The Graduate Management Admission Test (GMAT, pronounced G-mat) is a computer adaptive standardized test for measuring aptitude to succeed academically in graduate business studies. Business schools commonly use the test as one of many selection criteria for admission into an MBA program. It is given at various locations in the United States, Canada and around the world. Throughout North America and in many international locations, the GMAT is administered only via computer. In those international locations where an extensive network of computers has not yet been established, the GMAT is offered either at temporary computer-based testing centers on a limited schedule or as a paper-based test (given once or twice a year) at local testing centers. As of 2007, the fee to take the test is U.S. $250 worldwide.
The exam measures basic verbal, mathematical and analytical writing skills that the examinee has developed over a long period of time in his/her education and work. Test takers are given 3.5 hours to answer questions in each of the three tested areas, and there are also two 10-minute breaks; in general, the test takes about four hours to complete. It does not measure specific knowledge of business, job skills, or subjective qualities such as motivation, creativity, and interpersonal skills. If a test taker’s first language is not English, he or she may still perform well on the exam; however, the GMAT exam may not accurately reflect the abilities of someone whose first language is not English. Business Schools with a high proportion of non-native English speaking students tend to have a lower average GMAT score.
Scores are valid for five years (at most institutions) from the date the test taker sits for the exam until the date of matriculation (i.e. acceptance, not until the date of application). The Stanford University Graduate School of Business website offers a “test results calculator”  useful for determining the expiration date of test scores.
The maximum score that can be achieved on the exam is 800, and the 2005/2006 mean score was 533.
The verbal section consists of 41 multiple choice questions, which must be answered within 75 minutes. There are three types of questions: sentence correction, critical reasoning and reading comprehension. The verbal section is scored from 0 to 60 points with a current mean of 27.3/60.
This tests grammar and expression. Sentence correction items consist of a sentence, all or part of which has been underlined, with five associated answer choices. The test taker must choose the best way of rendering the underlined part. This question type tests the ability to recognize standard Written English. The task is to evaluate the grammar, logic, and effectiveness of a given sentence and to choose the best of several suggested revisions. Choice (A) repeats the original; the other answer choices vary. It tests the ability to recognize correct and effective expression. It follows the requirements of Standard Written English: grammar, word choice and sentence construction. The goal is to choose the answer that results in the clearest, most exact sentence and does not change the meaning of the original sentence.
This tests logical thinking. Critical thinking items present an argument that the test taker is asked to analyze. Questions may ask test takers to draw a conclusion, to identify assumptions, or to recognize strengths or weaknesses in the argument. It presents brief statements or arguments and ask to evaluate the form or content of the statement or argument. Questions of this type ask the examinee to analyze and evaluate the reasoning in short paragraphs or passages. For some questions, all of the answer choices may conceivably be answers to the question asked. The examinee should select the best answer to the question, that is, an answer that does not require making assumptions that violate common sense standards by being implausible, redundant, irrelevant, or inconsistent.
This tests the ability to read critically. Reading comprehension questions relate to a passage that is provided for the examinee to read. The passage can be about almost anything, and the questions about it test how well the examinee understands the passage and the information in it. As the name implies, it tests the ability of the examinee to understand the substance and logical structure of a written selection. The GMAT uses reading passages of approximately 200 to 350 words. Each passage has three or more questions based on its content. The questions ask about the main point of the passage, about what the author specifically states, about what can be logically inferred from the passage, and about the author’s attitude or tone.
The quantitative section consists of 37 multiple choice questions, which must be answered within 75 minutes. There are two types of questions: problem solving and data sufficiency. The quantitative section is scored from 0 to 60 points and the current mean score is 35.0/60.
Most international MBA programs take only the quantitative section into account, as the degrees they offer will not be taught in English. These areas normally demand a higher quantitative score and ignore the verbal sections.
This tests the quantitative reasoning ability. Problem-solving questions present multiple-choice problems in arithmetic, basic algebra, and elementary geometry. The task is to solve the problems and choose the correct answer from among five answer choices. Some problems will be plain mathematical calculations; the rest will be presented as real life word problems that will require mathematical solutions.
- Numbers: All numbers used are real numbers.
- Figures: The diagrams and figures that accompany these questions are for the purpose of providing useful information in answering the questions. Unless it is stated that a specific figure is not drawn to scale, the diagrams and figures are drawn as accurately as possible. All figures are in a plane unless otherwise indicated.
This tests the quantitative reasoning ability using an unusual set of directions. The examinee is given a question with two associated statements that provide information that might be useful in answering the question. The examinee then must determine whether either statement alone is sufficient to answer the question; whether both are needed to answer the question; or whether there is not enough information given to answer the question.
Data sufficiency is a unique type of math question created especially for the GMAT. Each item consists of the questions itself followed by two numbered statements. The examinee must decide whether the statements — either individually or in combination — provide enough information to answer the question.
- (A) If statement 1 alone is sufficient to answer the question, but statement 2 alone is not sufficient.
- (B) If statement 2 alone is sufficient to answer the question, but statement 1 alone is not sufficient.
- (C) If both statements together are needed to answer the question, but neither statement alone is sufficient.
- (D) If either statement by itself is sufficient to answer the question.
- (E) If not enough facts are given to answer the question.
Analytical Writing Assessment
The Analytical Writing Assessment (AWA) section of the test consists of two essays. In the first, the student must analyze an argument and in the second the student must analyze an issue. Each essay must be written within 30 minutes and is scored on a scale of 0-6. The essay is read by two readers who each mark the essay with a grade from 0-6, in 0.5 point increments with a mean score of 4.1. If the two scores are within one point of each other, they are averaged. If there is more than one point difference, the essays are read by a third reader.
