One major aspect to truly understanding science news is reading the articles the news items are derived from. This may be daunting for those who aren’t used to reading scientific papers, so we at Copernicus Called decided to post a guide on the best way to read scientific papers.
Before we get started, we need to introduce exactly what a scientific paper is and their relevance.
Scientific papers are the main way scientists convey their research, not only to colleagues across the world, but also to the general public. With that in mind, research papers are usually very dense with terminology and complex research methods that can be quite a challenge for someone outside of the field of study.
Most papers are also not submitted in magazines you find at your local doctor’s office. Papers are typically submitted by the researcher (usually for a fee), peer-reviewed for quality assurance by colleagues in the same field, and then if deemed worthy, are accepted and published in a journal specifically for research articles. For example, someone who researches plant cells and how they interact to make complex tissue structures would likely seek a journal that is well known for publishing articles on plant cells or plant development.
In addition to picking the right type of journal, researchers also prefer to publish in journals that are considered high quality. If you spend years working on research that is long and complicated, you do not want it to be placed in a journal that is hardly read by scientists in your field.
Therefore, journals themselves are usually given a number that correlates with the “quality” of the journal and the impact this journal has on the scientific or general population (i.e. how many people read it). This number is called an impact factor.
To give you a little context, journals Nature and Science currently have the highest impact factors (40 and 37 respectively). Most journals however fall into a much lower range (my experience is to say 5-10 impact factor is within a decent range). So if you happen to be a researcher, receiving a submission in Nature is like winning an olympic gold medal.
The impact factor can also indicate the value of research that must be done in order to be included in the journal. If a researcher does low quality work, and does not prove the point they are trying to make with their research accurately, it will most likely not be in a high impact factor journal. This however does not always mean that the research is fantastic in Nature, but crappy in an impact factor journal of 3. Most researchers in universities or other learning institutions are judged off of the work they have done and the impact factor journals they have had paper submissions in. If you have done high quality of work that is in a decent impact factor journal, you likely get to keep your job. However, if you are constantly submitting to impact factors of 1 or 2 (or not publishing papers at all) then your time teaching at an institution may be short. In the world of academics, we call this publish or perish.
It is also important to note, that papers typically published in Nature and Science are groundbreaking in nature, and tend to answer scientific questions that have been asked for a long time.
Types of Scientific Papers
Most people who discuss scientific papers (especially media outlets) refer specifically to primary articles. These articles include research that the scientists and their team have worked on and collected new and unique data in their lab. Primary articles discover new things in science and provide us with new insights than what we had previously. These articles typically have several authors, and most show collaboration efforts from multiple labs, even spanning across countries.
The other main type are called review articles. These articles typically have no “new” data, but rather discuss and combine several primary articles on a specific topic. These articles are typically written by a few people within the same lab, and are a good way to inform readers of current literature that focuses on a specific topic.
Note: While these are the most common types of scientific papers, there are other versions that we will not discuss today.
Parts of Scientific Papers
Now that we have talked about what scientific papers are, and how they differ from regular news items, we are going to go through the individual areas of a scientific paper. We will present the areas in the order in which they typically appear, but they may not be in the same order that you want to read each piece of the paper.
Title & Author List
This may seem obvious, but the very first thing you should read is the title of the paper. If you are researching a specific topic, the title will give you clues on whether or not this paper is relevant to your literature search.
If you are reading your paper due to a news item, or were otherwise instructed to read the specific article, then the title will tell you what you are about to get into. I recommend that if you are new to reading scientific articles, take a moment to read the title very thoroughly. If you do not know understand a word or two in the title, look them up. This will give you some insight to what specifically the paper is focusing on.
Additionally, the author list will provide you with all the members of the team, and which institution they work in. The first author is considered to have contributed the most work and is likely that scientists main subject of interest. This person may be an undergraduate student, graduate student or a post-doc. The last author is usually a PI (principal investigator). This person usually is a tenured professor at an institution and most likely is the mentor for the first author. When looking at authors, the first and last authors are usually the most relevant.
Scientists do not care about plot twists. We do not make you wait for our grandiose conclusions. In fact, the very first paragraph of the paper, the abstract, gives away everything that a research team did over the project and the conclusions that they can draw from it. This may seem weird to many of you, but for scientists who need to read hundreds of papers in order to work on their own project, this provides a quick way for them to favorite articles that are necessary for their research, or pass to read articles more relevant.
For those of you who are not trying to read hundreds of papers for your Ph.D dissertation, this paragraph may provide you with some interesting information, but it will likely be too dense with terminology and data analysis to completely understand. This section can really be skipped for new article readers. Everything in the abstract will be discussed in the rest of the paper.
This section may also be described as background. If you are new to the topic, I recommend that you begin with this section. It is here that you will learn acronyms and important terms that will likely be discussed throughout the whole paper. The introduction will also provide knowledge on the topic as a whole, and will likely highlight the “hole” missing in the scientific body of knowledge (and the paper usually tries to fill that hole). The intro will give you an idea of why the researchers decided to pursue this particular project.
If you are an experienced reader and are familiar with the topic, glazing over the introduction for any new information is recommended, but all in all may not need to be read in depth.
