ISTQB K3 Practice Questions
K3 is the level where the exam stops asking what a technique is and starts handing you data to run it on. Roughly 8 of the 40 questions on the real paper are K3, and they are the questions candidates most often run out of time on - not because they are conceptually harder, but because each one takes two or three minutes of actual work rather than ten seconds of recognition.
The eight K3 objectives
Every apply-level question on the real exam maps to one of these. This drill draws across all of them, weighted towards chapter 4 where five of the eight live.
- FL-4.2.1 - Apply equivalence partitioning to derive test cases
- FL-4.2.2 - Apply boundary value analysis to derive test cases
- FL-4.2.3 - Apply decision table testing to derive test cases
- FL-4.2.4 - Apply state transition testing to derive test cases
- FL-4.5.3 - Use acceptance test-driven development (ATDD) to derive test cases
- FL-5.1.4 - Apply estimation techniques to calculate the required test effort
- FL-5.1.5 - Apply prioritization to create a test execution schedule
- FL-5.5.1 - Prepare a defect report
Most wrong answers here are the neighbouring concept. See the 2 concepts this drill separates before you start, or work the drill first and use them to check your misses.
Start the drill
Certified Tester Foundation Level (CTFL v4.0)
Practising K3 questions only: every question in this set matches that filter. There is no clock - a drill is shorter than a real paper and covers one topic, so timing it would tell you nothing. Work at your own pace and submit when you are ready; you get a full answer review when you finish.
- Questions
- 0
- Time limit
- None
- Pass mark
- 65%
- Cost
- Free
No timer on a drill - take as long as you need.
How this level is examined
There are only eight K3 learning objectives in the entire Foundation syllabus, and they are listed above. That is a small enough surface that you can genuinely close it. Six of the eight are techniques with a procedure you can rehearse until it is mechanical: partition the input domain, take the values either side of each boundary, fill in the rule columns, walk the state model. If you can execute those procedures without hesitating, you have secured the marks that separate a comfortable pass from a marginal one.
The two that resist rehearsal are FL-5.1.4 (estimation) and FL-5.5.1 (preparing a defect report). Estimation questions give you historical figures or a set of expert judgements and expect an arithmetic answer, so the risk is a slip rather than a gap. Defect report questions give you a report and ask what is wrong with it - usually a missing element that would stop a developer reproducing the failure, such as the actual and expected results, the test environment, or the steps themselves.
A practical warning about how these are examined: the wrong answers on a K3 question are usually the results of plausible mistakes, not random numbers. A boundary value question offers the answer you get if you take only the valid boundaries, or if you count the boundary and the value beyond it as one. That means an answer that 'looks right' is weak evidence. Work the question, then check which mistake each remaining option corresponds to - if you cannot account for the distractors, you may have made the mistake they were built from.
Concepts this drill separates
Separate these and the distractors stop working, because there is nothing plausible left for a wrong answer to be built from.
Equivalence Partitioning versus Boundary Value Analysis
Equivalence partitioning divides data into partitions whose members should be handled alike and takes one value from each; boundary value analysis tests the values at the edges of those partitions, and can only be used where the partition is ordered.
How to tell them apart: The two are used together, so read what the question asks you to count. A representative from inside each partition is partitioning. For boundaries, check first whether the syllabus's 2-value form (the boundary and its nearest neighbour outside) or the 3-value form (the boundary plus its neighbours on both sides) is being asked for - they give different counts.
Decision Table Testing versus State Transition Testing
Decision table testing derives tests from combinations of conditions that hold at one moment; state transition testing derives them from the sequence of events over time and the states they move the system through.
How to tell them apart: Ask whether order matters. If the same set of inputs in a different order would behave differently, it is a state model. If the outcome depends only on which conditions are true together, it is a decision table.
Worked examples at K3
These 5 come with the answer and the reasoning already shown, so you can see what this kind of question looks like before you start. They are drawn from the questions-and-answers page and are deliberately held out of the drill above, so the drill still tells you something you did not already know.
1. Which module should be prioritized for additional testing based on defect density analysis?
K3Module KLOC Defects Found Authentication 12 30 Search 8 28 Reporting 25 20 Notification 10 12 - A.Search
- B.Authentication
- C.Reporting
- D.Notification
Correct answer: A
Defect density is defects divided by size, which allows modules of different sizes to be compared fairly. Authentication is 30/12 = 2.5, Search is 28/8 = 3.5, Reporting is 20/25 = 0.8 and Notification is 12/10 = 1.2 defects per KLOC. Search has the highest density, so it is the strongest candidate for additional testing.
2. Which requirement presents the highest testing risk and should therefore receive the highest testing priority?
K3Requirement Likelihood Impact Online Payment High High User Profile Medium Low Search Filter High Medium Help Page Low Low - A.Search Filter
- B.Help Page
- C.User Profile
- D.Online Payment
Correct answer: D
Risk level combines the likelihood of a problem occurring with the impact if it does. Online Payment is rated high on both, so it carries the greatest risk level and should be tested first when effort has to be prioritised.
3. Which test case should be executed third when considering both priority and dependencies?
K3Test Case Priority Depends On TC1 3 None TC2 2 TC1 TC3 1 TC2 TC4 2 TC2 TC5 1 TC4 - A.TC3
- B.TC4
- C.TC1
- D.TC2
Correct answer: A
After TC1 and TC2 are complete, TC3 has the highest priority among executable tests.
4. An input field accepts integer ages from 18 to 65 inclusive. Using equivalence partitioning, how many partitions exist for this field?
K3- A.Two
- B.Three
- C.Forty-eight
- D.Four
Correct answer: B
One valid partition (18-65) and two invalid partitions (below 18, above 65). Equivalence partitioning covers the whole value domain, invalid values included.
5. What is the MINIMUM number of test cases for 100% statement coverage and for 100% branch coverage respectively?
K3if (a > 0) { doSomething(); } finish();- A.2 and 2
- B.1 and 2
- C.2 and 1
- D.1 and 1
Correct answer: B
One test with a > 0 executes every statement. Branch coverage additionally needs the false outcome, so it requires two.
Other drills
By cognitive level
- ISTQB K2 Practice Questions
- ISTQB K1 Practice Questions
- What K1, K2 and K3 mean - how the levels map to the exam
By syllabus chapter
- Chapter 1: Fundamentals of Testing - 8 of 40 on the exam
- Chapter 2: Testing Throughout the SDLC - 6 of 40 on the exam
- Chapter 3: Static Testing - 4 of 40 on the exam
- Chapter 4: Test Analysis and Design - 11 of 40 on the exam
- Chapter 5: Managing Test Activities - 9 of 40 on the exam
- Chapter 6: Test Tools - 2 of 40 on the exam
Level and chapter combined
- ISTQB K3 Practice: Chapter 4 Test Design Techniques - the densest block of applied marks on the paper
Full-length papers
- ISTQB Practice Test 1 - 40 questions with answers
- ISTQB Practice Test 2 - 40 questions with answers
- ISTQB Practice Test 3 - 40 questions with answers
- Timed mock exam simulator - a fresh random draw each time