Since the inception of this blog, we have tried to provide strategies to help others conduct research more rigorously and transparently. As we learn new things along our travels, we identify gaps we can address and we target those areas. Writing for this blog also helps us track and consolidate our own learning. But perhaps the strongest motivation for these labours is that we believe science should be done well. It may seem obvious, but why is it important to do good science?

Because we have a formal responsibility to do good science

In many countries, the conduct of science is governed by a national code of research conduct. These codes outline the principles that underpin an honest and ethical research culture, and responsibilities of the research community to ensure rigour, credibility and trust in the research endeavour.

In Australia, we are governed by the Australian Code for the Responsible Conduct of Research (2018). The Code articulates eight broad principles of responsible research conduct (top three: honesty, rigour, transparency) and 29 responsibilities of institutions and researchers to uphold these principles. Related documents provide specific guidance in other areas, including managing data and information in research, publishing and disseminating research, and describing conflicts of interest. National codes of research conduct also apply in the UK, Canada and the Netherlands. Under these codes, all scientists and researchers are responsible for ensuring that research is conducted rigorously, ethically and with integrity.

Because good science should be the only kind of science

However, even if codes of research conduct were not formally documented, good science just should be the only kind of science. This is a view held by many working scientists and is arguably the strongest motivation (beyond curiosity, fame or money) behind some of the most influential scientific endeavours. We might think of examples such as the search for evidence of the expansion of the universe, the discovery and mass production of penicillin, and mapping the human genome. In each of these examples, people have sought to answer big questions by applying the most rigorous methods of their time.

But it is clear that the view that all science should be good science can’t be tested by science. We won’t find what "good science" means in a test tube. Although there are many horrible examples of how science can be abused (e.g. the Nazi experiments in human physiology, and the eugenics movement), no experiment can tell us why science should be done with respect and integrity, having regard for the intrinsic value of human life. Thus, not only is it clear that we are obliged to uphold research integrity and ethics, it is also clear that such values and duties exist objectively (i.e. they are not a matter of personal or collective preference).

We could summarise that:

1. Objective moral values and duties exist.
2. Good science should uphold objective moral values and duties.

where (1) is necessary for (2). Perhaps what (2) looks like in practice is something to be discovered or worked out. But this is open to discussion.

Because good science should be done, even when we don’t feel like it

As scientists, we are called to high standards, and we will all face challenges and disappointments in our research endeavours. This could be because the cost of doing reproducible research sometimes outweighs the returns, because we feel the pressure to "publish or perish", or because we are simply too tired. This piece was written to all of us, including my discouraged future self. To remind us of our great privilege as scientists in discovering truth, and also of our responsibilities to do science well.