Difference between revisions of "Writing an outline"

From Sustainability Methods
m
 
(8 intermediate revisions by one other user not shown)
Line 1: Line 1:
Writing an outline
+
 
 
With the completion of an outline you take one first step of what your research could be all about. One should never forget that research is an evolving and iterative process, though. Still, writing an outline makes you settle for what you want to focus on in this moment, and more importantly, also allows your supervisors as well as your peers to give you structured feedback. This is why any research project should start with the landmark of writing an outline. Different branches in science have different focal points and norms that an outline is build upon. Here, we present an approach that tries to do justice to the diversity of approaches that are out there, yet it is always advisable to ask your supervisors for modification if need be.  
 
With the completion of an outline you take one first step of what your research could be all about. One should never forget that research is an evolving and iterative process, though. Still, writing an outline makes you settle for what you want to focus on in this moment, and more importantly, also allows your supervisors as well as your peers to give you structured feedback. This is why any research project should start with the landmark of writing an outline. Different branches in science have different focal points and norms that an outline is build upon. Here, we present an approach that tries to do justice to the diversity of approaches that are out there, yet it is always advisable to ask your supervisors for modification if need be.  
  
All research starts with a  title. Personally, I read hundreds of titles of research papers each month, and only a small portion are appealing to my specific focus and interest to invite me to read further. Titles are the door-opener for most researchers, which is why titles should be simple and to the point. If a title is too long it will lack clarity and crispiness. If a title is too short, it will surely not give away enough information needed to know what the research is all about. Often people try to make a fancy title that is supposed to be witty or funny and contains some sort of wordplay or inside joke. Avoid this. You may go for such a title later once the research is done and the paper is written, yet such titles need to be earned. Hence especially in an outline it is best for a title that is walking the fine line of going away enough information but not too much.  
+
===== Working Title=====
 +
All research starts with a  title. Personally, I read hundreds of titles of research papers each month, and only a small portion are appealing to my specific focus and interest to invite me to read further. Titles are the door-opener for most researchers, which is why titles should be simple and to the point. If a title is too long it will lack clarity and crispiness. If a title is too short, it will surely not give away enough information needed to know what the research is all about. Often people try to make a fancy title that is supposed to be witty or funny and contains some sort of wordplay or inside joke. Avoid this. You may go for such a title later once the research is done and the paper is written, yet such titles need to be earned. Hence especially in an outline it is best for a title that is walking the fine line of giving away enough information but not too much.  
  
 +
===== Participants =====
 
In the end, it is going to be you who will do the research. You will sit at your desk, you will gather the data and look at the literature, you will get deeper into the topic, thus it is you who will basically write the research. To this end, it has proven of immense value to have a network of peers. More often than not, this is an informal network for critical reflection, but also for support. Such a peer network will not be mentioned here if it is not actively involved in conducting the research. Beside your supervisors actually very few people will be involved in your research. You may have someone helping with the analysis or being experienced in the topic if you are a PhD student, yet in a Bachelor or Master thesis the focus is stronger on proving that you can conduct independent research. While in a PhD this is also the goal, it is on a more sophisticated level, where because of the longer timeline and thus deeper focus collaboration may be of greater importance. It is also quite important to clarify roles and expectations at the beginning. Remember that a thesis is your work.  
 
In the end, it is going to be you who will do the research. You will sit at your desk, you will gather the data and look at the literature, you will get deeper into the topic, thus it is you who will basically write the research. To this end, it has proven of immense value to have a network of peers. More often than not, this is an informal network for critical reflection, but also for support. Such a peer network will not be mentioned here if it is not actively involved in conducting the research. Beside your supervisors actually very few people will be involved in your research. You may have someone helping with the analysis or being experienced in the topic if you are a PhD student, yet in a Bachelor or Master thesis the focus is stronger on proving that you can conduct independent research. While in a PhD this is also the goal, it is on a more sophisticated level, where because of the longer timeline and thus deeper focus collaboration may be of greater importance. It is also quite important to clarify roles and expectations at the beginning. Remember that a thesis is your work.  
  
 +
===== Background and topic =====
 
The topical focus and its background are often what drives people. Most researchers are very exited about their topics, and it is valuable to have something that can fuel your energy while you work on your thesis. There will be ups and downs surely, yet it is still good to focus on something that not only drives you, but also current research. Timely topics are ideally embedded into a longer development that led to the emergence of the topic. You do not want to work on a topic where thousands and thousands of paper were already published, but ideally you should also not work on something that is so new that there are no shoulders to stand on. Researchers do really stand on the shoulders of giants, and despite the thrive to make innovative and timely research, we should never forget that we are part of a bigger movement. Within this movement, we will add one tiny step. A friend once said that we are drops in a wave - a fitting allegory when looking how research evolves. Still, our contribution matters, and may move the research landscape forward. What it does however more importantly is that it moves us forward. A thesis is first and foremost a prove that you can conduct research, and that you are thus able to contribute to the scientific community. Because of this, the topic is less important than most people think it is, because research is a transformational experience for the research. Do not misunderstand me, all topics of past people I supervised were really important to me. What is however even more important is that they learned to overcome themselves, and slay the dragon that is their thesis.  
 
