Stakeholders’ concern about frequent paper leaks has brought competitive examinations under scrutiny, and an intense debate has begun to change the exam modalities. Amongst various changes on the anvil, the shift from pen-and-paper-based test (PPT) to computer-based test (CBT) is a likely major shift.Undoubtedly, CBT offers ease of exam conduction, evaluation, scalability, and better logistics, but the finer details ought to be explored. Granular analysis of the CBT brings forth the drawbacks associated with it, which may become quite critical in the high stake national level examinations.Technical aspectsGenerally, the efficacy of CBT depends upon the performance of computers, servers, and impeccable network during examinations. In online CBT, the quality of internet connectivity also becomes a concerning factor. Alongside, the uninterrupted power supply and software reliability and security provide basic functionality to CBT. The failure in one or more of these enablers due to any of the unforeseen reasons yields the collapse of the respective examination/test. There have been a series of issues faced by CBT, and the evidence shows that these are still not flawless from both a technological and human intervention perspective. A closer look shows that there have been numerous incidents of lapses in CBT such as power failures, server crash, slow response, freezing of screens, responses not saved, auto logout and sudden computer shutdowns, mouse/keyboard malfunction, system replacement during exams, incorrect questions/options display, unauthorized server access, remote access cheating, delayed commencements, navigational errors, accidental submissions, inadequate technical support, etc.A few real instances from last few years are, the CUET-UG examination – 2026 suffered technical glitch at a few examination centres and retest was conducted for the affected candidates; in JEE Main – 2026, the crashing of systems, difficulty in login and computer shutdowns were reported at some examination centres; the CBT of UGC-NET of 2024 was redone due to alleged paper leak; CBT of SSC Selections – 2022 suffered server slowdowns, delayed logins and technical glitches at some centres and the re-test was done for affected candidates, etc. As a result, the number of candidates seeking redressal of their grievances of CBT keeps increasing, and the organizers get burdened with litigations, and enquiries.Besides the technical aspects, there are also challenges emanating from the implications of digital proficiency of the examinees. The prevailing digital divide limits the computer handling exposure of the students from rural / economically weaker sections and puts them at a loss as compared to those proficient in the use of computers; thus, the digital proficiency predominates subject knowledge for certain examinees. Such candidates are subjected to added pressure of adapting to on-screen reading instead of reading of printed question papers for answering, increased test anxiety and fear of technical failure, which affects their performance. Quite often, the candidates find it tiring to read on screen and scrolling through the long paragraphs on different pages on screen makes it much more cumbersome. Also, flipping the questions is not easy in CBT as compared to PPT, as the former requires switching screens and later has all questions available at a glance in printed form. CBT also restrict scribbling, underlining, annotation, and rough work as compared to PPT. Thus, the answering and answer reviewing process is constrained in the CBT.From an infrastructure requirement point of view, the costs associated with setting up computer facilities, software updates, networking, servers, suitable ambience, cybersecurity protocols, reliable power supply / back up, securing from unauthorized access, and continuous maintenance are quite substantial and do not make CBT economical. Usually, the requirement of a large number of computer centres for conducting CBT results in varying experiences for each examinee due to the variations in respective computer specifications, screen resolution, quality of keyboard/mouse, and hardware performance, yielding no level playing field.Further, the remote proctoring has perpetual limitations in respect to identity verification, AI assisted cheating, using hidden devices, breach of individual privacy due to continuous audio-visual overseeing/recording. Sometimes, genuine concerns get raised regarding likely misuse of the collected biometric data, audio-visual & keystroke recordings, personal information data storage & retention.Assessment aspectsCBTs in general have inherent limitations of asking questions with objective type answers instead of long descriptive answer type questions. CBT favour multiple choice questions (MCQs) with options like true/false, match-making, etc. This restricts the higher-order capability assessment in the context of creativity, originality, analytical ability, ability to explain & argumentation, use of illustrations, synthesis, practical competencies, research aptitude, etc. Thus, the purpose of assessing knowledge, understanding, analysis, reasoning, application, reasoning, communication skills, originality in thinking, etc. gets defeated to certain extent in CBT. However, this may not be impacting much for the objective-type question-based examinations meant for merely putting examinees in order of merit, provided it is devoid of technical failure, completely secure, and no governance failure occurs.The CBT being conducted in a number of sessions instead of single shift PPT for all examinees does call for normalization as the questions asked in every session keep changing. The idea of fetching questions from a question bank having questions classified as easy, moderate, and difficult does not augur well due to the variations in the understanding and problem-solving abilities of the appearing candidates. For example, a question could be difficult to someone while it could be easy or moderate for the other or vice-versa. Mathematical normalization may be technically fair, but it cannot be used to equate the unequals (examinees exposed to different questions in different sessions make them unequal). Although, it may not be possible to argue against mathematical normalization exercise technically, the variable assessment yardsticks in different CBT sessions make it unethical for inter-se merit list preparation. At the same time, there is an absence of a physical record of the performance of examinees in the test, which makes CBT vulnerable to the allegations of manipulations by altering the databases in the custody of the conducting body.Way forwardIn light of the above, a deeper contemplation is required for choosing between CBT and PPT, specially in high stake examinations where the failures have massive impact on the examinees simultaneously. Due consideration must be given to the equality of opportunities in reference to digital proficiency and testing conditions. The primary objective should be to assess the domain knowledge and competency of examinees fairly without any bias towards technology usage. A deeper thinking is required for evaluating equivalence of multiple-session CBT with single-session PPT from an equality of opportunity viewpoint.With the growing awareness & concerns of the examinees over the credibility of examinations, all efforts should be directed towards restoring faith in the examining bodies by designing fair, ethically correct, and transparent assessment tools following the same yardstick. Most importantly, there should be complete transparency to the extent that the examinees can verify their performance after the examination and no panic is felt at the time of result declaration. Timely interventions for bringing in transparency and equal treatment of all would not only enhance accountability but also eliminate uncertainty and anxiety engulfing results, thereby strengthening trust in the examinations.
Disclaimer: Views expressed above are the author’s own.