« Prenatal exposure to a common pesticide linked with brain abnormalities | Main | This Week in Telescopes »

Journal articles: The loneliest literature


by Sarah Scoles

Let's say you're a scientist. You spend years of your life researching a topic. You get bone spurs from too much pipetting and carpal tunnel syndrome from too much coding. Then you have results, and you write them up into a paper full of informative and well labeled figures, and you send it off the be refereed. If you're lucky the editors wont use your manuscript as a coffee coaster, and once (if) it is accepted and printed, who reads it? 

Well, your collaborators (hopefully). And the peer-review panel (fingers crossed). And some woman who studies pretty much the same thing as you (will skim the figures and discussion). 

But 50% of peer-reviewed journal articles that are published are never cited, which not only gives graduate students across the world another reason to be depressed, but means that more than half of scientific research disappears into obscurity without adding another wall or floor to the building of scientific knowledge.

Either most papers do not describe useful or good science, or people do not read most papers.

This post deals with the latter options and asks the question, "Why don't people read scientific journal articles?"

I made up four reasons, and we'll see if you agree with them.

1. Jargon 
Webster Man says that jargon is "confused unintelligible language;  the technical terminology or characteristic idiom of a special activity or group; obscure and often pretentious language marked by circumlocutions and long words."

Jargon allows specialists in a field to communicate with each other in an efficient way. A single word of jargon can sum up all components of a complicated concept, so that instead of explaining the whole concept every time she needs to refer to it, a scientist can just use the word or phrase. That saves a lot of time. However, it excludes people who could understand the concept but do not know the secret passphrase for it. 

2. Passive Voice
Passive voice is when plans are made, experiments are done, conclusions are come to, and results are written up, all magically without someone to do them (or they are done by someone, rather that someone just doing them). Passive voice takes the action and the actors away, or at least deemphasizes them. The idea behind this removal of the person is that science is impersonal, so the people should be not take center stage in the write-ups. That makes sense, except that, you know, they are the ones doing the science and that does kind of matter.

However, paper authors also tend to use passive voice in sentences that don't involve people, and it obscures the cause-effect relationships (what is doing what) that are the whole point of the science.

3. Lack of authorial enthusiasm
The tone of journal articles is flat, suggesting both that everything in the paper has equal importance and that none of it is interesting to the people who wrote it. And if the people who did all that work aren't excited, why should the reader be excited? It's more fun to read words that someone else had fun writing. The ideas behind this problem are
a) the importance of the data and the conclusions should speak for itself, and you should both recognize and care about that importance because it is self-evident.
b) science is about objective truth-seeking, so the papers should be 100% dispassionate.
To read a scientific paper: dive in, get the information, get out as fast as you can; whatever you do, do not enjoy reading it.

4. Isolated topics
As the name of this blog implies, many of the questions being answered in science right now are "smaller." In good papers, authors relate small, specific problems back to larger issues, whether they be sturdy scientific umbrellas (like "this helps us figure out how galaxies came to exist")  or human-centered problems (like "this adds to our understanding of how psychiatric medicine actually works in the body"). In bad papers, the new, pure knowledge is enough (like "The beam shape of pulsar J1045+3958 looks like this plot") and is isolated from contexts into which it could be put. All papers would be better if authors explained the importance of the specific problem to the general, underlying principles of the field.

Science journalism exists to make science more palatable than it is in journal articles. However, science journalism doesn't always focus on the science that is coming out in journals right now; it certainly does not cover all the science in journals (not that anyone actually wants that). The point is, the science that reaches the public sphere is different from the science that is actually going on, and journalists decide which discoveries and which parts of those discoveries are interesting. The public sphere view of science, though, is based on science as portrayed in the popular media.

Brooke and I decided to find out how "science as science sees itself" versus "science as the rest of the world sees it" look different by making word clouds from different science news sources, progressing from closest to Scientist World to farthest removed.
The point of these word clouds, and all the words above them, is that the sources people actually read for science news do not necessarily reflect what the "news" in science is, or at least not what science would call its news. Scientific journals do not do a particularly good job of making their subject matter appealing to read, and most of the information in scientific journals doesn't make it out anywhere (even into other scientific papers). The scientific knowledge that reaches people is filtered--by which papers are accompanied by press releases, by which topics are trendy, by which topics are politically heated, by which topics are easily applicable to the average American's life. Here at Smaller Questions, our goal is to go straight to the sources--the journals--and to eliminate Mistakes 1-4 without eliminating the depth of coverage.

1. Astro-ph on arxiv.org, where scientists post astronomical science papers before they are published (if they are ever published). Here, the most common words are process- and observation-based and are rarely object-specific. Many of these words would never be printed in popular media because they are specialists' language for how they investigate rather than what they are investigating.

2. ScienceDaily's Astronomy news feed. Here,there are more object words, so you can actually determine what the articles are about. Additionally, the words "first" and "new" provide the first hints that anything exciting is being discovered.

