Hi Dinesh,

before interpreting the size of loadings (and other parameter) the more essential question is whether the model fits which should be tested with the chi-square test (if you use R / the lavaan package you can also add model-implied instrumental variable tests for each indicator, see for an application this study (if you allow the shameless self-ad ;) )

Rosman, T., Kerwer, M., Steinmetz, H., Chasiotis, A., Wedderhoff, O., Betsch, C., & Bosnjak, M. (2021). Will COVID‐19‐related economic worries superimpose health worries, reducing nonpharmaceutical intervention acceptance in Germany? A prospective pre‐registered study. International Journal of Psychology. https://doi.org/doi.org/10.1002/ijop.12753

Indicators can have low loadings if the measurement model is misspecified and these indicators are simply mis-aligned to the wrong latent. I remember my first experience of this, when I had a single-indicator measure of "intention" and threw it into a factor model together with some goal commitment items. Result was a (wrong) one-factor model and the intention indicator had a loading of .5.

Hope that helps,

Holger

I agree with Holger Steinmetz . Small factor loadings are often a sign of factor misspecification (i.e., small loadings often indicate that a variable does not measure the same factor as the other indicators). Notice that a factor can still be identified even with just two variables (indicators) as long as the factor is correlated with at least one other variable in the model.

Obviously, in simple words you have to test alternative paths or models