I would need help on which statistical analysis I should use to evaluate convergent validity. We have translated an instrument (Likert type) and did test retest along with other instruments to measure validity of the translated instruments? We are not sure which analysis to assess.
Asked
Active
Viewed 35 times
1
-
Tell us what convergent validity means. It is not a statistical term that I am familiar with. – Michael R. Chernick Mar 30 '17 at 18:35
-
Thank you. It is a subtype of construct validity (convergent and discriminant). The first one refers to the degree to which two measures of constructs that theoretically should be related, are in fact related. We want to see if translated instrument is measuring the construct as well as the instrument that we used and has been validated in previous research. Many thanks! – Jane Zhu Mar 30 '17 at 18:42
-
Our sample is small, 29 participants. We cannot use Factor Analysis. Can we use Spearmen test? Thank you very much. – Jane Zhu Mar 30 '17 at 18:50
-
I think you mean Spearman's correlation coefficient which is one of two nonparametric methods that I know about. The other is Kendall's tau. – Michael R. Chernick Mar 30 '17 at 18:54
-
Your help is greatly appreciated. Which one do you recommend, please? Our data has been gathered on same sample (two time within 15 days). Both instruments are Likert type (translated and second one, which will help us evaluate the validity of the translated instrument). Best of Regards. – Jane Zhu Mar 30 '17 at 19:01
-
Given that the data is not numerical either Spearman's rho or Kendall's tau would be appropriate. I think you can decide which to choose. I have no preference, – Michael R. Chernick Mar 30 '17 at 19:05