Scores derived from variance specific for a factor


New Member

I received a cooment from one of the reviewer but I do not understand what I should do. Here is short explanation of the study:
This study investigated the influence of lectures about the Renaissance and abstract art on ratings of paintings from these two periods in art history. The study included two sessions. In the first, 72 naive participants rated on 21 seven-point Likert type scales the figural and abstract paintings. In the second session participants were divided into three groups and they rated a new, parallel set of paintings (on the same scales) after attending a lecture on Renaissance or abstract art or without lecture attendance. Three dimensions were extracted: Aesthetic experience, Relaxation tone and Arousal.

I received following comments:
Since comparing pre- and post-lecture scores per factor showed generalized
effects across factors, I am wondering whether these effects originate from
the (substantial) shared variance of factors, i.e., from the G factor of
subjective experience. It would be interesting to test effects of lectures
on scores derived from variance specific for a factor
. Should the lecture
effects disappear one might argue that lectures influence the G rather than
each factor separately.

I am not sure how to get perform what is suggested.

Thanks in advance,


Can't make spagetti
i completely agree with Lazar. you'll need to run a bi-factor model on this dataset to address that concern, preferably using lavaan