The first reader is Intellimetric, a proprietary computer program developed by Vantage Learning, which analyzes creative writing and syntax of more than 50 linguistic and structural features. The second and third readers are humans, who evaluate the quality of the examinee’s ideas and his or her ability to organize, develop and express ideas with relevant support. While mastery of the conventions of written English factor into scoring, minor errors are expected, and evaluators are trained to be sensitive to examinees whose first language is not English.
Most business schools don’t weigh the AWA as heavily as the verbal and quantitative sections of the test. Some schools ignore the AWA altogether.
Each of the two essays in the Analytical Writing part of the test is graded on a scale of 0 (the minimum) to 6 (the maximum):
- 0 An essay that is totally illegible or obviously not written on the assigned topic.
- 1 An essay that is fundamentally deficient.
- 2 An essay that is seriously flawed.
- 3 An essay that is seriously limited.
- 4 An essay that is merely adequate.
- 5 An essay that is strong.
- 6 An essay that is outstanding.
The “Total Score”, comprising the quantitative and verbal sections, is exclusive of the analytical writing assessment (AWA), and ranges from 200 to 800. About two-thirds of test takers score between 400 and 600. The score distribution resembles a bell curve with a standard deviation of approximately 100 points, meaning that the test is designed for 68% of examinees to score between 400 and 600, while the median score was originally designed to be near 500. The 2005/2006 mean score was 533.
The quantitative and verbal sections comprise a computer-adaptive test. The first question may be difficult. The next few questions in each section may be around the 500 level. If the examinee answers correctly, the next questions are harder. If the examinee answers incorrectly, the next questions are easier. The questions are pulled from a large pool of questions and delivered depending on the student’s running score. These questions are regularly updated to prevent them from being compromised by students recording questions.
The final score is not based solely on the last question the examinee answers (i.e. – the level of difficulty of questions reached through the computer-adaptive presentation of questions). The algorithm used to build a score is more complicated than that. The examinee can make a silly mistake and answer incorrectly and the computer will recognize that item as an anomaly. If the examinee misses the first question his score will not necessarily fall in the bottom half of the range. However the first 5 questions are important as a whole because they go a long way to determining the score potential.
Also, questions left blank (that is, those not reached) hurt the examinee more than questions answered incorrectly. This is a major contrast to the SAT, which has a wrong-answer penalty. Each test section also includes several experimental questions, which do not count toward the examinee’s score, but are included to judge the appropriateness of the item for future administrations.
Verbal and Quantitative Section scores range from 0 to 60. Analytical Writing Assessment scores range from 0 to 6 and represent the average of the ratings from the two GMAT essays. Because the essays are scored so differently from the verbal and Quantitative sections, essay scores are not included in the total score.
Most schools do not publish a minimum acceptable score or detailed statistics about the scores achieved by applicants. However, schools do generally publish the average and median score of their latest intake, which can be used as a guide.
At nearly all of the top business schools that are commonly listed in popular magazines and ranking services, the scores will average in the upper 600s or low 700s. The Wharton School of the University of Pennsylvania, commonly regarded as one of the top business schools in the US, reports an average score of 713; Harvard Business School, another top tier U.S. business school, reports a 2006 average of 707. Northwestern University’s Kellogg School of Management reports an average GMAT of 700, with approximately 75 percent of enrolled students scoring between 650 and 740. INSEAD, one of Europe’s leading business schools with a highly multinational student body, reports a 2005 average of 705.
It may be possible to overcome a low test score with impressive real world accomplishments, good undergraduate performance, outstanding references, or particularly strong application essays.
History of the Graduate Management Admission Test
In 1953, the organization now called the Graduate Management Admission Council (GMAC) began as an association of nine business schools, whose goal was to develop a standardized test to help business schools select qualified applicants. In the first year it was offered, the assessment (now known as the Graduate Management Admission Test), was taken just over 2,000 times; in recent years, it has been taken more than 200,000 times annually. Initially used in admissions by 54 schools, the test is now used by more than 1,500 schools and 1,800 programs worldwide.
Until the end of 2005, Educational Testing Service (ETS) developed and administered the exam. On January 1, 2006, GMAC transitioned vendors to a combination of ACT Inc, which develops the test questions and CAT software, and Pearson Vue, which delivers the exam at testing centers worldwide.
On June 23, 2008 a cheating scandal was acknowledged by GMAC involving some 6,000 prospective MBA students who subscribed to website ScoreTop.com and may have viewed “live” questions in-use on the GMAT. GMAC has announced severe measures that include invalidating the scores of subscribers, notifying schools who have received their scores, and banning them from future resits of the test. On June 27, GMAC reassured applicants that only those who knowingly cheated using Scoretop’s website would be affected. 
Also, in response to cases of “proxy” test-taking, where students pay somebody else to take the test on their behalf, GMAC is going to be introducing Fujitsu PalmSecure (the palm vein scanning technology) at testing centers this year. Centers in Korea and India will be getting the palm scanning devices first, followed by the United States in Fall of 2008. GMAC plans to have them integrated at all testing centers by May of 2009.
Registration and preparation
The test taker can register in either of the following two ways:
To schedule a test, an appointment must be made at one of the designated test centers. While it is possible to make the appointment even just a few days before you would like to take the test, it is better to schedule a few weeks in advance to ensure an appointment that is convenient for the student.
Companies such as Kaplan Inc., Princeton Review, TestMasters, Veritas Prep, and Manhattan GMAT have different test preparation options available, which may include self-study using GMAT books, classroom GMAT preparation courses (live or online), online preparation courses, or private tutoring.