The topical focus and its background are often what drives people. Most researchers are very exited about their topics, and it is valuable to have something that can fuel your energy while you work on your thesis. There will be ups and downs surely, yet it is still good to focus on something that not only drives you, but also current research. Timely topics are ideally embedded into a longer development that led to the emergence of the topic. You do not want to work on a topic where thousands and thousands of paper were already published, but ideally you should also not work on something that is so new that there are no shoulders to stand on. Researchers do really stand on the shoulders of giants, and despite the thrive to make innovative and timely research, we should never forget that we are part of a bigger movement. Within this movement, we will add one tiny step. A friend once said that we are drops in a wave - a fitting allegory when looking how research evolves. Still, our contribution matters, and may move the research landscape forward. What it does however more importantly is that it moves us forward. A thesis is first and foremost a prove that you can conduct research, and that you are thus able to contribute to the scientific community. Because of this, the topic is less important than most people think it is, because research is a transformational experience for the research. Do not misunderstand me, all topics of past people I supervised were really important to me. What is however even more important is that they learned to overcome themselves, and slay the dragon that is their thesis.  
 
Once you found a topic that excites you, it is advisable to iterate it with your peers. Write down why your topic is timely, how it contributes to the wider research, and what  the current state of the art is. You want to make a thorough assessment yet have to be careful because reading too much may confuse you. Do not expect that everything is coherent and there are no contradictions. Research is discourse, and these discourses evolve over time. Often researchers disagree, and not everything makes sense. Be prepared to be confused. It is your job to evolve a critical perspective of the state of the art of the literature, and identify landmark papers. Also, try to find previous research that aimed in the same direction and learn how they approached the topic. What were the methodologies, where did the researchers struggle, and what were their recommendations. Lastly, try to develop an elevator pitch on why this topic is important to you. If you can explain in short why you think this research needs to be done, you are onto something.
 
Once you found a topic that excites you, it is advisable to iterate it with your peers. Write down why your topic is timely, how it contributes to the wider research, and what  the current state of the art is. You want to make a thorough assessment yet have to be careful because reading too much may confuse you. Do not expect that everything is coherent and there are no contradictions. Research is discourse, and these discourses evolve over time. Often researchers disagree, and not everything makes sense. Be prepared to be confused. It is your job to evolve a critical perspective of the state of the art of the literature, and identify landmark papers. Also, try to find previous research that aimed in the same direction and learn how they approached the topic. What were the methodologies, where did the researchers struggle, and what were their recommendations. Lastly, try to develop an elevator pitch on why this topic is important to you. If you can explain in short why you think this research needs to be done, you are onto something.
  
Just as topics are embedded in the current scientific discourse, sch research looks at partials of reality. In order to generate such a specific view of reality, research builds on theories. Theories in science can be operatiatnlized at different scales. Some theories are on a conceptual level, which is basically very theoretical and often restricted to non-empirical research such as much of philosophy. Other researchers may build on frameworks, and is very applied. In between are paradigms, which are the types of theories that branches of research are for some time built upon. Examples for concepts would justice or peace, examples of paradigms would be effective altruism or ecosystem services, and examples for frameworks would be concrete assessment schemes or the sustainable development goals. Much confusion in theory work is because these three levels are confused, which is why it is important to localise your theory in your work clearly.  
+
===== Key supporting theories =====
Another common problem when writing an outline is to settle for a low number of theories. If you ask me, one theory can be enough, and two can be exiting. More than two theories are more often than not impossible to tame. Hence you need to remember that you look at a version of reality, and that theories enable you to either test hypothesis or create research questions that are open and specific at the same time. You may not want to know how everything works, but how it works under a specific viewpoint or theory of reality. This is what theories are all about. Researchers theorise how mechanisms in the world might work, how patterns emerge, why people act, and why societies fail. There is a plethora of theories, and it is important to be literature in the theories that are within your realm.  
+
Just as topics are embedded in the current scientific discourse, such research looks at partials of reality. In order to generate such a specific view of reality, research builds on theories. Theories in science can be operationalized at different scales. Some theories are on a conceptual level, which is basically very theoretical and often restricted to non-empirical research such as much of philosophy. Other researchers may build on frameworks, and is very applied. In between are paradigms, which are the types of theories that branches of research are for some time built upon. Examples for concepts would justice or peace, examples of paradigms would be effective altruism or ecosystem services, and examples for frameworks would be concrete assessment schemes or the sustainable development goals. Much confusion in theory work is because these three levels are confused, which is why it is important to localise your theory in your work clearly.  
 +
Another common problem when writing an outline is to settle for a low number of theories. If you ask me, one theory can be enough, and two can be exciting. More than two theories are more often than not impossible to tame. Hence you need to remember that you look at a version of reality, and that theories enable you to either test hypothesis or create research questions that are open and specific at the same time. You may not want to know how everything works, but how it works under a specific viewpoint or theory of reality. This is what theories are all about. Researchers theorize how mechanisms in the world might work, how patterns emerge, why people act, and why societies fail. There is a plethora of theories, and it is important to be familiar with the literature of the theories that are within your realm.  
  
Most topics are associated with certain theories, such as mainstream economics often resolves around utilitarianism. While you can operate on safe ground if you follow suit with these associations, it can also be exiting yet daring to superimpose a theory onto a topic that has never been exposed to it. Yet this should ideally be discussed with your supervisor, because while it can be exiting, it may be prone to challenges that are hard to anticipate by the uninitiated. This is one of the key points where you should build on the experience of others. Lastly, while it may be exiting to work on a scientific theory, always remember that you look at a snapshot. Theories are often proposed, and then in subsequent years applied. Based on this, they are revised, scrutinised, and often the research community becomes divided about specific theories. While many researchers question or even reject evolutionary altruism for example, others may work based on this theory for their whole career. Choosing a theory is a normative responsibility, you decide to take a specific perspective, and this is a key responsibility as a researcher. While you should choose wisely, I would also say: Do not overthink it. Most early researchers tend to think way too much about it, in fact I never saw anyone thinking too little about it. And after all, at this point your supervisors should give you feedback.  
+
Most topics are associated with certain theories, such as mainstream economics often resolves around utilitarianism. While you can operate on safe ground if you follow suit with these associations, it can also be exciting yet daring to superimpose a theory onto a topic that has never been exposed to it. Yet this should ideally be discussed with your supervisor, because while it can be exciting, it may be prone to challenges that are hard to anticipate by the uninitiated. This is one of the key points where you should build on the experience of others. Lastly, while it may be exciting to work on a scientific theory, always remember that you look at a snapshot. Theories are often proposed, and then in subsequent years applied. Based on this, they are revised, scrutinized, and often the research community becomes divided about specific theories. While many researchers question or even reject evolutionary altruism for example, others may work based on this theory for their whole career. Choosing a theory is a normative responsibility, you decide to take on a specific perspective, and this is a key responsibility as a researcher. While you should choose wisely, I would also say: Do not overthink it. Most early researchers tend to think way too much about it, in fact I never saw anyone thinking too little about it. And after all, at this point your supervisors should give you feedback.  
  
