Information Line: 519-453-3198
HOW TO READ SCIENTIFIC PAPERS
(for the layperson)
WRITTEN BY: Corinne, March 14, 2019
Because scientific data should be accessible to everyone.
Wouldn’t it be great if scientific papers were written so that everyone could understand them? We could refer to factual scientific data when someone makes a false claim, or we could use scientific papers to sort out fake news. Unfortunately, reading a scientific paper is not the same as reading a news article or blog post. They are confusing and hard to read, even scientists need to go through a lot of training to be able to read papers effectively, but this guide will make them easier to read.
If you just want to quickly scan a paper without diving deep into the study, I recommend you start with reading the abstract to get a brief summary of the study, then look through the figures and look at the data, and then the discussion/conclusion.
If you want to take a deeper look at the paper to understand the study, here’s a simple guide for how to read scientific papers for the non-scientist:
Start with the Abstract
Don’t just read the abstract and ignore the rest of the paper. The abstract is great because it will provide you with a summary of the whole paper, but it probably won’t give you all the information you need. Skim it to get an idea of what you’re about to read and move on.
Read the Introduction
This part is usually easier to read because it is written like a story. It gives some background information on the topic and will sometimes include other studies that are similar to the one you’re reading so you can find more info on the subject. The most important thing about the introduction is that it explains the purpose and importance of the study. After reading the introduction, you should be able to summarize why the scientist is studying this particular topic and what they are trying to find out.
The most important thing to look at here is the sample size and diversity. Would you trust a drug that was shown to work in 10 people or 10,000 people? A study done on 10,000 people will produce more accurate results than a study done on 10 people. It’s hard to be able to draw a conclusion from a limited sample, so make sure the study you’re looking at is using has a large sample size.
Skip it for now and come back. This section is for laying out all the data, but personally I like reading the explanation of the data in the discussion/conclusion first because it’s written in a way that’s easier to understand. After I read the discussion/conclusion, I’ll come back to the results and look at the figures to make sure it all makes sense.
In my opinion, this is the most important section, just be careful what you read because this section typically includes the researchers opinion on the data. This isn’t necessarily a bad thing, but always go back and check to make sure the results match what the researcher is saying because sometimes people will exaggerate the results to help support their claims. This section is also important because the scientist might tell you about any errors that occurred in the study and whether or not more research is needed to confirm the results.
Back to the Results..
When looking at the figures, it’s important to take a look at error bars. A graph typically represents the average value of all the data collected. The error bars will show you how far the data points were from the average. A small error bar indicates that most of the data points were close to the average value and a large error bar means some of the data points were very different from the average value.
Look at the graph below. At first glance, it seems that participants who did not take medication experienced greater pain than those who took medication. However, if you look at the error bars (the thin black vertical lines sticking out of the middle of the bars), you’ll notice that they are pretty long and overlap with each other. The overlap between the error bars means the data values were similar between the two groups. This indicates the data may not be significant and patients could have experienced no change or little change in pain after taking medication.
The graph below shows no overlap between the error bars. This means the results are most likely significant and the medication probably helped with pain levels.
Don’t be afraid to go back and read the paper again
Science papers are REALLY hard to read so don’t be discouraged if you read through it once and still don’t understand, you might need to read through a few times to fully understand it. Go back, read it again, and identify any questions you may have and where you are getting confused. Google any technical jargon you don’t understand.
It is important that not only everyone has access to scientific data, but that they are able to understand it. Most scientific information is communicated through news articles by journalists who may not be trained in science. Being able to look into their sources and read through the scientific literature is important for sorting out fake news.