How To Read Papers Like A Scientist

How To Read Papers Like A Scientist

By Paige Marsland, PhD

While scientific papers are challenging to read, they can be broken down into smaller sections that are easier to digest, and knowing the vocabulary can help simplify things. The vocabulary used in papers often have very specific and precise meanings. When scientific articles are reported about in the popular press, the paper and findings can become decontextualized or misinterpreted, and the findings may lose nuance when they are reported on by the popular press. However, reading scientific articles is a skill that can be learned, and gets easier with practice! This guide was designed to break down the ways that scientists read papers and how they look for information. After reading it, you too can read and interpret scientific papers like a scientist! 

ABSTRACT

Scientific articles are driven by key concepts, such as an overarching hypothesis, then individual experiments are performed in order to answer the experimenter’s questions. Most papers follow the same order of abstract -> introduction -> methods -> results -> discussion. The abstract is the summary of the paper, which tells the reader what to expect. The abstract introduces the concept or topic of the paper, briefly describes why the reader should be interested in the topic, reports the main findings of the paper (results), and finally should have a conclusion that states the outcomes of the experiments and how the experimenter is interpreting the outcome. 

INTRODUCTION

The introduction is where the author lays the foundation of what you, the reader, need to know in order to understand 1) why the author asked the experimental question that they have asked and 2) should give the reader the context they need in order to understand the outcome of the experiment. This background information will also generally uncover questions that are still unanswered about the topic (known as “gaps in the literature”), why it is important for us to find the answer to this question, and how that answer will be able to help us move forward. 

METHODS

The methods section of a paper is where you can find all the details about the experiment, including things like: who participated in the experiment (people, animals, cells, etc), what chemicals may have been used, what scientific equipment was used, and pretty much everything else you might need in order to replicate the experiment. A good methods section is like a good cooking recipe - it should tell you all the necessary steps in order to replicate the experiment yourself. The methods will vary significantly depending on what kind of experiment was being performed. If you’re not planning on replicating the experiment yourself, you don’t necessarily need to pay attention to all the details in the methods section. Instead, scientists will often read the methods section after the results, in order to answer questions they might have about the results. 

RESULTS

The results section is usually directly after the methods section. This section goes hand-in-hand with any graphs that can be found in the experiment. Oftentimes graphs are referenced in the results section, and then a written description of the results will follow. The results are often written about in a lot of difficult jargon, making it hard to interpret the findings. While reading results section, look for things that are noted as “statistically significant” (what scientists use to determine if a finding is “real” or not) or look for corresponding graphs which have asterisks (*) or pound symbols (#), as these are commonly used to indicate whether a finding is significant or not. The written description should then tell the reader if 1) any of the findings are statistically significant and 2) if there are any differences, what is it that makes the result different. 

Graphs are one of the best ways for people to understand data, because you can see the differences between groups yourself! Take a look at graphs (“figures”) in the paper and look for the “figure captions” as well. This section will often give a summary of what was found to be significant in the results section and will highlight the key findings, as well as explain what symbols are used to indicate whether a finding is significant or not.

DISCUSSION

The discussion section generally begins with a restatement of the original hypothesis, as well as a description of the results and whether the results matched the experimenters predictions or not. Note - it’s ok if the results did not match the expected outcome! The whole point of running an experiment is to ask a question, and it doesn’t matter if the answer is yes or no - what matters is that you ran the experiment to ask your question and now you have your answer! It’s not “bad” if the prediction and the outcome are not the same - this happens often, and can be due to a number of different reasons. The possible reasons that the hypothesis and the outcome don’t match are usually discussed in this section, along with potential other questions that this experiment has uncovered. A good experiment answers 1 question but opens the door to ask 3 more questions. We’re always learning more about what we don’t know.

The discussion section also includes more information from the literature (from other experiments scientists have performed and published already), similar to the general introduction. The purpose of this is to try and put the results of the experiments into the context of the general background literature. Experiments generally answer very specific questions, but we can use the context of other people’s work to better understand our own results and conclusions. 

Discussion sections often will also talk about limitations to the experiments. No one is perfect, and no experiment is perfect, either. In this section, we talk about any mistakes that may have been made, any information that is helpful for interpreting the results, and other factors that we may have thought about but were not able to control for or test. Limitations can include ethical concerns (for example, you may be unable to add a control group to a cancer study because it is unethical for a scientist to withhold cancer treatment from a patient), or may have to do with physical limitations of the experimenter (might not have access to equipment). Other limitations may have to do with the experimental subjects themselves (not every experiment includes every gender, race, age, socioeconomic status, etc.). These factors can make it harder to draw conclusions about the experiment, but should be acknowledged to help us interpret the results.

Many papers include a paragraph or two at the end of the paper discussing “future directions”. This refers to experiments that haven’t been performed yet, but that the author proposes as studies which could be useful to help us further understand the subject. These often include caveats from the limitations section for future research to take into consideration, or may include novel hypotheses. Sometimes, the results we get from experiments don’t match what we think was going to happen at all - in that case, this section might be used to propose more experiments to help us figure out why our results didn’t match our prediction. There might be crucial information missing (“gaps in the literature”) that would help us explain why we got the results that we did.

CRITICAL ANALYSIS

Finally, we need to understand the broader picture of the paper or set of experiments. Sometimes authors claim to find a result, or sometimes the media will make bold or sweeping statements about an experiment, but they don’t always match the actual results found in an experiment. This can be due to lots of reasons - it could be purposeful misrepresentation for political or economic reasons, it could be due to lack of scientific analysis skills, it could be due to a misunderstanding of the background literature or field of study, etc. Not every misrepresentation of science is done on purpose, but you should be wary when headlines make grand statements about the results of a study - oftentimes, the results are much more nuanced.

The very end of the paper should have a statement about any “conflicts of interest” and financial support of the experiments. Experiments cost money, and scientific equipment and materials can be very expensive, so this is where scientists report how they received the money in order to run these experiments. Many of the published experiments you will read are supported by grant funding - this is funding that the researcher has applied for, sometimes years in advance, in order to complete this research. This funding is usually supplied by the government, and is supported by government agencies (in the USA) such as the NIH. However, sometimes the authors note that there is a conflict of interest, or that the financial support was provided by a company with an economic interest in the research. In that case, be very careful while reading the paper - see if you can identify areas where they may be overreaching their claims to support their own interests, rather than just reporting on what they found.

Congratulations, you're reading papers like a scientist now!