Ideally, you can at this point already identify the main references that your research is being based upon. Make sure to reference the sources that are most recognised by the research community, but also closest to what you think are the best sources. Often this may be one and the same, but sometimes it differs. What is crucial is to focus on the main sources. Especially early career researchers are so happy that they read so much stuff that they want to quote it all. While this sounds like a great idea, remember that not everybody is as deeply embedded as your are, thus focus on the most important sources. On a less dramatic note, make sure that your quotations are in the sea style. Settle early on one citation style, use a software to guarantee coherence, and never change it from then on.
+
===== Main references =====
 +
Ideally, you can at this point already identify the main references that your research is being based upon. Make sure to reference the sources that are most recognized by the research community, but also closest to what you think are the best sources. Often this may be one and the same, but sometimes it differs. What is crucial is to focus on the main sources. Especially early career researchers are so happy that they read so much stuff that they want to quote it all. While this sounds like a great idea, remember that not everybody is as deeply embedded as your are, thus focus on the most important sources. On a less dramatic note, make sure that your quotations are in the same style. Settle early on one citation style, use a software to guarantee coherence, and never change it from then on.
  
 +
===== Research question or hypothesis =====
 
The next point is a very delicate one: Research questions vs. hypotheses. Let us start with the more simple point. How much of those should you have? It is really difficult to answer one question. This is something that philosophers may do, but most research demands more questions that build on each other. See it from a structural point of view, answering one question pre-structures your introduction into one big blob of text. That is certainly not convenient. 2-5 questions seem ideal, because -let's face it- it is really hard to remember more than 5 things intuitively. We then tend to forget, and it also has the drawback to make your research seem über-structured. Hence it can be a good heuristic to have 2-5 research questions or hypotheses. Now let us move to the more troubling part, which is the question whether it is research questions or hypotheses. From a philosophy of science standpoint, the two should be mutually exclusive. While hypotheses are surely deductive and demand a clear confirmation or rejection, research questions are somewhat more open and hence inductive. Thus while the latter has never been clear cut, it can be helpful to draw such a clear line. Some folks may now raise the streetwise question whether we should not all be thinking abductively anyway? In a sense this is where all science moves anyway, because we need both inductive and deductive approaches, and ideally combines the benefits of both in the long run. Yet this is mainly a questions of both experience and temporal grain. Working abductively means that you can clearly remark between hypotheses and research questions, because the first demand a very clear structure, while the latter are way more open. There is an underlying relation that hypotheses tend to be more quantitative, while research questions are often more qualitative. Still, this is a mere correlation and not a really causal relation, but it confuses a great many people. Chasing whether to write hypotheses or research questions is in addition often a question of the tradition of the respective scientific discipline. Many folks in natural science are still proud of their razor sharp hypotheses, and other fellow within socials science lean clearly towards research questions. Ideally, find out what your supervisors demand, which is the most simple heuristic to this end. Still, this point underlines that from a philosophy of science standpoint the silo-mentality of disciplines has its reason, but does not always make sense. What is most important for you is that an adductive procedure may be most desirable, yet only in the long run. It is part of the tradition of many sciences the postulate, test, and adjust. This works more on a time-scale of years or decades, but thus demands a temporal grain of a longer research agenda. Hence an abductive is not suitable for a shorter project such as a thesis.  
 
The next point is a very delicate one: Research questions vs. hypotheses. Let us start with the more simple point. How much of those should you have? It is really difficult to answer one question. This is something that philosophers may do, but most research demands more questions that build on each other. See it from a structural point of view, answering one question pre-structures your introduction into one big blob of text. That is certainly not convenient. 2-5 questions seem ideal, because -let's face it- it is really hard to remember more than 5 things intuitively. We then tend to forget, and it also has the drawback to make your research seem über-structured. Hence it can be a good heuristic to have 2-5 research questions or hypotheses. Now let us move to the more troubling part, which is the question whether it is research questions or hypotheses. From a philosophy of science standpoint, the two should be mutually exclusive. While hypotheses are surely deductive and demand a clear confirmation or rejection, research questions are somewhat more open and hence inductive. Thus while the latter has never been clear cut, it can be helpful to draw such a clear line. Some folks may now raise the streetwise question whether we should not all be thinking abductively anyway? In a sense this is where all science moves anyway, because we need both inductive and deductive approaches, and ideally combines the benefits of both in the long run. Yet this is mainly a questions of both experience and temporal grain. Working abductively means that you can clearly remark between hypotheses and research questions, because the first demand a very clear structure, while the latter are way more open. There is an underlying relation that hypotheses tend to be more quantitative, while research questions are often more qualitative. Still, this is a mere correlation and not a really causal relation, but it confuses a great many people. Chasing whether to write hypotheses or research questions is in addition often a question of the tradition of the respective scientific discipline. Many folks in natural science are still proud of their razor sharp hypotheses, and other fellow within socials science lean clearly towards research questions. Ideally, find out what your supervisors demand, which is the most simple heuristic to this end. Still, this point underlines that from a philosophy of science standpoint the silo-mentality of disciplines has its reason, but does not always make sense. What is most important for you is that an adductive procedure may be most desirable, yet only in the long run. It is part of the tradition of many sciences the postulate, test, and adjust. This works more on a time-scale of years or decades, but thus demands a temporal grain of a longer research agenda. Hence an abductive is not suitable for a shorter project such as a thesis.  
  
