The lure of the P value

The Lure of the P-value versus the Lore of Statistics

If mathematics is the handmaiden of science, statistics is the whore: all that scientists are looking for is a quick fix without the encumbrance of a meaningful relationship. ―Stephen Senn: Dicing with Death: Chance, Risk and Health

And the p-value is the lure. A recent paper by ANU researchers [1] observed a phenomenon called "p-hacking" is not uncommon in published literature. Ironically the quest for the lowest p-value is not what researchers really care about, nor is the p-value what statisticians think much about. But the misplaced fixation with the p-value emanates from the desire to publish in a "high-impact" journal, to further a research career, and obtain research grants.

As a research student, you care principally about your academic discipline. This is a time in your life when you can think deeply and develop your methodological skills that will allow you to make new discoveries in your area of study. The methodological challenges of any academic discipline are particular and formidable, but there are elements that common to many research investigations. In quantitative research, one learns how to develop a research question, how to design an experiment or test that addresses the question, how to avoid bias in the investigation, how to collect data, how to analyse the results, and how to interpret the results of the investigation as evidence for or against the hypothesis of interest.

Statisticians think about research: experimental design, survey design, potential sources of bias and how to avoid them. They think about what counts as evidence for or against a hypothesis and whether the study is appropriately designed to assess the question of interest. It's not at all about getting p-values, it's about doing good science. As Senn went on to say:

Statisticians are engaged in an exhausting but exhilarating struggle with the biggest challenge that philosophy makes to science: how do we translate data into knowledge?

The fun of being a statistician is working with researchers who love their discipline, who can communicate their research interests and who are committed to sound methodology to address the big questions in the sciences and the social sciences.

So don't miss out on what could be a wonderful, meaningful and long-lasting relationship. Engage with the statistical community at the ANU. Request a consultation with a professional statistician at the Statistical Consulting Unit at the beginning of your investigation and start an unforgettable dialogue.