Why didn’t psychology lose its credibility?
Failure to replicate has not reduced interest in or the use of psychological concepts.
By Nina Powell
It’s well known that psychology has failed to reproduce the findings of major field-defining research. Consequently, as The Atlantic put it, “the stuff of textbooks and TED Talks might not be real”. The explanations for psychology’s problem with reproducibility range from the occasional outright fabrication, to more common exaggeration and poor methodology. Remarkably, despite the failure to produce reliable research, psychology as a whole still continues on as though nothing happened. By that I don’t mean that nothing has been done to address this issue — there are major movements to ensure that psychology research holds up to stringent peer review and methodological checks (e.g., pre-registration of predictions, journals dedicated to publishing replications, failed or otherwise, and increasing demands for multi-study papers with a replication check built-in). What I do mean is that the reputation of psychology has remained largely intact. Instead of a declining interest in psychology, we see increasing student numbers at the undergraduate and graduate level, and the continued societal thirst for psychological interpretations for human behaviour. The field seemingly lost none of its popularity and little of its credibility even though more than half of its foundational research doesn’t hold up.
I recently asked students taking my introductory psychology course why psychology is so popular in spite of its failure. The answers were rather heartening. The students pointed out that psychological questions will continue to interest us so long as we are curious about ourselves. Another student argued that failure is interesting — when things fail, we’re motivated to understand and figure things out. Success means that things just become accepted as fact, and then we stop trying to understand. In this sense, psychology’s failure might be a good thing. I agree with the sentiment that failure is interesting and does not, by itself, undermine the project of trying to understand ourselves through psychological research. But I don’t see the current changes in psychological research practice as ones that commit the field to accepting and pushing past these failures in an effort to access deeper and more meaningful questions and understanding.
Psychology needs to drop the defensiveness
In nearly all introductory courses in a psychology curriculum, you will find an opening gambit about what differentiates psychology from other disciplines. You will also find this as part of the standard rhetoric at university open days if you visit the psychology booths — how is psychology different from social work, sociology, philosophy, etc.? The answer that the textbook and the open day representative will typically give is that “psychology is a science!” We’re the same as biology and physics and chemistry — we use the scientific method, which means we are committed to the pursuit of scientific fact.
It is true that psychological experimentation uses the scientific method. Psychologists test predictions based in theory and existing research by measuring something of interest and eliminating confounding factors. Many other considerations go into designing a study to ensure reliable and valid findings, but the central point is that these considerations are a part of an overarching scientific method shared with other STEM fields. Yet just because psychology uses these methods does not mean it is anything like these other fields. Psychology deals with the human subject and human subjects are capricious in a way that objective physical bodies are not. Instead of accepting the separation of psychology from the physical sciences, psychology has been excruciatingly defensive and has insisted it become part of the “hard sciences”.
In rejecting its reputation as a “soft science,” psychology has rejected its origins and is now trying to be something that it cannot be. Psychology was born out of a meeting of philosophy and the natural sciences. Questions about the human condition have existed for centuries before psychology ever existed, yet today, you will find psychologists openly disinterested, and even outright disdainful of philosophy as though it were a weaker subject. You’ll see the same attitude in psychology academic circles towards sociology, anthropology, and other “soft or social sciences”. I’ve heard this belief articulated at major psychology conferences from big name academics. Undergraduate and graduate students are advised to avoid anything associated with “soft science” and to always adopt the empirical methods of the hard sciences. Tragically, colleagues in the “soft disciplines” are reluctant to collaborate with psychology researchers because they are discouraged by that narrative.
Perhaps it is this defensiveness that makes the response to the reproducibility crisis all the more unsurprising. The response was largely to clamp down on methods and take a heavy-handed approach to becoming an even harder science. It was as though psychologists amped up their defensiveness to a whole new level — we can fix this mess by becoming an even harder science! But it’s not going to work, and here’s why.
Psychology is not like the natural sciences
Psychology has defended its scientific reputation over the years by attempting to use stringent scientific methodology that works to eliminate confounds when investigating a variable of interest. But some things which psychologists attempt to understand are inherently confounded. To remove the confounds in many cases is to deconstruct and dilute the subject or variable of interest to the point of meaninglessness. Morality, for example, which is my area of interest, is confounded by time, place, ideology, politics, religion, culture, etc. because it is made by us, for us, and therefore not separate from us in any objective sense. You cannot make morality un-confounded from everything that makes us, “us”, because then it would no longer be morality.
To think that the stricter the methodology the better psychology will fare is to entirely miss the point of what psychology is. And this may explain why psychology has, and will continue, to face issues with reproducibility. A great deal of psychology is inherently a social science and you can’t turn a social science into a hard science because they are different beasts. Psychological research can be informative, it can describe and illuminate what we are, but psychological research is also frustrating. Findings can be local in space and time and dependent on a context that may never repeat, and so the findings can be inherently non-replicable. Psychology will never be as definitive as the hard sciences, which will make psychology less interesting to some and more interesting to others.
Psychology needs to worry less about being a hard science and using more stringent methods, and more about acceptance of how the nature of what we aim to investigate is variable, ever-changing and ever-dependent on other aspects of humanity. A failure to reproduce in many cases is not a failure, it’s just an indication that psychology is not like the natural sciences, and if we can get over that, maybe we can have an honest conversation about what the field can and cannot do.
Why didn’t psychology lose its credibility? was originally published in Cognitive Handshakes on Medium, where people are continuing the conversation by highlighting and responding to this story.
I don't think it is wise for psychology to let go of finding better methods. And to continue to strive to become more rigorous. That much of psychology is nothing more than the weak methodology of social science is also good to keep in mind and to accept that, as much of it is not generalizable, but it is also not wise to stop trying to find out if this is due to confounder, bad statistics or methods, or if this is intrinsically due to conceptual or situational unique factors that make the findings unique to those specific situations. You need to be able to seperate the two to even know if what you claim is true or valid. Just saying, accept that psychology is a social science and therefore intrinsically is very situational and weak and loose construct based is just lazy reasoning. I think psychology should do both. Accept that most of psychology is not better than sociology and anthropology in its methods and constructs but should also strive to find and use better methods and statistics to become better at making any form of models/theories from which actual predictions can be made and followed up on.