 This is in continuation of what Professor Sarnamurti told us in the previous session. That session was about how to read a research paper. In practice, it is not one research paper that we read, but multiple research papers have to be read. Typically, such a reading forms part of literature review for any research activity. In this lecture, Professor Sarnamurti is going to give us some tips on how to carry out literature review effectively and efficiently. So, could we have her lecture please. Welcome to the IST workshop on Introduction to Research Methodologies. This session is on tips on literature review. What I will do is, pose a series of questions and try to answer it by means of some tips or guidelines on various aspects of conducting and writing and literature review. So far, you have gone through the session on how to read a research paper. So, you are now, if you were not earlier, you are by now familiar with what to do when you have to read a single research paper. But in the course of your research, you have to read multiple papers and then you have to do something with them. You have to put them together, write about it, evaluate it and what we will discuss in the session is how to take, how to read multiple papers, which papers to select, what to do once you read them and so on. So, the first question we will ask is, what exactly is a literature review? And very briefly, it is both a process as well as a part of a research paper or a part of a thesis. And as a section or a part in a thesis or research paper, a literature review is a summary of related work. It is not a simple, just a direct listing. Instead, it has some descriptions, but it also has some evaluative components. So, there is both descriptive components of related work, what other related work is about and evaluative components, how they compare against each other, what works and what does not and so on. So, there is a lot of analysis involved here. And even as a process, when a researcher reads multiple papers, gets ideas, tries to find out what is missing, what is common, what is good and all, the entire process is highly analytical in nature. Why do we even need to do literature review? That is, I am doing research, why do I need to look at what others have done? And you may laugh if I say that, but there are actually several reasons here. One of it is that, one should give credit to others work, because the kind of research one does is never in vacuum, it is always based on something maybe a few years ago or maybe a few hundred years ago that somebody else has done. So, we need to give credit to others work and in the academic community, it is equivalent to a law. So, if a researcher writes a paper and does not give credit, people assume one of two things, does not give credit to other similar work, people assume one of two things. One is that the researcher is not aware of what others have done, others have done. That is not a good reputation to have as a researcher, it just comes across as if the researcher is naive or just does not have an idea of the field. The second option is even worse, the possibility is that the researcher knows that there is a related work, but does not want to write about it for whatever reason. That in fact carries a very ugly name, which we will talk about in the end. So, essentially it is equivalent to giving, it is equivalent to a law, one needs to give credit to others work. Now, from a scientific perspective, that was from more from a practice or from a ethical perspective. From a scientific perspective, the literature review gives the background on which the piece of research that is being reported is based on. So, it gives background to the work, it gives motivation to the work. So, in a way it lays the foundation to the main piece of research that is being talked about. And it situates the main work in the broader context, that is it is like think of a painting where there is a lot of background and maybe mountains and the sun and the clouds, but the central thing is a house and a tree in front of it, without the background the house and the tree are not very meaningful. So, it gives context, it gives meaning to the specific work that you as a researcher is doing. And finally, very importantly, science builds in a community as a series of steps. So, the progress of scientific knowledge happens when new work is built on old work and that is exactly what all of us are doing. We take something that is existing and we either extend it a little bit in one direction or we broaden its scope or we compare it against some other work or we contrast it. It does not have to be our work does not simply have to be an extension of somebody's work. It can be the opposite of somebody's work, but we still need to know what the original work is or what the related work is in order to extend or contrast our work with it. So, there is a lot of reasons why literature review and again this is mostly to set the stage. So, you are a researcher and you are embarking on a new research project. When should you do the literature review? At the beginning when you start looking for a problem, why? Because this is related work may be a place where the researcher gets ideas for the research problem. You may in fact get an idea as to whether you should extend the piece of work or you should try to argue against it, but that is not all. You need to do literature review in the middle of your research process once your work is underway and this kind of research, this kind of literature review may be slightly different from the first kind. In the beginning typically the review is very broad. One is sampling from various areas, one tries to cover, look in different places. The visual image I will give is one does that or one does that, but once your work is underway your problem is a little more focused and you know what you are looking for. So, at this point your literature review might become more focused. You may have chosen to work in a problem in this area and you want to find out more deeply what is happening in that area. So, that is when you need a more focused literature review. That is not as open or wide as the previous one, but you also need this process towards the end when you have, let us say you do an experiment and you have found some results and you have used your results to explain the solution to your problem, but this needs to be situated in the context of the broader work, the broader problem. So, do your results agree with the previous results or related results or do they contrast it? Do your results, are your results very surprising because related work said something else? All these questions you will have once you have your results somewhere towards the end of your research process, the research project. Even at that point you would need some amount of literature review, again there it is very focused. So, from here what you see is that the literature review is not a distinctive standalone step that you do once and that is it. It is iterative, it is a feedback loop based on what you find in your literature review at the beginning, you may choose a research problem, you may make some progress there and then read a little bit more literature and realize that that is not the right path to go and you may go back. So, it is an iterative feedback loop you do it multiple times, it is not one step, but it is a part of the entire research process. The first question that we ask as researchers is what type of papers should I look for? And