← All topics

Manual QA: Midterm Test

A mid-course check: bug reports, test cases, test design, testing types, client-server and DevTools. No SQL.

Question 1 of 27Correct: 0Incorrect: 0

Questions and explanations

Find an explanation for each question here. To test yourself, take the quiz above first.

  1. Which fields are mandatory in a bug report?

    Without steps and expected and actual results, a developer can't understand or reproduce the bug. Most templates include the environment, but trackers often make it optional — both answers count. Blaming someone or writing the fix is not what a bug report is for.

  2. What is wrong with this bug report?

    Summary: Payment doesn't work
    Steps: 1. Pay for the order
    Expected result: —
    Actual result: Error
    Environment: Chrome 129, Windows 11
    Severity: Critical

    A developer can't tell what to fix from this report. A good version: the summary “Card payment: error 500 after clicking ‘Pay’”, step-by-step actions with test data, the expected result “the order confirmation page opens”, plus the error text and a screenshot. The environment and severity are filled in correctly — they are normal bug report fields.

  3. Which bug report summary is the best?

    A good summary answers “where?”, “what?” and “under what condition?” while staying short. “Bug in the cart” and “Nothing works” explain nothing, and a long story with “please check” belongs in the steps and description, not the summary.

  4. Compare the requirement with the system's behaviour. Which statement is correct?

    Requirement: the password must be at least 8 characters long.
    
    Steps: enter the password Qwerty1 (7 characters), click “Sign up”.
    Actual result: the account is created.

    A bug is any difference between the actual and expected result, and the expected result comes from the requirements. If the requirement seems outdated, discuss it with the analyst or product manager, but still file the bug instead of silently accepting the behaviour.

  5. What is the difference between bug priority and bug severity?

    Severity is how badly the bug breaks the system (from a typo to a crash). Priority is how urgently it must be fixed. A typo in the company name on the home page has low severity but high priority.

  6. The “Pay” button doesn't respond to clicks for any user, and there's no workaround. What is this bug's severity?

    A key feature is completely broken for everyone with no workaround — that's the highest severity. Major is a serious failure that has a workaround or doesn't affect everyone. Minor and Trivial are small issues such as uneven layout or a typo.

  7. A developer fixed a bug, but it still reproduces when you recheck it. Which status should you set?

    A typical bug lifecycle: New → Assigned → Fixed → Retest → Closed. If the bug still reproduces on recheck, it is Reopened and sent back to the developer. Rejected means it isn't a bug, Deferred means the fix was postponed.

  8. What is the difference between retesting and regression testing?

    Retesting answers “is this bug fixed?” by repeating the steps from the bug report. Regression answers “did anything nearby break?” by checking functionality that already worked. After a fix, you usually do both.

  9. Which of the following is not part of a test case?

    A test case is written in advance: what is tested, the steps, and the expected result. The actual result only appears when the test is run — it goes into the test run results or a bug report.

  10. The requirements say: “The catalogue page must load quickly.” What is wrong with this requirement?

    A good requirement is unambiguous and verifiable. Everyone understands “quickly” differently, so the tester asks: how many seconds, on what connection, with how many products? Catching this at the requirements stage is testing too — and the cheapest kind.

  11. The “Age” field accepts whole numbers from 18 to 60 inclusive. Which values should you check using boundary value analysis?

    Bugs most often hide at the boundaries: a developer may have written < instead of <=. So you check the boundaries themselves (18 and 60) and the nearest values outside them (17 and 61): the former must be accepted, the latter rejected.

  12. The “Age” field accepts whole numbers from 18 to 60. How many equivalence classes are there for numbers?

    The classes are: below 18 (invalid), 18 to 60 (valid), above 60 (invalid). The system treats every value within a class the same way, so checking one from each is enough, e.g. 10, 35, and 70.

  13. Below is the order status transition diagram. An order can change status only along these three paths — there are no other transitions. Choose the transition that is not allowed and must be blocked by the system.

