RESEARCH METHODS AND DISSERTATION TRAINING (RMDT) and DISSERTATION GUIDELINES 2022/23
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
GRADUATE CENTRE FOR ECONOMIC AND FINANCIAL STUDIES
RESEARCH METHODS AND DISSERTATION TRAINING (RMDT)
and
DISSERTATION GUIDELINES
2022/23
CONTENTS
Course aims, intended learning outcomes and course outline 2
The nature of the dissertation in CEFS 4
Dissertation planning: time schedule 8
Format of dissertation and submission 11
Support and supervisor’s role 12
Research ethics and ethical approval 13
RESEARCH METHODS AND DISSERTATION TRAINING
Starting in Week One there will be a programme of research methods and dissertation training designed to introduce you to research methods and information sources, and to equip you to research and write your dissertation. .
Aims
1. To provide training and support that will enable students to research and write a dissertation which fulfils the requirements of the relevant PGT programme.
2. To engage students in independent research and study.
3. To teach students how to construct an argument and maintain it over an extended piece of written work.
4. To allow students to explore a topic of interest in some depth.
5. To enable students to share ideas and experience in developing research proposals and methods.
6. To develop students' oral presentational skills by facilitating group presentations and discussions.
7. To enable students to develop and deploy library and internet skills in literature search and data source identification.
8. To develop students' IT skills including manipulation and presentation of statistical data.
9. To help students develop a clear understanding of various econometric techniques that should be used to analyse different economic, financial, political and social datasets.
10. To analyse actual economic data so as to produce a statistically adequate model;
11. To check the validity of the statistical assumptions underlying the model, using the sample data and revising the model specification as needed;
12. To use the model to obtain reasonably valid statistical test of economic theory and to address a research question.
Intended Learning Outcomes
In completing their dissertation, students should be able to demonstrate an ability to:
1. Undertake a substantial, focused piece of research using research methods and analytical techniques that are appropriate to their programme of study.
2. Research the literature relating to a specific topic or issue relevant to their programme of study, and present a synthesis and critical review of that literature.
3. Identify and formulate hypotheses to be tested or problems to be addressed deriving from their analysis of the literature.
4. Confront these hypotheses or problems by mobilising relevant evidence and analysing it using statistical or other appropriate methods.
5. Evaluate findings and draw out their implications.
6. Present a substantial piece of written work that is clearly motivated and structured, and conforms to the requirements specified in the course documentation with regard to the quality of written work and to presentational features such as length, layout, citations and bibliography.
Course Outline
The following topics will be covered in the course:
• Introduction to the philosophy and methodology of research in the social sciences and economics
• Planning, researching and writing your dissertation
• Narrowing down the dissertation topic and formulating an appropriate research question
• Accessing literature and data: Glasgow University library services
• Using sources effectively and avoiding plagiarism
• Analysis of Economic Data
• Stata practical workshop
• Analysis of research example
• Analysis of dissertation proposals and dissertation
The Nature of the Dissertation in GCEFS (Compulsory Reading)
It may be helpful to think of the dissertation as a slightly extended academic article, rather than something more substantial such as a book. Therefore, the consultation of published articles is an invaluable aid not only during the process of deciding the topic of the dissertation but also during the preparation of the structure of the dissertation.
a) A specific and interesting focus
The key point is that the dissertation must have a specific focus on a central hypothesis or problem which is confronted, just like an article. The dissertation must be well motivated, i.e. there should be a convincing statement of why the dissertation is of interest. What makes a topic interesting? Some combinations of something that (i) matters (when X rises or falls, individuals are better off or worse off); (ii) is puzzling (it defies easy explanation); (iii) is controversial (some researchers/policy makers argue one thing while other argue another).The emphasis on a “specific focus” intends to highlight the fact that the dissertation should not consider a broad generic problem which cannot be addressed through an empirical analysis. For example, a dissertation should not be on the causes in general of financial crises; the effect of World Bank programmes; the consequences of the reform programmes adopted in a certain country; corruption and economic development; etc.
Then, the dissertation must have a specific focus on a specific issue which deserves to be researched, and for which there are available data that students can analyse to provide support for the arguments that are put forward. Thus, to continue with the examples made above, the dissertation should analyse not the causes of financial crises in general but, for example, whether the currency crisis in a certain country caused the banking crisis or vice versa, or could present an estimation of the cost of crisis in terms of output losses. It should analyse not the effects of World Bank programmes in general but, for example, whether they led to de-industrialisation or restructuring of the industrial sector, or the effect of these programmes specifically on economic growth, or poverty reduction, or food security, etc. It should not be on the general link between corruption and economic development but should analyse whether, for example, the eradication of corruption is a pre-condition for economic growth, or whether a specific country’s aid giving is affected by the presence of corruption in recipient countries. It should not be “forward-looking” and discuss the future prospects of the adoption of an independent central bank (as no data could be used) but should be based on something for which there are data that can be analysed: for example, do countries with independent central banks have faster economic growth (or lower business cycle volatility or lower inflation, etc.) than countries where the central bank is not independent? It is the answer to this question (which is the focus of the dissertation) that can then have policy implications concerning the prospects of the adoption of an independent central bank. The dissertation should not discuss generic effects of reforms in a country but should analyse, for example, whether the liberalisation of the domestic banking sector has led to an increase in the amount of financial intermediation, or whether large and small firms have benefited equally (in terms of access to credit, or the cost of credit) from the financial sector reforms.
