• You do not know what you are doing. This is normal, after-all we are talking about research and is impossible to “jump into the unknown” knowing exactly what you have to do.
• You do not know what you are doing (version 2). This is less normal but happen often. If you are a first timer, the master, hasn’t prepare you in all the minor details of your sub-domain of research and you may need to compensate this by studying a bit on your own.
• You are subject to the shiny object syndrome. Again, this is normal. You are human and as human we are attracted to “new things”. This aspects may play against you in this phase as you are risking to jump on too many things without investigate deeply in any.

Factors that may #Prevent you in #Advance towards the #research aspect of your #thesis
Share on X

Start by recognizing these aspects in yourself. Make sure to put some attention on what you are doing and on your thinking process. Over time you should be able to recognize these aspects and how they are effecting you when approaching to doing your thesis.

What follows is a step by step process that can help you in “putting your first foot out of the door” and leverage the literature review that you are probably already doing, followed by “a quick reading technique”. If you read the article on how to read a paper quickly and effectively you know that I always suggest to start by asking yourself “why should I read this paper”. If properly used, this approach can help you in this stage in several ways starting from the fact that you are trying to put some order into “the chaos of what you are reading”. But there is more; let’s start by considering the following categorizations:

• Objective. What is the problem that the authors are trying to solve?
• Methodology. What is the approach that authors are using for validating their research hypothesis?
• (sub)Domain. What is the expertise of the authors and from what angle are approaching the problem?
• Tools. What technique is used by the authors and what software is adopted for doing the “hands on aspect” of their research?
• Data. What kind of data set is used by the authors, are they proposing or adopting a standard benchmark for evaluating certain aspects of their research?
• Proximity to the problem that you are addressing. How is the article that you are reading related to your problem? Maybe it is trying to address something similar or maybe it is addressing something different but the approach of the authors could be useful for your case.

Possible ways to categorize a #scientific #paper for helping you during your #research
Share on X

Categorization is the first step on how to use the literature review for your research. Every thesis must contain a literature review part, normally in chapter 2 or chapter 3. I recommend you to take note of these aspects and to summarize them using one or more tables. This will give completeness to your work. In particular it will shows that you haven’t only read a few articles but that you are capable to navigate the landscape of cutting edge literature that is around the problem that you are trying to solve.

You are also welcome to propose your own categorization. Different domains will probably have some particularity that can’t be easily generalized. Nevertheless, let’s see how we could use each one of the categories for addressing the particular problem that is related to our thesis.

Finally, please note, that when you are approaching a literature review you should “remove your ego from the equation”. In other words, you are looking at the work of others and you want to be as objective as possible in evaluating it. It is not about your problem and your research questions; it is about their problems and the solutions that are proposing. Your objective is to understand and categorize. There will be plenty of room for expressing your opinion and your ideas in other parts and phases of your thesis.

1: Objective and Proximity to the problem that you are addressing:

This is probably the most strait forward of the proposed categorizations. When addressing a scientific problem you probably have a particular objective in mind. You are considering a literature review for helping you in specifying the problem further and/or you would like to evaluate possible solutions that have been proposed.

You want to categorize the work of others, taking into account, that aspect in mind. In particular considering if the work of the authors may be relevant or not. A reasonable categorization should consider the following aspects:

• Is the work of the authors trying to solve the same problem that you are? In other words you have the same objective of the authors.
• Is the problem a subset or a super-set of your objective? I.e. you are aiming at an objective that is smaller or bigger.
• Is the objective of the authors similar to yours but developed in a different domain? For example, you are both interested in improving the speed of delivery but you are focusing on door to door milk distribution and the authors are considering newspapers.

This categorization will help you in understating who is trying to solve a similar problem. You will also have a feeling of the level of deepness of their analysis. Finally, you will know and what has been already done in your area of research. As your goal is not to replicate the work of others but to propose (and evaluate) an innovative idea, this will help you in preparing and clarifying the landscape where your future work will be positioned.

This is the first categorization that will help you reasoning around how to use the literature review for your research.

