By Boris Litvin
I. “How Do You Even Read This?!”
At last—you found it. A research article that, at first glance, can precisely answer your questions. After Googling for hours and swimming through comment sections, anecdotes, videos, and the occasional heavy debate, you’ve come across what looks like the Holy Grail. Fifteen pages whose beginning can be traced to years ago, when a team of top experts at a top university recruited countless participants who gave their time for a scientific study. Pages written by these top experts over the course of months, which were then reviewed by a panel of other top experts over the course of more months, and which you now feast your eyes upon. The title is confident. The reference section has dozens of entries. There are plenty of graphs and tables.
With a smile, you scroll through the article for a minute…and close the webpage with resignation. “How do you even read this?!”
Whether you’re looking for quality research about OCD, related disorders, or anything else in psychology—and whether it’s for yourself, a loved one, or a therapist—this feeling of confusion and giving up is sadly too common. You found a research article that’s relevant to your question, but as you read it, you may soon realize that this was not written with you in mind.
Unfortunately, research articles are written for scientists by other scientists in peer-reviewed reputable journals. These are not texts like novels, blog posts, or even articles in scientific magazines and websites. There is a different language, full of specific terminology, parentheses, and symbols. This lingo is used by scientists to quickly communicate relevant information and allow them to test the study on their own, challenge or agree with the findings, and use it to further understanding of OCD and related disorders in their own way. However, these scientists have spent years, if not decades, in academia, where understanding this lingo was taught and practiced from the start. Apart from the terminology, there are numbers, statistics, and Greek letters (α, anyone?), which serve as the backbone of psychological research. These numbers and stats are ultimately the roots of this language, and all that terminology and formal verbiage become heavy flourishes that interpret these. What?!
To non-scientists, this is obviously frustrating. You may ask, “If these articles and studies are supposed to lead to a better understanding of OCD (or any psychological disorder), and inform treatment, and so on—shouldn’t they be accessible to all of us? Shouldn’t everyone be able to glance at them and gauge which course of action is best to treat their symptoms, or understand how a specific subtype of OCD works?” You may also ask, “If words in these articles are meant to interpret numbers and symbols and tests, why not just use simpler language? Statistics are complicated as it is!”
Scientists across all disciplines learn their lingo from early on; over time, it becomes internalized through reading tons of these articles and eventually writing their own. But all have found themselves in your shoes—as a student in an intro course receiving a 15-page print-out for homework (“I can definitely read this last minute…”) and then going into full-blown panic when they realize this reads like nothing they’ve seen before (“I finished War & Peace in 10th grade! Why is this so dense?!). Over time, they learn the tricks of the trade—what terminology means, what a Greek letter stands for, what numbers represent. Most importantly, they learn how to actually read a research paper—how to find what they need without reading the whole thing in linear order in a matter of minutes.
I started out as an overwhelmed student in an intro course, worked in labs, and contributed to a few research papers of my own. While I understand the terminology and am quite comfortable with reading a series of articles, I’m also mindful that most people can’t do that and would have to give up a solid amount of time by practicing this skill, going to classes, working as a researcher, and so on. So I’m writing this for anyone who wants to learn and know how to read a research article like the experts do. But before we dive in, please understand that this will require practice and comfort with some frustration before you can quickly skim these texts like an academic. It’s just like building a muscle or practicing a hobby.
II. The Structure of a Research Article
A research article can be about a comparison of ERP vs. ACT, how a cluster of neurons interact because of a specific chromosome, a summary of interviews with patients, or any other topic—but the structure of each paper is generally the same.
A summary of the entire paper in about 250-500 words that ideally points out the purpose of the study, what the study found, and what the results mean. 5-7 keywords are listed at the end that tell the reader the topic of the paper.
A text that explains the previous research that has been done on the topic of the paper, which finishes with a paragraph or two explaining the purpose of the present study. (Sometimes the “Purpose of the Study” will be its own section.)
Introductions are full of terminology and references, and are usually the most verbose parts. You’ll see a lot of parentheses with names and years (e.g. Smith & Jones, 2023). These are citations and tell the reader whose study is being referenced.