this is something we discussed a little bit in the session on how to read a research paper, a lot of it overlaps with what we discussed there that we are looking at engineering and scientific research papers. We are looking at published research papers in established journals and peer reviewed conferences and I have put established in quotes because this question always comes up how do I know a journal is established that is a question there are many answers to that. Sometimes in a community of researchers a certain journal has or is regarded as a sound journal or a reputed journal perhaps because it is been there for a long time perhaps because high quality papers get published in it and so on. Once you get more experience in a research area you will be able to identify these easily. If you are a novice you can ask your guide or you can if there is a paper you read and you feel that it is of high quality that it is a sound paper see where it is published that might be might turn out to be an established journal. At this when you are doing your literature review you should definitely look at survey or review articles where which talk not about one single research problem and how it was solved but a survey article is a literature review of a problem. So, in a way if you find a survey article your work is very easy because somebody else has done a lot of the work that you may have needed to do by looking at several papers. Sometimes you will find it in a book chapter we again discuss this in one of the earlier sessions that several papers on a research topic may be put together in a book. So, you can look at papers in all these places what I would say is that you avoid papers downloaded from websites which contain no source information I am not saying avoid the internet in fact most of the information today is available on the internet but let us say you find a paper on a website you do not know if it is published or not you do not know which conference it was presented at you do not know when it was presented you may not even know who wrote it it just is someone's or a bunch of opinions on some topic which is there on a website typically try to avoid those at least till you are sure you are able to evaluate the quality. Another way to answer what type of papers usually it is recommended that you read recent papers for let us say the past few years five years or so there are some papers which are important even if they are old they are the classic papers or the seminal papers which have key results that are applicable even today or there is a classic technique that was used then and it is explained very well there and everybody uses the same technique today. So, there are some papers in every field which are classic which everybody refers to. So, even if they are old those are papers which one should put in one survey where should I look for these papers I know now what type I am looking for but where should I look for them and here there is a there are number of things one can do. There are things called databases or indexes these are not a single research paper they but you can search in the database what it contains is links to collection to collections of several papers several hundreds of several thousands of papers some of these may be familiar to you for example, IEEE explore is an index or it contains you will be able to find all the papers published by IEEE by most of the IEEE journals within in the past several years. So, you can either browse by topic or you can search like that for science and engineering the ones on the slide are they are fairly commonly used InSpec is one of them all the ones posted here are commonly used databases and indexes for engineering engineering and science and if you look at the notes in italics there it says check your field because if you are a social science professor trying to do research these indexes may not help you there are alternate ones there. So, it is a good idea to check your field as to which databases and indexes you should be searching in whereas, one can find papers are websites of the actual journals. So, if you know that you want to look at the IEEE transactions on something go to the journal and if your institution has access to it has obtained access to it you might be able to see the entire article, but even if not most journal websites at least allow people any anybody to read the title and the abstract. So, journal websites are a good way to at least see what exists and then one can think about how to in fact obtain the particular article. And of course, now these days there are several journals which give open access that is another movement which is catching on quite soon. You can go to conference web pages and see the topics that were presented in the conference and the papers belonging to it. Conference proceedings sometimes are a little harder to find, but at least you will as we mentioned earlier you will be able to get the title and the abstract. And one thing you can do is suppose you know that there is a certain paper in which was presented in a conference in January of this year by such and such author, but you cannot find the actual paper copy. Look at go try to find the authors web page and see if it is uploaded on the authors web page that is another way another guideline to try to locate the paper. And what I have put at the end is what the senior members amongst you would be very familiar with because that was all there was let us say 15 years ago. The library, a good library, a good academic library in a university would have bound volumes of journals where you can go and photocopy read the papers and photocopy them and so on. The first several are all electronic sources. So, one question which gets asked at this point is what about Google? Can't I simply put a Google search for I am going to use the same example that we used in the previous session. Suppose I put a Google search on use of clickers in distance education what would happen? So, the answer is you may get something, but it is slightly it would be slightly better if you use Google scholar that is more appropriate than a simple Google search which is extremely broad. You can also try sites here which is another focused search engine mainly for academic papers. So, when you are trying to locate papers you want a large number of papers, but you want them to be relevant. If you go to plain Google it is going to be extremely broad may not be very relevant. Google scholar is better focused so is this one. More focused are the ones we talked about earlier, but they may not contain all the papers that exist and so on. So, it is a balance there. Let us look at the next question. Once I have the paper what information of the paper should I record? Especially when you start reading more and more papers and let us say you have read 50 papers for your thesis. It is daunting to make sense of the 50 papers. In a few minutes we will see what information within the paper as in what to do about the content of the paper, but this is even before reading the paper itself. This is called bibliographic information. It is the identifiers of the paper. So, we need to record a spreadsheet is usually a good way to record these or any other way of keeping track. We need to record the title of the paper, names of the authors, year of publication, very importantly the source that is where was it published with journal or which conference proceedings or which book? Who is the publisher? If it is a book what are the page