The writing of doctoral thesis is the main work to be completed by doctoral students. Due to the higher difficulty, longer writing cycle, and relatively high requirements in terms of innovation, writing standards, practical and theoretical significance, etc., the completion of a doctoral thesis is generally quite difficult. A good doctoral thesis is not only a good academic monograph, but also an academic exploration result with theoretical innovation value. From enrollment to graduation, a doctoral student should transform from a student to a scholar and become an expert in the field of study.
Although there are such high requirements for doctoral students, there are still some rules to be found in the writing of doctoral thesis. Below, based on my own experience (including my own experience as a doctoral student and my experience in supervising doctoral students), I will briefly talk about some ideas on how to prepare a doctoral thesis for your reference.
Study the literature
Generally speaking, a doctoral student has already determined the field or direction of his doctoral thesis after admission, because this is determined at the time of application and admission, unless there are revisions after admission, which will be done separately. Never mind.
After completing the doctoral program in the first year and obtaining the appropriate credits, you will naturally enter the process of collecting and studying literature. The collection and study of literature is not only the beginning but also very critical for a doctoral thesis. You have not conducted research in the past, and you have not carefully studied the research results of others in this field now. It is impossible to come up with anything decent. Therefore, before this job, you can only have a blank piece of paper, have nothing, know nothing, let alone what the cutting-edge research topics in this field are, and you will not be able to use common technologies in this field. It is impossible. It's just that you can make something decent with just one piece of white paper.
▍Collect literature
Collecting documents can be obtained from the Internet, libraries, bookstores, and friends and classmates. The main purpose is to highlight some (even just one) "recent" review literature that appeared in this field, generally papers published no more than three years ago.
What is a review literature?
Generally speaking, English review documents have the words “…..review…” (generally a more popular review) or “survey” (generally a more professional literature review) in the title.
The content of a review article is not to study a specific problem, but to review and comment on the research status of a certain field in the recent or past period, and to comment on the research characteristics of different researchers and different periods. Finally, It is also necessary to make predictions for future research directions and innovation possibilities.
In a review literature, the author will list a large number of references and comment on the relative importance of these references. You can further search for more documents based on such comments and the listed documents, and then find more references from the references there, and so on. In this way, along a route, you can basically master the research situation in this field within a period of time, for example, within a year, and know what the frontiers of this field are.
▍Collect authors and keywords to search
Of course, there are still many research issues even in the same field. You can search with questions in advance, such as corporate governance issues, sustainable development issues, economic growth issues, etc. This requires the use of keyword search methods.
How to use the Internet to search for literature?
The development of modern network technology allows people to obtain information resources more conveniently. In the process of document collection, making full use of network resources can often get twice the result with half the effort.
There are roughly two types of using the Internet to collect resources:
-
① You are in the topic selection stage and have no paper direction yet. If you want to find a suitable direction to work on, you can search for "review", "survey", etc. online, read some literature reviews, and find the direction you are interested in;
-
② The general direction has been determined and you want to understand the progress of this research field. At this time, you should consult experts in this field to find out who are the best in this field and search for their articles - foreign traditions, many magazines want When introducing achievements and progress in a certain field, talented people will be invited to write reviews - only by knowing who are outstanding representatives in this field can you experience the exciting developments in this field from the works of these people.
Let me also talk about Internet search skills: First of all, www.Google.com is definitely a good library. It is the largest search engine in the world and can support Chinese and English searches. If you already know the name of an article, you may wish to enter the name and take a look - one more thing here, because the full text of foreign papers are usually in pdf format, so you may wish to add "[pdf]" before the entered paper name. ", for example, you can use "[PDF]+A Survey of Corporate Governance" as a keyword search, which will help limit your search to the full text of the paper in pdf format. You can easily search for the Gang of Four Corporate Governance Research LLSV ( The full text of "A Review of Corporate Governance" by 1997).
However, the full text retrieved by Google is often only Working Paper without magazine page numbers, so it is inconvenient to cite it in the paper. It is also easy for us to search for published papers. We can purchase foreign language databases such as EBSCO. These databases provide a large number of full text papers. Documents written after the 1990s, especially after 1995, can basically be easily queried.
