Those little sciences grow up so fast!

Question for the day: is psychology a science? Yes¹. But I think people are made very uncomfortable by psychology, so they’d rather believe otherwise. I will explain.

This recurring line of thought was reawakened during a recent argument conversation with some friends, when one of them implied that the results of psychological research could not be applied to the subject at hand, despite the fact that the research I was referring to was addressed directly toward this same subject. The implication of the comment seemed to be that the results were not applicable because they were based on psychological research. We were not actually talking about the viability of psychology as a science, so what I say from here on out isn’t directed to the people I was having this dustup cool, collected exchange of ideas with. See, this has come up many times in other conversations with other people, so this instance was a trigger to remind me of the whole ball of wax.

It’s never fun to have one’s chosen profession dismissed outright, but I believe one must always be ready to admit, if necessary, that one’s activities may have been based on misguided assumptions. As a psychology guy, I’ve thought long and hard about the validity and viability of psychological science. The results of this thinking follow. Feel free to disagree or tell me I’m a total genius.

First question: Is psychology a science? Yes. I laugh heartily at anyone who says it’s not². Science is a method, not a field of study or a set of results. I could study the multicolored spirit auras over psychic tarot readers’ heads, and if I did it with the scientific method, color-aura-ology would be a science (even if it produced no useful results, but that’s a separate issue entirely). I am acutely aware that many psychologists — especially non-researchers– either avoid or willfully ignore the scientific method, but this is a problem endemic to all scientific fields. There are always some wackos, nut jobs, idiots and charlatans³. Many of them have PhDs.

–> Funny anecdote: I once had an MD friend explain the term “scientist” to his child by saying “I am a scientist. Even Darrin, here, is a scientist… in a way.” Ironic, because I was conducting my third or fourth empirical study at the time, while he had never conducted research in his life (like most MDs). But there are medical researchers, in addition to the legions of nonscientist practitioners, and medicine is a science.

Second question: Is psychology a useful science? This is an open question. I say yes, but this one seems to stick in people’s collective craw more than the first question. How do you know if a science is useful? By their fruits ye shall know them. Of course, like other fields (especially in their infancy), psychology has had its share of frivolous, dead-end, mistaken and even offensive research lines. But it has also had plenty of lines of research producing results with methodological strengths and real-world importance equal to anything in the physical or biological sciences (in some areas, it’s hard to distinguish between psychology and biology, but you know what I mean). The psychological study of memory, perception, sensation, basic cognition, metacognition, learning processes, social cognition, social influence, decision making, neuropsychology, neurobiology, psychopathology, relationships, testing and measurement (and on and on) are being profitably developed, with results that have informed - and sometimes revolutionized - many areas of daily life (usually for the better ;).

To answer the dozens of people who have told me over the years that psychological research is (1) not a science, (2) not a useful science, or (3) cannot produce results relevant to the daily lives of us humans, I say the evidence seems to firmly contradict these statements.

–> Objection: Humans are so Complex! The complexity of human behavior is, of course, a great difficulty to be overcome in psychology. However, those who think biology or physics doesn’t have to contend with similar levels of complexity are seriously mistaken. Cell biology? Genetics? Quantum physics? Paleontology? These are not simple pursuits. The scientific method was developed to help researchers tease apart such complexity, layer by layer. It’s a slow process, but it’s definitely happening. Psychological researchers use the method to tease apart the complexity of human behavior, just as the other sciences do. Complexity is not a reason to throw up one’s hands and dismiss a field of inquiry or its results.

–> Objection: Humans are Constantly Changing! The criticism that psychological reasearch doesn’t apply to human behavior because of human variability, especially the way humans react to being studied, is important, which is why these factors are almost always considered in research. There are many methodological elements to address these effects, and there are those who study these factors, exclusively. There are estimates of reactivity, its effects, how to minimize it, how to incorporate it into research, etc. This is not a phenomenon unique to psychology. If you have some illusion that the physical and biological sciences don’t deal with dynamic, reactive systems and phenomena, then perhaps you could tell a climatologist or marine biologist how much easier their job is. Let me know how that conversation goes. From one point of view, reactive effects slow the progress of the field, but from another, these are the meat and bones of what we study. The scientific method is adaptable, and in the past few decades researchers have continued to develop greatly improved methodologies for addressing the dynamic, interactive nature of human behavior.