numbers? So, all these are identifying information that one needs to do and just record it and keep it. I am mentioning it here it is a little boring to say or even to do, but it is invaluable if you have. I would like to intervene at this day to point out that what Prasasana Murti pointed out as a little boring and very preliminary information etcetera. How many serious mistakes that we make? I have for example, while reviewing papers invariably we come across situation where at the end of the paper the references are quoted in an extremely haphazard manner. The source is not identified properly, the title and authors are not written properly, the year of publication is not known, ISBN number for books is not included, most well established journals like for example, IEEE ACM journals and I am sure the top journals in every field have a single well defined mechanism of quoting references or writing references. People do not write them in that fashion in many of the research papers that I see here coming out of India and that is perhaps because they have not made a habit of writing down these details when they read a paper. So, I have seen many researchers including teachers at many engineering colleges who come to do their PhD here or who are collaborating in some of us. When you ask them about details of the paper they will invariably scribble something, some journal, some year something, author, if there are three authors they would have written only one name and they would have written copy of the abstract. Now, that is a very shoddy way of going about reading a research paper. I am going back to reading a research paper because in that lecture Prasanna Murthy did not emphasize this fact. There we are all concentrating on how to make sense out of the contents of the paper, but this bibliography information it is correct recording while you are starting to read a paper is vital. Indeed, it becomes more important these days because many papers are now read off the screen of a computer. So, when you are reading from a screen you generally do not write anything. I would suggest one fundamental rule and which has been stated by Prasanna Murthy mentioned it, never ever read a research paper, never ever without a notebook in your period and this is sacrosanct. This is not to be violated. If you do not have paper and pencil you do not read it. Second, when you start reading a research paper may be after preliminary investigation of whether it will be worthwhile for you or not or even if at all you or anybody else is going to relook at that paper then the first thing that you must write down in your own hands meticulously is the bibliographic information of that paper. Without writing that if you proceed to write your understanding of abstract or your understanding of contents that is bunk up. You have violated the fundamental discipline of a researcher and I will tell you when this will become important. You would have a large notebook at the end with lots of abstracts, information, etcetera, etcetera. In the end when you are preparing your final ME or MTech dissertation or a PhD report thesis or a research paper and you have to compile the references which you have cited in your work you will suddenly find that you do not have adequate details. And I will tell you the best of the work meets with extreme disgust by reviewers if the references are not written properly. In IIT Bombay routinely the best MTech student who would otherwise get an AA grade routinely gets an AB grade just because references are not written properly. Now that is the discipline we try to foster. That is the discipline you must try to foster both amongst your PG students and amongst yourselves first. I cannot enforce a discipline in somebody else if I am not following that discipline. But I will tell you this will also encourage you to discipline yourself to write something when you read something. Without writing there is no sense in reading and you must start by reading this particular by writing this particular information meticulously. 100 papers and then you want to try to make sense of it. Secondly, sometimes if we go back to our literature review after few months we may have forgotten what papers we have already seen. So, the identifying information is useful in such a case. How to select which papers to read or how to select the papers? Again here are these are all tips they are not it is not the entire process that I am talking about. There is really no single process, but these are useful tips that any experienced person would tell you. I am starting with ask your guide and the reason is if you are a novice researcher, let us say an MTech student beginning on your first research project. Usually your guide is a first person who suggests to you the paper that you should start reading and then it is up to you to find the next paper to read and the next paper and so on. So, you can always start with an experienced colleague or a guide. As I mentioned earlier a survey or a review article is a very good starting point because it has been written in such a way that a large number of papers pertaining to a given research problem have already been read, listed, analyzed, summarized, synthesized and so on. So, it gives you a very good idea of what the field is about. Sometimes survey articles do not exist for the topic that you are looking for and at that point your guide might ask you to write one yourself, but that is a different topic. Next point here, how to select papers? Every field has certain key papers or seminal papers. Our job is to try to identify them. Once we have identified let us say 2 or 3 or 5 key papers in a field, then their papers we must read and their papers we will go back to over and over again. Some ways of trying to identify these key papers is that a lot of other papers would be referring to this one paper. So, if you see a name of a paper or a particular author or a research group being mentioned several times in many different papers, you can assume that this seems to be a key paper in this field because everybody seems to be going back to it. Sometimes it is the first paper that does some groundbreaking work in the area, but sometimes it is not necessarily so. Once you find the key paper what you can do is look at some follow up work from the same paper. So, find the key paper and see what the same authors have done after it by searching for the authors in one of the databases that we talked about or by going to the authors web pages and seeing if their research groups have done any further work. So, what you are doing here is really building up the collection of papers you are reading rather than going about it in a very random fashion. You look at let us say one survey article, two or three key papers, some related work and there is one more point I did not put here, but it was there on a previous slide that look at some very recent papers say five recent papers in the field that would give you a very good starting point to begin your literature survey. Then when you get more into the process what you can do is let us say you read an interesting paper. You like the paper for whatever reason it has a good technique or its results are interesting. Follow references from within the paper. So, see what papers are present in the references of the paper you like and backtrack through them. That is another way to select which papers to read. Sometimes it may happen that one does not find too many sources there are very few papers one finds what