All research needs to stand on prior research, theories, and evidence, and is typically a new look into something or a reassessment of a prior study. The introduction is also called a “literature review” because it looks at what has been done before and gives credit to the authors. It’s super important—a scientist may think they’ve come up with a completely new idea, but may find that someone else already had that and wrote about it. Credit must be given where it’s due.
The “Purpose of the Study” at the end (or the last few paragraphs) will state what the current study is trying to achieve. It will present a hypothesis (an “educated guess” based on prior research), how it will test this hypothesis, and why this study is important.
A description of the sample—people whose characteristics and responses were used to carry out the study. The participants section tells the reader who the “experimental group” is: the people chosen for the study, and how and why they were chosen (e.g. diagnoses, age). It will also usually describe a “control group” (a group of people who do not have a particular diagnosis or similar metric, whose responses will act as a comparison to the “experimental group”). This section will usually reference a table that describes demographic characteristics in terms of full numbers and/or percentages.
A description of the tools used to carry out the study with the participants. Here, tools can range from the complicated machinery used to carry out brain scans to the simple questionnaires that scientists use to ask participants about their symptoms, thoughts, and feelings and even to the location of where the study took place (a laboratory vs. somewhere else). The methods section will also describe the statistical tests used to analyze the data that is gathered using these tools.
This section is super important for scientists. It serves as an instruction manual of sorts for how to conduct the study, and would ideally allow other researchers who wish to copy the experiment in the future to do so. For the non-scientist, keep in mind that the methods section is very technical by design.
A description of results based on the analyses of data gathered from testing participants.
Although usually shorter than the others, this section is by far the most technical and complicated to understand for non-scientists. It’s common to see numbers, Greek letters, abbreviations, and other “lingo” here. Most tables and figures (charts and graphs) in an article are tied to the results section, and are also technical.
Although each paper has its own approaches, you will see a few common phrases throughout. Here are a few that are the “backbone” of each paper, without getting into too much “stats class” detail:
(Statistically) Significant: The results of the analysis show that there is a connection to something that is not attributable to chance or randomness.
(Statistically) Nonsignificant: The results of the analysis show that there is a higher likelihood of the results being attributed to chance or randomness, and not necessarily to the treatment, intervention, or interplay between the things studied.
α: The Greek letter alpha (α) stands for significance level, a metric used as a threshold for whether the analysis is significant or not. α is usually set at .05, .01, or .001, depending on the needs of the study. These numbers are also percentages—0.05 here means that there is a 5% risk of the study showing that the results are correct when they’re actually not, 0.01 means a 1% risk, and so on.
p: The letter p stands for probability, and the number that follows either an equals sign (=), a greater-than sign (>), or a less-than sign (<) is the probability number that is compared to the α value set by the researchers. If a p value is less than α set at 0.05, this means that the result is statistically significant and is not attributed to chance. If it is greater than the α value, the result is not statistically significant.
—Think of α as a threshold and p as the number you compare to the threshold.—
Let’s look at this with an example. “Treatment A is statistically significant at reducing symptom X, p=.03, α<.05.” In this study, with the group of participants who were tested at the location where this study took place, Treatment A reduced symptom X at a solid enough level where the results can be soundly not attributed to chance or randomness, because the p value was less than the α value.
The first part of that sentence is emphasized for a reason: this study can be done again by the same set of researchers or by a different group, and find that Treatment A actually doesn’t work so well (or works even better). As authoritative and confident a results section can look, all results should be read with the context of who the study participants were, where the study took place, and other important circumstances.
A less technical description of the results, usually written without many numbers and statistics. The discussion section shows in plain(er) language whether the results of the study supported the hypothesis outlined in the introduction, and why these results are important. A good discussion should also show what the researchers believed to be the strengths of their study (it’s good to do something innovative), as well as admit to limitations and weaknesses that could be addressed in future research (no study is perfect, and scientific humility goes a long way). The final paragraphs can suggest what future research on this topic can do, and neatly summarize the study (sometimes this can be a separate “Conclusion” section).
III. How to Read a Research Article and Get What You Need
With the descriptions of each section in mind, you’ll see that this is nothing like reading a novel, a magazine, or a web article. Texts and articles like that usually follow a pattern that requires reading from start to finish, or else there will be much confusion.
