Luc Hens (Last updated on 20 November 2024)
For a master’s thesis I expect you to write a polished research paper of 12,000-15,000 words (excluding the list of references). The research should draw on up-to-date academic work published in refereed scholarly journals: a master’s thesis typically refers to 10 to 15 articles from peer-reviewed scholarly journals. For a bachelor’s thesis or a paper for the “Onderzoeksmethoden / Research Methods and Skills” courses at the VUB I expect a polished research paper of 5,000-6,000 words (excluding the list of references), typically referring to 5 to 8 articles from peer-reviewed scholarly journals. To have a better idea of the skills you should demonstrate in a research paper, read Hansen, W. L. (2001). Expected Proficiencies for Undergraduate Economics Majors, Journal of Economic Education, 32(3), 231-242. Available via the JSTOR database: go the VUB library web site, follow the link “databases,”” then go to “J”.
Check with your thesis promotor: different promotors have different expectations. I expect to see you at least two times per semester, starting end-September (if you plan to submit your paper or thesis at the end of the Spring semester) and not more than four times per semester. I hold office hours during teaching weeks, usually on Friday afternoons from 2 to 4:45 pm. Make an appointment here. Come prepared. Bring your complete research file: hard copies of your research proposal, research log, reading notes (one set for each source), annotated bibliography, and current and previous drafts. During holidays and exam and study periods I am usually not available. I usually take off during summer from begin-July to begin-August, so if you submit your thesis in the summer session (with as deadline usually begin-August) you should submit a draft to me no later than begin-June. Meetings are in person or via Microsoft Teams. Microsoft Teams is part of the Microsoft 365 subscription the university offers you.
Submit written work as a word processor file (.docx) (or, only if you use LaTeX, as a pdf); I use the Comments and Reviewing functions in Microsoft Word to comment on your work and suggest editorial changes. Name the file as yourlastname_yymmdd.docx (e.g. if your name is Jones and you submit a draft dated March 22, 2024, use: jones_20240322.docx). I only give feedback on work that is in APA Style. Always include a standard APA Style title page with the date of the draft and a properly formatted list of references. The turnaround time typically is two weeks but tends to be longer near due dates, for instance in April and May when I get many drafts. If you submit a draft in the two weeks before a due date you probably won’t get feedback in time. I don’t give feedback during my summer break (usually from begin- or mid-July to begin- or mid-August).
APA Style is a set of formatting and editorial rules for scientific papers, developed by the American Psycholigical Association (APA). APA Style is used by most academic journals in the social sciences (including economics). Use APA Style for all written work assigned by me, for term papers, and for your master’s thesis. The *Little Seagull Handbook** (Bullock, Brody, and Weinberg, 2014, pp. 109-157) explains the basics and contains an example paper. You can find the Style and Grammar Guidelines and sample papers on-line:
One ECTS credit represents between 25 and 30 hours of work, so the workload for an 18-credit master’s thesis is 450 to 540 hours. Spread over 30 weeks (one academic year) that is 15 to 18 hours of work per week.
If you want to write a research paper on an economics subject, you’ll need a basic reference library. Here’s a list of reference books that you need during your research. Make sure you have a copy of these books at hand when you’re working on the paper.
Essential:
site:https://apastyle.apa.org your search termRecommended:
Other useful resources:
Dowload, print, and read the following:
You’ll need (access to) a computer with an internet connection. A laptop computer is recommended. For most term papers and master’s theses, a word processor will do (e.g., the open-source LibreOffice Writer, Pages (macOS), or Microsoft Word). Microsoft Word is part of the Microsoft 365 package that you get as a student of the VUB (link > e-tools). To take notes, better use a simple text editor such as TextEdit (macOS > Applications) or Notepad (Windows), and save your files as plain text files (.txt). Format your notes with Markdown (Markdown cheat sheet)). A reference manager such as Mendeley (and the Mendeley Cite add-in for Word) may be useful but is not essential. For graphing and data analysis, I recommend to use Python in Jupyter Notebook. See “How do I perform statistical analysis?” below.
If you write a doctoral dissertation or if your paper requires a lot of mathematical typesetting, I recommend that you use LaTeX instead of a word processor. See: “What is LaTeX?”.
A good on-line bookstore is amazon.de. You can change the site language to English. As for bricks-and-mortar bookstores, the Brussels branch of Waterstones at Adolphe Maxlaan / Boulevard Adolphe Max 71-75, 1000 Brussels is well stocked, staff is competent and friendly, and they can back-order books (link).