to do then. So, there are few things one can do we may have to change tracks a little. So, I would go back to discuss with some expert see if the expert is able to suggest some paper within some article within in that problem area. One more thing you can do is look for related topics or synonyms. So, for example let us say you try to search for clickers in distance education and you do not come up with anything very few. What you can do there is realize that the word clicker is sometimes also called student response systems make half the world calls it a student response system. So, may we just look for a synonym or look for some related topics. So, instead of distance education you can think of online education which is not exactly the same, but related. One more thing you can do is this is important you can broaden your scope a little bit. So, instead of looking at clickers in distance education look for benefits of clickers that that becomes very broad and you will get a lot of articles on clickers in face to face education also, but then narrow back sift through those and narrow back to clickers in distance education that is one more way you can try to if you have too few sources you can try to increase them. But what experts would recommend is look very hard try very hard before you conclude that no work has been done in this topic. If not this topic some work would have been done in a related topic and you have to be able to find it. I am not saying it is impossible, but it is very unlikely probability wise that the problem I am working on is such that nobody ever has thought of anything related to it. If that is the case then the person who does that might indeed be the next genius around him which is a good thing, but most probability wise most of us are not do not come up with ideas of that nature. What if you have the opposite problem which is more common actually well both may be common. What if you find too many sources what to do then? That means you find so many high quality good articles in established reputed journals that your guide approves of and recommends in the topic that you want to research. At that point you have to find a way to scale down your literature review to scope your literature review. So, there are some tips there you can choose to in the beginning look at only the recent papers. So, you can restrict by here you can restrict by source. So, for example, you will you can say that I look at only this these top two journals in the field I will restrict it only to those two journals or you can restrict it by topic that is you can further narrow down or scope the topic to give you an example. We will go back to clickers in distance education if you get too many papers in that you can look for distributed architecture for clickers in distance education or you can look for clicker use in computer science in distance education. So, you can find words to scope your search further sometimes you use all of these together. So, you may look at only last year stop conference in a narrow field that is also possible. Again here an expert's help is useful in trying to sift through or scale down your search. So, now let us look at until now we were looking at how to conduct the review how to find the papers what to read and so on. So, now let us say you have done all this and you have found some manageable number of papers that you have read what to do then and how to report it. So, what I am going to start off with is an example that we see all these are many novices do. Let us say you have read three papers on the use of clickers. I am just taking clickers because you have read this paper. So, we keep the same topic no other and it is also a general enough topic which is useful that can be understood by people, faculty members and students of all different domains. So, here is a typical novice way of reporting their literature review. Take a moment to read this you see that there are three paragraphs here or three sentences. The first one talks about what paper one did the second one talks about what paper two did and the third one is about paper three. The first one another way of looking at it is that the first line talks about student motivation and attendance. The second talks about student learning and also talks about how many students there were and the first one does not say anything about how many students there were. The third the second and third papers both talk about the topic of the class. The second one was in a CS class and the third one is in a health science class. The first one does not say anything. This is not the way to write a literature review section. It is the easiest thing to do when you have papers there is room or there is a place there you will have opportunity to list papers one by one there is value in doing it, but when you write a literature review section in a thesis or a paper an article you are writing this is not the way to write it and the reason is a literature review or a literature survey or the related work section goes by different means is not a listing of papers. It is not only a listing of papers it is not even a descriptive listing of papers. It is not a summary of one paper after another even if it is done well. So, so far we have talked about what a literature review section should not how it should not be written. So, I am sure all of you are wondering so what should I do. So, let us look at some tips there again. The main idea here is one has to be able to analyze and synthesize and what this means is what analysis means is there is something large and complex just like an engineering system which all of you are very familiar with. Analysis is breaking it down into parts and finding out the relationship between the parts. So, when you read research articles what are the little pieces of the parts of a research articles? It could be the themes, it could be the variables in the research article, it could be the problems faced. So, there are many themes issues factors and variables which are let us say the pieces or the sub pieces of a research article those need to be identified and then they need to be organized. So, one way to do it I will just give you one example that which might be useful in organizing these various themes and factors. So, I am going to show you a table and this is these are papers on clickers. Each row is a new paper this is paper 1, paper 2, paper 3 and so on. So, the list is long what has been done is themes have been identified and that is where the skill of the researcher comes in. What are these themes? Some are very obvious such as the paper title and year and maybe even the authors must go in this table. So, the broad topic here was use of clickers. So, the researcher in fact this was a PhD student in her first year what she said is ok first I am going to try to see which subject each paper or what is the subject in which the clicker user clickers have been implemented. The first one was physics, the second one was some numeric modules, the third one was physics again. So, even just by looking at 3 papers you might get an idea that perhaps we can start looking for trends or patterns and similarly we want to develop more and more categories or themes and try to classify each paper according to these categories. Let us look at one more example. This category here talks about the methodology used in the paper. One uses surveys, the second uses a qualitative questionnaire and the third uses something. In terms of what is measured the first