Why is sql server storing question mark characters instead of Japanese characters in NVarchar fields?

This cannot be a correct answer given your example, but the most common reason I’ve seen is that string literals do not need a unicode N prefix.

So, instead of

set address = N'スタンダードチャ'

one would try to write to a nvarchar field without the unicode prefix

set address="スタンダードチャ"

See also:
N prefix before string in Transact-SQL query