Try to #understand what you are doing by looking at the objectives of other #Researchers. Devise analogies and differences.
Share on X

2: Methodology

This approach for categorizing relevant works suggests to forget completely about the objective and to focus “on the how”. There is not a single way to solve a problem and probably each author is specialized in a adopting different techniques. Consequently, different works will approach the problem using different methods. For example they may propose a Qualitative or Quantitative approach and within that type they may focus on something different.

Let me clarify with an example and lets’s consider the problem of predicting the weather. In this case the problem is clear and well defined but each researcher may have different approaches in proposing a solution:

• Maybe they are asking to 10 different groundhogs their opinion,
• Or are considering to model the data that they have using linear regression,
• Maybe are using a <Insert Your Favorite AI technique here> for building a predictor.
• Alternatively they are just focusing on a sub-problem like, for example, on the quality of the data that their sensors are collecting.

As you may imagine the list of possible methodologies can become very big. However, each domain of research has their “preferred styles” for solving their challenges. In practice you will find that authors tend to focus on a subset of 1-10 different approaches. This is due to the fact that researchers are humans and their knowledge is specialized.

If you want to understand how to use the literature review for your research a categorization regarding the methodologies used by researchers is an important aspect.

Look at the methodologies that other #Researchers are using. It may help you in finding one that you useful for you as well.
Share on X

3: (Sub) Domain of Expertise

As anticipated in the previous section, researchers are humans and they approach a problem with their expertise. Some problems (usually the most fun) transcend a single domain and are called “multidisciplinary”. If the problem that you are considering is in this category you want first to clarify with your supervisor a few aspects of your research. In particular you may want to be sure that your research question(s) are focus enough and not too broad. Topics like global warming or world poverty are fascinating problems but are a bit to ambitious to solve for a single thesis.

For the rest of this section I will assume that your research topic is narrow enough and has multidisciplinary aspects. Let’s, for example, consider the domain of “Internet of Things”, in particular the problem of “Privacy in Internet of Things“. Clearly this domain is analyzed by several types of researchers.

• Economists are trying to understand how these changes will impact businesses and the economy in general.
• Lawyers are considering regulatory aspects.
• Big Data Experts see this as an opportunity for collecting an analyzing data.
• Engineers are considering how to build new devices
• Computer Scientist focus on how to develop applications on top of that infrastructures
• Social Scientists study societal adoptions of these technologies

Each single expert is approaching the same problem from his/her own domain of expertise. Consequently you may find articles that are relevant but very different in therm of styles. If that is the case a categorization that take into account these dynamics can help you in understanding (and probably narrow down) your research questions and your future research.

Are your #problems multidisciplinary problems? Categorize the #literature review taking into account the various #discipline that are approaching the problem
Share on X

4: Tools

What are the tools that the researchers are using for solving the problem? Here I am explicitly not talking about the methodology. I am referring to tools and technology used in order to implement the validation of the research question. This categorization vary a lot in each domain of research. It can range from tools for building online surveys and questionnaires to software and/or libraries for implementing a particular proof of concept.

If you are considering more types of paper, i.e. papers that are using different methodologies, a single categorization maybe difficult to build. If that is the case you may want to consider a categorization per methodology that you are considering.

Let’s consider the example of Artificial Intelligence in particular, AI applied to weather prediction. You may find that researchers may focus, for example, in Neural Networks and in Data Mining techniques. If you want to survey both, you may want to consider two different tables. One dedicated to Neural Networks tools and the other to Data Mining tools.

If you are envisioning too many tables you may want to question the scope of your research. Check with your supervisor if your research question is focused enough.

#understanding the tools that are used by #Researchers can help you quantify the time that you will spend doing your #research. Thus #speeding the #learning curve
Share on X

5: Data

What information are the researchers using for validating their work? Many of the considerations and concerts that we made for the “Tools” category also apply here. In particular, if you find yourself considering several different types of data you may want to question the scope of the research. Typical sub-categorization of this categories includes:

• Data from questionnaires and/or surveys.
• Data collected from a physical device and/or use case.
• Simulations.
• Use of standard data-set known and accepted among researchers.
• Data collected from a particular company.

