The three main parts of the GMAT are designed to assess the test taker’s ability with analytical skills, quantitative abilities and verbal competence. The first section is a writing assignment, the following two sections are completed by answering multiple choice questions.

The multiple choice sections are made up of a pool of questions ranging from a low to high level of difficulty. In computer adaptive tests (CAT), each correctly answered question will usually be followed by a more difficult one. For every incorrect question the computer will present an easier question next, until the section is completed, so that an accurate assessment of the candidate’s abilities can be made.

The sections are structured as follow:

The GMAT is purposefully designed to assess a candidate’s abilities under time constraints. Keeping within the allocated time limits is crucial and candidates must pace themselves accordingly as there are severe penalties for not completing a section. Test takers have on average just under two minutes for each question in the verbal section and two minutes for each quantitative questions. If you are uncertain about a question, it is generally better to make an educated guess by eliminating answer choices that you are certain are wrong.

The computer adaptive format (CAT) is based on an algorithm that adjusts the level of difficulty of the questions to the ability of the test taker. In the GMAT context, the first few questions are therefore crucial. The level of difficulty of each question is based on whether the previous question was answered correctly or not. A correct answer will yield in a subsequent question that is more challenging than the previous one. The test thus assesses what level of difficulty the test taker can manage, which is reflected in the candidate’s overall score.

Questions in each section appear in random order, i.e. in the AWA section, there is a not pre-determined order as to whether the Issue or the Argument analysis is presented first. Similarly, the different types of questions asked in the quantitative and verbal section may appear randomly, however, they tend to be bundled in a sets in the exam.

A certain percentage (typically approximately one third) of the questions asked in the exam will be trial questions. This means that the test taker takes part in an ongoing assessment of the appropriateness of the level of difficult of certain questions and / or other characteristics. These trial questions are not counted and test takers will not know which questions are experimental and which ones are ‘real’.

The final score of the GMAT is based on three factors: the number of questions answered, the number of correct and incorrect answers and the level of difficulty and other properties of each question. In the CAT format, the questions are weighted by the level of difficulty and other statistical characteristics. The AWA score is not included in the combined 200-800 scale score but is reported separately on a scale of 0-6.