(urgent) Sample Size

anugrah · Post by **anugrah** » Tue May 10, 2011 1:54 am

Thank you for answering Prof Bido.
I hv another problem: composite reliability vs cronbach alpha

"The first criterion which is checked is internal consistency reliability.The traditional criterion for internal consistency is Cronbach’s a (Cronbach,1951),

The composite reliability takes into account that indicators have different loadings, and can be interpreted in the same way as Cronbach’s a.[Ringle et all (2009), p299]
No matter which particular reliability coefficient isused, an internal consistency reliability value above 0.7 in early stages of research and values above 0.8 or 0.9 in more advanced stages of research are regarded as satisfactory (Nunnally&Bernstein,1994), whereas a value below 0.6 indicates a lack of reliability."

When i calculate composite reliability, the result is above 0.7 for all. But when i see that cronbach alpha, just 1 which not fulfill that criterion (only 0.45)

So, which should i use?

Best regards,

Diogenes · Post by **Diogenes** » Tue May 10, 2011 9:42 am

Hi,

Cronbach’s alpha is used in the summated rating scale context (the LV score = sum or mean of the indicators).

C.R. is used in SEM/PLSPM context, for this reason I will use it.

Best regards,

Bido

anugrah · Post by **anugrah** » Wed May 11, 2011 7:40 am

Thank you very much sir..

This is my first time i use SmartPLS. I just learn in this forum >.<

Im still have problem sir about Blindfolding.
Based on my model before, when using Blinfolding:

- Select omission distance is 7? (is that ok?)
- When i select construct "Rel" >> i see Construct Crossvalidated Redundancy (total), 1-SSE/SSO of construct "A" is the Q2-value of construct "A" , is it right?

- then how to calculate q2?
Is it Communality is q2?

- When i choose "Rel", my Q2 (Redundancy) is 0.112692. but in Communality there are empty number. Please explain to me about this.. and what should i do?