3. Scientific American's space news feed. Here, object words abound, but they are perennially popular topics--exoplanets, Pluto, life, black holes. Regardless of what astronomers are studying right now, the popular scientific press tends to return to the same kinds of trendy topics that are guaranteed to garner readership. But we also see the first evidence that anyone has a positive opinion of this science--"great." Additionally, this word cloud makes reference to the people who do science! Astronomers, international ones, perhaps from Earth.

4. New York Times space news feed. SPACE. It dominates. The NYT wants to make sure you know this stuff is cool because it is out there, not on earth. Other than that, most of the words refer to the people, organizations, and nationalities of scientists, and to spacecraft. The NYT coverage is clearly about the human endeavor.

5. ScienceDaily's microbiology news feed. Mechanisms and animals dominate this feed. Discovering mechanisms gets you Nature and Science papers, so it's interesting that a popular news feed picks up on this focus. I would have imagined that more buzz words would have appeared on this feed. But the focus on animals is telling of where the biology community is heading. Since funding has been cut short, the way to ensure grant-renewal is to relate your research back to the organism. Want money? Show it's important in killing an animal.

Unlike the space news feed, it's dead-on to what is currently happening in the research world. Perhaps this is because microbiology as a topic is more specific than "space"? 

It is also interesting to note that, unlike the space feed, there are tons of similarly sized words, and most of these involve urgency like "hypervirulent," "rapidly," "disease-causing," "outbreaks," "infections," "poisoning," etc. 

6. Scientific American's microbiology news feed. People and bugs dominate this news feed. This is more what I expected from a popular science publication - bugs and how they affect humans. Additionally, the buzz words in popular microbiology appear: "drugs," "vaccines," and "resistance." This genuinely reflects the state of microbiology and how the field is veering away from mechanism and more toward clinical application. 

The preoccupation of geographical location (Soviet, US, Cuban, California, East, West, Central, etc.) is beyond me. Generally microbiology tries to veer away from specific regions to make their science more applicable to everyone (again, helps with the grants). 


EmailEmail Article to Friend

References (3)

References allow you to track sources for this article, as well as articles that were written in response to this article.
  • Response
    Journal articles: The loneliest literature - Blog - Smaller Questions
  • Response
    Journal articles: The loneliest literature - Blog - Smaller Questions
  • Response
    Journal articles: The loneliest literature - Blog - Smaller Questions

Reader Comments (4)

Love the word clouds, quite telling!

Things I find interesting:
- Scientific American space feed features Christmas....perhaps due to the intrigue of the Star of Bethlehem portion of the Christmas story?
-Did something happen with Chagas Disease recently? SA's micro feed seems to portray it quite active? CDC has some older reports of increasing prevalence in CA, AZ, and LA...but nothing very recent.

April 25, 2012 | Unregistered CommenterTimS

I think your assessment is correct on a number of levels. Jargon can prevent people from reading papers - particularly if the authors themselves have come up with their own TLAs (Three Letter Acronyms - which in biology and microbiology may mutate into FLAs as well) and use terminology which they think their audience will already know (when they do not). But the fact that 50% of these papers are not read and lost knowledge is tragic - among these papers may very well lie useful gems which would advance our understanding of the universe or ourselves; be a new treatment for some disease; contain a solution for problems we already face.

Audience may play a role in this, and basic readability of the content. It's my observation that papers are often written at the last minute, sometimes put together hastily before a conference after the abstract has been submitted to the conference organizer in preparation for one's 15 minutes of conference fame. The audience here is others who are researchers in the same field, typically. And naturally they should understand the significance of one's findings first. But the real audience one needs to sell to are those paying the bills - those supplying the grant money. And they may not always understand what on earth one is talking about.

So one has to write to convince them to continue reading from a more generalist perspective, and at the time, with potential financiers in mind. To do so and still maintain a clear materials and methods section is possible, and that part is more important to others in the field who may want to duplicate one's experiment and see if they get the same results.

I get the impression that what makes a readable and interesting paper is field specific as much as it is about how engaged the authors are and how well they can sell the importance of their findings. And to some degree, relating one's research to existing research that further strengthens one's position. The quality of one's existing citations are often important - especially in online research where one is looking at relationships within a body of work. If citations are of poor quality or overly self-referential, I have to wonder how often that contributes to unread publications?

April 28, 2012 | Unregistered CommenterCamp Other

I definitely agree that at the very least, authors should take the time to write papers that people in their fields will want to read (so the results can be out there!), even if they don't want to write for 'the public.'

April 29, 2012 | Unregistered CommenterSarah Scoles

Mrs. Scoles, I am one of your students from PSC and I was hoping that you could check my 2nd test that I did for pulsars. I really need it before tomorrow and I corrected it after you sent me that second hint, though I am not too sure whether you saw it or not. I'm really sorry for trying to contact you through something non-PSC related, it's just that I created a large problem for myself and there really seems to be no other option for me and my grade may end up in ruins.

April 30, 2012 | Unregistered CommenterPraveena M

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>