The next thing you want to focus on is the study area. This does not necessarily need to be a concrete space, but can also be something of a less concrete system. In a systematic review or critical content analysis this can be a branch of the literature. Within an image analysis this can be a set of paintings. In an ethnographic study this can be a specific group of people. Within a lab experiments it can be a completely artificially designed study, such as a set of planting pots that are irrigated, shaded and endure different temperature settings. This the study area is the specific system or setting that is being investigated. Again, there is a certain tradition that deductive studies are more tamed or have a deeper control or understanding of their study area, while inductive studies are more open minded and less clear cut when it comes the the pre-study understanding of the study area. Yet do not be fooled, inductive studies can be very clear when it comes to looking at something specific, these studies do it just with a different kind of open-mindedness. What is the benefit for you is to be clear in defining where you want to work in, may it be inductive or deductive. It is helpful to have a clear definition, because otherwise you will be either overworking yourself or have a sample set that is too small. The study area should ideally be chosen to make it very clear how your research represents some dynamics, pattern or mechanisms that can either serve as a basis to approximate reproducible knowledge, or at leats knowledge where the path towards it can be clearly documented. Hence the study area is something that allows you to make your research specific and thus tamed. Your choice remarks a start as well as an end. Ideally, it should be exciting and represent a clear knowledge gap, but in addition it also demands to represent a bigger picture that is well reprpesneteed by the chosen system. If you want to work on small business dynamics you need to work with organisation that can be seen as representing the overall dynamics. If you want to make a survey, you certainly do not want to focus on outliers if you try to represent a bigger group. Ideally you build on the experience or already established researchers to learn to make the right choice concerning your study area.  
+
===== Study area =====
 +
The next thing you want to focus on is the study area. This does not necessarily need to be a concrete space, but can also be something of a less concrete system. In a systematic review or critical content analysis this can be a branch of the literature. Within an image analysis this can be a set of paintings. In an ethnographic study this can be a specific group of people. Within a lab experiments it can be a completely artificially designed study, such as a set of planting pots that are irrigated, shaded and endure different temperature settings. This the study area is the specific system or setting that is being investigated. Again, there is a certain tradition that deductive studies are more tamed or have a deeper control or understanding of their study area, while inductive studies are more open minded and less clear cut when it comes the the pre-study understanding of the study area. Yet do not be fooled, inductive studies can be very clear when it comes to looking at something specific, these studies do it just with a different kind of open-mindedness. What is the benefit for you is to be clear in defining where you want to work in, may it be inductive or deductive. It is helpful to have a clear definition, because otherwise you will be either overworking yourself or have a sample set that is too small. The study area should ideally be chosen to make it very clear how your research represents some dynamics, pattern or mechanisms that can either serve as a basis to approximate reproducible knowledge, or at least knowledge where the path towards it can be clearly documented. Hence the study area is something that allows you to make your research specific and thus tamed. Your choice remarks a start as well as an end. Ideally, it should be exciting and represent a clear knowledge gap, but in addition it also demands to represent a bigger picture that is well represented by the chosen system. If you want to work on small business dynamics you need to work with organization that can be seen as representing the overall dynamics. If you want to make a survey, you certainly do not want to focus on outliers if you try to represent a bigger group. Ideally you build on the experience or already established researchers to learn to make the right choice concerning your study area.  
  
