*Abstract*: Recent events have challenged psychological scientists to
profoundly reconsider our research practices in the service of a shared
goal: to maximize the evidentiary value of our science. Given that there
are multiple possible means to achieve this goal, how do we choose between
them? Proposing a one-size-fits-all set of best practices for the entire
field risks inadvertently marginalizing certain subfields or researchers
who face different sets of constraints and complexities in their particular
research context. An alternative strategy is to start local, focusing first
on what practices we could implement in our own lab that would best enable
us to maximize the informational value of our research. The start-local
approach affords two clear benefits. First, our own research gets better
and it gets better right away—we don’t have to wait for field-wide change
or agreement on new shared standards to reap the benefits of improving our
science. Second, we begin to accrue much-needed information about what
works—and what doesn’t work—in each particular research context, which we
can then share with others to inform the thoughtful development of
field-wide standards that are appropriately nuanced. After discussing these
themes, I describe how I have implemented this approach in my own lab to
construct a set of research practices and supporting infrastructure that
facilitate the goal of “getting it right” (rather than just “getting it
published”). Throughout the talk, I highlight the importance of finding a
balance between the abstract ideals of a perfect science and the practical
realities of limited resources and messy data.
<http://ledgerwood.faculty.ucdavis.edu>