what is the order of setUp/tearDown in python unit tests?

I have a discrepancy in understand of the basic unittest methods in python. Given this test file below:

import unittest, sys


class TestStringMethods(unittest.TestCase):

    def setUp(self):
        self.mystring = "example string"

    def tearDown(self):
        del self.mystring

    def test_upper(self):
        self.assertEqual('foo'.upper(), 'FOO')

    def test_isupper(self):
        self.assertTrue('FOO'.isupper())
        self.assertFalse('Foo'.isupper())

    def test_split(self):
        s = 'hello world'
        self.assertEqual(s.split(), ['hello', 'world'])
        with self.assertRaises(TypeError):
            s.split(2)

My understanding based on what I've read ("Methods named setUp in a TestCase are run, automatically, before every test method.", http://ift.tt/1Jw9z3p, etc) I interpret the order of events like:

1. set up self.mystring
2. run test_upper
3. tear down self.mystring

4. set up self.mystring
5. run test_isupper
6. tear down self.mystring

7. set up self.mystring
8. run test_split
9. tear down self.mystring

My co-workers interpret the docs as saying unittest works as follows:

1. set up self.mystring
2. run test_upper
3. run test_isupper
4. run test_split
5. tear down self.mystring

this is a pretty important distinction, which one is right?