What if factor loadings are negative?
Some variables are shown to have values of negative loadings. A negative sign of loading does not indicate any meaning regarding the strength of the variable to the factor. However, it gives meaning that the variable is related in the opposite direction with the factor [10].
What do negative loadings mean in PCA?
In the interpretation of PCA, a negative loading simply means that a certain characteristic is lacking in a latent variable associated with the given principal component.
What do factor scores tell us?
Factor scores are composite variables which provide information about an individual’s placement on the factor(s). Once a researcher has used EFA and has identified the number of factors or components underlying a data set, he/she may wish to use the information about the factors in subsequent analyses (Gorsuch, 1983).
What do negative factor scores mean?
If an item yields a negative factor loading, the raw score of the item is subtracted rather than added in the computations because the item is negatively related to the factor.
What does a negative value for factor loading mean?
High factor loadings reflect the measurement of a very narrowly defined construct, which is not always desirable. Therefore, theory is important and not just chosing items because of the size of the factor loading. That is what you meant, I think. Descriptive model fit criteria can be very misleading.
When to reverse score a negative loading variable?
You are correct: the negative loading suggests a negative linear association between the latent variable and the observed variable, and if you were to compute an observed score to approximate the latent variable (e.g., by summing or averaging items), it would be appropriate to reverse-score the negative loading variables.
Can you treat all factors consisting of only negative loadings?
You can treat all factors consisting of only negative loadings as factors consisting of positive loadings. Just search the internet, you will find many articles explaining this phenomenon.
How to take negative loading into account when creating a score?
As for how you could take this negative loading into account when creating a score for your variable, I see you as having three options: Keep the model in latent space: whatever analysis you are attempting to do, if you can, you should try to do it using some form of latent modelling.