If you cannot find the paper you need through these full-text databases, you may wish to write down the page number of the journal in which the paper was published (for easy reference), and then search for its Working Paper on Google. If you can't find it on Google, you can also try searching the author's homepage, maybe it will be there. If you still don't have it, you can also try to write directly to the author's email address to request it. From my experience, foreigners are more enthusiastic. If all of these methods have been tried and still don't work, you can finally check the domestic collections at EBSCO and ask for help. Generally speaking, most of the more classic papers will be put online by good people. Therefore, if you try all the above methods, you will not be able to get a certain document. You will either give up reluctantly, or you will have to ask foreign friends for help, or indirectly quote from other documents.
There is another aspect worth mentioning about collecting literature. The publication cycle of foreign papers is long. The papers we see now may actually be works two or three years ago. Therefore, to grasp the research trends, it is best to read the Working Paper, so as to understand what others are currently studying. Moreover, the good Working Paper you read when you were developing your doctoral thesis will probably have been published by the time your doctoral thesis is written. At this time, you might as well check the journal in which it was published, and officially list the publication information as yours. References will make your paper annotation more standardized and take into account the requirements for annotating the latest literature.
Finally, let me list some of the better websites that I often use that provide working paper downloads for free:
IDEAS：http://ideas.repec.org/
NBER：www.nber.org
In addition, many universities and research institutes also provide free download addresses for their working papers, such as IDEI (http://idei.fr/presentation.php) founded by Professor Laffont. There are many websites, and everyone slowly learns to use them. No more details will be said here. Jestor, EconBase, ScienceDirect, etc. are also good databases, but they require payment.
Study literature and reading notes
In fact, studying documents should be carried out at the same time or interleaved with the document retrieval process. When the literature is retrieved, you should start to study the literature. At the same time, while studying the literature, you should search for new literature according to the new search requirements that appear during the reading process. Collecting literature is an iterative process. .
Therefore, "studying" rather than "reading" documents requires that when reading documents, you should capture the sparks generated for research, and it is not just a simple reading process, such as deriving new models and using new methods. Data calculations, etc. When reading other people's articles, you should develop the habit of thinking with a "critical" and "skeptical" eye. When reading other people's judgments, you should constantly ask in your own mind, "Really? Why? Maybe not" Is that so?", while trying to find opposite examples, and thinking deeply about it logically - this is "research".
According to my own experience, when reading literature, especially after reading many documents along a fixed line of thought, it is easy for new sparks and springs of thought to appear. But after that, it's easy to forget about them. Therefore, in order to make full use of the new ideas and clues that appeared when reading the literature when writing the paper later, in order not to forget them. Generally, the method of taking reading notes is used.
Reading notes are written as follows:
(1) What magazine, which issue, where is there any discussion on what issue? What reference does it have for my upcoming PhD thesis? Or what are my own opinions on this issue and what contribution I may make, etc. wait.
(2) Reading notes should be classified, such as into mathematical models; empirical research; Chinese literature; foreign literature; data; pure theory; policy analysis; and academic debates. ; Case materials, etc.
(3) Focus on analyzing or intensively reading several representative documents;
(4) It is important to write down your experience;
(5) Important data should be written down in time;
(6) The previous process may be repeated, and your own opinions may also change.
It is generally difficult to collect documents in China, especially foreign language documents. However, you can collect almost the same documents as abroad in the Beijing Library. Therefore, try to go to Beijing Library to collect materials; however, at present, Chongqing University Library also A considerable amount of foreign literature can be collected; visiting the library is a necessary process for studying for a Ph.D. This is true all over the world, and this link is indispensable.
Opening report
On the basis of document collection and study, it is time to begin the process of developing a topic. The essence of the proposal report is to report the preparation status of your doctoral thesis to the teachers, and at the same time allow the teachers to comment and make suggestions on their own research ideas, so as to further clarify the research goals, clarify the research ideas, and improve the literature and research methods. The purpose of getting more help from teachers.