These are just a few examples. The crucial aspect of these examples is that the dissertation must be about a well-defined specific issue which can be addressed through an analysis of appropriate economic/statistical data.
The dissertation should make an interesting read. You want to convince the casual reader early that (s)he will learn something worthy from your work. One of the purposes of the dissertation is to give you an opportunity to improve your communication skills in the context of your chosen field of study.
The format for the dissertation includes:
- An abstract which states on two sides of A4 or less the aim of your dissertation and what you have found. You can think of this abstract as the billboard for your paper;
it should stimulate the interest of the reader and inform him/her on what (s)he can expect from your dissertation.
- The body of the dissertation as described below. Many dissertations consist of an introduction, a chapter reviewing the literature, an empirical chapter, and a conclusion. However, there is no prescribed template; the structure of the dissertation should be adapted to the topic being addressed.
- The material presented in your different chapters should form a cohesive whole. Each chapter should contribute to the higher purpose being investigated in your dissertation. You would do well to explain, at the beginning of each chapter, how this chapter fits in with the previous one, what it adds to the dissertation and how it is leading towards the next one. If, for instance, you recommend some economic policies in your conclusion, your recommendation should be informed by the results from your empirical chapter. In turn, this latter chapter should explain where it stands in relation to the literature review, and the literature review should bring home to the reader the importance of the topic you introduced in your introduction, and showcase your knowledge of the field.
b) The introduction
The introduction must present the research question, the motivation for the research question, the antecedents and a road-map. The author is expected to clearly explain what the dissertation actually does. The reader should have an idea of a clean research question that will have a satisfactory answer by the end of the dissertation. The author should explain in very clear terms why the topic is worth researching and the dissertation is worth reading. Identifying the prior work that is critical for understanding the aim of the dissertation is a good way to motivate the research question. The motivation can be cast in “theoretical” terms, e.g. one theory says that financial crises (i.e. “twin crises”) are started by banking crises; another theory says that exchange rate crises are at the root of twin crises. Therefore it is interesting to analyse whether in a particular country it was the banking crisis or the exchange rate crisis which caused the financial crisis. As another example, one theory says that trade liberalisation would result in a stronger domestic industrial sector; another theory says that liberalisation would weaken the industrial sector. It is therefore interesting to analyse which theory is supported by the data.
The motivation can also be cast in “empirical” terms. For example, while in theory corruption leads to the waste of aid, do countries give aid only to low-corruption countries? Or, as another example in the area of corruption, is it empirically the case that fast-growth countries are the low-corruption ones?
The dissertation can never be motivated by the argument that the author wanted to know more
about a certain topic. The reader needs to be motivated to read the dissertation! Finally, the introduction should outline the organisation of the dissertation.
c) The literature review
Following the introduction, there should be a survey of the existing literature relevant to the topic. This literature should normally contain both theoretical and empirical elements. For example, in a dissertation that aims to explore whether liberalisation leads to de- industrialisation or industrial restructuring, the literature survey should report those studies which discuss how in theory liberalisation could strengthen the industrial sector (e.g. by increasing the competition of domestic firms which are open to international competition). The survey should also review those studies which are critical of the benefits of liberalisation for the domestic industrial sector (e.g. because liberalisation could kill domestic infant industries). In fact there are several channels through which liberalisation could affect – either positively or negatively – the industrial sector and these should all be discussed in the literature survey.
The literature survey should not be limited to theoretical contributions. Given that the dissertation contains an empirical chapter, the empirical literature should also be reviewed. Continuing with the liberalisation and industry example, the literature review should examine the empirical findings of papers concerned with the effect of liberalisation on the industrial sector, no matter whether in a cross-country, time-series or case study setting.
The literature survey can be done in one single chapter or in two chapters (one for the theoretical literature, one for the empirical literature), as appropriate to the nature of your specific dissertation topic.
It is important to remember that all the different parts of the dissertation must be tightly integrated with each other. Therefore, the literature survey must cover exclusively those issues that are relevant to your discussion of the dissertation topic. This consideration suggests that a good literature review is not simply a list of published papers summarised one after the other. Rather, the survey must be structured, i.e. organised around concepts/issues, summarising the points of view emerging from the literature about those concepts/issues.
d) The empirical chapter
The dissertation must include a contribution that is your own. This contribution is an applied, empirical chapter which directly addresses the main hypothesis of the dissertation. In this chapter, students are expected to make use of and analyse economic/statistical data. This is meant to be the key chapter of the dissertation.
