How to read scientific articles: advice from scientists
“Reading a scientific article is something that makes you feel dumber nowhere,” writes scientist Adam Rubin in his humorous column for the journal Science. Adam is right: reading an article from a peer-reviewed scientific journal, many of us start with optimism (“In the end, it's just a few pages”), but end up with a desperate desire to urgently change the scope of activity to one where there are no scientific articles or they write somehow different. In fact, the problem is not in the articles - but in the way we select and read them.
The fact that scientists themselves are advised to do about this, will be described below.
Photo by Nic McPhee CC-BY
')
The basic advice that all, without exception, from scholars to students and journalists, is to break a big task (reading an article) into small blocks. Another question is how to do it in order not to sit at the material for years and at the same time understand the subject. To answer it, you have to “rewind back” - namely, to understand what article you are going to read and (most importantly) whether you need it.
There may be several goals that encourage reading an article: in some cases this is a necessary material, for example, to prepare for a seminar — then you have no choice whether to read it or not. In other cases, there is still such a choice - much depends on why we want to read something. The objectives of reading a scientific article may be as follows:
This is not an exhaustive list, but you can catch a tendency from it: either you want to understand a common question, or you are looking for a solution to a particular, narrowly-specific problem (sometimes this and the other together - but this will have to read several articles of different types ). And the type of article, in turn, helps to understand how your choice of material corresponds to the goals. It does not always make sense to read everything - sometimes it is not less useful to decide what to give up and what to return later: scientists are not at all embarrassed by such a pragmatic approach.
In a general sense, sources can be divided into “primary” (primary) and “secondary” (secondary). "Primary sources" in the scientific world can be reports on research results, case studies, in some cases - editorial articles and conference proceedings. In other words: materials in which the description of the new research (primary research article) is given. “Secondary sources” - review articles, book reviews, guidelines, comments, systematic reviews and meta-analysis. The word “secondary” does not mean anything bad: it’s about the fact that these materials are prepared on the basis of already existing articles.
The easiest way, of course, is to focus on the type of material (often it is placed in the subtitle of the article). In addition, a good "primary" article, in which the result of the research is given, can be distinguished by structure - it usually contains such sections as: "Introduction", "Methods", "Research" and "Interpretation" / "Discussion" (introduction, methods , results, and discussion, IMRAD).
However, the set of components may vary depending on the journal - in some additional subsections are required (for example, “Research hypothesis”), in others the “Discussion” section is reflected in the “Conclusions” block.
But the logic remains unchanged: the research article necessarily contains information about the research topic, about how and with what it was carried out, as well as what results were obtained and what it all means.
Let's go back to our goals. If we need to dive into the topic and get some general idea about it, it is useful to study the “secondary” sources: reviews, meta-analysis. They will allow to decide in which direction to move with your own research - which areas in the chosen topic are already well studied, and in which additional research is required.
If you already have general information, you can refer to primary sources - for example, research reports. They will help to understand exactly what has been done in the chosen field, in which direction you can “push” the results achieved by other scientists, how the related tasks were solved (this can be convenient if, for example, you want to apply new methods).
Tasks like “finding a solution to a specific problem” or “clarifying something in a selected area” are also closer to the primary sources — assuming that the general information that may be contained in the review materials is already well known to you, and you know exactly what to look for. . And to confirm his point of view, cases may turn out to be useful, among other things. Here, as in the previous case, we are talking about the fact that you are searching for quite specific, specific keywords.
Suppose you have chosen an article for yourself, the type of which corresponds to the task before you. Now the most important thing is not to try to read it from beginning to end as a whole. The most simple algorithm is proposed by Charles Darbin (Charles G. Durbin)
In his article "How to read a scientific research paper":
As we noted earlier, scientific articles consist of sections of several fixed types. Therefore it is necessary to deal with what is contained in each of them.
As in the case of the post on Habré, the title of the scientific article should be informative - that is, such that the reader immediately had an understanding of what the material will be about. Ideally, the reader should have an understanding of what type the text belongs to - whether it is research or meta-analysis.
Unfortunately, not all good scientific articles possess these characteristics. Editorial columns, book reviews and other materials that are not dedicated to describing new research may sometimes contain lengthy headlines. In this case, for starters, you can navigate by the author - if you know who in the chosen field is an authoritative scientist (review articles will help to deal with this). However, a vague headline may be the first signal — perhaps, as long as this material is worth deferring.
