 Hi, I'm Carrie Wade, the Health Sciences Librarian here at University of Wisconsin-Milwaukee, and today I'm going to talk to you about some tips for successfully reading and engaging with a scientific article. So this is just kind of an 11-step program for how you can make sense of a scientific article because sometimes it can be a little difficult. You might have find that experience a little intimidating. So a little bit about scientific articles. They're written by and for scientists. It's part of their scholarly universe and what we would call the scholarly conversation, where scientists are engaging with each other about the topics that they're experts in. And what that means is that it's full of scientific language and they're going to assume that the audience has the knowledge of that terminology and vocabulary used in their fields and that they're already experts. Similarly, the articles also assume that the audience is familiar with a scientific research methods and article formats, and they follow a specific format. And that's part of the secret to understanding one of these articles is that they do follow this very specific format. Most scientific articles can be broken down into a few sections here. So typically they have an abstract, an introduction, a method section, a result section, and a discussion or conclusion section with some references at the end. This can help us kind of unlock and figure out how to make sense of these articles. So here are some quick tips for how to read an article and how you can best interpret this. If you've ever sat down in front of one of these before and you try to read every single word from start to finish, that may have been really difficult. But here's a fun surprise. You don't have to read every word. There are parts that you can skim. You also should write down and define unknown words such as scientific jargon. That's going to help you make better sense of what they're talking about. Also write down acronyms and what they stand for. That will also help you as you're interpreting these works. Also know that some parts are going to be more important than others. So we'll talk about that as we go through articles. When we approach an article to read it, we want to read the introduction first. This is what's going to tell us the background of what an article is and what the researcher's problem that they're addressing is. This helps me identify a few different things and this gets to some of our other steps along the way. But first you might say, why am I not reading the abstract first? Well, technically you've probably already read the abstract because when you're searching for articles, either in a library database or even just out on the internet, you've probably already at least skimmed the abstract to figure out if it's going to be a good article to read. So that's why I'm typically saying we're not going to read the abstract. And it also comes into play a little bit later on. It can help us make good decisions about which articles to read in full. But when reading the full article, it's actually more useful to skip the abstract upon my full text reading and go straight into the introduction. So that's why we're not technically reading the abstract first because we've actually probably already read it a little bit. Our second step is when reading through the introduction, we really want to try to identify what the big question is here. So what is their main problem that they're trying to solve or their hypothesis if you want to use the scientific term? And you really want to take special note of that and keep that in mind as you read the rest of the article. So let's do that with an example here. I have this article from the Annals of Epidemiology, which is a public health journal. And I have this introduction here where I have my introduction about poor diet. And if I'm skimming through this, they give me this information about this HHFKA, which is the Healthy Hunger-Free Kids Act of 2010. And what it is, is they're giving me this overview of standards when it was implemented, which is 2012, when it was implemented. But when we can see down here in this last paragraph, they talk about the goal of their study. And that is to estimate the dietary effects of HHFKA implementation on days when school foods are eaten and on the annual population basis. So that's going to be our big question there. Next we want to do is summarize the introduction background information. So that would be the information that I talked about earlier, where I want to try to put it into my own words rather than using their words. That's going to make it easier for me to summarize. It's going to make it easier for me to remember and recall what they're talking about rather than using their words. So for this article, it's about a piece of legislation that was signed and the information about it is that it was signed in 2010, it took effect in 2012, it affects school children, and it's about nutritional outcomes. So that's kind of the information that I need because it's about policy and its impact on dietary outcomes in school-aged children. And then are there any specific questions? So I noticed one specific question in my article that goes along with my big question about the dietary impacts of this policy. So there might be related factors or other ideas that emerge as a point of interest. And so again, we'll want to point them out and make note of them. So there's this other one right here, and this is my more specific question. So we also considered whether it's effect differed by age and household income. So there's two specific questions around that. Is if the dietary effects of school foods eaten were differed by age and household income. So they're also going to be looking at those specific factors. So that's important to look at too. Now what we want to do is figure out how they decided to approach this data. And so we want to try to figure out this. This may appear in a few