one measured student learning and the researcher has also written how student learning is measured. For example, they say that she says that there is a concept test performance which is measured. Second paper claims to measure perception and learning and so on. Now, as you see this table is evolving you would not be able to write all these headings after writing only the first paper. So, your number of columns in the table goes on increasing so do the number of rows. After reading two papers the researcher, student researcher realized that she should write down what are the main results. So, she wrote down results and then she wrote down a column called comments and ideas for future. So, as she was reading the paper she was getting some ideas what she could do and she was jotting it down here. So, let us go back to our main slides. So, one of the first things you have to do is identify these themes and find a way to organize them. This is just one way to organize the literature review section and organization is important mainly to identify the trends and the patterns. Then what one does the table will become very handy now is analyze these papers further analyze on the basis of categories which we have already learned to some extent, but on the basis of strengths and weaknesses. So, for example, using this table one can look at the results section and see which papers report positive results and which report neutral or negative results. One can read the notes one has made under methodology and comments and decide whether what are the strengths and weaknesses of the paper. Perhaps after you read a few papers you may want to come up with a new column called strengths. So, the analysis has to be done on the basis of strengths and weaknesses and here is an example from again based on the clicker paper. The categories that the researcher decided. So, this is a researcher reading the paper that you read the distance clicker use and distance education. It is some student who read the paper and what he decided to do is he came up with mainly two categories. One is one category I will call as effectiveness and his summary was that clickers in face to face classrooms are highly effective for the following reasons. So, this has been written after reading several papers. Another way of analyzing was on the basis of benefits and drawbacks. So, if you read the paper you will see the if since you have read the paper you may be able to recognize these that the authors write about both the benefits and the drawbacks. So, they are analyzing the literature review section on the basis of benefits and drawbacks of clickers and of related work to improve interactivity. What do you know? What can you do next? At this point you have to put all the little pieces you have found out together you have all the little pieces that you have analyzed you have to put them together or you have to synthesize and create something new. For example, you can find a pattern you can say that I will first look at the graph at the bottom. This was also a student what she decided to do is plot a histogram a bar chart actually of the number of papers she read and the subject in which it belonged and how she did it was based on counting, categorizing and counting the entries under this topic of subject. This immediately gives us insights you will see that number of research papers on whatever topic is very high in health professions and psychology not so high in others. So, synthesis gives us deeper insights. How does one do synthesis? What you can do is try to look for anything that is common or something that is different. So, analyzing and synthesizing on the basis of commonalities and differences is a very often used technique. You can use representations that you use in engineering charts, graphs, tables to represent the analysis. Once you have these graphs you can write your literature review section in terms of maybe three paragraphs. You can talk about one paragraph on what are existing solutions for your problem on related work, what are the common things in all of them and how do they compare. So, each of these questions can become one paragraph in the literature review section. Another important purpose one has to identify the gaps in the existing work. In fact, that is one of the key reasons why one does literature review. The example this example in blue has been taken from the Clickers in Distance Education paper. Just read it and the last sentence here clearly identifies the gap in the existing work. This paragraph comes from the paper itself. It comes somewhere towards the end of the somewhere in the introduction section where they are talking about existing work. So, the authors have concluded here is that no one has identified Clickers in Distance Education so far. The gap is clear. This is important because this leads to the clear need for the research or the thesis or the paper that one is writing. So, now let us look at just some minor points to note two or three of them. As I might have mentioned earlier at a broad general level all this does hold in engineering and science maybe when in other fields. But in spite of that you need to be aware of the conventions in your discipline because in certain disciplines it may be possible that the literature review section has to be written in a very specific way. If that is the case please follow it. In the absence of strict guidelines like that you can use the broad guidelines that we discussed right now. This is one minor point. The second point and this is really the last point I want to make here. There is a tricky issue. It is a conflict in a way that you have read about all these papers. You have read about or you have read all the existing related work. So, ideas already exist and they have already been written by other people. But you have to write about something that already exists but at the same time you have to write it in an original way. You have to say something original. What this means is we will do an entire session on this. You cannot copy a sentence from another paper and put it in your literature review section. You cannot copy a graph and put it in your literature review section. You have to say something original. So, this is tricky. It comes with experience. I know that seems to be a cure for everything. But if you follow some of these guidelines. For example, how do you provide a synthesis? How do you do analysis? How do you identify themes and so on? These are guidelines that help a researcher come up with something original based on something that has been already said. What one should not do precisely that there will be most likely there will be a session on how to avoid plagiarism. This is the ugly word that I was mentioning earlier. I do not want to talk about it here. Let us end here because what this session was supposed to be is some tips and guidelines on conducting and reporting the literature review section of one's research. I hope you will be able to apply some of this to your own work. Thank you. Fine. You would have a worksheet for an activity. It says worksheet for activity one and you have for example, the paragraph in purpose, paragraph in purpose. So, there are two sheets like this and you say paragraph and purpose. Now, the worksheet that you have looks something like this. I hope this has been circulated to you all. Now, this worksheet says that you have been given two papers. Actually there are not two papers really, but there are two copies in the print out. So, one is