    1) Created → Paid → Shipped → Delivered — the normal order path
    2) Created → Cancelled — cancelled before payment
    3) Paid → Cancelled — cancelled after payment

    This is state transition testing: you check every allowed transition and that forbidden ones are blocked. An order can't become delivered without being paid and shipped. Forbidden transitions are easiest to check via the API — the UI often simply hides such a button.

  14. In what order is it recommended to run tests?

    There is almost never enough time for testing. Starting with the most important checks finds critical bugs sooner, and if time runs out, only minor scenarios are left untested.

  15. What are the main stages of the Software Testing Life Cycle (STLC)?

    STLC describes the testing process only: first you analyse the requirements, then plan what and how to test, write test cases, run them, and finally sum up the results in a report. Sales, marketing, and user training are not part of testing.

  16. When is testing considered completed?

    Exit criteria are agreed in advance in the test plan: for example, 100% of tests executed, no open critical or blocker bugs, regression passed. Fixing every bug is impossible, and “time is up” or “everything passed” are only parts of such criteria, not the criterion itself.

  17. You check that the cart module works correctly together with the payment module. Which test level is this?

    Test levels: unit tests a single piece of code, integration tests how modules work together, system tests the whole application, acceptance checks that the product suits the customer or user.

  18. You enter −5 into the “Age” field and check that an error message appears. What kind of test is this?

    A positive test checks behaviour with valid data, a negative one checks that the system handles invalid data properly: it doesn't crash, doesn't save garbage, and shows a clear error. A negative test passes when the error appears.

  19. Which checks of this sign-up form are negative?

    Sign up
    
    Email:                [                ]
    Password (min. 8):    [                ]
    Repeat password:      [                ]
    [ ] I agree to the terms
    
    [ Sign up ]

    A negative check submits data the system must reject: an invalid email, a short password, mismatched passwords, unticked consent. Valid data and a long password are positive checks: sign-up must succeed.

  20. Which of these are non-functional testing?

    Functional testing answers “what does the system do”: adding to cart, field validation. Non-functional testing answers “how well does it do it”: performance, security, support for different languages and regions.

  21. A new build has arrived. You need to quickly check the main features to decide whether further testing makes sense. What type of testing is this?

    A smoke test is a short run of the most important scenarios (launch, login, the key feature). If it fails, the build goes back to developers. Regression is deeper: it checks that changes didn't break what already worked.

  22. Which HTTP method is usually used to create a new resource, such as a new user?

    POST creates a new resource. GET reads data, PUT/PATCH update existing data, DELETE removes it. HEAD returns only the response headers without a body.

  23. What is the difference between the PUT and PATCH HTTP methods?

    To change only a product's price, a PATCH with {"price": 25} is enough. With PUT you send the whole product: fields you leave out may be reset. That's a separate check for a tester: what happened to the other fields after PUT.

  24. What do 4xx response codes (e.g. 400, 401, 404) mean?

    2xx is success, 3xx is a redirect, 4xx is a client error (400 bad data, 401 unauthorized, 404 not found), 5xx is a server error (500 internal error).

  25. Where is the new user's data (name, email, password) sent in a POST registration request?

    The request body holds the data the client sends to the server, e.g. {"email": "user@test.com", "password": "..."}. The body format is declared in the Content-Type: application/json header.

  26. Where are the page and limit parameters sent in this request?

    GET /api/products?page=2&limit=10

    Query parameters go in the URL after “?” and are separated by “&”. They usually set filters, sorting, and pagination: page=2&limit=10 means the second page of 10 products. A tester should check boundary and invalid values: page=0, limit=-1, limit=abc.

  27. The API response is correct, but the product page shows a different price. Where is the bug most likely?

    GET /api/products/42  →  200 OK
    
    {
      "id": 42,
      "name": "Headphones",
      "price": 19.99,
      "currency": "EUR"
    }
    
    On the product page: Headphones — €199.90

    The server returned 200 and the correct price 19.99, so the backend and database worked fine. The error appeared in the browser when the frontend formatted the number (e.g. shifted the decimal point). Attach both the API response and a screenshot of the page to the bug report.