–> Objection: You Don’t Know What a Different Person Would Do! A potent criticism of psychological (or any) research is the generalizability of results. Unfortunately, a large number of studies⁴ are done with “convenience samples,” usually university students in Western nations. However, there are also researchers who are meticulous about sampling, and an increasing diversity of populations are being sampled, due to political, transportation and technology advances. Sometimes the results of the new research upset models developed with White College Kids, and sometimes they support them. None of this makes psychology fundamentally different from other sciences. Nobody is surprised that biological results from European plants, animals or humans have been reconsidered after data was collected in other parts of the world, or geological theories derived from Old World exploration have been overturned by samples from the New World, etc. Sampling is important in every domain, and it’s never possible to do it perfectly. You can’t have the whole population, so you always compromise. You therefore never know with absolute certainty whether your observations will be replicated in a different population. Scientists are good at thinking about sampling.

–> Objection: You Can Never be Certain! I have heard many times that “you can never be sure” about psychological results, as if there were a fundamental difference between the level of certainty in psychology versus other sciences. I suggest that a belief in the certainty of any scientific endeavor is seriously misguided. Science does not deal in certainties. Results are probabilistic, with effect sizes constantly tempered by estimated statistical significance levels. Replication is critical, as individual studies are frequently nonrepresentative.

Probably due to some of the factors already listed (and others we surely haven’t encountered yet), many fields in psychology (e.g., personality theory) have fairly low average effect sizes; but other areas (such as social psychology and cognitive psychology) consistently return extremely high effect sizes. There is similar variation of effect strength in the physical and biological sciences. At the end of nearly every biological research report you will find a probabilistic statement of significance (p < .01) and some measure of effect size⁵, just like in psychology research articles. Some areas of psychology are dominated by correlational studies. Just like in some other sciences⁶. There are subfields of psychology⁷ that emphasize rigorous experimentation almost to a fault. Just like the other sciences. There are phenomena that will probably not be conclusively tested for years, or decades. Just like the other sciences. My point? It’s the same process. Uncertainty is constantly present, and constantly quantitatively estimated, leaving the consumer free to evaluate the meaningfulness of results.

I could go on⁸. I cannot think of any fundamental deficiencies in the science of psychology. To the extent that psychologists pursue their research with methodological rigor, the results represent our best scientific understanding of the human processes being studied. Therefore, they should be evaluated with the same set of rules that govern evaluation of results from physics, cell biology or astronomy. But for many people there is an extra rule for psychological research: Disregard When Inconvenient.

The Big Problem Here: We are supposed to evaluate scientific output without bias, by looking at methods and results. This is basic. But psychology (and perhaps other social sciences) are just different in the minds of many people outside the field. As mentioned above, I continue to encounter the same recycled objections, the function of which seems to be to preserve the objector’s right to summarily dismiss the results of psychological research, no matter how scientifically rigorous, while still accepting “science” in general. Answering each criticism with logic leads to the presentation of the next purported indictment (though rarely any specific refutation of the logic). When the end of the list of objections is reached, psychological research is still seen as fundamentally less “scientific” than research from other fields. It’s puzzled me, because it does not add up. If the objections were the reason for dismissing psychological research, then resolving them would change the evaluation of said research. But this does not happen.

I can think of two general reasons for the trend to avoid giving the results of psychological science the same level of credibility as those of other sciences⁹:

1. Ignorance
2. Prejudice

#1 - Ignorance: I have found that people sometimes think of science as being defined by truths or absolutes. Perhaps, therefore, the scientific validity of a field is perhaps seen as being linked to its perceived “absoluteness”¹⁰. This sense of “absoluteness” may be related to the perceived level of certainty of results in the field and the overall public prestige of the field.

–> Science = Uncertainty: We have all seen school films of Galileo, Copernicus, Newton, etc. doing experiments without any statistical analysis, and with clear, obvious results. The balls rolled down the incline. The feather dropped in the vacuum tube. But these visions of the scientific greats are highly oversimplified. Additionally, clear and obvious results are never as clear and obvious as they may seem at first. They also don’t happen very often, now that many of the the easily-observable phenomena have been studied. In the physical and biological sciences, research involves uncertainty, sometimes in large doses. Results are inferred from secondary, tertiary or even quaternary sources. Observations do not all agree with one another. Studies are mutually contradictory. Publication bias, prestige and other human factors filter the results we have access to. Scientists in all fields deal in averages, margins of error, statistical assumptions and estimates.