It is somewhat beyond an introductionary entry to tame the whole world of data, which includes gathering data, analysing it, and interpreting it. Yet we may take a peek at what is relevant, and can and should be achieved in the realms of an outline. The core goal to this end is probably the sweet spot between innovation and feasibility. Innovation in terms of your research design means that you read the relevant literature and, integrated it into your considerations on how to gather and analyse data. Start with textbook knowledge, and then go from the most highly cited general papers on the subject into more specific papers that are closer to the context you focus on. There is a lot of experience out there that is relevant to your specific methodological design, you just need to adapt it to your focus. This is exactly the reason why divergent thinking is a key skill in academia. Instead adding to the pile of literature you read you may want to consult an expert, however only with specific questions. You can get the basics on simple methods from textbooks, yet your supervisor may know best if there is anything else to consider. More often than not this may not be the case, yet there are things to consider. For instance is the sample size often relevant on both quantitative and qualitative science, and other questions such as heterogeneity of the sample are often of considerable importance. Questions of bias often already play a central role when gathering data, and then there are the simple mechanics. You to you document the gathering process, are there any machines or tools needed, and are there other technical aspects to consider such as software solutions? Yet the key question when getting data is whether if it is feasible or not. No data can be gathered if it is impossible to get it. Early career researchers often overplay their hand to this end. Make a pretest and see how long it takes you. Also remember that over time you may get faster concerning some mechanics, but also new and unanticipated problems may arise. Plan at least a third more time that the most conservative estimate. If you do not need it, fine, yet much research already took longer at this stage. Lastly, research needs to add to existing knowledge, hence it should be innovative. We all stand on the shoulders of giants. At an early stage of your career this is often contextual knowledge, not fundamental one. You may not want to reinvent the wheel, but will certainly not revolutionise levitation. You hover somewhere in between, floating between the solid ground of existing knowledge and the stars overhead. Make sure that you always remember that we all started somewhere. If you read this text about how to write an outline, it is most likely that you are only starting in your career. Remember that we all started somewhere, and in research that is often done by learning to gather data.
+
===== Data gathering =====
 +
It is somewhat beyond an introductory entry to tame the whole world of data, which includes gathering data, analyzing it, and interpreting it. Yet we may take a peek at what is relevant, and can and should be achieved in the realms of an outline. The core goal to this end is probably the sweet spot between innovation and feasibility. Innovation in terms of your research design means that you read the relevant literature and, integrated it into your considerations on how to gather and analyze data. Start with textbook knowledge, and then go from the most highly cited general papers on the subject into more specific papers that are closer to the context you focus on. There is a lot of experience out there that is relevant to your specific methodological design, you just need to adapt it to your focus. This is exactly the reason why divergent thinking is a key skill in academia. Instead adding to the pile of literature you read you may want to consult an expert, however only with specific questions. You can get the basics on simple methods from textbooks, yet your supervisor may know best if there is anything else to consider. More often than not this may not be the case, yet there are things to consider. For instance is the sample size often relevant on both quantitative and qualitative science, and other questions such as heterogeneity of the sample are often of considerable importance. Questions of bias often already play a central role when gathering data, and then there are the simple mechanics. For you to document the gathering process, are there any machines or tools needed, and are there other technical aspects to consider such as software solutions? Yet the key question when getting data is whether if it is feasible or not. No data can be gathered if it is impossible to get it. Early career researchers often overplay their hand to this end. Make a pretest and see how long it takes you. Also remember that over time you may get faster concerning some mechanics, but also new and unanticipated problems may arise. Plan at least a third more time than the most conservative estimate. If you do not need it, fine, yet much research already took longer at this stage. Lastly, research needs to add to existing knowledge, hence it should be innovative. We all stand on the shoulders of giants. At an early stage of your career this is often contextual knowledge, not fundamental one. You may not want to reinvent the wheel, but will certainly not revolutionize levitation. You hover somewhere in between, floating between the solid ground of existing knowledge and the stars overhead. Make sure that you always remember that we all started somewhere. If you read this text about how to write an outline, it is most likely that you are only starting in your career. Remember that we all started somewhere, and in research that is often done by learning to gather data.
  
The next step is analysis. Analysing data equals experience. There may be some simple recipes how to analyse data, after all this is what normal-science text books are made for. Yet, from a methodological standpoint this does not seem to be very innovative. Yet again, also when it comes to learning methods, you need to start somewhere. One would not start to learn making food with a lasagne. Making simple pasta is a good start, yet the multilayered mixedness of a lasagne takes some experience. With methods it is quite the same. Learn the basics first, and then move on to get into the more complicated stuff. You need to get the initial stuff under your belt in order to go into advanced analysis, may it be qualitative or quantitative. No one can take this away from you, you will have to learn this yourself. Hence, doing analysis is first and foremost a matter of experience. Your supervisors and others are not there to help you to this end, because this would take the experience away from you. They may stop you if you made a mistake, but learning to make a content analysis -as an example- takes a long time. Read other research and then get at it. You can only evolve into an experienced researcher yourself, because experience is tough to teach. This with your outline you just need to write the main steps, and then your supervisor will make you aware of any biases, flaws or sources or problems. If you hear nothing, great. Then all should be fine, otherwise you will find out on your own.  
+
===== Data analysis =====
 +
The next step is analysis. Analyzing data equals experience. There may be some simple recipes how to analyze data, after all this is what normal-science text books are made for. Yet, from a methodological standpoint this does not seem to be very innovative. Yet again, also when it comes to learning methods, you need to start somewhere. One would not start to learn making food with a lasagne. Making simple pasta is a good start, yet the multilayered mixedness of a lasagne takes some experience. With methods it is quite the same. Learn the basics first, and then move on to get into the more complicated stuff. You need to get the initial stuff under your belt in order to go into advanced analysis, may it be qualitative or quantitative. No one can take this away from you, you will have to learn this yourself. Hence, doing analysis is first and foremost a matter of experience. Your supervisors and others are not there to help you to this end, because this would take the experience away from you. They may stop you if you made a mistake, but learning to make a content analysis -as an example- takes a long time. Read other research and then get at it. You can only evolve into an experienced researcher yourself, because experience is tough to teach. This with your outline you just need to write the main steps, and then your supervisor will make you aware of any biases, flaws or sources or problems. If you hear nothing, great. Then all should be fine, otherwise you will find out on your own.  
  
Data interpretation is the toughest point to anticipate in an outline, because this would take a deep command of the available literature, and the cross-sectional mechanisms that are relevant within this literature. One should fall into the trap of writing a generic critique before the study was conducted. There is a tendency that research write such simple truths as "this research is biased by a perspective of the global north", which may be true, but is a bit of a one-size-fits-all category that helps now one, least any suppressed groups. Hence make sure to be specific if you write anything at all. Yet this is a section often best led intentionally left blank.  
+
===== Data interpretation =====
 +
Data interpretation is the toughest point to anticipate in an outline, because this would take a deep command of the available literature, and the cross-sectional mechanisms that are relevant within this literature. One could fall into the trap of writing a generic critique before the study was conducted. There is a tendency that researchers write such simple truths as "this research is biased by a perspective of the global north", which may be true, but is a bit of a one-size-fits-all category that helps no one, the least any suppressed groups. Hence make sure to be specific if you write anything at all. Yet this is a section often best led intentionally left blank.  
  