different sections in the article, and some it appears in the introduction. In this one it actually does it in the, their approach appears in the method section. So they looked at the NHA and ES or NAHANA's survey, which is the, it stands for the National Health and Nutrition Examination Survey. And it's nationally annually collected data that is done on an annual basis. It's widely available. It's publicly available records. So they're doing a record collection and examination where they're looking at publicly available data questionnaires. And they try to identify different numbers of records. And so that's their approach where they're looking at data that's already been collected to identify certain information to see if trends emerge. So that's their approach to this problem. Is that they're trying to find trends that emerge from past data that's been collected from 2007 to 2016 to see if there's a difference in the implementation of this program. So from 2007 to 2016 so that they have this kind of benchmark point to see when things made the difference in 2012. So that's their approach. Now when you get out of that you can get into the method section and really figure out what they're doing. So this is where we want to really figure out the really important information such as the population in this article, their sample size which is also known as N. They'll give us an N number and that's really important to know what kind of data the authors have collected, the time span over which they collected it, and what they compared it to and what kind of outcome they're looking for. And if it's more involved like if they're doing a scientific study or an experimental study this is going to be more involved. But for this article because they're collecting past data it's going to be a little bit less involved. But this would be one where I would want to draw out some more details here because there is a lot of details because they're looking at specific parts of the survey. So they're looking at questions around this healthy eating index which is they're looking at a dietary assessment, a healthy eating index, exposure to school cafeteria foods, effect modifiers, and then any kind of adjustments and demographic factors. So this gets a little complicated here and I'm going to talk about that. Some of this is really confusing. You may have seen a lot of math there. Yes, there was a lot of math there. This is where I talk about the bits that you can skim. Some of these things like the statistical methods, the feasibility paragraphs, anything that delves really deep into mathematical formulas for what our purposes is at our level. So if we're doing papers that are just going to be turned in where we're talking about if we're just writing overview papers about the effects of cafeteria foods on school nutrition and lower income households, we don't need to know all about these statistical methods. A lot of that exists for good reasons like for PhD students and even master's level students like for study reproducibility, but it's really not relevant for our purposes. So some of this stuff we can skim over for our purposes today. So then we want to go to results and try to summarize them and this is why tables are great. So you may have seen that table that we encountered in our study there and I love this table a lot. So they gave us a really great demographic breakdown of who all was included in this data collection, the household characteristics like people who were included for free and reduced meals, that's going to answer some of that specific information. There's also going to be some more in the results section. So if we go down to our results past some of these skimable sections. So let's go into our results here. They talk about the demographic characteristics which is really significant where they talk about the average age of participants. So that's 11 and a half years old at the point that we really want to record. They also want to say half of participants were eligible for free and reduced lunch. School foods were reported and at least one recall by almost half a participant. So almost half of them were eating school food. We get a little mathy here, that's a little hard. But then we get into kind of our exposure to school foods and that's a really important result here. So average across the school year including non-school days 9.9 students energy intake came from school cafeteria foods. So that tells us something really important. So across a year, so that's 365 days, think about even the days kids are not in school. 9.9% of US students energy intake comes from school cafeteria food. That's a good number to know about. On days when students reported eating school foods 33.5% of daily energy was from school cafeteria foods. And so this is just across all students that they're talking about. And so they had it broken down into three time periods. They say across those three time periods their energy intake from school foods didn't change significantly. Now they'll look at the association between school foods and dietary quality. So I'm going to take notes on this. And this is where we get a really big chart with lots of stuff. Get a lot of numbers. It's really hard to kind of make sense of this a little bit. But if we take the time to read it, we can get some good information here. Now we have to summarize them a little bit. But we don't necessarily have to know what they mean, but we're summarizing. So this is where we talk about how do their results answer the specific questions. So for this article, they really wanted to talk about how do those, how does this act from 2010 about school cafeteria food? How does that affect people at different or children? How does this affect children of different age groups and of different household income levels? So for household income level, they're using free and reduced lunch as a comparison point. And for age, they're looking at, you know, school grade levels. So