In very broad terms, you have to demonstrate that you master the following skills:
For more on these skills, read Hansen, W. L. (2001). Expected proficiencies for undergraduate economics majors, Journal of Economic Education, 32(3), 231-242 (available via the JSTOR database: VUB library > Databases > A-Z: J). A good research paper usually identifies a problem, explains why the problem is interesting, reviews the existing literature, formulates an interesting hypothesis related to the problem (or a set of hypotheses), and tests the hypothesis. A test of a hypothesis can take different forms; empirical papers may use statistical tests (using published data or data gathered by the researcher, for instance through a questionnaire), while theoretical papers may construct a model and prove theorems.
If you have no idea what to write about, here’s one approach. Start listing th course(s you particularly liked. Pick one (for instance: International Trade). Now you have defined the field of your paper. Take the textbook of the course. Which chapter or topic did you like most? (for instance: trade integration and free trade areas). Trade integration is too broad as a topic, so you’ll have to narrow the topic further down. Read the chapter on trade integration from your textbook. Make a short list of the specific problems that were addressed in the chapter. For instance: is it better that many countries simultaneously agree to remove obstacles to trade (multilateral trade integration), or that two countries agree to do so (bilateral trade integration)? That’s a useful starting point to narrow down the topic. What were the main points about this issue made in the textbook? Talk to the teacher of the course. Check whether the textbook chapter has a list of recommended readings at the end. Also check the textbook web site: often there will be internet exercises using actual data. These exercises can suggest interesting angles to the problem, that you might expand in your paper.
A second approach is to read a good newspaper (I recommend the The Economist or the Financial Times) for a couple of weeks and collect the articles you found particularly interesting in a file. Make a list of topics covered in the articles, and try to link them to subjects covered in courses you took. Then narrow down the topic as described above.
A third approach is that of the “natural economist”: look at the world around you and wonder why things are the way they are, or look for interesting problems in everyday life. Why is food more expensive inside a theme park? Why do some hairdressers reserve one day a week for customers without appointment? Are artists hurt when I illegally download their music? Think about the economics of the problem: relate it to concepts your studied in your economics courses, or try to develop a theory.
One you have decided what will be the the broad topic, you’ll have to narrow down the topic to a set of rather specific questions that you can handle in an undergraduate paper. At this stage you should be able to write up a first broad thesis statement. A thesis statement (or a hypothesis) is a testable statement, or question that is sufficiently specific so that you can verify whether the evidence supports it or not. When you start your research, you don’t know whether the thesis statement is true or false. Although you may have an intuition of the right answer, it’s your job to do your research with an open mind; you should be willing to revise your preconceived ideas in light of the evidence. You should with equal fervor look for evidence that supports the thesis as for evidence that does not support the thesis. You might first come up with a statement like “some scholars argue that bilateral trade negotiations are good.” This is too vague—you should be more specific. Specify what you mean by good. Better than something else? Better than what? And better in what sense? After a while (and discussion with your teachers), you may end up with: “bilateral negotiations are a better way to liberalize trade, because multilateral negotiations usually take very long and liberalize trade in very small steps, compared to bilateral negotiations”. You can use this as your first broad working hypothesis. When you’ve read more about the subject, you’ll be able to refine the working hypothesis. Maybe you’ll be able to formulate a set of related hypotheses. But first you should read more: if the problem is relevant, probably other scholars have thought about it, too. They may have discovered pitfalls, interesting angles, data, techniques –all of which will help you to tackle the problem by yourself. Try, right from the beginning, to summarize your main point in a one paragraph “research thesis.” The shorter the better—and difficult indeed!
Read The Little Seagull Handbook (2/e, 2014): “Proposal abstracts” (p. 72) and “R1: Doing Research” (pp. 80–86), especially “Posing a research question” and “Drafting a tentative thesis”. This is what you should minimally do in a short (300-word) research proposal:
If you are proposing a master’s thesis topic, for me a 300-word proposal with the elements above will do. Like all written work, the research proposal is in APA Style, with a proper APA Style title page and APA Style references (* American Psychological Association. (2021). APA Style: Style and Grammar Guidelines. (link: Instructional Aids > Sample Papers). A more extended proposal typically reviews five (rather than two or three) scholarly sources. In such a more extended proposal you can also:
First read The Little Seagull Handbook, chapter W-12: Annotated Bibliographies. If you used to first method to find a broad topic for your paper, you already have the start of a working bibliography: the textbook chapter, the recommended readings, the textbook web site. As you expand your bibliographical research and look for materials using the catalogs, your working bibliography will get larger. Add annotations for each source. Annotations may include: the date when you consulted the work, the call number (that is, the library code) of the work and the library where you found it, the location (on your hard disk, in your files) where you keep a copy of the work, a short personal appreciation, the subject descriptors and any other remarks you may find useful. A typical entry for an annotated bibliography may look like this:
Anderson, K. (2004). Subsidies and Trade Barriers. In: B. Lomborg (ed.). *Global
Crises, Global Solutions.** Cambridge University Press, pp. 541-577.