While a research article can certainly be read from start to finish (and it is encouraged to do so when trying to understand a topic at an expert level), it doesn’t have to be. In fact, most researchers don’t have time to read multiple studies word-for-word, and the structure of research articles allows them to jump around an article and get the information they need quickly and efficiently. Everyone has their own way of quickly tearing apart an article in 10 minutes, and I’d like to give you a technique which works for me if I want to casually understand a study. (For scientific purposes, I definitely read them in more detail.)
Read the abstract first. It’s a short summary of the whole study, and you’ll go into it with a general idea of what’s happening.
Read the “Purpose of the Study” or the last paragraphs of the introduction. With the abstract in mind, you’ll now understand in more depth what the study is about and trying to achieve here.
Read the “Conclusion” or the last paragraphs of the discussion. Yes, skipping all the way to the end is allowed—now you’ll know the basics of what the study found and if more research is needed.
Briefly read the “Discussion” in full. This will give more detail to what you just read, will explain why the study found what it found, and its strengths and limitations.
Skim the “Participants”. Look for demographics, the location of the study, how long it took, and so on—now you’ll understand in what kind of experimental group these results were found.
Skim the “Methods”. It’s technical; just pay attention to what kind of tests were run and whether they’re something standard in psychology or something new.
Skim the “Results”. If you can understand the statistics and read the charts, great! If not, follow the “common phrases” I outlined above and skim.
Skim the “Introduction”. By this point, you’ll already have a general idea of what the study’s purpose was, what the results found, and whether this topic needs further research. But if you’re interested in seeing how the study came to be based on prior research articles, then skim the “Introduction”. You’ll also find some definitions for what certain terms mean (and if not, you can look them up at a reputable source).
This should take about 10 minutes and give you a casual but solid understanding of what, when, where, why, and how a group of scientists did what they did.
IV. Some Final Words of Advice
I hope that the last sections made it easier to read a research article and know how to understand it. Keep practicing these skills by reading more articles, especially challenging ones. Over time, you’ll see that you will follow along with greater ease and (dare I say it?) enjoy reading them.
Before this ends, here are a few final words of advice.
1. You may realize it’s hard to fully trust statistics, especially when they are difficult to understand.
First, a good research article is published in a peer-reviewed academic journal, meaning that groups of other scientists rigorously review an article before it is published. These reviewers pay special attention to the statistics and results, as publishing correct data is essential for good and flourishing science. If an article made it to the point of being published, there is a high chance that you can trust the statistics.
Also, being healthily skeptical and asking questions about statistics—or anything else—is the mark of a scientist. No scientist knows absolutely everything, and always asks questions, reaches out to fellow scientists, and reads and watches relevant content. Ask questions and be curious!
2. Don’t worry if you don’t understand everything (something that I recognize is a struggle for some folks with OCD). No one understands everything—not even top scientists. They ask questions all the time, and know that having a definitive answer is usually impossible when it comes to psychology.
3. Most things in psychological research are approximations, based on the sample that was chosen, where the study was conducted, and the strengths and limitations of a particular approach. The same framework can lead to different results with different samples and locations—and all of those results are approximations. By their nature, social sciences—those that deal with people—are more dynamic and have more nuance than the hard sciences (physics, chemistry, and biology). Nothing is perfect here and there is always more work to be done in OCD and related disorders research; it’s perfectly fine to take things with a grain of salt.
4. Google and YouTube are your friends—if you know how to use them properly. If you have questions, you can definitely search things up, as long as you know what you’re searching for and where to look. If you want to understand statistics in more depth, you can check out channels like CrashCourse for easy-to-follow explanations and Khan Academy for detailed lessons.
5. You can reach out to researchers directly if you have questions. Though busy, they’re usually quite happy to talk about their studies and point you in the right direction! And you can always email us at email@example.com or firstname.lastname@example.org as well!
Thank you for reading this! I’ll follow up with a second part that will get more into the technical aspects of research articles, but hope that this will be enough for a start. In the meantime, please visit https://iocdf.org/research/ to learn more about research at the IOCDF, the studies we have funded, and our 2023 Research Grant Program!
Boris Litvin is the IOCDF Research Communications Specialist.
The post <strong>How to Read a Research Article: A Kind Guide for Non-Scientists</strong> appeared first on International OCD Foundation.