MrPedantic wrote:

First, we need to know how many items there are in the set {constructions}, of which {non-standard constructions} is presumably a subset. The figure "100" is meaningless, otherwise. MrP

Just think about all the ones that are continually labeled as non-standard, day after day, on this forum and maybe we can begin the topic discussion.

I suppose that the issue is that even if there are not that many differences between standard and non-standard English, many non-standard forms are considered to be a sign of lack of education, and that can have an effect on people's job prospects for example, or the type of treatment they may receive from others. They may be more readily forgiven in a native speaker whose accent shows that a non-standard form is standard for their dialect/area, but a non-native using non-standard forms may have a few additional problems. If they should find themselves living in situation where non-standard forms are actually the standard, it's quite simple to pick them up if they want, whereas if you learn them in the first place as your 'base' English you might find them hard to shake off if you find yourself living in a situation where they are causing you problems.

I think very few people speak 100% standard English. There are always a few area/dialect/class variations. However, that is then the natural language for that environment. To deliberately learn a random 'non-standard' English, possibly with elements from different areas/dialects/classes, seems a little foolish. Fine if you know you are going to be in the environment where that version is 'standard' but that's awfully hard for an overseas learner to predict.

It is also a very good idea to know both the local non-standard variations and standard English, as you might move location, or go for a job where standard English is pretty much required (even if it isn't spelled out on the application form, there are plenty of jobs where you won't have a chance if you don't speak in the correct way).