I’m going to be honest with you. Psychology is in shambles.

Only 30% of research is reproducible.

There is no unified theory in psychology (or even a pair of theories as there is in physics). There are as many theories as there are possums.

Over the 100+ years we have been a “science,” very little technology has emerged. (Though teaching pigeons to peck a lever is a notable exception).

We pretend to be scientific and yet our science has little to show for decades of research.


I think it’s methodology. We attempted to copy physics but failed to transfer some of the most important characteristics of what makes a discipline scientific, including:

  • a focus on frequent replications
  • building and refining existing theories rather than proposing new ones
  • precise tools for measuring constructs

And some things we hoped would transfer well have failed when applied to human subjects, such as

  • preferring convenience sampling over random sample (after all, an atom in geneva is just as good as an atom in portland….not so of humans)

It’s easy to throw one’s hands in the air and surrender our career paths to an easier science, like chemistry of physics. But I’m content being stubborn. I’d rather repair the sinking ship than abandon it.

So how can it be fixed? I don’t have all the answers, but I think it starts with…

  • an adamant and persistent insistence on replication
  • a greater focus on the visual presentation of data (not just reporting p-values)
  • a preference for theory-driven experimentation over sexy headline-making discoveries

And I’m sure there’s more.