The use of econometrics is not required while the use of at least some basic techniques of descriptive statistics is a requirement (for this purpose your work on the dissertation is supported by lectures on descriptive statistics in the “Research Methods and Dissertation Training” course). Students must show that they are not frightened by numbers and can, to the contrary, use and manipulate (through appropriate techniques) data to support their argument.
Normally, the empirical chapter is expected to contain a description of the data and a motivation of the data used. A practical problem that any researcher faces is that the data which are available are only a very imperfect reflection of the theoretical concepts they are supposed to measure. For example, concepts such as de-industrialisation or liberalisation cannot be measured directly and the available economic data give only an approximate indication of de-industrialisation or liberalisation. Therefore, this chapter should describe which data are used (including the source of the data) and why they are appropriate (reference to other empirical studies that have used the same data for the same purpose as the dissertation would be useful in this context). The limitations of the data should also be clearly pointed out. In the example of liberalisation and industry, an additional complication is that some economic data may indicate an industrial contraction but may not easily discriminate between the alternative hypotheses of de-industrialisation or industrial restructuring. How one intends to test one hypothesis against the other should be clearly explained.
Following the description of the data and the motivation for their use, the chapter should clearly explain your testing strategy (i.e. how you are going to assemble and manipulate the data in order to answer the particular research question which is at the core of your dissertation) and the techniques of analysis that you are going to use (without merely reproducing a few pages of a statistics textbook). For example, if the dissertation aims to analyse whether trade liberalisation has a contractionary or expansionary effect on the industrial sector, you could take a group of countries that have liberalised their trade regime and test whether their industrial performance after liberalisation is statistically different from their performance before liberalisation.
Subsequently, the empirical results are presented and discussed. For those students using econometric methods, regression tables should include (i) the coefficients estimated, their corresponding standard errors, the number of observations and the r-squared, (ii) the dependent variable should be reported in the table’s caption or as a column heading, (iii) self-explanatory labels should be used for the explanatory variables (abbreviations or symbols force the reader to page back and forth to understand the results), and (iv) superscripts to show statistical significance should be included: "a" or “***” (1%), "b" or “**” (5%), and "c" or “*” (10%).
The empirical chapter should normally close with a discussion of the limitations of the analysis and possible alternative interpretations of the results
A dissertation where the student makes no use of economic/statistical data would not be acceptable.
e) Conclusions
The final chapter presents the conclusions. These must be based on the analysis performed in the dissertation, rather than generic conclusions with policy implications that are unrelated to the rest of the dissertation. The conclusions put forward must be justified and supported by the analysis performed. This does not mean that broader statements are not admissible. However, the student must make it very clear to what extent the dissertation can shed light on a certain topic, which of the conclusions and policy implications can be justified by the work presented in the dissertation and which conclusions and policy implications are, by contrast, more tentative and less grounded on the dissertation.
The dissertation structure described above is one that a reader would expect to find. Nevertheless, it is important to note that there is no prescribed formula for the form which a good dissertation should take. For example, depending on the specific topic of a dissertation, there may be other chapters, such as a chapter on certain policies adopted in a specific country. Also, the form of the empirical chapter will depend both on the nature of the hypothesis that one intends to test and on the interests/strengths of the student.
Dissertation planning: time schedule
a) Choosing the dissertation topic
The section above on “The nature of the dissertation in CEFS” has clearly stated and explained that the dissertation topic must consist in a precise and focused research question which must be addressed through the analysis and manipulation of economic/statistical data.
It is also suggested that the choice of the topic, which could arise from material or issues studied during your courses, would be helped by the consultation of published literature, especially articles in specialised journals. Thus, the starting point for any search for a suitable topic can only be a search of the literature (not only through Google but, more importantly, through bibliographic search engines such as EconLit or through Repec). You should aim to find papers which analyse a topic you are very interested in. This topic must be amenable to analysis through the use of economic/statistical data.
Once you have found a topic you are interested in and relevant literature on this topic you are not at the end of your search for a suitable research topic yet. First, you must determine how your dissertation would not simply replicate published papers. Your dissertation is intended to represent an original contribution to the literature. There are a number of possible modifications which might lend originality to your dissertation. For example, while the published papers you found might include cross-country analyses of the effects of a certain policy/institution, your dissertation may focus on the effects of that policy/institution in a specific country, or vice versa. Or the published papers might concentrate on one country, while you intend to focus on another. As another example, the published papers may compare some countries along a certain dimension, while you are interested in comparing those countries along a different dimension. For example, while the published studies may investigate whether countries that have liberalised their trade regime have achieved faster economic growth, your dissertation may analyse whether those countries have seen an expansion (or a contraction) of the industrial sector.