When I was a kid, dinosaurs were lizards, human evolution was linear, dark matter was silly, quantum theory was unsupported, and string theory was a hippie halucination. All that has changed, sometimes through reinterpretation of the same data that seemed conclusive in a previous era. Science is a specific, pragmatic tool for a certain domain of knowledge. It does not tell us what it True, even when it is working perfectly. Further, the system of Darwinistically winnowing ideas from hypothesis to theory to law is wide open to problems¹¹. I have a suspicion that many people want to treat science as a religion, when it’s anything but. It seems likely that the media has a role in people’s misunderstanding of science; it is not uncommon to see the results of research presented as if they were clear, incontrovertible, and obviously supporting or refuting a particular hypothesis. But, as I’ve said, it rarely goes that way, no matter what area you study.

Prejudice: This one is more insidious, because simple education doesn’t usually eradicate prejudice. Psychology seems to have been the redheaded stepchild of science for the last hundred years for a variety of reasons, some of them based in reality¹². However, I think people have a general negative reaction to psychological science that is independent of the variability and flaws in the field. In fact, I believe people sieze upon these flaws and fluctuations, using them to justify their knee-jerk avoidance reactions. This is not a process specific to evaluating psychology as a science; rather, this is part of how people think.

I guess I can’t blame anyone for being freaked out at the act of turning the scientific method — developed for studying rocks, stars and atoms — on them, with resultant claims of understanding humans better than humans understand themselves. The results have been unpleasant. Freud (notably not a psychologist, and not a very good scientist, but he is sometimes believed to be both) painted a horrifying picture of human motivations, which he reiterated, with no possibility of appeal, from his self-appointed position as High Priest of his psychoanalysis movement. Skinner and Watson (whose work has stood the test of time in amazing ways), used their research to support a view of humans as programmed automatons in a world devoid of God or any meaning outside the illusion of consciousness¹³. There have been a myriad of swindlers and loonies, too, but even the good scientists have told us we’re mean, vicious, stupid, unpredictable, silly, deluded, or just bad. Nobody wants to hear that.

I also think politics figures heavily in the prejudice reaction to psychological research. Social science has always been more firmly associated with the Left than the Right, and as a member of a conservative family, I think I can safely say that conservatives sometimes feel that the attacks from the left use social science as weapons.

The existence of psychology threatens people’s senses of control and identity, two critical aspects of our continued well-being. And if psychology has taught me anything, it’s that such threats lead to prejudice.

Have I solved anything with this overly-long essay? Probably not. The processes I have discussed are deeply ingrained in an awful lot of people. They are not the kind of thing that will be resolved by an online essay. Even if this blog were widely read, I would not expect anyone who dismisses psychological research to make it past the first section. Humans don’t like to expose themselves to ideas they disagree with. But writing this has certainly made me feel better. It has organized some of my thoughts. Next time someone tells me that the research results don’t apply because they’re from psychology, maybe I can find enough Zen to change the subject, knowing nothing I say is going to make much difference, anyway.

1. As much as anything else is [↩]
2. Ha ha ha! [↩]
3. Notably, there is research suggesting that there are more of these in psychology than in the general public; see Maeder T: Wounded healers. Atlantic Monthly, Jan 1989 [↩]
4. Including most of mine [↩]
5. Hopefully [↩]
6. e.g., evolutionary theory within biology [↩]
7. e.g., social psychology, cognitive psychology, judgment & decision making, etc. [↩]
8. seriously; I’m not kidding [↩]
9. Feel free to add to the list [↩]
10. Note the general belief that nonapplied mathematics is a science in the modern sense of the word, when most thinkers–even mathematicians–tend to think otherwise [↩]
11. Often due to the humans who practice said science [↩]
12. I am painfully aware that psychology probably has more than its fair share of quacks [↩]
13. kind of like Stephen Hawking and Richard Dawkins, [↩]