Potential pitfalls is again something that is better if aimed at concrete problems, yet should not be lofty or speculative. What may be challenges that are potentially relevant for your very context? Are these about you (e.g. concerning time scheduling issues), about the research (such as sample availability), or about the broader context (i.e. a global pandamic)?Write only what you know that has been proven as a problem in the past, do not speculate way beyond your current sphere of knowledge. This may be a moment to consult experienced researchers again, because they may have bumped into problems that are rooted in experience, which is a much more solid knowledge base compared to speculative anticipation. Always remember that research is also adaptation, you face problems, and then you overcome them.  
+
===== Potential pitfalls =====
 +
Potential pitfalls is again something that is better if aimed at concrete problems, yet should not be lofty or speculative. What may be challenges that are potentially relevant for your very context? Are these about you (e.g. concerning time scheduling issues), about the research (such as sample availability), or about the broader context (i.e. a global pandemic)? Write only what you know that has been proven as a problem in the past, do not speculate way beyond your current sphere of knowledge. This may be a moment to consult experienced researchers again, because they may have bumped into problems that are rooted in experience, which is a much more solid knowledge base compared to speculative anticipation. Always remember that research is also adaptation, you face problems, and then you overcome them.  
  
Time frame
+
===== Time frame =====
For many people making a time planing is beneficial, as they claim that they need the pressure. Others make a timeline is just for breaking it, much in the spirit of Douglas Adams: "I love deadlines, I love the whoosing sounds they make when they fly by."
+
For many people making a time planing is beneficial, as they claim that they need the pressure. Others make a timeline that is just for breaking it, much in the spirit of Douglas Adams: "I love deadlines, I love the whooshing sounds they make when they fly by."
I wish I could claim that you just make a time line and then make it work, but this is also not really functional. Ultimately, I believe we make deadlines and timelines to learn about our own imperfections.
+
I wish I could claim that you just make a timeline and then make it work, but this is also not really functional. Ultimately, I believe we make deadlines and timelines to learn about our own imperfections.

Latest revision as of 14:16, 18 April 2023

With the completion of an outline you take one first step of what your research could be all about. One should never forget that research is an evolving and iterative process, though. Still, writing an outline makes you settle for what you want to focus on in this moment, and more importantly, also allows your supervisors as well as your peers to give you structured feedback. This is why any research project should start with the landmark of writing an outline. Different branches in science have different focal points and norms that an outline is build upon. Here, we present an approach that tries to do justice to the diversity of approaches that are out there, yet it is always advisable to ask your supervisors for modification if need be.

Working Title

All research starts with a title. Personally, I read hundreds of titles of research papers each month, and only a small portion are appealing to my specific focus and interest to invite me to read further. Titles are the door-opener for most researchers, which is why titles should be simple and to the point. If a title is too long it will lack clarity and crispiness. If a title is too short, it will surely not give away enough information needed to know what the research is all about. Often people try to make a fancy title that is supposed to be witty or funny and contains some sort of wordplay or inside joke. Avoid this. You may go for such a title later once the research is done and the paper is written, yet such titles need to be earned. Hence especially in an outline it is best for a title that is walking the fine line of giving away enough information but not too much.

Participants

In the end, it is going to be you who will do the research. You will sit at your desk, you will gather the data and look at the literature, you will get deeper into the topic, thus it is you who will basically write the research. To this end, it has proven of immense value to have a network of peers. More often than not, this is an informal network for critical reflection, but also for support. Such a peer network will not be mentioned here if it is not actively involved in conducting the research. Beside your supervisors actually very few people will be involved in your research. You may have someone helping with the analysis or being experienced in the topic if you are a PhD student, yet in a Bachelor or Master thesis the focus is stronger on proving that you can conduct independent research. While in a PhD this is also the goal, it is on a more sophisticated level, where because of the longer timeline and thus deeper focus collaboration may be of greater importance. It is also quite important to clarify roles and expectations at the beginning. Remember that a thesis is your work.

Background and topic

The topical focus and its background are often what drives people. Most researchers are very exited about their topics, and it is valuable to have something that can fuel your energy while you work on your thesis. There will be ups and downs surely, yet it is still good to focus on something that not only drives you, but also current research. Timely topics are ideally embedded into a longer development that led to the emergence of the topic. You do not want to work on a topic where thousands and thousands of paper were already published, but ideally you should also not work on something that is so new that there are no shoulders to stand on. Researchers do really stand on the shoulders of giants, and despite the thrive to make innovative and timely research, we should never forget that we are part of a bigger movement. Within this movement, we will add one tiny step. A friend once said that we are drops in a wave - a fitting allegory when looking how research evolves. Still, our contribution matters, and may move the research landscape forward. What it does however more importantly is that it moves us forward. A thesis is first and foremost a prove that you can conduct research, and that you are thus able to contribute to the scientific community. Because of this, the topic is less important than most people think it is, because research is a transformational experience for the research. Do not misunderstand me, all topics of past people I supervised were really important to me. What is however even more important is that they learned to overcome themselves, and slay the dragon that is their thesis. Once you found a topic that excites you, it is advisable to iterate it with your peers. Write down why your topic is timely, how it contributes to the wider research, and what the current state of the art is. You want to make a thorough assessment yet have to be careful because reading too much may confuse you. Do not expect that everything is coherent and there are no contradictions. Research is discourse, and these discourses evolve over time. Often researchers disagree, and not everything makes sense. Be prepared to be confused. It is your job to evolve a critical perspective of the state of the art of the literature, and identify landmark papers. Also, try to find previous research that aimed in the same direction and learn how they approached the topic. What were the methodologies, where did the researchers struggle, and what were their recommendations. Lastly, try to develop an elevator pitch on why this topic is important to you. If you can explain in short why you think this research needs to be done, you are onto something.