Under the guidance of this task goal, the proposal report should include topic selection, literature review (in addition to the proposal report, doctoral students must also write a special literature review report, which can be combined), research goals, research content, methods, Provide a comprehensive and in-depth introduction to innovation, technical routes, feasibility of research, etc., and accept criticism and suggestions from teachers.
Here, there are some issues to pay attention to:
(1) The nature and task of the project proposal report is to report on the preparation of one's own research work, including the collection of literature, study status, preliminary ideas of the research, feasibility, etc., and at the same time accept the suggestions of teachers. Therefore, the proposal report is different from the graduation defense. The student's attitude towards the teacher is not to debate, but to "consult". Students should use this opportunity to ask as many teachers as possible to get more help and information. Based on past experience, this process will be of great help to future research and paper writing, and students should pay extra attention to it.
(2) In the research content, all aspects to be studied should be written down, as well as the problems, methods, preliminary ideas, possible difficulties in studying these aspects, and preparatory ideas on how to deal with and solve these difficulties. , but the text should be concise, because it is just a preliminary idea and does not need to be lengthy.
Many people write this part of the research content like a book or a table of contents for a graduation thesis, which does not meet the requirements.
Generally speaking, the research content should be written in the following order and scope from a broad perspective:
First, the significance of the topic selection, including theoretical and practical significance
First of all, it is necessary to explain what contribution the topic may make in theory. To make this clear is actually very demanding for students, because only if they have fully mastered the relevant literature and have a certain understanding of the theoretical development in this field Only with a full understanding of the situation can we explain this issue clearly.
Then, to explain the practical value of studying this problem, this requires an understanding of the actual usefulness of the problem being studied; generally speaking, the topic selection must have "importance", "cutting edge", "operability", and It must have methodological significance; "importance" means that the research question is related to important breakthroughs in theory and practice, and has a key role in promoting the country's socio-economic development and people's well-being.
Furthermore, "operability" means that it is expected to be completed within the prescribed deadline for the paper, that is, it can be done. Therefore, the difficulty must be evaluated and neither choose a topic that is too easy nor one that is too difficult; "methodological" "Significance" means that some new research methods are expected to be used to study this problem.
Finally, in scientific research, the use of new methods is very important. In fact, the results obtained by some studies may be trivial (that is, what people already know, or know intuitively), but using new methods to study These questions are also valuable; for example, old farmers may have many completely empirical methods to predict the weather, but this does not prevent scientists from using new aerodynamic methods, including the Stocks equation, to predict the weather. It is difficult to say that this new method is necessarily more accurate than the old method; the application of methods occupies a very important place in doctoral thesis. Therefore, the proposal report requires a specific explanation of the method to be used;
Second, the theoretical and practical background of the research, theoretical basic research and review, etc.
Third, a review of domestic and foreign literature
Fourth, analysis of the current situation, root causes and causes of the problem
Fifth, model or index system research (construction)
This part is the core of the paper, where innovation occurs, and is a key part to measure the academic level of the paper. It usually requires a quantitative model or quantitative indicators. It is also the content extraction part for publishing high-level papers alone.
Sixth, case analysis or empirical analysis part
Econometric models are usually used to test the theoretical innovation results obtained previously or to verify them with cases.
Seventh, policy recommendations
If the previous theoretical research is a diagnosis of economic management problems, here we will prescribe prescriptions based on the diagnosis results and put forward policy suggestions based on our own theoretical research to solve the studied problems. it's here. The problem that many students like to make is that there is little relationship between the policy suggestions they put forward and their previous theoretical research.
Of course, this does not mean that all doctoral dissertations should be written in the above order. Here is a reference. However, the content organization of most papers is basically based on this framework, or is carried out after making some changes to this framework.
The innovation part is the highlight of the paper, and the key points of innovation must be written "specifically" and "appropriately". Don’t be too abstract. You should tell how your ideas are different from existing research, and where your ideas are better than existing research. Otherwise, it cannot be called “innovation.” .
The technical route generally refers to the process of preparing, initiating, conducting, repeating, and achieving results for research. It does not refer to the process of writing a graduation thesis, nor does it refer to the process of preparing and conducting the defense. Many students will make these mistakes.