[VUB call number: 330.191.6 G LOMB 2004; photocopy filed in "Trade and Growth" folder;
"(...) argues that putting effort into phasing out wasteful subsidies and trade
barriers should be ranked highly among then opportunities being addressed in this
project [the Copenhagen consensus project discussed in this edited volume] (...)"
(p. 541). Table 10.4 p. 568 summarizes costs and benefits. Uses GTAP model,
discussed in appendix (p. 570). Check the (extensive) list of references later.]
Once you start reading the article, copy the reference and the annotation to a new word processor or text document. Give the document a title such as Anderson-2004-notes.docx. Expand the annotations (summarize the main points, collect verbatim quotes that you may use in the paper, indicate links with other sources, …). Usually not all items in the working bibliography the list will appear as references in the final paper: sources you don’t cite should not be in the references list (“References cited in text must appear in the reference list; conversely, each entry in the reference list must be cited in the text (…)”, American Psychological Association, 2001, p. 215).Include the list of references with each draft you submit.
You are required to do an extensive search of the EconLit catalog using EconLit subject descriptors. The EconLit subject descriptors are here. The EconLit database is accessible via the VUB library web site (databases > A-Z > E). List the subject descriptors you used in your research log. Store the results of your search in a file. The VUB library web site also gives access to the Social Sciences Citation Index via Web of Science (VUB library web site databases > A-Z > W > Web of Science). You will be able to find most economics-related materials in the VUB university library. You can use the inter-library loan services of the VUB university library to get works from libraries in Belgium and the rest of the world. The ULB university library is a 20-minute walk from the VUB campus.
A research log is a diary where you keep track of the progress in your research, report problems, and write down ideas. Your research log allows me to give recommendations to change your research strategy. Each entry in the research log is dated. List the database you searched, the descriptors you used, and the file where you saved your search results. Here’s an example of an entry
Friday 4 August 2006
- Checked https://www.aeaweb.org/econlit/jelCodes.php for the JEL subject descriptors
related to my subject. Result: F130 - Commercial Policy; Protection; Promotion;
Trade Negotiations; International Trade Organizations.
- Searched EconLit database (via VUB Library web site) using DE= F130. Yielded 14517
works. Narrowed search down by searching for
DE=F130 AND bilateral (anywhere) AND multilateral (anywhere)
Yielded 104 works. Saved list in file econlit20060804.doc. Will explore references
later.
- Checked VUB library catalog (VUBIS) for Lomborg (Ed.) (2004) (which contains
Anderson (2004)). VUB call number is : 330.191.6 G LOMB 2004. Went to VUB library,
located Lomborg (2004) and photocopied Anderson (2004). Photocopy filed in "Trade
and Growth" folder. Carefully check the list of references in Anderson (2004) later.
EconLit is the main bibliographical database for research in economics. In order to get acquainted with EconLit, go to here and read the information. Follow the links at the bottom of the page and read. Browse through the List of Journals Indexed in EconLit to get a feeling for the type of journals covered by the database. Browse through the Subject Descriptors. Carefully read the section on Search Hints (you may want to print this page for future reference).
You can access the EconLit database through the site of the VUB library web site databases > A-Z > E > EconLit). Click the “Information” icon next to EconLit and read the fact sheet describing the database. Try out a search (see: How do I search a bibliographical database?)
The bibliographical database EconLit and the JEL Classification System use a detailed system of subject descriptors covering all fields of economics. Most other economics databases (such as RepEc) also use the JEL subject descriptors. Each JEL subject descriptor corresponds to a subject code, starting by a letter (A through Z) describing the broad field, followed by a two- to three-digit number describing the sub-field. For example, F stands for International Economics, and F140 stands for Country and Industry Studies of Trade. Usually, authors attribute one to three subject descriptors to their paper; the subject codes (only the codes, not the full descriptors) are listed within brackets at the end of the abstract (don’t start a new line), e.g.: (JEL F140, F230) (check an article in a recent issue of the American Economic Review in the library for an example). This allows other researchers to retrieve the paper by doing a search in a database (such as EconLit or RepEc) using the subject code as the search key. You can find the list of subject descriptors here: JEL Classification Codes Guide. I require students to list two or three JEL subject codes for their master’s thesis (below the abstract).