Key supporting theories

Just as topics are embedded in the current scientific discourse, such research looks at partials of reality. In order to generate such a specific view of reality, research builds on theories. Theories in science can be operationalized at different scales. Some theories are on a conceptual level, which is basically very theoretical and often restricted to non-empirical research such as much of philosophy. Other researchers may build on frameworks, and is very applied. In between are paradigms, which are the types of theories that branches of research are for some time built upon. Examples for concepts would justice or peace, examples of paradigms would be effective altruism or ecosystem services, and examples for frameworks would be concrete assessment schemes or the sustainable development goals. Much confusion in theory work is because these three levels are confused, which is why it is important to localise your theory in your work clearly. Another common problem when writing an outline is to settle for a low number of theories. If you ask me, one theory can be enough, and two can be exciting. More than two theories are more often than not impossible to tame. Hence you need to remember that you look at a version of reality, and that theories enable you to either test hypothesis or create research questions that are open and specific at the same time. You may not want to know how everything works, but how it works under a specific viewpoint or theory of reality. This is what theories are all about. Researchers theorize how mechanisms in the world might work, how patterns emerge, why people act, and why societies fail. There is a plethora of theories, and it is important to be familiar with the literature of the theories that are within your realm.

Most topics are associated with certain theories, such as mainstream economics often resolves around utilitarianism. While you can operate on safe ground if you follow suit with these associations, it can also be exciting yet daring to superimpose a theory onto a topic that has never been exposed to it. Yet this should ideally be discussed with your supervisor, because while it can be exciting, it may be prone to challenges that are hard to anticipate by the uninitiated. This is one of the key points where you should build on the experience of others. Lastly, while it may be exciting to work on a scientific theory, always remember that you look at a snapshot. Theories are often proposed, and then in subsequent years applied. Based on this, they are revised, scrutinized, and often the research community becomes divided about specific theories. While many researchers question or even reject evolutionary altruism for example, others may work based on this theory for their whole career. Choosing a theory is a normative responsibility, you decide to take on a specific perspective, and this is a key responsibility as a researcher. While you should choose wisely, I would also say: Do not overthink it. Most early researchers tend to think way too much about it, in fact I never saw anyone thinking too little about it. And after all, at this point your supervisors should give you feedback.

Main references

Ideally, you can at this point already identify the main references that your research is being based upon. Make sure to reference the sources that are most recognized by the research community, but also closest to what you think are the best sources. Often this may be one and the same, but sometimes it differs. What is crucial is to focus on the main sources. Especially early career researchers are so happy that they read so much stuff that they want to quote it all. While this sounds like a great idea, remember that not everybody is as deeply embedded as your are, thus focus on the most important sources. On a less dramatic note, make sure that your quotations are in the same style. Settle early on one citation style, use a software to guarantee coherence, and never change it from then on.

Research question or hypothesis

The next point is a very delicate one: Research questions vs. hypotheses. Let us start with the more simple point. How much of those should you have? It is really difficult to answer one question. This is something that philosophers may do, but most research demands more questions that build on each other. See it from a structural point of view, answering one question pre-structures your introduction into one big blob of text. That is certainly not convenient. 2-5 questions seem ideal, because -let's face it- it is really hard to remember more than 5 things intuitively. We then tend to forget, and it also has the drawback to make your research seem über-structured. Hence it can be a good heuristic to have 2-5 research questions or hypotheses. Now let us move to the more troubling part, which is the question whether it is research questions or hypotheses. From a philosophy of science standpoint, the two should be mutually exclusive. While hypotheses are surely deductive and demand a clear confirmation or rejection, research questions are somewhat more open and hence inductive. Thus while the latter has never been clear cut, it can be helpful to draw such a clear line. Some folks may now raise the streetwise question whether we should not all be thinking abductively anyway? In a sense this is where all science moves anyway, because we need both inductive and deductive approaches, and ideally combines the benefits of both in the long run. Yet this is mainly a questions of both experience and temporal grain. Working abductively means that you can clearly remark between hypotheses and research questions, because the first demand a very clear structure, while the latter are way more open. There is an underlying relation that hypotheses tend to be more quantitative, while research questions are often more qualitative. Still, this is a mere correlation and not a really causal relation, but it confuses a great many people. Chasing whether to write hypotheses or research questions is in addition often a question of the tradition of the respective scientific discipline. Many folks in natural science are still proud of their razor sharp hypotheses, and other fellow within socials science lean clearly towards research questions. Ideally, find out what your supervisors demand, which is the most simple heuristic to this end. Still, this point underlines that from a philosophy of science standpoint the silo-mentality of disciplines has its reason, but does not always make sense. What is most important for you is that an adductive procedure may be most desirable, yet only in the long run. It is part of the tradition of many sciences the postulate, test, and adjust. This works more on a time-scale of years or decades, but thus demands a temporal grain of a longer research agenda. Hence an abductive is not suitable for a shorter project such as a thesis.

