotional Stability”. These four items should have a positive correlation between them.

Convergent and Discriminant Validity1

Discriminant Validity

To confirm discriminant validity, the results of correlation should show that there exist less or no correlation between the constructs of the problem studied.

Discriminant Validity2

we have two constructs named as “self esteem” and “locus of control”.  Self Esteem and locus of control has two items each (SE1,SE2, LOC1, LOC2). The correlation between the items in Self esteem and Locus of Control is very low ie near to no. As we know the value of correlation ranges between -1 to +1. Zero means no correlation. All the values in the above example are positive and close to zero. This shows that the construct “Self esteem” is different from the construct “locus of control”, hence discriminant validity proved.

Reference:

https://www.socialresearchmethods.net/kb/convdisc.php

Visit Saru Publications – Online Journals

List of Journals:

1. 
   

   INTERNATIONAL JOURNAL OF SCIENCE & ENGINEERING RESEARCH – IJOSER

   
   


2. 
   

   INTERNATIONAL JOURNAL OF COMMERCE, ECONOMICS & MANAGEMENT – IJOCEM

   
   


3. 
   

   INTERNATIONAL JOURNAL OF SERVICE QUALITY – IJSQ

   
   


4. 
   

   INTERNATIONAL JOURNAL OF WEB COMPONENTS, DATA & ANALYTICS – IJWCDA

   
   

(0)

Construct validity, Convergent validity, Discriminant validity, Multiple Correlation, #ConstructValidity, #ConvergentValidity, #DiscriminantValidity, #MultipleCorrelation, Deepa-Enlighten
https://www.de.sarupub.org/convergent-discriminant-validity-using-correlation-coefficient/