Merit versus diversity in NHMRC grants

A more sensible analysis would be based on the sex of the researchers who benefit, not of the grant holder. But the gender skew would not then be so stark – and would not generate so many angry tweets. How boring.

What about diversity?

Many claim that diversity is an important driver of better science. This is based on various artificial laboratory experiments that manipulate team profiles and measure their success on well-defined 'decision' tasks. There is less evidence of a diversity dividend outside the laboratory.

Diversity, in any case, has little to do with Y-chromosomes. Where is the evidence that gender predicts different medical research approaches and frameworks that would lead to a diversity dividend? The wisdom of crowds is real, but it is not about identity-defined diversity. It is about actual diversity.

Moreover, in medical research in particular it is hard to see how diversity is a key driver of excellence. Researchers are specialists with a deep understanding of a narrow topic. They are not a random independent sample of the population brain-storming a question that a social scientist has just lobbed in front of them. Newton did not develop his three laws by consensus and committee. He thought about it alone for most of this life. There was no diversity involved at all. Research success requires you to develop a unique combination of expertise and then be single minded in pursuing your research agenda, regardless of setbacks and scepticism.

Moreover, why does the diversity argument not apply to vocations like primary school teaching, where it is becoming increasingly difficult to recruit males? For example take the profession of psychology in the US where 71% of psychologists are female and, amongst under 30s, 95% are female? It is clear where this trend is heading. And the gender of your teacher or psychologist is surely more relevant than your unseen medical researcher. It's medical research, to improve health outcomes for us all. I don't care about their identity. I only care about the research.

And the elephant in the room? Those who assess grants are far from diverse. And by this I do not mean white and male. These are the survivors of a competitive, decade long vocational process that is as much social and cultural as scientific. The assessors are flattered to be assessing the grants of others, including their betters, and they are only human so they also have their own incentives. With the best will in the world, a researcher in malaria will think that research in malaria is more important than any other research. And researchers who are part of the current consensus will tend to receive more favourable assessments than the maverick researcher who wants to disprove the prevailing norms of the assessors.

Undermining female achievement

Affirmative action inevitably leads to the perception that female achievements are not merit based. This is not just my opinion. Below is an email I received from a very accomplished female data scientist who wasscared to have her name published. She is one of several folk, most of whom were female, who thanked me for challenging this thoughtless NHMRC decision.

"As someone who has received awards designated for women, I can tell you first-hand what a terrible impact they have. Others couldn't help but assume that I'd won them because I'm a woman, not because I deserved them. Worse, I myself didn't know whether I deserved them! Early career researchers in particular do not need additional reasons to doubt themselves.

I have children of both sexes, and I fear for them. I want them to be judged on their merits alone, but that wish seems unlikely to come true in the current climate."

What should the NHMRC have done, if anything?