Functionality

Item Builder

Creating tasks is the most important part of the testing platform. When creating tasks, you have 37 interaction types at your disposal. These include textual, graphical, file-based, and interaction types integrated with other systems. You can see our sample tasks to explore the wide range of interaction types. When creating tasks, you can combine different types of interactions in a single task.

As a test organizer, you can create tasks only for use in your own tests. To manage tasks, you can group them into work collections.

In an exam center, you can also create public tasks, either for everyone or for test organizers.

A task always has one primary language. PRO package users can also add translation layers in other languages to the task.

Test Builder

As a test organizer, you can compile a test from tasks and decide how many points each task is worth. You can also share the right to modify a test by granting the owner role of that test to a colleague.

Tests can be one-way (tasks solved in a fixed order, with no possibility to return to previous tasks) or multi-directional (tasks can be solved in any order, and you can return to earlier tasks later).

You can set a time limit for taking the test.

The basic test organizer package includes the ability to create tests with a wide range of interactions but a simple structure, making it easy to use for those who do not create tests daily. The PRO package allows you to create more complex test structures with multiple test sections and subsections. Each section of the same test can have a different response format (oral, written, or interview). In test sections with subsections, you can separately determine whether tasks are solved one-way or multi-directional for each subsection. You can also set an individual time limit for each subsection.

As a PRO package user, you can also create tests with multiple sets of tasks. Depending on the test organization, the system can choose a set of tasks randomly, or the test taker can choose which set to solve.

A special case is diagnostic tests, where the e-testing software individually selects the next task as you move on, taking into account the knowledge the test taker has demonstrated in the tasks solved so far.

Registration

As a test organizer, you grant access to a test by creating a test list. You can create multiple test lists for a single test. The same list contains those test takers who generally solve the test at the same time. If you are a teacher and work with specific classes, you can group students by class and then assign the test to the entire class at once. Alternatively, you can generate a link to the test and share it via external channels, so you do not need to add test takers individually, as anyone with the link can take the test.

In centrally organized tests, there are many registration options: a test taker may register themselves, the school may register students, and the exam center may always register test takers. Test organizers can decide which registration method is allowed.

Organizing centrally managed tests

Organizing centrally managed tests and exams starts with describing dates and various settings in the exam center. For a multi-section test, each test section is given its own organizational settings, and test-taking locations are determined—usually schools. In each test-taking location, you can specify the rooms where the test will take place. Once test-takers are registered, they are assigned to test-taking locations and rooms. For oral tests, you can schedule individual start times for each test-taker. Local configurations are usually performed by local administrators at the test-taking locations, but they can also be done by the exam center’s test organizer.

Conducting the E-test

Test takers log in to the system. If they are registered for the test and the test is about to begin, a link to the test is displayed right on the main page.

At least one e-test administrator is assigned to each test room. For a teacher’s test list, you yourself are the administrator, but if desired, you can also assign colleagues as administrators for your test list. The test administrator sees the statuses of all test takers in that room and may have the authority to grant permission to start, remove someone from taking the test, generate passwords if needed, or give extra time.

For security reasons, it is possible to register the computers in the computer lab before the test takers arrive, ensuring the test can only be taken from those specific computers.

Computer-based Grading

Questions may be computer-graded, human-graded, or hybrid.

The automated grading process is based on grading matrices. Each question has its own grading matrix, which includes correct answers (if needed, partially correct or incorrect answers) along with the points awarded. With textual interaction, the grading matrix can include an exact text answer, a regular expression, or a formula that interprets the response. The formula can take into account answers to other questions. For graphical interactions, the grading matrix can include coordinates. In addition to the question-specific grading matrices, there are so-called calculated values, which calculate a result based on answers or points from other questions and may award points or generate feedback.

The grading rules for computer-graded questions are defined before the test is administered, but they can also be supplemented afterward, and the test takers’ results can be recalculated.

In hybrid grading, the system performs computer grading first, and only if it cannot assign points that way does the work go to a human for grading.

Human Grading

In the test organizer’s own tests, tasks that are not computer-gradable are graded by the organizer. As a test organizer, you can also assign other users the right to grade the works in your test list. If you wish, you can assign graders on a per-task basis.

In centrally organized tests, tasks are grouped into grading sets. Each grading set is assigned graders. If a test has multiple grading sets, each work will be graded by multiple graders. Depending on organizational settings, graders may be assigned centrally or by the school.

In the grading set settings, you can specify whether single or double grading is needed. In the simplest case, single grading is used (this is grading level I). For subjectively graded questions, double grading is used — graders I and II, who may be assigned in pairs or separately. In double grading, the result is typically the average of both graders’ points. If these differ more than allowed, the work goes to a third grader (grader III, a more experienced grader). If graders I, II, and III all give different results, grader IV (an expert grader) steps in. In the case of appeals, there is also grader V.

Feedback & Statistics

Detailed feedback and statistics provide very valuable information. For tests created in the exam center or with the PRO package, you can define feedback forms that generate written feedback for both the test taker and the test organizer. In these forms, you can use formulas and functions that take into account the points received and the answers given.

Certificates

It is possible to develop certificate templates that generate certificates for those who pass a centrally organized exam.

Appeals

For centrally organized exams, the organizer can set up the possibility for examinees to submit an appeal about their result. Appeals are submitted directly in the system. A user with the appropriate role must then process the appeal, forwarding the work to expert graders for re-assessment. The expert graders make a recommendation, which may increase, decrease, or leave the points unchanged. Based on the recommendation, the expert committee makes a decision.

Additional Development Work

Our dedicated team is happy to provide additional development work to create new functionality, integrate the system with other information systems, export or import data, or — for a company installation — adapt workflows to a company’s practices, etc.