called design and evaluation of Oscar physics learning objects. You have a bold title here and it has paragraphs marked as A 1, A 2, A 3, A 4, A 5 and you have exactly a similar paper but written slightly differently where the paragraphs are marked as A B 1, B 2, B 3, B 4, B 5. So, there are two sides. There is a two page paper say it is not a very long paper. The worksheet requires you to look at the difference between the first draft and the final version. So, obviously A is the first draft and B is the final version. You are expected to form group. One suggestion is that if you have already formed teams for doing some work together, does not matter what is the area of work that you have chosen for the team assignment post workshop, but those teams could ideally be deployed even today for this particular activity. If nothing else, the team members will get to know each other better. So, it is ok if you form teams of four for example, or five or three. Now, you are required to read the paper A because each paragraph is marked with 4, A 1, A 2, etcetera, etcetera and use the worksheet to write the purpose of each paragraph in the paper. Repeat the same thing with paper B. Now, while you form the team, the activity that you do for paper A and paper B first must be done individual. Each one of you should read these papers and write down your own observations about the purpose of each paragraph. After you do that, collect these sheets together and the team coordinator should read out what purpose has been written by each individual member of the team and then come to a conclusion through a small discussion. Please do not try to summarize or write the contents of the paragraph, but you have still to write complete sentences. So, there are the first open activity one worksheet. You will find that this worksheet, the first page of the worksheet that is A 1, A 2, A 3, two samples are given. I will read that out A 1 says problem context introduced importance mentioned contribution stated. The second paragraph A 2, the purpose is described as related prior work described. Now, I tell you something, writing down a single sentence describing something is actually far more difficult than writing down four or five sentences. If you remember that you are required to state the purpose of that paragraph in the context of the discussion which was done in the previous lecture. This exercise is deliberately being given at this juncture. You should spend a good half an hour before the tea break on this. My session is you do this very, very seriously because this emphasizes certain aspects of reading the paper which was studied in the, which we, which Prasalamuthi discussed in the previous session. I was telling you why I decided to choose this occasion after discussing the literature survey. And that is because the literature survey notion is a continuation of a reading of a single paper. And that is why I thought we will first complete our lecture on the literature survey so that you can move from a single paper to a multiple paper set and how to handle it. But before you begin this exercise, I have an extremely important point to make and that relates to plagiarism. I am not talking about a plagiarism of the worst kind where people are tempted to quote results arrived at by hard work of somebody else reported in their paper and you try to put it as your own work. Now that is a academically worst criminal activity. It is like a murder, academic murder. There is no other words for that. And if we commit murder in real life, we are hanged. Academic community does not physically hang you, but if you are found out, you are academically hanged. Your career is practically ended. That is the policy of the entire academic world. The policy is not very apparent and very obvious here, but if we are going to involve ourselves, a developed world and doing great research, this is something that we will have to add up. You would have noticed that very senior people including the vice chancellor of a university was castigated and in fact, he had to resign because a paper of one of his students in which he was a joint author was found to contain results claimed to be done by the vice chancellor and the student, but it was actually done and reported elsewhere. Sadly, my own colleagues report regularly from time to time that they have received letters from somebody who is a reviewer of some other paper saying process so and so. While I was reviewing this paper, I came across a result which I remember you had reported one year ago in your paper so and so. The implication is either you lied and you copied this fellow's result which this fellow has not reported or this fellow is lying and it becomes a very, very messy affair. So, undoubtedly such a thing is not to be done, period. Now, this is something that we would like to emphasize to our students when they do research, but this is something which we must imbibe no matter what urgency, no matter what the requirement of my having to write some papers and publish them because of my career and not, no matter what. This is not to be done. Having said this very, very empathetically, I would like to now go back to the literature survey and say that even when you do literature survey, when you summarize somebody's SS work and you are quoting it, let us say you are citing in reference so and so. Dr. So and so and so and so reported in reference five or whatever the following. Now, if that following is not written in your own words, but you are quoting exactly the sentences and paragraphs which are actually appearing in the paper that you are referring to unless you put quotation marks that is plagiarism because whenever you summarize whenever you write a literature survey or whenever you report related work, it is expected that the reporting is done in your own words. That is the norm. Any related work must be reported in your own words. In case there are certain sentences of phrases to be quoted, they must be explicitly enclosed in quotation marks. Anything which is not in the quotation marks is assumed to be your own writing, your own wording. So, you must necessarily do paraphrasing. You must necessarily learn to write in your own words. I will tell you, I am not blaming people for being tempted to doing this because our traditionally the Indian habits are not oriented towards strong writing in our own words. We have long traditions of were long verbal traditions, but we do not have strong written traditions. We read very little and we rarely write. That is not acceptable. The very fundamental requirement of research is we read extensively particularly in the methodical fashion that Professor Sarnamurti indicated and we write. What I mentioned in the at the end of the previous lecture is necessary to have a notebook and and write starting with bibliographic information. That is not sufficient. So, when you write an abstract or when you refer to the material the work that has been done by somebody else and reported in that paper, you must automatically learn to rephrase it, paraphrase it in your own words. If you do not do it and if you do not enclose it in quotation marks, then it will be considered as a plagiarized reporting. It is it will be considered that you are actually stealing words, stealing sentences from somebody and it is as bad as something that you ought to write in your own words. So, this is a point to be noted that you have to write about what has already been done written, but write something original. This is