Descriptive vs. Experimental Research

Because I have this handy soapbox, I’m gonna use it. Here’s the thing. There is descriptive research and there is experimental research. Descriptive research on its own is not enough. You’ve got to get in there, change something, and see what happens. Just reporting on what you see under the microscope or on a blot is NOT hypothesis-driven science. Descriptive science is a starting point, it sets the baseline, the control state, what is known. Experimental research tests a hypothesis, which means altering a variable in the known system and seeing what happens – the result will lead you to reject or fail to reject either prove or disprove your hypothesis. Of course you’ll repeat the experiment in exactly the same way several times so you can be confident your results are statistically true. But then you’ll need to try changing something else, repeat, repeat, repeat, and so on.

In a research grant proposal (and I’m coming from the NIH perspective here), each aim should independently test your central hypothesis from different angles. Angles meaning using different methods or combinations of methods, or working at different levels (biochemical, molecular, cellular, tissue, organism, ecosystem, etc). What you learn in each aim will come together to shed light on the system you are studying.

Now, one of those angles might be descriptive, but I would argue that a purely descriptive aim is going to be your weakest aim. Devoting an entire aim to descriptive science breaks two rules in scientific grantwriting – descriptive science is not able to test your central hypothesis, and your aims must not depend on each other. (Because if one aim fails, there goes the entire proposal, and no agency will be interested in funding something so risky.) Any aim that is descriptive will be dependent on what you find in the other two aims.

The same descriptive vs. experimental idea applies to journal articles too. If your article is just descriptive, you’ve got half a manuscript. Sorry, but it’s true. The best, most compelling, field-advancing, paradigm-shifting articles are those that have a clear hypothesis, describe what is known (from descriptive science), and then describe a logical progression of changes made to the known and what happened. I know you’ve heard this before, but the best paper tells a story, leads the reader into the known system and the hypothesis, and then through each question, discovery, question, discovery, until the Discussion section brings the reader back around and gives some context. I know, some journals will accept purely descriptive articles, but in my experience, those are the smaller, second-tier journals. Not the Cells, Sciences, Natures, etc.

It’s getting more and more competitive out there – for research grant funding and publishing articles. So get in there. Get your hands dirty. Know your system then change it and see what happens.  Then change it again and see what happens. And if you need help telling your story, getting other people to understand exactly what it is you’re doing, I’ve got your back.


2 responses to “Descriptive vs. Experimental Research

  1. First – One cannot “prove a hypothesis”. Hypothesis can only be supported or rejected. A hypothesis is supported if carefully designed experiments failed to reject it.

    Second – your NIH grant should not be about a “central hypothesis”. It should be about a central question in mind, and you should supply several alternative hypotheses, that is tentative answers to that central question. Testing and rejecting alternative hypotheses leaves one hypothesis not-rejected and that one can be considered supported.

    Third – descriptive science is not necessarily observational. It can also be experimental. The difference between descriptive and hypothesis-driven research is somewhere else – descriptive science does not start with an apriori hypothesis supplying an explanation for a given phenomenon.

    Good luck in your research efforts!
    Dr. Res Scientist

    • All very good points. Perhaps the post would be better titled “Descriptive Research vs. Research Where You Go In And Change Things.” My conclusion stands, regardless of the terminology, that a purely descriptive aim in an NIH grant will be the weakest aim and a purely descriptive manuscript, although it might be interesting, may not be considered as strong. One person’s opinion, though.

      As for the central hypothesis of an NIH grant – I still believe that it is important to state a central hypothesis, or if it is more appropriate to the RFA, a central problem to be solved. Other, more specific hypotheses will be stated within each aim, and will either be rejected or fail to be rejected based on the experiments in that aim. But all results from all aims should feed back into testing the overall hypothesis or solving the overall problem. That said, every proposal is organized and presented differently depending on the question being tackled.

      Thanks again for your comment!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s