Cleaning up research conduct in India

Recently, several papers published from various Indian universities were reported to contain either re-used, duplicated, or fabricated images in research manuscripts1. A total 980 papers have been retracted from India so far, and of them, 33% were retracted due to plagiarism, while 13% were retracted due to image duplication and manipulation.

A sample of papers taken from predatory journals across the world in 2017 found that most articles came from India.

Mukund Thattai, a genomicist at NCBS Bangalore recently looked at the percentage of papers retracted from India, China, and US between 2000 and 2017 using Retraction Watch , an online database of retracted papers. In the informal analysis,Thattai found that around 2005, India’s retraction rate suddenly jumped to double that of the US (Figure 1).

These reports have sparked debates and discussions to understand the rise in misconduct in India.

The impact factor

“I was writing an application for a grant in India, and one of the questions was what is the cumulative impact factor of your publication,” says Karishma Kaushik, a scientist in the Institute of Bioinformatics and Biotechnology, University of Pune. To calculate the cumulation impact factor, one needs to add up the impact factors of all your publications and give the gross number.

Kaushik cites the example of her colleague who works on pea fowl in India, a subject that may not be interesting to a high impact international publisher but could be very relevant to the country's conservation efforts for of its national bird. "But she struggles to find significance for her work because she doesn’t publish in high impact journals.”

It is very difficult for work in many niche fields to become published in very high-impact journals.

Also, the funding and facilities in Indian institutes are poorer compared to western universities, so the science does not progress at similar rates. This leads to a pressure to fudge data to get into the numbers race.

“The question they should be asking is about the impact of your work and how it has advanced your field,” Kaushik says.

Kaushik compares applying for a tenure track position in India to doing so in the US, where factors beyond papers — courses taught, student evaluations, curriculum changes made — also matter. Having a holistic evaluation, including teaching, science outreach, contribution to the institution, such as managing facilities and organising meetings, spreads the importance of a single parameter.

However, Thattai, is sceptical of such evaluations working in India. “If we add points for outreach. What is going to happen is very bad outreach just so people can get points for it.”

Risk factors for misconduct

A recent study 4 found that the pressure to publish was not a significant risk factor for scientific misconduct. However, what did matter was whether a country had legally enforceable misconduct policies and a socio-cultural academic environment where a scientific work is scrutinised by peers, mentors, and society.

Interestingly, countries where publications are rewarded with cash incentives were at a higher risk for misconduct, reflecting a “certain value system which might incentivise negligence and/or misconduct”. Recently, a similar scheme was proposed in India early this year, where students would receive 50,000 and 20,000 rupees upon publishing in “reputed” international and national publications, respectively.

Rahul Siddharthan, a scientist at the Chennai-based The Institute for Mathematical Sciences says enabled by the lack of legal restrictions, some scientists may be more prone to misconduct because of their desire to publish more and in higher impact journals.

One rule to bind them all

“While policies should not dictate the day-to-day activities of a scientists, they do have a role in dictating that lines should not be crossed when it comes to scientific misconduct,” Thattai says.

Recently, the government of India issued a national policy on academic ethics which lays down broad guidelines and norms to ensure ethical research practices.

Additionally, in response to misconduct allegations in its laboratories, the Council for Scientific and Industrial Research (CSIR) formed a committee to draft guidelines on scientific ethics. CSIR director general Shekhar Mande told Nature India that the guidelines will be implemented in all the CSIR labs. A three-member committee looking into the allegations of scientific misconduct at CSIR's Indian Institute of Toxicological Research in Lucknow and other laboratories has submitted its interim report. “Necessary action will be initiated, once the committee’s final report is submitted,” he says.

Another point addressed by the policy on academic ethics is how misconduct should be dealt by institutions. It recommends that “all institutions should set up a standing committee which ensures timely and impartial redressal of all grievances alleged to arise out of policy violations.”

For many of the institutes recently in the spotlight for misconduct, the process of redress and the outcome of investigations seem unclear.

Ethics training

The national policy on academic ethics also states that institutions must adopt suitable ethics documents and run direct ethics training to increase awareness among the staff. While many research institutes in India do have a research ethics course, in most cases, such courses are a token effort. “We do have a research methodology course. But we do not have any regular class, there is just an exam,” says Siva Shakthi, a PhD student at the Indian Institute of Science Education and Research, Trivandrum.

As institutes in India have different scientific ecosystem and finances, coming up with a national solution will be difficult.

However, experts have been seeking urgent changes in how India evaluates science. “When a scientist’s survival is determined by impact factors, ethics becomes the privilege of a few who can afford to be ethical,” says Kaushik.

References

1. Prasad, R. 127 papers from India retracted for image duplication, manipulation. The Hindu (2019) Article

2. Fanelli, D. et al. Testing hypotheses on risk factors for scientific misconduct via matched-control analysis of papers containing problematic image duplications. Sci Eng. Ethics (2019) doi: 10.1007/s11948-018-0023-7