Assessment glossary

Assessment operations covers the full activity of designing, delivering, marking, and analysing an examination. It is broader than running the exam itself: it includes deciding what the exam is meant to measure beforehand and interpreting the results afterwards.

In a digital system this is usually split into four stages: content preparation, session delivery, marking, and result reporting.

A test blueprint is the table that allocates items across topics and cognitive levels. It expresses assessment intent before any specific item is chosen, so the paper reflects the intended weighting rather than whatever happens to be in the item bank.

A well-built assessment system generates the paper from the blueprint rather than assembling items and checking the distribution afterwards.

An item bank is a store of questions classified by topic, cognitive level, and other attributes, used to generate examination papers. Its value comes from the quality of that classification, not from the raw number of items it holds.

Items that are not correctly tagged cannot be used for blueprint-driven paper generation.

Assessment technology transfer is the process of moving operational ownership of a system from the supplier to the using organization. It covers documentation handover, team training, and a support period until the organization can run exams itself.

Distinct from licensing: the goal is that the receiving organization no longer depends on the supplier to run an exam.

Cognitive level classifies items by the kind of thinking they demand, for example recall, comprehension, application, and higher-order application. It forms one axis of a test blueprint, alongside the content topic axis.

Item difficulty is the proportion of candidates who answer an item correctly. Items that are very easy or very hard contribute little to discrimination. The figure is only meaningful when calculated over a large enough and reasonably homogeneous candidate group.

Item discrimination measures how well an item separates higher-performing candidates from lower-performing ones. Low discrimination usually signals confusing wording or an item that is measuring something other than what was intended.

A rubric is a table describing quality levels for each assessment criterion, used when marking work that has no single correct answer. A rubric is what allows two different markers to reach similar results on the same piece of work.

Essential for written subjects and for the speaking and writing skills in language assessment.

Formative assessment happens during learning and exists to adjust teaching and study. Summative assessment happens at the end of a period and exists to certify what was achieved. The two call for different paper design, frequency, and reporting.

Competency assessment tests the ability to apply knowledge and reason with it, rather than to reproduce studied material. In Vietnam, a number of higher education institutions use competency assessment results as part of admissions.

Featured formats include HSA, TSA, SPA, BCA, QDA, and VACT.

HSA is the abbreviation for the competency assessment administered by Vietnam National University, Hanoi. It is one of the competency assessment formats used by a number of Vietnamese higher education institutions in admissions.

TSA is the abbreviation for the thinking assessment administered by Hanoi University of Science and Technology. The format focuses on reasoning ability and is typically used for admissions into engineering and technology programmes.

VSTEP is the Vietnamese six-level English proficiency assessment framework. The test covers all four skills, listening, speaking, reading, and writing, and is used in a range of academic and employment contexts in Vietnam.

A computer-delivered test is an examination in which candidates respond on a digital device instead of on paper. It allows automatic marking of objective sections, audio capture for speaking, and faster result turnaround.

Distinct from remote at-home testing: a computer-delivered exam can still be sat in a supervised computer room.

An online exam room is the delivery unit inside a digital assessment system: the candidate roster, the assigned paper, the timing, and the session control rules. It is where everything specific to one sitting is configured.

Online proctoring is the set of controls used to protect the integrity of a remote examination. The level applied should match the stakes of the exam: a classroom quiz and an admissions test have very different requirements.

Adaptive testing selects each next item based on how the candidate answered previous ones. It shortens the sitting while still estimating ability, but it requires an item bank that has been properly calibrated first.

An LMS manages the delivery of learning content, tracks progress, and records learner results. Most treat testing as an attached module, unlike an assessment platform where evaluation sits at the centre of the design.