What is a Qscore?

A qscore is the QuantHub scoring metric.

The test results are scored according to a Bayesian-based scoring algorithm that leverages the IRT methodology, thus creating a qScore that is used throughout the platform. Because the questions vary in difficulty based on the test taker’s performance on the test, QuantHub is able to evaluate test-takers based on their actual skillset instead of simply reporting the number of questions the test taker answered correctly out of the total number of questions asked. A weighted score is calculated based on the relative importance of the various skills assessed, as specified by the hiring manager or recruiter. A Bayesian algorithm is used to score the assessment by making score approximations after each item is answered and tracking the margin of error in the responses after each subsequent score to measure confidence in the score approximation as the assessment taker progresses through the assessment

Picture1

Each assessment is comprised of an individual skill-specific qScore and an overall qScore reported on a 0.0 to 5.0 range. Each qScore indicates the skill level of the individual based on the following criteria:

0.0       Does not meet base level ability.

1.0 – 1.9 Demonstrates basic understanding or general familiarity with data science/engineering terms and concepts.

2.0 – 2.9 Demonstrates conceptual understanding and ability to describe or explain components of data science/engineering tools and methods.

3.0 – 3.9      Demonstrates practical understanding and ability to apply data science/engineering concepts and tools to solve data challenges.

4.0 – 4.9      Demonstrates expert-level understanding and ability to adapt data science/engineering methods and tools to solve various use cases.

5.0     Demonstrates mastery-level understanding and ability to generate new approaches.