This is not just an introduction to the text: academic annotation is made according to strict rules. First, it must be compressed (in different journals there are different requirements for the exact volume of the abstract - but, as a rule, it takes one paragraph). Secondly, the abstract should answer the main questions of the article: provide background information about the research context, briefly describe its goals, methods, results and conclusions.
The introduction, as a rule, answers the question: why was the study conducted. In this section, the context is most fully reflected: what is happening in this industry now, what research was conducted earlier (literature analysis) and which questions scientists have not yet answered. All this should lead us to what, in fact, the author was engaged in as part of his scientific work: what is his contribution to resolving the problems and contradictions existing in this field.
By the way, the introduction is another marker for the reader. If it is hard for you to “wade” through its content, and you hardly grasp the meaning of what has been written, perhaps this material should be postponed. And this is not about the fact that it is poorly written: perhaps, for the time being, you just need to switch to review articles or better study the theoretical basis. After that, the text will seem more understandable, and you can extract much more benefit from it.
This section describes what the author did - how he collected data, formed a sample, or prepared samples for research, which calculations and experiments he carried out. On the one hand, a detailed indication of the methods helps to understand what has not yet been implemented in this area: which areas were consciously (or unconsciously) not affected by scientists. On the other hand, a good description of the materials and methods is the key to reproducing such an experiment on your own.
The obtained results are placed in a separate unit. In this section, as a rule, no conclusions are made about what the author was able to achieve - only specific data, graphs, and charts. Scientists usually do not have the opportunity to paint every number they receive - so if you want to thoroughly understand the content of the material, you will have to carefully study all the illustrations and tables.
But in this section you can see the author's opinion about the results: what was done and what was not, what place the study takes in relation to other works, what were the limitations and assumptions of this study, and why the author received this result. It is important to keep in mind: this block, unlike the previous two, contains the point of view of the author, and not just dry facts - and you do not have to come to exactly the same conclusions by reading the material.
So, the “anatomy” of a scientific article is clear to us, the main markers that are worth paying attention to when reading, too. Now it is necessary to understand how the scientists themselves read such materials.
Jennifer Raff, a bioanthropologist and geneticist from the University of Kansas, provides some practical tips to help the reader:
Jennifer advises using a critical approach and not trusting researchers “on the word” - so she advises to pay less attention to the “Interpretation” section, to give yourself answers to the questions “What have the authors achieved?”, “Have they answered specific research questions?” and also to study the opinions of other members of the scientific community about this article.
Moreover, according to Jennifer, one should not start reading the article from the abstract, but from the introduction - this, in her words, helps her to keep herself in a critical mood and evaluate the material more impartially - without relying on the authors' opinion.
True, this approach is not practiced by everyone: the journal Science surveyed 12 scientists (among them was the editor-in-chief of a scientific journal) to find out how they read articles. It turned out that the majority still begins with annotations - especially in cases where they are not very well acquainted with the subject matter of the study.
At the same time, about half of the scientists surveyed practice the approach of Jennifer and study every word - from their point of view, it is impossible to sort out a scientific article without having “worked out” every obscure term. The other half first reads the material fluently - and only then (if necessary) delves into the details. And here are some more tips that practicing researchers gave Science journalists:
1. Do not hesitate to ask questions: google incomprehensible terms, use “not enough scientific” sources like Wikipedia or blogs, and also discuss the difficult moments of articles with colleagues. After all, scientific texts are not fiction. In a pinch, you can even write a letter to the authors - and ask to explain an incomprehensible point.
2. Break the article into blocks and read it in parts - it is quite possible that it will take you several days to read one article. This is normal - the scientists are doing the same thing.
3. Print the article on paper and highlight the most important points for you. This will make it easier to perceive the material and quickly restore knowledge when you need to return to the text. Pay special attention to those moments that change your perception and understanding of the research topic.
4. Try to be an “active reader” - that is, before you start reading, determine what exactly you want to get from the article: a list of the most significant research in this area, a list of authors you should pay attention to, a plan for conducting a similar study, confirmation or refutation of your point of view. In some cases, information that answers your question is worth saving separately. For example, Lina A. Colucci of the Harvard-MIT Health Sciences and Technology program stores all relevant information that relates to the context of the study in a separate text file. And prefers to accumulate information on materials and methods of research in Excel-tables.
1. Think about why you need to read a scientific article? It depends on what kind of goal you pursue whether you need research materials (primary sources) or survey materials (secondary sources).
