Confession: I love Social Science research, even though it's often deeply flawed.

I spent the weekend with a ribald group of lake-loving friends, among them several empirical research scientists. Believe me, in this crowd you don't throw a soft but compelling research finding into the conversation unless you're prepared to dissect the whole study. For years I received a free subscription to Nature, the preeminent interdisciplinary scientific journal which dates from 1869 and claims 3 million readers per year. It was a freebie, included in my membership to NY Academy of Science. I thought it would be good for me, broaden my horizons and deepen my knowledge on subjects I knew too little about. A few issues made it clear that a lot of empirical research is over my head. For example, one study:

Transport evidence for Fermi-arc-mediated chirality transfer in the Dirac semimetal Cd3As2

Philip J. W. Moll,​ Nityan L. Nair,​ Toni Helm, Andrew C. Potter,​ Itamar Kimchi​ + et al

Dear Investigators,

I'm talking about your study as a type of research. Nothing personal.

Now I'm sure that article is right up the ally of my lake friends. The study no doubt thoughtfully designed, rigorously executed, data ethically analyzed, and its precise results important. I haven't read it because it would be Greek to me. That's the kind of research that is grant-funded, researchers refer to, respect, and believe in. In studies like this the research question is so narrow that small manipulations can have measurable impact (or lack of impact) on specifically defined outcome variables. That precision is what makes it research.

Now, let's consider the other kind of research. For clarity, I'll call it Social Science research. In doing so I'm demarcating these studies from hard core, (officially known as Basic Science) research. As a point of comparison, here is an article title from The Annual Review of Clinical Psychology:

The Efficacy of Exposure Therapy for Anxiety-Related Disorders and Its Underlying Mechanisms: The Case of OCD and PTSD

Edna B. Foa and Carmen P. McLean

Vol. 12: 1–28

In fairness, I have not yet read this study. So again, guys, nothing personal.

My first blush impression is the level of resolution, ie: the size of the research question, is comprehensible and matters to me. This study used people as subjects. As a clinician working with people, I can use this stuff. I can tell people compelling and useful pointers for healthy living like: "If you're depressed, isolate yourself, focus on how bad things are, research suggests you'll get more depressed." Which I believe to be true.

So what's the rub here?

This softer kind of research can be overdone (take all the compelling little articles in Cosmo) and plainly put--bullshit. For example:

Study Says Napping Can Cause Serious Health Issues

A "study finding" can be found to support almost any belief, even directly contradictory conclusions. Like seriously, is coffee good for you or bad for you? Is butter healthier than margarine or vice-versa? Are eggs heart healthy or death in a oval? Are carbs good, or poison?

Not surprisingly, these softer research studies often don't conform with the demands of research design or the scientific method. They can't possibly conform. The rub is: if a research question is interesting, applicable to humans, relevant to life as lived---it is generally unmanageable as a topic of rigorous research methods. If the research question does remit to empirical methods of investigation, it generally is very difficult to relate to, not useful in a clinical context, and comprehensible to experts alone.

Are there just two categories? Empirical and Bullshit? No, of course not. The above mentioned article from The Annual Review of Clinical Psychology lies somewhere in the middle. Studies like it explore a dimension of human experience, identify a phenomenon, manipulate it, measure it, etc. There are not two categories, in 21st century parlance "it's a spectrum."

But there are kind of are two camps. The proponents of hard-nose science are generally critical and dismissive of Social Science research. The clinicians providing direct service to humans find empirical research dense and clinically useless. Of course, Basic Science research provides the building blocks for important scientific breakthroughs--like diagnostic clarification, validating effective treatment regimens and drug development.

So if you're depressed, research suggests you shouldn't isolate yourself and dwell on your negative feelings, but BEWARE it isn't very respectable research.