Ixly Tips: Why an Unanswered Question Is the Worst Answer

Ixly is the Dutch publisher behind a large share of the capaciteitentesten sent to applicants in the Netherlands, and you may never see its name: the tests run in the Test-Toolkit, so that is usually what the invitation says. Most advice about timed aptitude tests tells you to pace yourself and not to waste time on hard questions. Ixly has one rule, printed on its own instruction screen, that changes what that advice should be.

The rule that inverts everything: a blank is a wrong answer

Ixly's instruction screen tells candidates, in plain Dutch, that each question allows 45 seconds and that if they do not enter an answer in time the question is scored as wrong: “Doet u dit niet, dan wordt de vraag fout gerekend.”

Read that carefully, because it is not the same as “there is no penalty for guessing”. On a test like the Criteria CCAT, a blank scores zero and a wrong answer scores zero, so guessing is free. Here the vendor has gone further and removed the blank altogether. Leaving a question unanswered and answering it wrongly produce the identical mark, which means hesitation is pure loss: you spend the time and get the wrong-answer outcome anyway.

The practical habit that follows is specific. Commit your best answer at around 35 seconds, whatever state your reasoning is in, and move on. Not at 44 seconds, because the clock will occasionally beat you; not at 20, because on an adaptive test the spare time is worth spending.

Walking away does not pause anything

The second Ixly-specific rule is the one nobody expects. If you close the test in the middle of a question, the question's clock keeps running. When you come back, the time has moved on and so has the test: you resume on the next item, not the one you left. The item you walked away from is scored, unanswered, as wrong.

That has an obvious implication for anyone thinking about a second window, and a much more useful one for everybody else: sit the test somewhere you will not be interrupted. On most online assessments an interruption costs you some seconds. On this one it costs you the question.

It is adaptive, so half right is the designed outcome

The Adaptieve Capaciteiten Test is a genuine computerised adaptive test rather than a fixed paper with adaptive branding. Ixly publishes the specification: it starts deliberately just below average difficulty so your first item is likelier to go well, chooses each next item by how much information it expects to gain, re-estimates your ability after every answer, and stops when the standard error of measurement drops below 0.39. After the second calibration that takes between 10 and 17 items per subtest.

Two things follow. First, the number of questions you get is not a signal: a shorter run means the test reached a confident measurement sooner, not that you did badly. Second, and more usefully, your raw percentage tells you almost nothing. The test escalates until you start missing, so a strong candidate and a weak one both finish somewhere near half right on completely different material. What carries information is how hard the questions got before that happened.

That is why practising on easy items is close to worthless here. Practise at the difficulty where you start getting things wrong, because that is the band your score is built from.

Three item formats worth meeting in advance

Cijferreeksen: watch for two series in one row

A row of numbers with one replaced by a question mark. The rule families Ixly documents include a constant multiplier, a step that changes by a constant amount, and, the hardest one, two series interleaved through a single row so the pattern skips a cell. On an interleaved item the instinct to read consecutive gaps produces nonsense, and the only way through is to read every other number. The gap is also not always at the end of the row, so check where it is before you start.

Figurenreeksen: expect both a grid and a series

Ixly's product page for Abstract Reasoning describes progressive series where you choose the right successor. Ixly's own ACT manual, for the same item bank, shows a 3x3 grid with eight cells filled, the bottom right empty, and four options labelled A to D. A Dutch preparation vendor lists both. The most likely reading is that the bank contains both, so prepare for either. On a grid, the thing to find is which properties appear exactly once per row and once per column.

Verbale Analogieën: you pick two words, not one

This is the format that catches people, because nothing else works like it. Two words are shown together in a square, linked in some way. You choose from six options exactly the twothat are linked to each other in the same way, and the mark is for the pair. Ixly's own example: walking is done with your feet, reading with your eyes, so the answer is the pair.

There are fifteen possible pairs among six options, so a blind guess is worth about a quarter of what it would be worth on a four-option question. Work out the relation from the two given words first, name it to yourself, and only then look at the options.

Cognitive Ability Pro is a different test, not a shorter ACT

Ixly sells both, and you may be invited to either. The ACT is three consecutive blocks in a fixed order. Cognitive Ability Pro is about 15 minutes with no blocks at all: eleven question types interleaved in random order, so the type changes almost every item.

That switching cost is the thing worth preparing for, and it is trainable. The most common loss on a mixed test is carrying the previous item's method into the next one. Before you start solving, name to yourself what kind of item you are on. It costs a second and it stops you applying a number-series instinct to a word analogy.

One honesty note: Ixly publishes neither the names of the eleven types nor how many items the test contains. Four types are named by a Dutch preparation vendor, and anyone claiming to know all eleven is guessing.

Reading the result: a 5 is not an onvoldoende

Ixly reports four numbers: a sten from 1 to 10, a T-score, a percentile, and an IQ score with a mean of 100 against the Dutch working population. Each comes with an 80% confidence interval, because one sitting measures you to within a range rather than to a point.

The sten is the one that causes trouble. It has a mean of 5.5and a standard deviation of 2, so 5 and 6 are the middle bands and roughly 38% of the population sits in them. Ixly's manual addresses this head on because the confusion is predictable: on a Dutch school report a 5 is a fail, and candidates read a sten 5 the same way. It is an average score. It is not an onvoldoende.

It is also a sten and not a stanine. A stanine runs 1 to 9 with a mean of 5. If a document uses the two words interchangeably, it is not describing the scale you were measured on.

Some Ixly customers are not hiring anyone

Worth knowing if you are trying to work out what an Ixly invitation means. Ixly's largest published deployment by volume is UWV WERKbedrijf, at 10,000 to 15,000 assessments a year, and that is career guidance for jobseekers rather than screening for a job. So is USG Restart's outplacement work. Transavia and Thebe use Ixly for 360 feedback on existing staff.

The customers where Ixly is genuinely used for selection, according to Ixly's own case studies, are Dienst Justitiële Inrichtingen, Nederlandse Spoorwegen for external traineeship applicants, Sogeti, and Sectorinstituut Transport en Logistiek for driver recruitment. If you are being assessed as part of a career or reintegration programme rather than an application, the stakes are genuinely different and nobody is comparing you against other candidates.

And one thing Ixly is not

Ixly is not the TMA method. TMA, Talenten Motivatie Analyse, is a separate instrument from a separate publisher. Both are Dutch, both are based in Utrecht, several Ixly products use the word Talent in their names, and Dutch practice-test aggregators list the two side by side as interchangeable assessment bureaus, which is enough to confuse anyone. Ixly's own catalogue contains no TMA product and its about page never mentions TMA. If you have been told you are sitting a TMA, practising Ixly's tests is not what you need.

What to do this week

Sit one full run of whichever test you have been invited to, timed, uninterrupted, with paper beside you and no calculator. Then look at two numbers: how many items expired, and how many arrived on time and were wrong. Those are different problems and they need opposite practice. Expired items are a pacing habit and fixable in a session. Wrong answers are the reasoning, and the way to move those is to work at the difficulty where you start missing rather than at the difficulty where you feel comfortable.

Our Ixly practice tests run at the real 45 seconds an item, score an unanswered item as wrong the way Ixly does, and cover Cijferreeksen, Figurenreeksen, the two-answer Verbale Analogieën, Cognitive Ability Pro, Stroomdiagrammen and the WPV. They are in English while the real test is most likely to be in Dutch, which transfers almost completely on the number and figure papers and least well on the word analogies.