Science of testing

With Zmartests, skills assessment is scientific, not intuitive.

According to the World Economic Forum, skills-based hiring is five times more predictive of job performance than education-based hiring, and twice as predictive as work experience-based hiring. With Zmartests, your candidates can stand out from the crowd by highlighting their skills.

Two illustrated characters shrugging with a large blue question mark between them.

Normalized questions

At Zmartests, question creation is at the heart of our competency-based approach, and is carried out by experts in the field. New questions are tested, rated (difficulty level) and used in test design.

Zmartests questions types:

  • Objective questions that prompt candidates to choose the correct answer from a number of selections, or to provide a word or short phrase to complete a statement (Linguistic tests).
  • Subjective questions which require candidates to structure and formulate an answer and write an essay on a given topic (Linguistic tests).
  • Lifelike simulations that plunge candidates into a virtual experience with the Microsoft Office software. Each exercise validates a skill (e.g. AVERAGE), and the score awarded during grading varies according to the approach used by the candidate to complete the task.

Powerful Benchmark Tools

Zmartests is more than 600,000 tests administered and 12 million questions weighted and marked. All this statistical data is processed by our algorithms and incorporated into our reports and dashboards.

Benchmark your candidates against Zmartests indicators and never make a bad hiring decision again!

A man in a white shirt and dark tie interacts with a tablet next to a detailed dashboard featuring various charts and percentages.
Illustration of a person in a gray suit interacting with blue bar graphs using a puzzle piece.

Adaptive testing (CAT)

These tests automatically adapt to the candidate’s level. We compare them to a project of climbing Mount Everest. All have the same starting point. Many stop at the base camp (beginner), while some go on to the following camps (intermediate and advanced), and only a few reach the summit (expert).

  • These tests start with beginner-level exercises, then the difficulty level gradually increases if the questions are answered correctly.
  • The exercises are randomly selected and will differ from one candidate to another.
  • The duration of these tests varies from 20 to 40 minutes, depending on the candidate’s level (more advanced candidates will progress to more difficult questions).
  • The final score establishes the candidate’s level and his or her position in relation to the population (percentile).

Fixed-item testing (FIT)

Are you looking for a customized test that assesses specific skills? For example, you want to confirm that your candidates are proficient in Excel pivot table functions?

Let our experts assemble tests that mimic the tasks your candidates will have to perform as part of their job. A scale (level of difficulty) is attributed to each test, and the results obtained by your candidates are compared against this scale.

With FIT tests, we can predict in advance the percentage of candidates who will obtain a passing grade. This feature is a major advantage for jobs covered by a collective agreement, or when you need to find the right balance between the level of difficulty and the success rate of a test.

Take advantage of Zmartests’ expertise to better recruit!

A person interacts with a large smartphone screen displaying the ZMARTTESTS interface, surrounded by three smaller screens with similar interfaces.

Établissement d’une cote objective

L’évaluation subjective laisse place à l’erreur car c’est un évaluateur et non la science des données qui décide si un exercice est de niveau débutant ou intermédiaire.

Zmartests réalise uniquement des évaluations objectives basées sur des mesures concrètes. Nous développons les exercices et analysons les scores. Puis, nous utilisons les résultats obtenus pour déterminer le niveau d’un exercice (débutant, intermédiaire, avancé, expert).

Comme les données ne mentent pas, vous profitez d’une vraie évaluation des compétences.

The Dunning-Kruger effect and biases in your selection processes

The Dunning-Kruger effect is a cognitive bias that translates into overconfidence. It corresponds to the tendency of people who are less competent or less familiar with a given subject to overestimate their true abilities, because they are less aware of what they don’t know. This discrepancy between perception and reality is often cited by recruiters. Who hasn’t heard of the candidate claiming to have a good level of mastery of Excel functions, when in fact their knowledge is not up to the task once they’re on the job?

Chart explaining the Dunning-Kruger effect

The Dunning-Kruger effect in numbers for Excel competencies

SELF-EVALUATION
According to a survey of 1,000 *
ZMARTESTS INDICATORS
Based on Excel test results
Beginner11%
of those surveyed rated themselves as beginners
36%
of people tested are ranked as beginners
Intermediate57%50%
Advanced27%13%
Expert6%1%

* Source : https://www.acuitytraining.co.uk/

Stop relying on your candidates’ self-evaluations and start measuring their actual skills.

Jumpstart with your evaluations today!

Open an account and benefit from a simple, accurate and efficient skills assessment process.

Black arrow in a black circle pointing to the right