Are the Questions Randomised?
Yes, and in more places than the sequence they arrive in. Which items your attempt contains is decided when the attempt starts, the set that comes back is then arranged, and every item arranges its own answer options. That arrangement is kept afterwards, so the review you read is the sitting you actually sat.
The question is asked to settle one thing: whether a second attempt at the same networking certification exam is a second reading of the same paper. That turns on two things together — how an attempt is put together, and how much material the mode you are sitting has behind it.
How a Practice Exam Attempt Is Assembled
-
The Items Are Drawn Rather Than Listed
An exam mode is attached to a pool of material, not to a fixed sequence of questions. Starting a session takes a set out of that pool for that attempt alone. Where the pool holds more than one sitting’s worth of material, that is why two attempts differ in what they ask and not only in the order they ask it. A demo is a short sample, so a second demo attempt may well be the same items newly arranged.
-
The Drawn Set Is Then Arranged
Once the set exists it is shuffled before it reaches you. The draw settles the content of the attempt and the shuffle settles its shape, and the two happen independently, so an item you met in the opening minutes last time is as likely to turn up at the end as at the start.
-
Each Item Arranges Its Own Options
The choices underneath a question are ordered per question and per attempt. Nothing about where an answer sat last time survives into this one, which matters more than it sounds: recognising an answer by the place it occupies is the easiest false confidence to build and the fastest to collapse in an examination room.
-
The Arrangement Is Kept for the Review
The order you were shown is stored with the attempt rather than generated again afterwards. When the review opens, each item is laid out the way it was laid out when you answered it.
Why the Review Shows You the Order You Saw
This is a small mechanism with a disproportionate effect on how useful a post-mortem is. If the options were reordered again for the review, going back over a sitting would mean hunting for your own answer in a list that had moved underneath it, and the one thing you most want to reconstruct, which is what you were looking at when you committed, would be gone. Because the arrangement is held, the item comes back the way it reached you. That is what makes it possible to work out whether you misread the evidence or judged wrongly between two options that both looked defensible at the time.
What Every Randomised Attempt Holds Fixed
Randomisation varies the presentation and not the standard, which is what keeps a run of attempts comparable with one another. The pool the items are drawn from does not change between sittings. Nor does the published objective list they were written against, the item count and allowance for the mode you chose, or the pass mark the attempt is measured against. What does vary is which corner of an objective you happen to meet. An objective covering stub and not-so-stubby areas can produce an item asking which advertisement is filtered at the boundary and, on another attempt, an item that simply hands you a router holding a default route and no external prefixes and asks why. Both come off the same line of the blueprint, and meeting one of them is not the same as having covered it.
Randomisation here is a property of the attempt rather than of a question list: your sitting is assembled at the moment you start it, and preserved as it was assembled so that you can go back over exactly what you sat.
