S6 ICT: System Testing and Acceptance (four test levels, equivalence and boundary values, exit criteria) Tutor Prep Guide with a 45-minute plan, teaching script and common errors

2026-11-15

Tutor lesson-prep guides · Secondary ICT

Students know the project must be tested, yet cannot say which level tests what, or when the product may be released. This lesson turns testing into four owners and one exit rule with numbers in it.

Objectives — By the end of the lesson students can: (1) name the owner and the reference document for each of the four test levels; (2) design a minimum set of cases for an input range using equivalence partitioning and boundary values, and explain why 17, 18, 60 and 61 are all needed; (3) turn 216 passes out of 240 into a 90 per cent pass rate, then work out that 95 per cent needs 228 cases, a gap of 12; (4) rank fixes by severity and state that 5 critical and 9 major defects must all be closed, with 38 of 40 UAT scenarios passing at 95 per cent.

01 Prerequisites (3-minute check)

If any of these fails, fix it before the lesson starts.

Skill	Quick check	Pass mark
Percentages	240 test cases, 216 pass: what is the pass rate	90 per cent
Fractions	To reach 95 per cent, how many of the 240 must pass	228 cases
Classification	24 defects split into groups of 5, 9 and 10: which group is fixed first	The group of 5, the critical ones
Reading a process diagram	Which test comes after unit testing	Integration testing

02 Lesson flow (45 minutes)

Time	Stage	Tutor does	Student does
0-6 min	Entry check	Ask for the pass rate, the 95 per cent target, the first group to fix and the next test level; keep the answers on the board	Spoken answers, working on paper
6-16 min	Act one: four levels, four owners	Draw four boxes and write who runs each level and which document it checks against	Students say acceptance testing is run by users
16-26 min	Act two: black box reads inputs, white box reads paths	Show the classes and boundary values for ages 18 to 60, then compute 15 of 18 branches	Students write 17, 18, 60 and 61
26-36 min	Act three: severity and exit criteria	5 critical, 9 major, 10 minor; 18 fixed is 75 per cent and 720 minutes; 90 per cent misses 95 per cent	Students compute 228 and the gap of 12
36-42 min	Question types with changed numbers	Switch to 300 cases with 270 passing and run the four question types	Students do one and write the unit (cases, per cent)
42-45 min	Exit check	Three spoken questions; note which act each error comes from	All three correct

03 Teaching script (read it out)

Scene 1: Act one: four test levels, four different owners

Testing is not one event but four narrowing layers. Unit testing is the developer testing each function alone: a discount function given 100 should return 90. Integration testing checks the interface between two modules: whether the order module hands the payment module the right data, and whether the amount changes on the way. System testing treats the whole product as a black box and checks it against the requirements specification: if the specification says an unpaid order can be cancelled, one case must try exactly that. Acceptance testing, or UAT, is run by the user or client against real business scenarios, and ends with a sign-off. So when a question asks who runs a level, the answer always depends on the layer: developers for unit, users for acceptance.

What goes wrong if developers run the acceptance test themselves?

Scene 2: Act two: black box reads inputs, white box reads code paths

Black-box testing never looks at the code, only at input and output, and two tools carry most of the weight: equivalence partitioning and boundary values. For an age field of 18 to 60, split the input into three classes: below 18 is invalid, 18 to 60 is valid, above 60 is invalid. One representative from each class replaces a thousand numbers. But classes cannot catch a less-than sign written as less-than-or-equal, so add boundary values: 17, 18, 60 and 61, each pair one just outside and one just inside. White-box testing is the opposite and counts branches: 18 branches with only 15 covered is 15 over 18, or 83.3 per cent, leaving 3 branches never executed, and a defect inside them stays invisible.

With 15 of 18 branches covered, how many are missing, and why are unexecuted branches the most dangerous?

Scene 3: Act three: severity ranking and exit criteria

A test run leaves a pile of defects, and they must not be fixed in arrival order. Rank by severity first: 5 critical (data corruption, crashes), 9 major (a function unusable but with a workaround), 10 minor (a typo on screen). Of 240 cases, 216 pass, so the pass rate is 216 over 240, which is 0.9 or 90 per cent. Eighteen are fixed, that is 75 per cent, at 40 minutes each, so 720 minutes or 12 hours, leaving 6. Exit criteria must be written as numbers: a pass rate of at least 95 per cent, and 240 times 0.95 is 228, so 12 more cases must pass; zero critical and major defects open, and since all 5 plus 9 are fixed that holds; and 40 UAT scenarios with 38 passing, 95 per cent, signed off by the client. All three at once means release; miss one and another round is due.