2. Try to formulate for yourself the main task that you want to solve. If your goal is to get a general idea of the subject, think about what exactly you want to know: who worked on this topic? How did the research tasks change? What is now in the sphere of interests of scientists? And so on. This will allow you to read “in focus”, keeping in mind the question posed.
3. If you figure out what type of article you need to read, and choose the material for yourself, you can start reading. Pre-worth an article - it will facilitate the process.
4. Do you think that the research topic and the content of the annotation / introduction correspond to your interests? Then you can continue reading.
5. Reading the introduction, determine for yourself the main issue (what scientists do in this area and what the main problem they solve) and specific issues that the authors of the material pose. Make a brief retelling of the research context. Try to write out unfamiliar words and clarify their meaning - even in "unscientific" sources.
6. Do not forget about breaks. Reading a scientific article is a long and difficult task, you may have to break it into several days.
7. Reading the “Materials and Methods” section, try entering information in a table (if you need to understand how research is conducted in this area, and you want to compare the approaches of different scientists) or draw a diagram of the study for yourself. This will help to better understand what exactly the scientists did.
8. In the "Results" section, pay close attention to graphic information: tables and figures. As a rule, they contain even more valuable data than directly in the text.
9. Re-read the discussion section. Think about whether the authors have answered the questions posed in the introduction? How consistent is your interpretation of the results?
10. Finally, examine the opinions of critics and editors: what other authors say about this work.
PS How technological entrepreneurship and science fiction are connected .
The fact that scientists themselves are advised to do about this, will be described below.
Photo by Nic McPhee CC-BY')
The basic advice that all, without exception, from scholars to students and journalists, is to break a big task (reading an article) into small blocks. Another question is how to do it in order not to sit at the material for years and at the same time understand the subject. To answer it, you have to “rewind back” - namely, to understand what article you are going to read and (most importantly) whether you need it.
We understand the goals
There may be several goals that encourage reading an article: in some cases this is a necessary material, for example, to prepare for a seminar — then you have no choice whether to read it or not. In other cases, there is still such a choice - much depends on why we want to read something. The objectives of reading a scientific article may be as follows:
- Understand basic issues / get a general idea of a certain direction in science
- Get ideas for your own research
- Find confirmation of your point of view
- Find a solution to a specific problem.
- Refine something in the selected area
This is not an exhaustive list, but you can catch a tendency from it: either you want to understand a common question, or you are looking for a solution to a particular, narrowly-specific problem (sometimes this and the other together - but this will have to read several articles of different types ). And the type of article, in turn, helps to understand how your choice of material corresponds to the goals. It does not always make sense to read everything - sometimes it is not less useful to decide what to give up and what to return later: scientists are not at all embarrassed by such a pragmatic approach.
What are the types of articles and how they differ
In a general sense, sources can be divided into “primary” (primary) and “secondary” (secondary). "Primary sources" in the scientific world can be reports on research results, case studies, in some cases - editorial articles and conference proceedings. In other words: materials in which the description of the new research (primary research article) is given. “Secondary sources” - review articles, book reviews, guidelines, comments, systematic reviews and meta-analysis. The word “secondary” does not mean anything bad: it’s about the fact that these materials are prepared on the basis of already existing articles.
How to distinguish different types of articles
The easiest way, of course, is to focus on the type of material (often it is placed in the subtitle of the article). In addition, a good "primary" article, in which the result of the research is given, can be distinguished by structure - it usually contains such sections as: "Introduction", "Methods", "Research" and "Interpretation" / "Discussion" (introduction, methods , results, and discussion, IMRAD).
However, the set of components may vary depending on the journal - in some additional subsections are required (for example, “Research hypothesis”), in others the “Discussion” section is reflected in the “Conclusions” block.
But the logic remains unchanged: the research article necessarily contains information about the research topic, about how and with what it was carried out, as well as what results were obtained and what it all means.
How to choose the type of article
Let's go back to our goals. If we need to dive into the topic and get some general idea about it, it is useful to study the “secondary” sources: reviews, meta-analysis. They will allow to decide in which direction to move with your own research - which areas in the chosen topic are already well studied, and in which additional research is required.
If you already have general information, you can refer to primary sources - for example, research reports. They will help to understand exactly what has been done in the chosen field, in which direction you can “push” the results achieved by other scientists, how the related tasks were solved (this can be convenient if, for example, you want to apply new methods).
