What it means to validate an algorithm or program
Name:
Date:
Class:
A
Quick Questions
Shade in the A or B circle for each question.
1. Which best describes what it means to validate a program?
2. A program runs with no error messages. Is it proven correct?
3. A program should find the average of 3 marks. It runs and prints (a + b + c) / 2 for 60, 70, 80. Which is true?
4. Which is an example of validating a program (not validating input data)?
5. A student tests a program with only one test case and it passes. What can they conclude?
B
Match the Pairs
Draw a line connecting each item on the left to its match on the right.
| Left column | Right column | |
|---|---|---|
| 1assert average([2, 4, 6]) == 4 | AThis only checks the input data. It does not prove the program produces the correct output for the problem. | |
| 2if age >= 0 and age <= 120: | BRunning without errors is not proof of correctness. The output must be checked against expected results from test cases. | |
| 3A program can run with no errors and still be wrong, so you must compare its output with the expected output | Cdef total(a, b): return a * b print(total(2, 3)) # expected 5, got 6 | |
| 4A program is only as validated as its tests: passing every test in a weak set proves little | DValidates one case: checks the function gives the expected output of 4 for this test input | |
| 5print('No errors, so the program is valid!') | Etests = [(0, 0)] for x, want in tests: print(double(x) == want) | |
| 6if int(age) > 0: # validates my program | FValidates input data, not the program: it checks the value is sensible before it is used |
C
Definitions
validate a program
D
Fill in the Blanks
Complete each line of code by filling in the missing parts.
1.
def double(n):
return n * 2
print(double(4) __ 8)
2.
if actual == expected:
print("PASS")
____: print("FAIL")
3.
def double(n):
return n * 2
______ double(3) == 6
4.
tests = [(2, 4), (5, 10)]
___ n, want in tests:
print(double(n) == want)
5.
tests = [(2, 4), (5, 10), (0, 0)]
print(___(tests), "test cases")
E
Spot the Error
Shade in the letter of the line that contains an error.
Question 1
A
def area(w, h):
B
return w + h
C
print(area(3, 4))
D
# expected output: 12
Question 2
A
expected = 12
B
actual = expected
C
print(actual == expected)
Question 3
A
def is_even(n):
B
return n % 2 == 1
C
print(is_even(6))
D
print(is_even(7))
Question 4
A
tests = [(2, 4), (3, 6)]
B
for x, want in tests:
C
print(double(x) == x)
Question 5
A
tests = [(0, 0), (5, 10)]
B
for n, want in tests:
C
print(square(n) == want)
Answer Key
What it means to validate an algorithm or program
Section A: Quick Questions
Q1
B — Prove it gives the correct output for its problem by testing it against test cases
Q2
A — No, it could still give wrong answers (a logic error)
Q3
B — It ran, but the output 105 is wrong. The expected average is 70, so it fails validation
Q4
A — Running a grade calculator with test cases and comparing results to expected outputs
Q5
A — Only that it works for that one case. A program is only as validated as its tests
Section B: Match the Pairs
Q1
assert average([2, 4, 6]) == 4 → Validates one case: checks the function gives the expected output of 4 for this test input
Q2
if age >= 0 and age <= 120: → Validates input data, not the program: it checks the value is sensible before it is used
Q3
A program can run with no errors and still be wrong, so you must compare its output with the expected output → def total(a, b):
return a * b
print(total(2, 3)) # expected 5, got 6
Q4
A program is only as validated as its tests: passing every test in a weak set proves little → tests = [(0, 0)]
for x, want in tests:
print(double(x) == want)
Q5
print('No errors, so the program is valid!') → Running without errors is not proof of correctness. The output must be checked against expected results from test cases.
Q6
if int(age) > 0: # validates my program → This only checks the input data. It does not prove the program produces the correct output for the problem.
Section C: Definitions
Q1
validate a program — To prove a program gives the correct output for its problem by running or tracing it against test cases. Running without errors is not proof; it is only as validated as its tests.Section D: Fill in the Blanks
Q1
def double(n):
return n * 2
print(double(4) == 8)Q2
if actual == expected:
print("PASS")
else: print("FAIL")Q3
def double(n):
return n * 2
assert double(3) == 6Q4
tests = [(2, 4), (5, 10)]
for n, want in tests:
print(double(n) == want)Q5
tests = [(2, 4), (5, 10), (0, 0)]
print(len(tests), "test cases")Section E: Spot the Error
Q1
B — The program runs without errors, but it prints 7 instead of the expected 12. Running without errors is not proof of correctness. It should be return w * h.
Q2
B — The actual output must come from running the program. Copying expected into actual means the check is always True and proves nothing. It should be actual = area(3, 4).
Q3
B — This returns True for odd numbers, so it gives the wrong output even though it runs fine. Testing against 6 and 7 would expose it. It should be n % 2 == 0.
Q4
C — The output must be compared with the expected output (want), not with the input x. It should be double(x) == want.
Q5
A — The expected output is wrong: 5 squared is 25, not 10. A correct program would fail this test. Expected values must be worked out correctly first.