Using list and dict comprehension is simpler and computationally quicker than calling list() and dict().

Comprehensions are more efficient than calling dict() and list().

Given that these are completely equivalent:

values = {item: get_value(item) item item in items}

values = dict([item, get_value(item)] for item in items)

values = dict((item, get_value(item)) for item in items)

values = dict([(item, get_value(item)) for item in items])

Why not choose the easiest to both read, write, and quickest to execute (the first one)?

Do not follow this advice with set comprehension though, as when set comprehension is performed on an empty iterable it returns a dict because Python cannot differentiate set comprehension and dict comprehension when working with an empty iterable:

items = [1, 2, 3]

one = {item: None for item in items}  # evaluates to dict

two = {item for item in items}  # evaluates to set

empty = []

three = {item: None for item in empty}  # evaluates to dict

four = {item for item in empty}  # evaluates to dict OH NO!

Notice that four evaluated to an empty dict, which was unwanted behaviour.