Materials & Methods
As you probably guessed, the methods section focuses on the experiments (sometimes called assays) that were performed in the research. A procedure for the experiments is typically discussed, as well as a list of the materials purchased. High quality papers usually list explicitly what was done and what reagents or tools were used. I recommend that this section saved for last. Once you have gone through the data and have developed specific questions on how a particular experiment is achieved, you should then refer to this section as a reference. You will also find in this section how the scientist performed the data analysis that is crucial for the next section.
The results section is arguably the most important in the paper, and thus is typically the largest. Within this section you will find results from the experiments performed and it is here where the scientists will tell you a story. Most papers have several figures (ranges from maybe 1-2 at the bare minimum to upwards of maybe 10-12) each with several components.
The example above gives you an idea on the typical size, layout, and components of a scientific figure. Most of them will have some pictures (of cells under special microscopes, diagrams, etc…) and then you will typically see a variety of graphs that are actual data. The graphs will likely show a trend that fits with the scientists narrative and supports their hypothesis. Each figure also has a legend explaining what every piece of data was collected from (what experiment was performed) and now many times an experiment was replicated.
In science, one experiment performed once is not enough to prove a trend. For research paper, an experiment must be replicated at least 3 times before it can be accepted (sometimes shown as an n=3, or n=x; x being the number of time the experiment has been replicated). If the results show a significant difference between the control group (the group that has been untreated) and the treated group, the graph will usually be marked by a star. This star in most cases means the p value is below 0.05, indicating that the trend we are seeing is not occurring by random chance.
I personally recommend that you spend the most time in this section. Look at every graph and every figure. Read the figure legends to understand what experiments are being performed. If you do not understand the experiment, then you can look it up online or read the methods section.
As you read through how the scientists are explaining the data and why they decided to follow the path they were on during the research, ask yourself questions. Is what they are saying make sense? Are their claims matching up with the data? Is there something they are missing that they should have done? It is in this section that you can judge whether a paper is legitimate or not. If the data does not make any sense, or does not seem to match up with their claim, something is likely wrong. Although researchers specifically write to remove bias from their work (that’s why these aren’t entertaining) there will most likely be some sort of over exaggeration. Make judgments from yourself. Just because it is a scientific paper does not mean that it is flawless. If you are uncertain whether or not you can trust the article, I recommend trying to read other articles discussing the same topic, and doing a search on Google to see what is known, and what remains to be discovered with the topic.
The discussion is a section that I typically read after the results. Once you have looked at the data for yourself and determined if the research is legitimate or not, you can move on to the discussion and see what more they have to say.
In this part of the paper, a summary of the most important figures is typically common. This is where the researchers can truly extrapolate from their data and determine what needs to be done further within their specific field.
Most of the time, scientists will discuss future directions for the project and any pitfalls that may have occurred during the whole project. Once you have finished this section, you can re-read anything that may not make sense, and determine if the scientists proved their hypothesis to be true.
If you are interested in the topic discussed, or just want to learn more, browsing the references for articles that may relevant to you is not a bad idea. Typically there can be up to a hundred or more articles. Reading all of them is probably unnecessary, but there are likely a few gems to enhance your knowledge, including previous work done by the first and last authors.
Other tips: Here are a few more tips for reading scientific papers in general
- Read and re-read: It is likely that you will not understand every piece of data and comprehend every step the researchers took in their work. True understanding can only come from reading multiple times
- Highlight important items: highlight what the scientists main points are in the text. This will help you take a step back and see the big picture for the research.
- Google: As I have already mentioned, you will probably have to Google a few terms. Don’t be afraid to. Even though you may not be experienced now, if you take the time to learn the language and the purpose for experiments; it will help you understand the results that much better.
- Don’t be afraid to criticize: If you feel like the paper is exaggerating, don’t be afraid to think so. Unfortunately, some studies are just plain wrong. When this happens, look up the authors and the journal to determine if it there have been discrepancies in the past. If it turns out to be true, then the paper is unreliable.
- Read review articles: For research topics new to you, it may be beneficial to read a review article on the subject. These provide great background information and will give you a little history on what scientists have done previously. It’s not uncommon for topics to evolve and change over the years. Informing yourself of what is known, and what isn’t, is never a bad option.
- Read other articles written by the author: Most of the time, authors are continuously releasing articles on the same overall topic, except with new information (for example someone who focuses on the growth and reproductive cycle of sagebrush most likely has already written previous studies on this topic.
Most of these tips were derived from my personal experience as a scientists and a student. However, some of the resources below are worth mentioning. If you would like other opinions on how to read scientific papers, please follow the links below.
- Segonzac C, Newman TE, Choi S, Jayaraman J, Choi DS, Jung GY, Cho H, Lee YK and Sohn KH (2017) A Conserved EAR Motif Is Required for Avirulence and Stability of the Ralstonia solanacearum Effector PopP2 In Planta. Front. Plant Sci. 8:1330. doi: 10.3389/fpls.2017.01330http://journal.frontiersin.org/article/10.3389/fpls.2017.01330/full
- http://www.nature.com/nature/about/index.html -Nature Impact Factor.
- https://en.wikipedia.org/wiki/Science_(journal)- Science Impact Factor.
- Title photo- https://www.sciencenewsforstudents.org/article/study-drugs-can-be-dangerous