With 216 passing, how many more must pass to reach 95 per cent, and may the 6 minor defects ship?

04 Question types and worked steps

Type	Example	Steps	Answer
Classify the test level	Which level, who runs it, which document it checks	Three questions: who executes, one unit or the whole system, which document is the reference	Unit is the developer against code; integration is module against module; system is the product against requirements; acceptance is the user against business scenarios
Design test cases	An age field of 18 to 60	Split into equivalence classes, then add boundary values one each side	Three classes plus four boundary values: 17, 18, 60, 61
Compute the pass rate and the gap	240 cases, 216 passing	Passes over total, then the target percentage times the total for the gap	90 per cent; 95 per cent needs 228, so 12 short
Rank the fixes	5 critical, 9 major, 10 minor, 18 fixed	Critical first, then major; minor may be deferred but only inside the exit criteria	All 14 critical and major closed is the hard line; 6 minor may ship with conditions

Sanity check (teach this): Four checks: (1) a pass rate never exceeds 100 per cent, so a larger figure means the denominator is wrong; (2) equivalence classes must tile the whole input range with no hole, and invalid classes are the ones most often forgotten; (3) boundary values come in pairs, 17 with 18 and 60 with 61, and testing only 18 and 60 cannot catch an off-by-one; (4) exit criteria carry numbers such as 95 per cent and zero critical, because roughly tested is not a criterion.

05 Independent practice (3 questions)

Question	Answer	What to watch
240 test cases with 216 passing: what is the pass rate	90 per cent	Watch for 24 used as the numerator
To reach 95 per cent, how many of the 240 must pass	228	240 times 0.95, not 216 plus 24
An age field of 18 to 60: which boundary values	17, 18, 60 and 61	Check both sides have one just outside and one just inside

06 Common mistakes and what to say

Mistake	Student writes	Say this
Treating the test levels as a slogan about order	Answering that acceptance testing is done by developers because it comes last	Acceptance belongs to the user or client, with developers only supporting. The level decides what is tested and against which document, not who holds authority.
Leaving a hole in the equivalence classes	Listing 0 to 17 and 18 to 60 while missing everything above 60	Classes must tile the whole input domain including every invalid class; a missing class is an untested place for a defect to live.
Writing only the boundary itself	Testing only 18 and 60	Test 17, 18, 60 and 61 so that one just outside and one just inside pair up; that is what catches a less-than sign written as less-than-or-equal.
Declaring done below the exit criteria	216 out of 240 gives 90 per cent, yet reporting that it broadly passes	The criterion is a hard line: 95 per cent means 228, so 12 more cases need fixing or retesting. Shipping on broadly passes hands the risk to the user.

07 Homework (paper, section and question numbers)

Secondary ICT prep guide: system testing and acceptance chapter (question types and chapter exercises) — Chapter exercises, questions 1 to 4

08 Exit check (last 5 minutes)

Question	Answer	Decision
Who runs acceptance testing?	The user or client, with developer support	Wrong: reteach act one, the four owners
216 of 240 pass: how many more to reach 95 per cent?	12 (228 minus 216)	Wrong: reteach act three, target times total then subtract
15 of 18 branches covered: what is the coverage?	83.3 per cent, 3 branches missing	Wrong: reteach act two, white-box coverage

09 Tutor FAQ

How does this connect to S5 databases and S4 spreadsheets?

System testing is the last gate. S4 spreadsheet formulas and S5 normalisation teach how to build a system; this chapter teaches how to prove it works. Case design uses the ranges and conditions met in S4 maths work.

Is 45 minutes enough for four test levels?

Yes, provided act one takes only ten minutes. Four boxes and one owner line settle the levels, and the time belongs to equivalence classes in act two and exit criteria in act three, because those are where the arithmetic lives.

Are 240, 216 and 24 real project figures?

No, they are teaching numbers chosen to divide cleanly and to be done mentally. Real pass rates and defect densities vary widely, and the method and direction of the judgement are the point.

Find a tutor — free Tutor Registration →

A TutorZone lesson-prep guide. Written by the TutorZone team; questions and lesson flow are original and are not taken from any school or publisher material.