In the process of thinking about your original modification of the published literature, you must assess the feasibility of the research, particularly with respect to two aspects, namely, the feasibility of the statistical analysis and the availability of data. Let us start with the feasibility of the statistical analysis. Nowadays, published papers tend to make use of very sophisticated techniques of analysis which may be too demanding for your capacity. You have to ask yourself how you can simplify the statistical approach in a way that still allows you to sensibly address your research question. Remember that you are not required to undertake advanced econometric analysis (though you may do so if you have the capacity and wish to do it): you are only required to show that you have some numeracy skills and can employ appropriate statistical techniques of analysis. Ask yourself this question: if I were asked to find an answer for this question, what kind of empirical evidence could I bring to give a sensible answer? If you feel that the required statistical technique is too demanding and a simpler approach would not be meaningful, your research topic is not feasible and you should think of a different one.
Feasibility should also be assessed with respect to the availability of data. Do you have access to the data that are required? Through the library website, you can have access to a very extensive dataset, namely, the World Development Indicators. Many other datasets can be accessed after simple registration (ESDS international is particularly useful, as it gives access to a large number of datasets in addition to the World Development Indicators). Datasets are also available through the World Bank website. Many datasets are freely available through the internet. It is imperative that, already at the stage when you are trying to determine your research topic, you ascertain that you have access to the data you require. Never assume that you will find the required data but always check that you have access and that the data are appropriate for your purpose.
Please note that, if you wish to consult previous years’ dissertations, you should be very careful in how you evaluate this information: first, you should not look too far back in the past, since over the years the dissertation requirements have changed and become progressively more demanding (do not look for dissertations that were written more than one or two academic years ago); second, you do not know the grade received by the dissertation you are consulting.
b) Dissertation proposal
Having determined your dissertation topic, your next step is to write your dissertation proposal. The section above on “The nature of the dissertation in CEFS” has clearly explained the typical structure that a dissertation in CEFS is expected to have. Therefore, your dissertation proposal must contain the title, an explanation of the background and motivation for the dissertation topic, an outline of the structure, a detailed summary of the research methodology and approach (NOT the textbook description of the statistical technique), a description of the data which will be employed, and an indication of the main bibliographical sources. How to write a good dissertation proposal will be discussed in the penultimate session of the Research Methods and Dissertation Training course.
Before sending your proposal, ensure that it conforms to the format described in the previous paragraph (check in particular that the topic is well focused and motivated and that the empirical approach is well determined, feasible and appropriate to the task).
The date by which dissertation proposals should be submitted will be updated in due
course. Late submission of the dissertation proposal without good reason is not acceptable
and supervisors may not provide supervision in the absence of a proposal. Please note that students will not normally be allowed to change their dissertation topic after the proposal has been submitted, though – with the supervisor’s agreement – a minor change of focus necessitated during the process of research is admissible. A change of topic will only be allowed under special circumstances after discussion with the CEFS director and the assigned supervisor. Supervisors will be allocated following the Summer exam diet and students should contact their supervisor to arrange their first meeting to take place at in Septemeber (after finishing the final exams).
c) Typical dissertation writing timetable
Work on the dissertation should be carried out throughout the year, including vacation periods.
Start thinking about the dissertation; begin research in view of choosing a dissertation topic (i.e. reading of the literature).
Extensive reading of the literature; check data availability; work on the dissertation proposal.
Submit your dissertation proposal (28th March)
Start the literature review by gathering and reading papers relevant to your topic; start to collect data.
Full-time work on your dissertation: 4/5 meetings with your Supervisor
3 Drafts to be Submitted .
Submit the revised and final version of the dissertation. (AUGUST 18TH, 12:00 PM NOON, BST)
Students are required to work full-time on their dissertation at the University of
Glasgow Part-time jobs or internships are in principle not allowed.
• As part of the overall assessment process, students may be asked to present their dissertation in a VIVA (i.e. an oral presentation of your dissertation)
If the nature of the dissertation makes it imperative to perform research abroad or outside Glasgow (such as collection of primary data), students are required to obtain the approval of both the research supervisor and the programme director.
Format of dissertation and submission
Dissertations must be presented in printed form, using a word processing package. The length is 12000- 15000 words and a word count must be included. The word limit is for the main text and does not include references, tables and graphs. Dissertations that do not conform to these limits will be penalised by the reduction of secondary bands at the discretion of the examiners.
. The dissertation deadline will be a strict deadline and late submissions without good
cause will be penalised by the reduction of secondary bands, as described in detail in the Programme Handbook.
Extensions will only be granted in exceptional circumstances with the supervisor's support. Any request for an extension to the deadline for submission of the&
2023-03-23