Study area

The next thing you want to focus on is the study area. This does not necessarily need to be a concrete space, but can also be something of a less concrete system. In a systematic review or critical content analysis this can be a branch of the literature. Within an image analysis this can be a set of paintings. In an ethnographic study this can be a specific group of people. Within a lab experiments it can be a completely artificially designed study, such as a set of planting pots that are irrigated, shaded and endure different temperature settings. This the study area is the specific system or setting that is being investigated. Again, there is a certain tradition that deductive studies are more tamed or have a deeper control or understanding of their study area, while inductive studies are more open minded and less clear cut when it comes the the pre-study understanding of the study area. Yet do not be fooled, inductive studies can be very clear when it comes to looking at something specific, these studies do it just with a different kind of open-mindedness. What is the benefit for you is to be clear in defining where you want to work in, may it be inductive or deductive. It is helpful to have a clear definition, because otherwise you will be either overworking yourself or have a sample set that is too small. The study area should ideally be chosen to make it very clear how your research represents some dynamics, pattern or mechanisms that can either serve as a basis to approximate reproducible knowledge, or at least knowledge where the path towards it can be clearly documented. Hence the study area is something that allows you to make your research specific and thus tamed. Your choice remarks a start as well as an end. Ideally, it should be exciting and represent a clear knowledge gap, but in addition it also demands to represent a bigger picture that is well represented by the chosen system. If you want to work on small business dynamics you need to work with organization that can be seen as representing the overall dynamics. If you want to make a survey, you certainly do not want to focus on outliers if you try to represent a bigger group. Ideally you build on the experience or already established researchers to learn to make the right choice concerning your study area.

Data gathering

It is somewhat beyond an introductory entry to tame the whole world of data, which includes gathering data, analyzing it, and interpreting it. Yet we may take a peek at what is relevant, and can and should be achieved in the realms of an outline. The core goal to this end is probably the sweet spot between innovation and feasibility. Innovation in terms of your research design means that you read the relevant literature and, integrated it into your considerations on how to gather and analyze data. Start with textbook knowledge, and then go from the most highly cited general papers on the subject into more specific papers that are closer to the context you focus on. There is a lot of experience out there that is relevant to your specific methodological design, you just need to adapt it to your focus. This is exactly the reason why divergent thinking is a key skill in academia. Instead adding to the pile of literature you read you may want to consult an expert, however only with specific questions. You can get the basics on simple methods from textbooks, yet your supervisor may know best if there is anything else to consider. More often than not this may not be the case, yet there are things to consider. For instance is the sample size often relevant on both quantitative and qualitative science, and other questions such as heterogeneity of the sample are often of considerable importance. Questions of bias often already play a central role when gathering data, and then there are the simple mechanics. For you to document the gathering process, are there any machines or tools needed, and are there other technical aspects to consider such as software solutions? Yet the key question when getting data is whether if it is feasible or not. No data can be gathered if it is impossible to get it. Early career researchers often overplay their hand to this end. Make a pretest and see how long it takes you. Also remember that over time you may get faster concerning some mechanics, but also new and unanticipated problems may arise. Plan at least a third more time than the most conservative estimate. If you do not need it, fine, yet much research already took longer at this stage. Lastly, research needs to add to existing knowledge, hence it should be innovative. We all stand on the shoulders of giants. At an early stage of your career this is often contextual knowledge, not fundamental one. You may not want to reinvent the wheel, but will certainly not revolutionize levitation. You hover somewhere in between, floating between the solid ground of existing knowledge and the stars overhead. Make sure that you always remember that we all started somewhere. If you read this text about how to write an outline, it is most likely that you are only starting in your career. Remember that we all started somewhere, and in research that is often done by learning to gather data.

Data analysis

The next step is analysis. Analyzing data equals experience. There may be some simple recipes how to analyze data, after all this is what normal-science text books are made for. Yet, from a methodological standpoint this does not seem to be very innovative. Yet again, also when it comes to learning methods, you need to start somewhere. One would not start to learn making food with a lasagne. Making simple pasta is a good start, yet the multilayered mixedness of a lasagne takes some experience. With methods it is quite the same. Learn the basics first, and then move on to get into the more complicated stuff. You need to get the initial stuff under your belt in order to go into advanced analysis, may it be qualitative or quantitative. No one can take this away from you, you will have to learn this yourself. Hence, doing analysis is first and foremost a matter of experience. Your supervisors and others are not there to help you to this end, because this would take the experience away from you. They may stop you if you made a mistake, but learning to make a content analysis -as an example- takes a long time. Read other research and then get at it. You can only evolve into an experienced researcher yourself, because experience is tough to teach. This with your outline you just need to write the main steps, and then your supervisor will make you aware of any biases, flaws or sources or problems. If you hear nothing, great. Then all should be fine, otherwise you will find out on your own.

Data interpretation

Data interpretation is the toughest point to anticipate in an outline, because this would take a deep command of the available literature, and the cross-sectional mechanisms that are relevant within this literature. One could fall into the trap of writing a generic critique before the study was conducted. There is a tendency that researchers write such simple truths as "this research is biased by a perspective of the global north", which may be true, but is a bit of a one-size-fits-all category that helps no one, the least any suppressed groups. Hence make sure to be specific if you write anything at all. Yet this is a section often best led intentionally left blank.

Potential pitfalls

Potential pitfalls is again something that is better if aimed at concrete problems, yet should not be lofty or speculative. What may be challenges that are potentially relevant for your very context? Are these about you (e.g. concerning time scheduling issues), about the research (such as sample availability), or about the broader context (i.e. a global pandemic)? Write only what you know that has been proven as a problem in the past, do not speculate way beyond your current sphere of knowledge. This may be a moment to consult experienced researchers again, because they may have bumped into problems that are rooted in experience, which is a much more solid knowledge base compared to speculative anticipation. Always remember that research is also adaptation, you face problems, and then you overcome them.

Time frame

For many people making a time planing is beneficial, as they claim that they need the pressure. Others make a timeline that is just for breaking it, much in the spirit of Douglas Adams: "I love deadlines, I love the whooshing sounds they make when they fly by." I wish I could claim that you just make a timeline and then make it work, but this is also not really functional. Ultimately, I believe we make deadlines and timelines to learn about our own imperfections.