I have a quick question I was hoping you could please help me with. I have run an EFA on ordinal items from several different scales. The items are measured using different Likert scales (e.g., some are 5 point, some are 7 point etc.) and most of the items are positively skewed. I have used both WLSMV and ULSMV estimators and get the same results. I am finding that the items from each scale are sticking together and all loading onto their own factor (e.g., all the items from Scale 1 are loading onto one factor and all of the items for Scale 2 are loading onto a separate factor). Could this be a result of using different Likert scales (e.g., some 5 point and some 7 point)?