Hello again! In Part 1 of this series, we learned how to conduct a scientific literature search and I assume you now have handfuls of research papers and reviews in your hands. Below, I cover the basics of what to expect and how to efficiently read a scientific paper.
What is the difference between research paper and a review?
This sounds like a simple question, but to interpret a paper or review, we need to understand the scientific process behind them.
A research paper is a short publication based on original scientific research that helps to answer a question. In science, you first ask a question, create a hypothesis or educated guess of what the outcome will be, perform a series of experiments to answer the question, interpret the results, and then draw conclusions from these results.
Even for a scientist to create one publication, there are many steps they follow and a lot of room for error. Error can come from the methods used in the experiment or analysis or from the interpretation of the results. Because of this, you should never read one study and use that to draw a conclusive conclusion. Read several papers on a subject to see what the general consensus is.
Over time, enough studies are conducted and published on a topic that a pattern emerges and you can draw more confident conclusions. Instead of having to read through dozens of papers yourself, look for reviews on topics.
Review papers are written by experts in the subject who are better equipped to interpret and explain the patterns observed across multiple individual studies. The downside is that everyone has biases and a review may be coloured with the bias of the reviewer, but I still consider reviews to be an excellent starting point. They may even be sufficient for your purposes without digging further into original research. If you do decide to stick to reading reviews, try to find more than one on the subject written by different authors.
Is this science any good?
You may have heard that an important skill in reading science is knowing if an article is credible. But this is a difficult task when you aren’t familiar with journals, authors, or what constitutes good vs. bad methodology in an experiment. The simplest method of establishing credibility of a paper is looking into the credibility of the journal.
In scientific publishing, every research article is reviewed by a panel of experts, and prestigious journals have stringent publishing requirements. You have probably heard of the most prestigious journals, such as Nature and Science. Because of this, a good journal is less likely to publish a bad article, while there are many bad journals that will publish very bad data.
One way of determining the credibility of a journal is by clicking ‘Metrics’ in the menu bar of Google Scholar. This brings you to a ranking of journals using citation-based metrics. You can search for your journal to determine how highly ranked they are. Highly cited journals are seen as better, which you can imagine has major limitations, but is still a good starting point for determining credibility.
An abstract is a summary of a scientific article. Some journals request abstracts as single paragraphs, while some require sections, but all abstracts, including review abstracts, contain roughly the following information in the following order:
- A sentence on the topic/importance.
- A sentence on the purpose of the research/what the researchers did.
- Sometimes, a sentence on hypotheses of expected outcome is included.
- One or two sentences on the methods used to address the research purpose/question.
- One or two sentences on the results achieved.
- A concluding sentence.
Abstracts are great for determining if an article or review is worth your time. Even if you have full article access, I suggest reading the abstract before you commit to the full article. Sometimes, an article will look useful based on the title, but after skimming the abstract, you realise you aren’t interested in it. This will save you time. If you can’t access full articles, you will still have access to abstracts, and they are essential for you to determine if it’s worth your energy to track down the full text.
As I mentioned earlier, I prefer to read reviews when I am researching a new topic and do an initial search dedicated to reviews. They are much more fluid in their structure than original research articles. Reviews are usually written more like an essay or mini-textbook, with an introduction, body, and conclusion.
If the whole review is of interest to you, you will probably find it easiest to read it from start to finish, as they are usually set up in a logical fashion. But if you find a review and are only interested in one section, don’t feel forced to read more than you want. When I am trying to find a specific piece of information, I skim the sub-headings or search for the term through the review to see if it is mentioned only in one section or the whole article.
If reading reviews answers your questions sufficiently, then you shouldn’t feel forced to dig into the original research.
Reading Original Research Papers
An original research paper is set up much like its abstract: Introduction, Methods, Results, and Discussion.
Even though almost all articles are set up in this order, you shouldn’t actually read them in this order. First, read the introduction. This provides background information on the topic, like a mini-review. In the introduction, you should be able to determine the purpose of the experiment. You might encounter terms or concepts that are new to you in the introduction, which you should look up. Wikipedia is your friend here.
Once you are comfortable with the information provided in the introduction, skip the methods and skip the results! Go straight to the discussion at the end. In some journals, the discussion is combined with the results section. The discussion highlights the trends and significance of the results and offers some interpretation of what they could mean.
At this point, the introduction and discussion might have all the information you want. You can consider reading the results section to complement the discussion, which explains the observations from the study with little interpretation.
I suggest not reading the methods section. In my experience, reading methodology is only useful if (1) you already understand the methods and are trying to determining the credibility of the paper or (2) you are trying to set up methods for your own research. Otherwise, it is too much pain for too little gain.
Obviously, this is not an exhaustive guide to every trick and tool into conducting research of scientific literature. However, using these tips, you should be well on your way to beginning your own research and answering your skincare questions. If you have any more questions, let me know in the comments and I will respond as best I can.