Tasks like “finding a solution to a specific problem” or “clarifying something in a selected area” are also closer to the primary sources — assuming that the general information that may be contained in the review materials is already well known to you, and you know exactly what to look for. . And to confirm his point of view, cases may turn out to be useful, among other things. Here, as in the previous case, we are talking about the fact that you are searching for quite specific, specific keywords.
So how do you read scientific articles
Suppose you have chosen an article for yourself, the type of which corresponds to the task before you. Now the most important thing is not to try to read it from beginning to end as a whole. The most simple algorithm is proposed by Charles Darbin (Charles G. Durbin)
In his article "How to read a scientific research paper":
- The first step is to make sure the title and keywords match the topic that interests you.
- Read annotation and (if necessary) discussion / conclusions
- If the abstract and discussion clearly describe the goals and objectives, the hypothesis is unambiguously defined, the conclusions are indicated - and all this information is relevant to your goals - then the article can be read in its entirety.
What is the article
As we noted earlier, scientific articles consist of sections of several fixed types. Therefore it is necessary to deal with what is contained in each of them.
Headline
As in the case of the post on Habré, the title of the scientific article should be informative - that is, such that the reader immediately had an understanding of what the material will be about. Ideally, the reader should have an understanding of what type the text belongs to - whether it is research or meta-analysis.
Unfortunately, not all good scientific articles possess these characteristics. Editorial columns, book reviews and other materials that are not dedicated to describing new research may sometimes contain lengthy headlines. In this case, for starters, you can navigate by the author - if you know who in the chosen field is an authoritative scientist (review articles will help to deal with this). However, a vague headline may be the first signal — perhaps, as long as this material is worth deferring.
Abstract (it's abstract)
This is not just an introduction to the text: academic annotation is made according to strict rules. First, it must be compressed (in different journals there are different requirements for the exact volume of the abstract - but, as a rule, it takes one paragraph). Secondly, the abstract should answer the main questions of the article: provide background information about the research context, briefly describe its goals, methods, results and conclusions.
Introduction
The introduction, as a rule, answers the question: why was the study conducted. In this section, the context is most fully reflected: what is happening in this industry now, what research was conducted earlier (literature analysis) and which questions scientists have not yet answered. All this should lead us to what, in fact, the author was engaged in as part of his scientific work: what is his contribution to resolving the problems and contradictions existing in this field.
By the way, the introduction is another marker for the reader. If it is hard for you to “wade” through its content, and you hardly grasp the meaning of what has been written, perhaps this material should be postponed. And this is not about the fact that it is poorly written: perhaps, for the time being, you just need to switch to review articles or better study the theoretical basis. After that, the text will seem more understandable, and you can extract much more benefit from it.
It is important to understand that some scientific articles are the result of many years of work of dozens of scientists. It is unrealistic to expect that you will be able to run through the text in the evening and understand everything. Jeremy Borniger , Ohio State University
Materials and methods
This section describes what the author did - how he collected data, formed a sample, or prepared samples for research, which calculations and experiments he carried out. On the one hand, a detailed indication of the methods helps to understand what has not yet been implemented in this area: which areas were consciously (or unconsciously) not affected by scientists. On the other hand, a good description of the materials and methods is the key to reproducing such an experiment on your own.
Results
The obtained results are placed in a separate unit. In this section, as a rule, no conclusions are made about what the author was able to achieve - only specific data, graphs, and charts. Scientists usually do not have the opportunity to paint every number they receive - so if you want to thoroughly understand the content of the material, you will have to carefully study all the illustrations and tables.
Interpretation / discussion
But in this section you can see the author's opinion about the results: what was done and what was not, what place the study takes in relation to other works, what were the limitations and assumptions of this study, and why the author received this result. It is important to keep in mind: this block, unlike the previous two, contains the point of view of the author, and not just dry facts - and you do not have to come to exactly the same conclusions by reading the material.
How do scientists read articles
So, the “anatomy” of a scientific article is clear to us, the main markers that are worth paying attention to when reading, too. Now it is necessary to understand how the scientists themselves read such materials.
Jennifer Raff, a bioanthropologist and geneticist from the University of Kansas, provides some practical tips to help the reader:
- Formulate the main question of the article (what problem are the scientists trying to solve in this area)
- Describe the research context in several sentences (Jennifer suggests using five sentences) - or, in other words, tell yourself what was done before the present study to answer the main question.
