Interaction effect

[Choukri]

Hi everyone,

To test a moderating effect, is it normal to use Product indicator (calculation method) with standardized (product term generation)?

Thank you

[cringle]

We suggest using the two stage approach or the orthog. aprroach (with stand. data).

[Choukri]

Hi,

Thank you for your response

Henseler and Chin (2010, p. 82) indicated that ((For prediction accuracy, the orthogonalizing and the product indicator approach provide a significantly and substantially more accurate prediction than the other two approaches. Among these two, the orthogonalizing approach should be used in case of small sample size and few indicators per construct. If the sample size or the number of indicators per construct is medium to large, the product indicator approach should be used.))

According to the last sentence, my sample size is large (N= 443), and the indicators per constructs are between 4 to 7.

Does this mean that I shoud keep using the product indicator approach or it is better to switch to the orthogonalizing approach?

Greetings

[cringle]

hi

the article also states some advantage of the two stage approach. again, i would either use the two stage or the orthog. approach.

best
cr

[jmbecker]

Moreover, results change if you have formative constructs in your model. Two stage approach especially gets better if formative constructs are included in your model.

[Choukri]

Thank you for your response

My model does not includ formative constructs. All constructs are reflective.

What should I do?

Greetings