for elementary school, middle school, and then high school. And they're looking at different periods of implementation. So they are addressing those with their results. And now we're going to read that conclusion discussion. And I have that right here. And it does a really great job of kind of nailing this home here. So they talk about those different time periods of the four years before of, you know, they did a really good job of those first four years of implementation, because that's all they could measure from this. But they did really well with income and grade level subgroups, which is what they were looking at. So they said, you know, as far as the subquest of, as far as the specific question goes, we found some really good results. We also had some findings that were concordant with prior assessments. This is not the terminology I would use, but I would have to summarize this to be in my own words. So I could say that this is, you know, I would say that this is this has a strong correlation. So, you know, a good dietary program at schools has a strong correlation to creating better out, dietary outcomes for children of younger ages and of lower income levels. So that's what we could say as a conclusion based on this work. So now we get to go back and read the abstract. So we get to see what their purpose, their intended purpose is, what their methods are, because this is just kind of a nice summary, actually, where their conclusion is that the implementation of this HHFKA, which stands for Healthy Hunger Free Kids Act of 2010, improve the total dietary quality of U.S. school students, U.S. students would benefit from eating school meals in the post HHFKA era, and HHFKA regulations should not be relaxed. So there's a policy recommendation that comes with this, which is kind of cool. They didn't go into too much detail about which age groups it had the most significant impact on, which it leaves out some kind of details here. But it is really great. And the fact that it does give us a nice summary. And I think it matches up with a lot of the details that we're looking for, and also giving us a good summary of those results there. So looking at that healthy eating index, and how they look at that. So what are others saying about this? Have other people cited this article? And I'll show you how that works here. So has it been retracted? Is there a publishing controversy? I'll show you how some of that works here. And then I can paste it in here. We can see that no one has cited it yet because it was published this year. So it was published in 2020, so probably no one's really had time to cite it yet. But we can see some related articles, things like that. We can also see who they've cited in their work cited list there, which is kind of nice. We can also check this site called Retraction Watch. So this is a website where you can check for journal retractions. And so it tells you which sites or which papers are not great. There's a whole bunch on COVID-19 right now because that's a very hot topic and which ones are having problems. And we can also search for our paper. Okay, nothing found. It hasn't been retracted yet. That is great. We can also go to websites like Web of Science or PubMed and pop it in there and see if anything comes up. So it's in there. It was very recently published, actually just a month ago. But we can see if there are similar articles like that. So there's a lot of information about this act that was implemented. So we can see some similar articles that are part of this conversation is what I would say. So that's pretty cool. And that's how you read a scientific article. Now let's just kind of summarize that up here. Let's talk about why you don't have to read every word and when you can skip. So like I said, anytime you hit something with lots of statistical or mathematical modeling because we're not PhD students, that's probably okay to skip over because we're not part of that discourse community. If we're just trying to get good information from these articles, that is not necessary because we're not trying to do the statistical work of these articles. And if we're not trying to reproduce these studies, that's not necessary. That often means that parts of the method section and results will likely have skimmable parts like regression analyses or confidence intervals or things like that. So those are kind of keywords you can look for that will be skimmable parts. But there are also other parts that are going to be really important. So trying to look at tables, any kind of graphs, those are parts that really help us try to understand an article and can really be meaningful in helping us understand an article. So those are the parts that I find really important. The most important parts of an article in my opinion are the introduction, probably the earlier parts of a method section. So the first few paragraphs of a method section are some of the most important parts. And then the discussion or conclusion section can be some of the most important parts of an article. And those are the ones that you really need to read deeply and really intensely to truly get the meat of what's going on in that article. And if you want to know more about how to read a scientific article, this was inspired by a blog post by Dr. Jennifer Raff and it's called How to Read and Understand Scientific Articles, a guide for non-scientists. And it's our really wonderful blog post that I use a lot for teaching folks how to understand scientific articles. And this is just specifically kind of how they work in the health sciences. They work a little bit differently in chemistry and engineering and some of the other sciences. But you know, the formula does stay pretty similar across different articles. And that's all I have for you right here. And if you have any more questions about how to access and use scientific articles, just contact us in the UWM Libraries. We are more than happy to help you. Thanks and have a-