- Identify specific questions that the authors are going to answer.
- In the process of reading, write out all unfamiliar words and look for their meanings.
- In the process of reading the section "Materials and Methods" draw an approximate scheme of the study.
- Formulate for yourself the results of the study in one paragraph, focusing on the facts, and not on the opinions of the authors.
Jennifer advises using a critical approach and not trusting researchers “on the word” - so she advises to pay less attention to the “Interpretation” section, to give yourself answers to the questions “What have the authors achieved?”, “Have they answered specific research questions?” and also to study the opinions of other members of the scientific community about this article.
Moreover, according to Jennifer, one should not start reading the article from the abstract, but from the introduction - this, in her words, helps her to keep herself in a critical mood and evaluate the material more impartially - without relying on the authors' opinion.
True, this approach is not practiced by everyone: the journal Science surveyed 12 scientists (among them was the editor-in-chief of a scientific journal) to find out how they read articles. It turned out that the majority still begins with annotations - especially in cases where they are not very well acquainted with the subject matter of the study.
At the same time, about half of the scientists surveyed practice the approach of Jennifer and study every word - from their point of view, it is impossible to sort out a scientific article without having “worked out” every obscure term. The other half first reads the material fluently - and only then (if necessary) delves into the details. And here are some more tips that practicing researchers gave Science journalists:
1. Do not hesitate to ask questions: google incomprehensible terms, use “not enough scientific” sources like Wikipedia or blogs, and also discuss the difficult moments of articles with colleagues. After all, scientific texts are not fiction. In a pinch, you can even write a letter to the authors - and ask to explain an incomprehensible point.
2. Break the article into blocks and read it in parts - it is quite possible that it will take you several days to read one article. This is normal - the scientists are doing the same thing.
3. Print the article on paper and highlight the most important points for you. This will make it easier to perceive the material and quickly restore knowledge when you need to return to the text. Pay special attention to those moments that change your perception and understanding of the research topic.
4. Try to be an “active reader” - that is, before you start reading, determine what exactly you want to get from the article: a list of the most significant research in this area, a list of authors you should pay attention to, a plan for conducting a similar study, confirmation or refutation of your point of view. In some cases, information that answers your question is worth saving separately. For example, Lina A. Colucci of the Harvard-MIT Health Sciences and Technology program stores all relevant information that relates to the context of the study in a separate text file. And prefers to accumulate information on materials and methods of research in Excel-tables.
So TL; DR
1. Think about why you need to read a scientific article? It depends on what kind of goal you pursue whether you need research materials (primary sources) or survey materials (secondary sources).
2. Try to formulate for yourself the main task that you want to solve. If your goal is to get a general idea of the subject, think about what exactly you want to know: who worked on this topic? How did the research tasks change? What is now in the sphere of interests of scientists? And so on. This will allow you to read “in focus”, keeping in mind the question posed.
3. If you figure out what type of article you need to read, and choose the material for yourself, you can start reading. Pre-worth an article - it will facilitate the process.
- If you are familiar with the topic and are critical, you should start by reading the introduction. This will allow you not to be distracted by the conclusions of the authors and independently form an opinion about their work.
- If you trust the source and the authors, and are not so well versed in the topic, you should start with annotations. If the information in the abstract is not enough - refer to the last section (“Discussion” or “Conclusions”)
4. Do you think that the research topic and the content of the annotation / introduction correspond to your interests? Then you can continue reading.
5. Reading the introduction, determine for yourself the main issue (what scientists do in this area and what the main problem they solve) and specific issues that the authors of the material pose. Make a brief retelling of the research context. Try to write out unfamiliar words and clarify their meaning - even in "unscientific" sources.
6. Do not forget about breaks. Reading a scientific article is a long and difficult task, you may have to break it into several days.
7. Reading the “Materials and Methods” section, try entering information in a table (if you need to understand how research is conducted in this area, and you want to compare the approaches of different scientists) or draw a diagram of the study for yourself. This will help to better understand what exactly the scientists did.
8. In the "Results" section, pay close attention to graphic information: tables and figures. As a rule, they contain even more valuable data than directly in the text.
9. Re-read the discussion section. Think about whether the authors have answered the questions posed in the introduction? How consistent is your interpretation of the results?
10. Finally, examine the opinions of critics and editors: what other authors say about this work.
PS How technological entrepreneurship and science fiction are connected .
Source: https://habr.com/ru/post/336672/
All Articles