what I would like to say as a concluding part of the second session of Professor Sarnamurti and I would like to repeat what I said earlier. Whenever you write an ME or intake dissertation or whenever you write a PhD thesis or whenever you write a technical paper, if you report somebody else's work as your own work and try to claim the credit for it, please note I am not referring to literature survey. I am referring to the main part of your own paper that you write or your report that you write or the thesis that you write. If you ever are tempted to report somebody else's work and claim as your own work, it is a criminal office equivalent to academic muller and you will be academically hanged. If you are not academically hanged immediately, it is merely because somebody has failed to notice it, but that is not the way we live. That is not the way academicians should live. You know, this is where the very, very fundamental issue of ethics. You buy a railway ticket because you are going to travel by train not because if you do not buy a TC may catch you and find you. That is not the objective of buying a railway ticket and that is so fundamentally ethical requirement of any societal behavior, ordinary societal behavior. Anybody who travels without a ticket, we call that person a thief. If you ever plagiarize like this, it is nothing short of theft, but in the opinion of all serious academic researchers, this theft, such theft where you claim somebody else's work to be your own work is tenth amount to academic muller. Sorry for using these strong words, but it is absolutely essential that you emphatically burn these words in your own heart and in your own mind and imprint these on every student who is attempting to do research and report that. So, I would like to take some questions if we can have some interaction. Two or more candidates can register to same guys because there are certain topics which are having two or more aspects. Like suppose I am talking about electrical engineering, when we do some topic, it comprises of power system field aspects as well as drive or control. So, it is very difficult for one candidate to pursue all the fields simultaneously. So, is it possible that two or more participants can work on the same topic under the guidance of the one guide instead of choosing two guides, two or three people can do research on same topic. It is a routine practice that multiple students and multiple researchers work under the same guide. When you talk about multiple people working on the same topic under the same guide, of course most people will be working on the same topic because that would be the topic of expertise of that guide. It is not the topic whose commonality is of any concern. It is the specific problem that you choose in the topic on which you are working, which is of concern. Two people cannot be working on identical topics under the same guide unless there is identical problem under the same topic under the same guide. The problems must be defined differently. Topic, field, subject, etcetera are nomenclatures which are broad nomenclatures and there is no topic or no field or no subject. However similar it might look between two students or two researchers who are working on it where two distinct problems cannot be defined. Indeed it is quite possible that in an emerging area initially a few people work together for six months to define a set of problems, but the initial ground work done could be common and could be commonly reported as part of the literature survey for example. Literature survey could be very similar, but if it is identical in the final thesis there is a serious issue. The problems defined must differ and the moment problems differ there must be something different in the literature survey even of the two people who are going to work on those problems. And of course the problem themselves and their solutions must be clearly discussed. Okay, we will go to Goa College of Engineering. Sir, while doing literature survey how do I know that I have exhausted all the papers related to my topic? Very good question. The question is I will repeat for others. The question is while doing literature survey how do I know that I have exhausted all the papers? The right answer is you will never know. In my field in software for example there is a thing called testing. So teams of people test my software to find out bugs or errors in the software. No testing team can ever claim that because they have not found any error the software is error free. In fact it can be proven for large software there cannot be any error free software. In exactly the similar fashion the number of research papers that are being published are so many in number that it is extremely difficult to conclude that I have exhausted all papers. That is the reason why two things are suggested. I will go back to Professor Salamurti's observations that the literature survey is not to be regarded as one time activity. Initially when you do the literature survey you go across the broad spectrum of the areas related to your problem and try to identify multiple papers then you choose from amongst those papers. Somewhere down the line when you have started working on the problem you once again do that kind of serious literature survey phase to find out whether there are more papers in a problem that you have specifically chosen. And finally when you believe you have completed your work you once again go back and do a very serious literature survey particularly of the research papers to find out whether something relevant to your problem has been ever noticed. It does not mean that there are exactly three phases beginning middle and the end when you do these things. Literature survey is a constant process you keep going to the library going to the web and finding out once in a while if something new has come. It is not uncommon for people to have found out after considerable hard work of three or four years when they believe they have solved a problem in a particular way they suddenly come across a paper which describes exactly a similar solution but then independently. Now if that has been reported in a paper your four years work is lost because nobody will give you credit for doing that work similar in which case you will be forced to continue your work and produce some different results better results etcetera. So this is a I would not say it is a never ending process but it is definitely never concluding process. In a nutshell I will say that is a good question and the answer is no you can never be sure therefore the correct strategy and approach is keep on looking at it while you are working keep on looking at other papers and so on. Thank you R. V. S. College of Engineering and Technology Coimbatore. I am Dr. Naganandini from R. V. S. College of Engineering Sulur Poimathur Remote Center. I have a question. Mr. Zagana Modi has given a wonderful speech on selecting the paper reading the paper and review of literature. Is it necessary to have a foreign literature in our PhD thesis and how we can write foreign literature and one more question is that any limit in number of writing reviews in thesis relating to Indian and foreign literature. Interesting question. Thank you for raising this question. This question comes out of certain set practices which this country has been sadly following for more than 60 years after complete independence. In my own personal opinion this is utter nonsense. What is foreign and what is so big and important thing