The Sciences

In India, IITs' Decision to Use Tenure to Improve Research Could Backfire

It may not make sense to expect that a Western solution can address a problem with uniquely Indian pathologies.

A PhD is a minimum requirement for a career with tenure in academia that centres on research. In October this year, the Council of IITs, India’s premier centres of engineering education and research, resolved to adopt a tenure-track system in the appointment of new faculty members. Once a faculty member receives tenure, she will have a permanent job at the institution.

Whether Indian institutions should have a tenure process at all has become a topic of debate.

The road to tenure

An aspiring faculty member usually completes their PhD by the time they are about 28 years old. After this point, there are requirements for further research experience. One of these is one or more postdoctoral fellowships, typically of two or three years each. In the biological sciences, four to five years spent in a single postdoctoral appointment is not uncommon. In other fields, particularly in engineering, time spent in industry is also okay.

Once a candidate has accumulated sufficient experience, they apply for academic jobs. They are now anywhere between 30 and 36 years old, sometimes older but almost never younger. Selection committees short-list candidates, sometimes picking only four or five people to interview from a pool of 400 or more applicants.

After a candidate is interviewed and has been offered a position, she joins a department in a so-called tenure track position – if the institution is serious about the tenure process. Under usual circumstances, she is likely to be 32-38 years old, sometimes older.

And then there is tenure itself. This is the last hurdle an academic scientist will face. After about six years, extendable for circumstances such as the birth of a child or debilitating illnesses, senior scientists both at the institution and outside will evaluate the newly minted faculty member on their contributions since their appointment. The evaluation criteria will favour research activities, and typically account for the papers or books the candidate may have published, the students they have supervised and the grant money they have brought in. Some evaluators also include additional components such as teaching and service responsibilities (e.g., membership of institutional committees). Proven professional or personal misconduct of any sort would disqualify a candidate from being considered for tenure.

The tenure system is not universally applied. Only a few academic institutions in India enforce tenure with the same rigour as a similar process in a high-profile American university, but this number is increasing. The IITs may soon be among them – but not, it seems, without a debate.

The principal argument against tenure is the varying quality of institutional support available to young faculty members. With honourable exceptions, such support in India has fluctuated between just about adequate to actively obstructionist. As a result, young researchers are often not as productive as they’d have to be to qualify for tenure in a typical high-quality Western institution.

It may not even make sense to expect that a Western solution can address a problem – of improving research output and quality, broadly speaking – with uniquely Indian pathologies.

The argument in favour of tenure is that the system has successfully extracted high volumes of good-quality research from faculty members in institutions in the West.

Traditional and progressive criteria

This background places a recent study, about how scientists are assessed for promotion, in context. The study’s authors examined academic criteria for promotion and tenure in biomedical sciences faculties in 146 universities around the world. They chose 12 criteria, split into the categories ‘traditional’ and ‘progressive’. The traditional criteria are peer-reviewed publications, authorship order, journal impact, grant funding and national or international reputation. The progressive criteria are citations, data sharing, publishing in open-access mediums, registration of research, adherence to reporting guidelines, alternative ways for sharing research and accommodation for extenuating circumstances.

The researchers started with the premise that academic scientists adapt the way they produce and publish their research based on how their work is going to be evaluated. According to the paper carrying the study’s results, “Reward systems within universities typically incentivise quantity of publications and novelty of findings rather than the reliability, accuracy, reproducibility and transparent reporting of findings.” The researchers stress that “inappropriate incentives can inadvertently contribute to research waste, with billions of dollars invested in non-usable research.”

Ensuring administrators implement the right incentives is particularly important given that the stakes in biomedical research are high.

Among other things, the researchers found that most institutions placed peer-reviewed publications at the top of their criteria; some even specified a number or reasonable range expected: from a modest 1 to as high as 53 over a 10-year period. (A smaller number of publications could be justified if they appeared in journals with a higher impact factor.) The ability to attract external funding ranked second.

However, only three of the seven progressive criteria were present in any significant way. These included, in order of importance, adjustments to be made for special circumstances, citations of research and – in a few institutions – openly sharing research and data. According to the paper, “mention of publishing in open-access outlets, registering research, and adhering to reporting guidelines were absent from all institutions.”

Some aspects of this study are unique to the biomedical sciences. Constraints on reproducibility of research are not particularly central to physics or chemistry, or even biology, where reproducing an older experiment is often the first step towards finding a new, unexplored problem.

Further, the financial stakes in biomedical research are often significantly higher than in other fields. Depending on the specific sub-field, scientists must also deal with working with real people as well as sensitive biological samples, as opposed to the inanimate matter that experimental physics and chemistry typically deal with.

And then there’s the constant pressure to produce novel results, and the urge to locate a potentially significant effect even though the study might not have been designed to best locate them. Such factors impose additional burdens on a young scientist hoping to get tenure.

Resistance to change

Then again, it’s not as if scientists don’t know how to fix these problems. For example, they could share the data from their experiments so others can check the results themselves. This also helps other scientists decide which lines of enquiries to pursue, and which to drop. Scientists can also publish their results in open-access journals so that other scientists and people at large can access the fruits of research for free. They can pre-register their research to ensure, among other things, that they will report their results irrespective of the outcome.

But as it happens, institutions around the world appear to be systematically failing at using the tenure system to advance these aims. While most scientists agree that the progressive criteria outlined in the study above count as best practices, it’s also hard to see how they might be more widely practised if the system of evaluation itself doesn’t promote them.

The biomedical sciences are special in some ways, but the study contains lessons that should apply to all of academia. Any scientist supported in their efforts by public funds has a responsibility to explain to the people as to what they’re up to. Indian institutions that are considering installing the tenure system should pay attention to this aspect.

Moreover, recent reports of scientists, including in prominent government labs, engaging in fraudulent research practices suggests that even without tenure, the perverse incentives that lead academics to unethical behaviour still exist. The underlying obsession with numbers – as in publishing X papers in a journal with Y impact factor – profoundly distorts academic sciences, and not just in India.

It would make more sense to judge a candidate for tenure not on the basis of the total number of papers they have published but on the two or three that the candidate considers will make a genuine difference to their field.

Gautam I. Menon is a professor at the Ashoka University, Sonepat. The views expressed here are his own.