Categorize the data used by researchers will help you understand how you can use yours and is an important aspect of how to use the literature review for your research.

Look at your #data in an objective way. Can they be used as the validation of your your #scientific #claims ? What are other #Researchers using?
Share on X

Wrap up: How to use the literature review for your research?

I am sure that these categorizations, as well as, the one that you will suggest are typical of your particular domain of research and will improve the quality of your literature review. However, this is not the point of the article. The goals is “to move forward” towards a successful experimental research.

How to use the literature review for your research? My suggestion is the following:

• Reflect around the Objectives of the researchers and consider if you would like to be doing something similar. Try to understand analogies and differences and what “they are proposing” and what “you would like to do”.
• Use the Sub-Domain of Expertise for reflecting around the scope of your thesis. Decide the angle that you are more knowledgeable for approaching the problem.
• Shop around for Methodologies, find the one that resonate more with you. After-all, you will have to follow it in detail for the coming months.
• Play with the Tools that are normally used by the researchers for validating the particular methodology that you selected. This will give you some mutuality for elaborating the practical part of your research.
• Ask yourself if the tool that you are considering is a good match with the Data that you are collecting or you have at your disposal.

This is how to use literature review for your research. If you start reflecting on all these aspects you will have a lot of questions in your mind. That is perfectly normal, and if you succeed in organizing your thoughts around these 5 categories you will have a clear objective in a well defined domain. In addition, you will have a methodology and a tool to use for analyzing your data.

If you find yourself with more question then answers. Try to organize your thoughts around the 5 mentioned categories. This will break down the problem in more digestible pieces such as:

• Is my objective clear enough?
• How deep do I know the methodology that I would like to apply?
• Are the data that I have good for the methodology that I would like to use for validating my results?
• Am I capable of using that particular tool? How much time will take me to learn it?
• Is the tool adequate for been used in conjunction with my data?

The goal is to find an objective that is achievable and can be evaluated with a sound methodology in the time-frame of your thesis.

You may want to take note off all your ideas and discuss them with your supervisor as some of them may be too ambitious or not sufficiently specified.

Use a #Literature review for #understanding what you are doing in the context of your #research. A proper categorization can help you in learning from #OTHERS
Share on X

This article (How to use the literature review for your research) is part of the miniseries on how to do a good thesis, you can see the full list of post at the following links:

If you like what you read, consider sharing it!

The post How to use the literature review for your research appeared first on Francesco Lelli.

• You have a supervisor and you discussed with him/her a possible topic for your thesis.
• You already browsed a few scientific articles related to the topic and you like the subject.

If that is not the case please look at How to take the most from your thesis and How to read a scientific article quickly and efficiently that will give you the basics for understanding what kind of thesis you would like to do and how to not spend an infinite amount of time reading scientific articles.

With all been said is now the time to understand what you have “your assets” and how you can start using them towards a successful thesis.

Count Your Assets

If you are thinking that you really have nothing, well think again. You have more than you can imagine. I am pretty sure that you can count on the following:

• A supervisor that is happy to mentor you and support you.
• Access to the facility at the university for reading scientific papers.
• Support for English writing via some university facilities.
• Guidelines on how to do a thesis from the thesis coordinator of your bachelor/master program.
• Examples of “good past thesis”.
• Knowledge of everything you have studied so far.
• Friend and family that will be happy to support you.

That is far from been nothing! On top of what I just listed you may also have:

• An Internship in a company.
• A part-time job in a company.
• Or no internship or job.
• Continuing the work of a previous student.

The first set of items are clearly “assets” that you can use for helping you in doing your thesis. The second set represents things that are both an asset and a liability at the same time. An internship or a work in a company can provide you access to information that are normally not available to “outsiders”; however, at the same time, it subtract you precious time that you could dedicate to your thesis. Please do not misunderstand me, I do not believe that an internship or a part-time job at a company is “a bad thing”: it really depends on what you would like to do when you will graduate and if you want to know more please consider reading this article on How to take the most from your thesis before warming up your research engine.

