Computer-adaptive CAT needs committed work

Updated on: Tuesday, June 23, 2009

Ever since the announcement that the CAT or Common Admission Test will be computer-based this year, speculation has been rife over the format of the exam.

“There are two kinds of computer tests – linear and computer-adaptive,” says S. Balasubramaniam, Director of T.I.M.E (Triumphant Institute of Management Education), Chennai.

Since the number of test-takers is greater than the number of computers available, students take the test in batches, or slots. In the case of linear tests, “all students taking the test in the same slot get the same paper,” says Mr. Balasubramaniam, adding that “different slots will be given different papers, with the process of normalisation making the scores comparable.”

An example of a linear test is BITSAT, the entrance exam for the Birla Institute of Technology and Science (BITS).

A computer-adaptive test, on the other hand, varies from student to student in the same batch. Essentially, the computer adapts the test to match the aptitude of the test-taker.

The scoring system is based on the difficulty level of the questions attempted. “Everyone will start with a medium level of difficulty,” says Shwetha Rajagopal, Product Manager of TOPGRE.

“The difficulty level will vary depending on how well you answer the previous questions and the more questions you solve with greater difficulty, the higher you will score.”

Usha S., who is a faculty member at The Princeton Review, concurs. “If the student answers the question right, the difficulty level goes up. The scoring depends on the number of questions answered right, the difficulty level of the questions answered and the number of unanswered questions.”

The GRE (Graduate Record Examination) and the GMAT (Graduate Management Admission Test) are both internationally administered computer-adaptive tests. The GRE became computer-adaptive in 1992, with the GMAT following suit in 1997.

Computer-adaptive tests are proven to give an accurate assessment of a candidate’s capability. “It is a statistically sound model and is reflective of the student’s ability,” says Mr. Balasubramaniam. He believes that it is also free of malpractice since people sitting at adjacent computers will be answering different questions.

Shwetha Rajagopal of TOPGRE says that this form of testing has the advantage of no negative marking, although “you cannot go back to your previous questions if you want to change an answer.”

Shreya Srinivasan, a student who recently wrote the GRE, says that she found the test quite challenging because “you have to spend a lot of time on the first question and make sure you answer it correctly, which puts a lot of pressure on you in the beginning.”

She feels that it is critical to try answering the first 10 questions correctly, since this will raise the difficulty level of the test significantly, which will result in a higher score. She also emphasises the importance of taking as many mock computer-adaptive tests as possible.

Ms. Srinivasan advises other test-takers to avoid analysing whether the difficulty level of the questions is increasing while taking the test. “Even though you want to know how you’re doing, it is better not to think about it too much, since you’ll only be adding to your stress,” she warns.

Ms. Usha adds that “the exam is time-dependent, so it is important to answer all the questions.”

In terms of the CAT, while it is unlikely that the content of the test will change significantly, a computer-adaptive testing system will present a whole new array of challenges for students.

Ultimately, just like with any other exam, the key to tackling a computer-adaptive test is commitment, diligence and preparation.

The Hindu

Computer-adaptive CAT needs committed work

More Education news