I have tried reversing the coding but now the loading previously negative has changed to positive but changing another previously positive loading into negative.
before you are too quick to recode consider the notion that they are simply poor items to indicate that particular factor rather than fine items that just need to be recoded.
Items of the same scale may be written in a positive way (i.e. In the same direction of the scale meaning) or in a negative way (in the opposite direction). For example you may find the following item in a typical depression scale: "I am satisfied with my life" which might has a negative sign when factorially analyzed with other items that measuring depression.
@Gary K. Leak, items with negative loadings are somehow relevant and are related to others. I can even understand the relationship in some factors but my question is that, can keep them negative in my end results? or do I need to have them all positive.
secondly, i tried recoding as well. but in that case, that particular items' loading become positive but another loadings gets negative.
@Hesham Fathy Gadelrab, thank you very much for the explanation. I understand it very well now. and I can relate it to my case also. but, please guide me with the practical application now. can you please suggest me, how can I start interpretation? should I explain the negative loadings with the negative sign or do i need to do any treatment?
I am a beginner in econometrics, your reply in accordance will be highly appreciated. thank you,
These are factor scores calculated once you run the factor analysis and appeared in the input file. I need to infer the trend of it and I don't know how to do it
how would you treat an item that is cross-loaded across factors, where in one factor it displays a positive value, and negative value across another factor?