Start your master or bachelor #thesis by #CountingOn #yourself and your #Assets !
Share on X

Count how much time you have

Now that you know on what you can count during your journey towards a great thesis you should get a real and tangible grasp of the time that you have. In counting you should also consider that during this journey you want to spend time with your friends and have fun. This is as important as doing the thesis itself, as you do not want to burn out during the process but, hopefully, you would like to enjoy the process!

Start by working backward from the deadline.
In other words do the following:

1. Figure out the deadline for submitting the thesis,
2. Count vacation periods and time that you plan to take off,
3. Count time that you have to spend in your internship as time that you will not dedicate to your thesis,
4. Count the time that you have to spend in doing the remaining exams (if any).
5. Do not forget to consider time for gym and/or beers!

If you do the math you will have an idea of an upper bandwidth of the time that you could dedicate to your thesis. Critically assess that number and ask yourself “Is this doable or you are overestimating/underestimating yourself?” If you are experiencing difficulties you may involve your supervisor in this discussion as he/she has more experience then you in this matter. However, if that is your decision start thinking at possible scenarios before having that discussion. You will have a set of solutions and gather valuable feedback instead of “sharing your problem” to you supervisor.

When #planning your #thesis , #measure your #time carefully: start by counting it #backwards
Share on X

Build a fuzzy but feasible timeline

Now that you have a clear understanding on how much time you can/want to dedicate to your thesis you can start envisioning a possible timeline. As is impossible to predict the future you can expect that you will deviate from your planned deadline. That is why is important that you have a flexible (fuzzy) goal-driven and action-oriented timetable. In defining the milestones for your timetable you can consider that the thesis can be divided in the following sub-tasks:

1. Understand the problem: this can be done both, via an extensive literature review, or using your internship or expertise of your supervisor.
2. Acquiring the basic knowledge of the subject: once that you know what is the topic of your thesis you want to supplement your current education with a one or two books about the topic in order to gather the big picture on which problem that you are trying to solve.
3. Extensive Literature Review: you want to survey and understand the cutting edge research that is perform around the topic in order to become an expert of the sub-domain in which your thesis is part.
4. Formulate a scientific hypothesis and related model: the more you will be analyzing the literature and spending your time in solving the challenge from a practical point of view at your internship the more you will start be able to formulate a scientific hypothesis that you can “cast some light” into the problem.
5. Building an experiment that validate/invalidate your hypothesis: use the data or use cases that you define to validate your theories from a qualitative or quantitative point of view.
6. Write the thesis: you should consider the task of writing a thesis as a an independent task. Being able to presents the finding of your work in a simple and understandable manner has its challenges and require time and effort.

There is a clear dependency among these tasks and the more you will advance in one the more you will feel the need to update and revisit what you have already done in the others. That is perfectly normal, and is the reason why you want to consider a “fuzzy” schedule: the more you will advance in building the thesis the more you will become expert in the material and the more you will see things clearly.

You may want to discuss with your supervisor all the aspects of your thesis and put a time duration on each task together with “your next first step” that is covered in the following subsection.

your #thesis need a fuzzy #schedule when you will set your #deadline . Have clear #goals and a #flexible #Mindset on how to #reachinggoals
Share on X

Think about your next step

If you reached this step you have a list of assets that you can use for building your thesis, a clear and feasible timeline and a set of related tasks. The only item that you are missing for warming up your research engine is “your first next step”. Every great journey start taking that first step. Your thesis is not an exception in this respect.

Figure out where you are in the journey of your thesis and what you could do for advancing it forward. Do it and book an appointment with your supervisor for discussing what will come next!

This is pretty much all you need to know for warming up your research engine!

Have a clear objective and try to be flexible in reaching it. After all, you are doing research and this activity has a serendipity component.

What are the first steps when starting your #thesis ?
Share on X

This article ( warming up your research engine ) is part of the miniseries on how to do a good thesis, you can see the full list of post at the following links:

If you like what you read, consider sharing it!

The post Warming up the Research Engine appeared first on Francesco Lelli.