The proposal report must be carried out after the literature collection and study process, because according to the logic explained above, it is impossible to follow the content of the proposal report if it is made before. Therefore, in order to develop the topic, literature collection and study must be carried out as soon as possible, and we must seize the time to do it.
The application of methods should be written more specifically. Many students wrote this item as "using the methods of dialectical materialism and historical materialism", "using methods that combine quantitative and qualitative methods", "combining macro analysis with micro analysis", " "Combining theoretical analysis with practical application" and other large frameworks are not specific at all. In fact, I don't know what kind of method he is going to use; here we need to write down specific methods, such as using game theory methods, econometrics The method, the case analysis method, is even more specific such as "by implanting non-renewable resource input variables in the Solow model, and then studying whether sustainable growth of per capita income can be achieved or under what conditions it can be achieved, thereby obtaining Approaches to new models of sustainable development” and more.
Research and key breakthroughs
Theoretical innovation in a doctoral thesis is the core and a symbol of the level of the thesis, and it is also the most difficult. It can be said that as long as this problem is solved, more than half of the paper will be solved. Therefore, we should focus on breaking through this point first. At the same time, because the entire paper is basically developed around the breakthrough or innovation of this issue, if this core issue is not resolved, there is no way to start writing the paper. Therefore, don’t be busy writing papers first. You must first conduct research. Only when you have results will you have something to write.
Therefore, the school requires doctoral students to publish papers, especially in first-level journals and core magazines. Because, without innovative research results, it is not easy to publish papers.
This process is the most difficult in the entire thesis writing stage. Don't rush, take your time. There must be a stage of hesitation and more attempts. This stage is also the most frustrating. You will find that you are "very weak" and inferior to others in many places because you will encounter many failures. You may even feel that your doctoral thesis cannot be completed, and you may even feel that you want to give up,…….. Note that this feeling is felt by everyone when doing a doctoral thesis, not just you, so you must persist. In my experience and that of many others, when the process goes on for a while, a spark develops, and the winner is the one who sticks around in the end. Goethe said: Winter has come, can spring be far behind? At this critical moment, don’t give up.
Innovation comes from the generation of new sparks, and how are new sparks generated? As a rule of thumb, it is mainly the process that comes from the multi-read model. Generally, after reading many models, you will have ideas about constructing new models; of course, the construction of this new model is generally based on modifying other people's models, and if you have not read many other people's models, Generally speaking, it is impossible to do innovative research without standing on the shoulders of predecessors. This process generally has the following characteristics:
(1) Propose the necessity of modifying and developing existing models by looking for phenomena that cannot be explained by existing models, or problems that cannot be solved or handled;
(2) Find the reasons for the shortcomings of existing models. This is a diagnostic process;
(3) After discovering the areas that can be modified in the existing model, we must start the modification work. This process is relatively difficult, mainly due to technical difficulty, that is, the technical difficulty of conducting model research;
(4) At this time, you should contact your instructor more and ask other experts for advice, but more importantly, you should think more and research more, and at the same time, you should also learn some technical research methods that you have not mastered in the past.
Writing and publishing papers
If the problem in the previous stage is successfully solved, this problem will be solved naturally. Submitting articles should be submitted to first-tier magazines because this is a problem that needs to be solved. Doctoral students are required to publish papers in first-level journals, otherwise they will not be able to defend themselves. In addition, you need to publish several papers in journals listed in the CISSCI catalog. These journals are wider and less difficult than first-level journals, but they are not easy either.
Some people say that it is difficult to publish papers in magazines now, especially in first-class magazines, and it is impossible without acquaintances. Because there are only a few first-class magazines, and there are so many doctoral students across the country who want to publish articles. At the same time, many of the articles published in magazines are articles about "human feelings" and "relationships", as well as articles about celebrities, celebrities, and high officials. The remaining space is fairly reserved for freelance contributors.
