Комментарии:
A lot of clarity used explaining this concept! I really appreciate the video you made. Thank you.
ОтветитьVery best video
ОтветитьReflective: construct-> items , generally similar scores, high correlation , reliability internal consistency alpha
Formative: items -> constructs, response can be completely different, reliability cannot be used
Cbsem: all constructs are reflective
Boxes: observed variables
Round: constructs
Reflective: factor loading (relationship bw construct and items)
Formative: factor weight
can a model have both formative and refelctive consturcts?
Ответитьvery useful sir
ОтветитьSir, construct cannot measure directly. We need to find variables that measures the construct ? How can I measure construct directly with 3 statements
ОтветитьCan we use formative measurements with AMOS? Thanks a lot.
ОтветитьPlease could i use amos with second order formative constructs because the purpose of the study is the confirm a theory already put forward. Please i need your answer
ОтветитьYour presentation is made for us non-statisticians. Hahaaah. Thank you so much
ОтветитьRespected sir, please tell how can we statistically know whether our 2nd order construct is reflective or formative?
Ответитьthanks for your time
ОтветитьHi, i need some help on my question to determine whether it is reflective or formative... I using SERVQUAL model
Dependant variables: Occupants satisfaction
1. I am satisfied with the services provided by the management.
2. The services meet my expectation.
3. I am satisfied with employees respond and prompt services.
4. Overall service quality provided by the management is excellent.
Thank you sir for this meaningful presentation! I have struggled a lot to understand and distinguish between reflective and formative models before watching your splendid video!
ОтветитьHello Sir,
Really it was excellent lecture. If possible then please take any examples of survey.
Thank you professor! One thing I could not figure out is the "reliability" vs "colineality". In sense that if indicators are "correlated", they mean higher reliability. However, high correlation between indicators cause colineality problem. It seems to me that these two are somehow contradictive. Hope that you could give some clarifications in one of your videos. Many thanks again!
Ответитьloved
ОтветитьThe best videos on SEM I have ever come across.
Ответитьloved it!
ОтветитьThank you so much.... It was a great help.
ОтветитьVery helpful Saed, thanks!
Ответить