Hi,

Cronbach's alpha test should be performed on unidimensional constructs (according to DeVellis (1991) "a unidimensional scale or a single dimension of a multidimensional scale should consist of a set of items that correlate well with each othet'' p.25). In other words, the test should be carried out for each of the factors or sub-factors (in multidimensional scales) in your model. Have a look at work of Cho for greater clarity:

Cho, E. (2016). Making Reliability Reliable: A Systematic approach to reliability coefficients. Organizational research methods, 19(4), 651-682.

Best,

Lukasz

Oshin Gulsia, for a start, I would be very cautious about using coefficient alpha (that's the more appropriate name, not Cronbach's alpha) for the metric you're interested in). I suggest you do some reading about it so that you don't fall into any traps. One of these traps is thinking that the value of alpha is unrelated to the number of items under consideration. By the time you have about 25 variables, alpha is almost inevitably high even if there is not much relationship between those variables. The following article as well worth reading:

Schmitt, N. (1996). Uses and abuses of coefficient alpha. Psychological Assessment, 8(4), 350–353. https://doi.org/10.1037//1040-3590.8.4.350

Taber, K. S. (2018). The use of Cronbach’s alpha when developing and reporting research instruments in science education. Research in Science Education, 48(6), 1273–1296. https://doi.org/10.1007/s11165-016-9602-2

With regard to your more specific questions, no, don't include demographic variables in any calculation of alpha. If you want to see how demographic variables relate to variables created from scales, conduct analyses such as regression instead.

You should also calculate a separate alpha for each factor. I confess, however, that I'm not sure what you mean by factors and subfactors. However, the alpha value is quite likely to differ depending on the set of items it is based on.

I hope that's helpful - and all the best with your research.

Thank You for your valuable inputs Sir Robert Trevethan Lukasz Derdowski