Still, I don't think it's as bad as it could be. There are no solutions to some problems, because academic corruption is a very prominent problem at present. However, as long as the article is well written, it is not impossible to publish an article, or even publish multiple articles, even in a first-tier magazine.
For example, some doctoral students have published multiple papers in first-tier magazines within one or two years of enrollment. Of course, it is also a reason why they participated in some fund-related topics. In many first-level journals, research papers on funding-related topics are encouraged to be published. Therefore, students should try their best to participate in research on some fund-related vertical topics. In addition, articles with models are easier to publish. The school requires at least one paper to be published in a first-level journal. However, just one article is just a passing grade. Generally, two or more papers will have better results.
Writing thesis
After completing the quantity and quality requirements for thesis publication, it is natural to enter the writing stage of the graduation thesis. In fact, the paper published in the magazine is also an important part of the graduation thesis. However, the published papers are topic-based and not systematic. It is necessary to connect different topic studies and link different parts to form a unified argumentation system. This is graduation thesis writing. At the same time, graduation thesis is also a way for schools to test students’ overall knowledge mastery, so it is not a simple research question. Therefore, in addition to academic innovation, the writing of graduation thesis also has certain requirements for the expression of written language, knowledge, structure and organization of the paper, and logical application.
When writing your graduation thesis, you should rely heavily on reading notes. Based on the reading notes that have been completed previously, you can achieve twice the result with half the effort. Especially the literature review part of the paper is basically derived from the reading notes, or it is just a more systematic arrangement, deepening and expansion of the reading notes.
Additional thoughts on “innovation”
Doctoral thesis has special requirements for "theoretical innovation". Many students are not clear about this. First of all, the requirements for innovation in a doctoral thesis are "innovation" in "theory", not improvement "innovation" in writing program software and a specific algorithm, let alone the writing of a case or specific policy analysis. "Theoretical" innovation should be said to be one of the fundamental differences between doctoral thesis and MBA thesis. Secondly, there are certain methods to be found in the innovation process, and it does not depend entirely on pie-in-the-sky. So, how can we achieve theoretical innovation? What is the method?
Words of experience:
1. Analogy
Applying methods that have been successful in a certain field to another new field; for example, applying game theory methods that have been successfully used in economics to human resource management in the management field; this is E P. Lazear Led an innovation in the field of human resource management theory based on game theory methods;
2. Hypothesis modification method
Modify the assumptions of existing theoretical models with good reasons according to the new environment to be studied, and achieve new results; for example, adding resource input factors to the production function that usually only contains two production factors: capital and labor, We can study economic growth issues under resource constraints to study sustainable development or sustainable growth issues; this is the research innovation of sustainable development or sustainable growth issues;
For another example, D. Kahneman and A. Tversky divided marginal benefits in economics into lost and gained marginal benefits, and based on the discovery of psychology that people always Give the lost a higher value than the upcoming gains, and use it as a new general hypothesis to replace the original assumptions in economics, thereby obtaining the "prospect theory" innovation of behavioral economics;
3. Combination method
Combine different fields to see if meaningful results can be produced; for example, consider the concept of transaction costs in institutional economics in conjunction with market size, then combine market size with the deepening of division of labor, and then consider the deepening of division of labor. Considering degree and economic growth together, we get the endogenous theoretical innovation of Yang Xiaokai's division of labor evolution and economic growth;
4. From special to ordinary
Everyone can go from general to special, but reverse thinking is often a successful method for creative thinking. Pay attention to the generalization of individual case observations. Seeing general rules from individual cases is "enlightenment"; for example, as early as 1962, K. Arrow discovered the "learning by doing" phenomenon that exists in some manufacturing industries (such as the aircraft manufacturing industry). ; However, in the 1980s, after Professor Paul Romer's generalization, this mechanism was assumed to be a general law in the long-term production process, thus obtaining his endogenous economic growth model and launching the so-called "new" economy. The curtain on growth theory research.
5. There are many works discussing the laws of scientific innovation . Those who are interested can read the book "Scientific Discovery Talk" by mathematician Professor Wang Zikun.
Source | Science Network, Peking University National Institute of Development Management Doctoral Program, Quantitative Research Methods