about foreign? In research what is important is good quality research. If it is done in a village 20 kilometers from Coimbatore and is reported then that must be very very respectfully recorded and referred to and used. And if it is an absolutely useless paper, bad quality paper published in some foreign journal by a foreigner it must be trashed with equal vehemence. I hope I am very clear on this. For us researchers what should matter most is the quality of the paper and the relevance of the paper. The origin of the paper is of absolutely no consequence. These methods that you suggest and even some prevailing norms in many even top universities and institutions that for a PhD evaluation there should be some examiners from outside the country and some examiners from inside the country. In my personal opinion that is bunkum that is a clear clear nonsensical statement we should be ashamed of making such statement. If your university states this rule somebody should raise this issue that why are you saying foreign and Indian. Why are you not saying good quality papers, good quality researchers established researchers. If the established researchers in a particular field and if the well good quality papers in a particular field are not available in Indian journals and good quality researchers are not available in India then wherever they are available they should be taken. Please understand that the academic community is always regarded as a global community. The notion of Indian and foreign or Chinese and Japanese or Mongolian and American are notions which have come because of completely different and wrong reason. So I would submit to you that what should be the requirement for a PhD thesis whether there should be so many foreign journal references and whether there should be so many Indian journal references is an irrelevant point. What is important is good papers and relevant papers must be referred to and good examiners must be examining our work. These good examiners they may exist here or they may exist anywhere. So I hope that answers the first part of your question. The second part was how many papers should be referred to in your literature survey. Is there a number required? I have personally come across a PhD thesis which has more than 300 references. In fact the bibliography was larger than any single chapter of that PhD thesis. I would regard that work itself as a good survey work provided it is backed up by substantial amount of original work as well. However I have also come across absolutely top quality PhD thesis which has which does not have more than 20 references 20. Now these are extreme. My own thought in this process and by the way my answer may not tally with the commonly accepted norm across the country even. But my own take on this is that if you consider that you have covered majority of good papers that are relevant to the problem that you are solving then that is the number of research papers that you should include. If those number 500 then be it you require all those 500. But if you can find some 8 or 10 extremely good and exhaustive survey papers which cover most of the background and then add to it some 20 or 30 papers which are very relevant to your work directly and nothing more is relevant you can stop it. I believe if you state so in your research report that this is how you have done it that will be completely accepted. So thank you very much for another good question let's go over to Rajiv Gandhi Prabhupadi with this is Bhopal. Hello good afternoon sir. I am Rupesh Devang from Rajiv Gandhi University. There is a one question regarding flag if somebody copy means 70% copy original material exactly how we can identify or how we can identify the originality we can say means totally copy the material of 70%. You are saying that two people 75 70% material is exactly identical you are saying how to determine who copied from whom. Well that problem is difficult because even in a classroom in a 3 hour exam when two people copy from each other how do you find out who copied from whom. So that is not a problem which is solvable so easily. The standard norm is that the material which was published earlier in time is considered the original the material which was published later is considered the copy unless otherwise the person who is accused of copying can prove through the notes or through other work or through other internal reports that no it is that person who has actually done the original work. It is sad that such problems are arising here with us. By the way we talked about India and foreign some time ago such problems do not generally arise in foreign country. They arise only in India and that is because we given to the temptation of getting a degree or getting something very quickly without doing work for it. So what you ask is a very sad question you are you are from Bhopal right, Madhya Pradesh. I also belong to Madhya Pradesh. My own formula would be go to that person who copies and bashes his head that is not a polite reaction in a lawfully abiding citizenship world so I would not recommend that but that would be my temptation. Thank you. We will go over to some other person. We will go over to national institute of electronics and management now. Sir can we create national level database for all the PhDs who have been registered in our country so that it would facilitate all the students to actually go to the particular person who has that PhD. Suppose I am interested in finance it is specialized area inside management so I do not want to locate throughout places where I can search people for example I am interested in stock market and there might be several persons throughout India who might be interested in stock markets because even those who are in finance and those who are in Tamil Nadu yes sir. I got your point in fact I had a discussion with Dr. Manta chairman of AICT very recently on two issues. One is to create a database of researchers who are working in different problems exactly as you are saying but more importantly another database which has become more pertinent today. What we are suggesting is that all the project reports of final year students starting with BE or MCA with ME and MTech and PhD thesis all these reports and their abstracts at least should be available online publicly. This is particularly relevant because while the problems of plagiarism might be occasional at higher levels of PhD and ME but at the BE and MCA level these are absolutely well spread out and they are causing a deep problem in our ethics of academics at the very root level. So the other advantage of such a database that you mention is to create a situation such as it prevails in most of the universities including IIT system and I believe in NITs and other institutions where every MTech student of mine for example when he submits a dissertation the dissertations of copy has to be uploaded in the institute database and the abstracts which are uploaded are available worldwide. So that people can know what subject who has worked on and they are searchable database. So I agree with you such a recommendation is being made I will personally follow it up very strongly and I will try to get this database implementation started and the other database I mentioned where every researcher when he or she submits the research report dissertation or as a PLD thesis or as a group project done at the BE or MCL level the abstract must be reported and must be publicly available and visible. Thank you very much over and out.