RepEc (Research Papers in Economics) is a freely accessible catalog of working papers and journal articles in economics (link). Google Scholar is a useful tool, too (link). Business Source Premier is a bibliographical database for business research (VUB library web site databases > A-Z > B > Business Source Premier; click the “More information” icon next to Business Source Premier and read the fact sheet describing the database. A comprehensive annotated bibliography of economics-related internet resources is: Goffe, B. (Ed.). RFE: Resources for Economists on the Internet (link).
Access the EconLit database via the VUB library web site (see: How do I get access to EconLit?). Use the Advanced Search link (the Quick Search is a free text search –-usually a poor search strategy). Check what some of your economics professors have published between 1990 and today. Show citation and abstract. How many hits did you get? Save a couple of the citations and abstracts you found to a file. Try out a search using a JEL Subject Descriptor. Open a new window of your browser and go to JEL Classification Codes Guide. Find the Subject Descriptor for the subfield of economics that deals with Commercial Policy; Protection; Promotion; Trade Negotiations. Use this subject descriptor to find citations & abstracts published from 1995 onwards and listed in Econlit. You will notice that the list of hits is very long, so Return to Advanced Search to add delimiters. Refine the search by limiting it to citations that contain the keyword “dumping”: the delimeters now are (Subject Descriptor = …) AND (Keyword = dumping). How many hits do you get? You’ll notice that the list of citations is much shorter! A search by subject descriptor(s) is strongly preferred over a search by keyword. In EconLit, use the Advanced Search link (the Quick Search is a free text search –usually a poor search strategy). Use Boolean operators (such as AND and OR) to refine a search. For more on searches in EconLit, see the entry on “What is EconLit?” In some cases, a free text search can be useful. For a primer on using free text search engines (Google or other), consult the Google Search Help page.
Consult the following dictionaries & encyclopedias:
There is no strict rule on how many entries in the references list a good paper should have; published research papers in refereed journals have anything from 10 to over 30 scholarly references. For a master’s thesis, 10 to 20 scholarly references, of which 10 to 12 articles from peer-reviewed scholarly journals, is usually a reasonable number (check with your thesis advisor). Scholarly references include: research articles published in refereed journals or refereed collected works (often proceedings of scientific conferences, but also professional encyclopedias); working papers (that is, scholarly articles before they have been accepted for publication by refereed journals, and that are circulated via web sites and at conferences); and books with serious academic content, written by scientific researchers. If you are not sure whether a source counts as scholarly reference, check whether it is listed in EconLit; if it is, it counts as a scholarly reference. The reference list should of course contain other sources you used (articles from newspapers and magazines, documents from web sites, etc.), but they don’t count towards the total of 10 to 20 scholarly references.
As you are doing your bibliographical research, you will notice that many economics journals are often too technical for undergraduate students (students in a bachelor program). Econometrica is one example. In some cases, struggling through such difficult material may be worthwile (check with me or your professor). With some dedication and discipline you should be able to understand most of the empirical papers in the mainstream economic journals (that is, journals indexed by EconLit). Consult your professor to find out what level of reading he or she expects. Here are some periodicals that are accessible to undergraduates:
Documenting you sources is essential: it allows your readers to verify the sources for themselves, and it gives credit to whom credit is due. Not or insufficiently documenting your sources is considered as academic dishonesty and can have serious consequences. I require APA Style for focumenting sources and formatting papers, that is, Harvard-style parenthetical author-year citations (no footnotes). For details consult The Little Seagull Handbook (Bullock, Brody, and Weinberg, 2014, pp. 161-165) and the APA resources cited above.
Format your paper in APA Style. Use the sample student paper as a template. (See: What is APA Style?).
A scholarly article is an article reporting the scientific research done by a scientist (a scholar). When a scholar’s research has yielded results that are interesting enough to communicate to the outside world, he or she writes them up in a working paper. Working papers are often work in progress, and are revised several times (sometimes substantially). In seminars, at scientific conferences, and via the internet researchers present the results of their work to their colleagues (peers) in order to get feedback: suggestions for improvement, extensions, corrections. Once the working paper has survived this stage and has been rewritten to take the comments into account, the researcher submits the paper to a scholarly journal related to the topic of research (e.g., the Journal of International Economics). The editor of the scholarly journal (usually a top-notch academic in the field) sends copies of the papers to two or three specialists in the field (that is, peers of the author) who will review the article: they will read the article, carefully check it for mistakes, make suggestions for revisions, and judge whether it is fit for publication in the journal (rejected, revise and resubmit, accepted). This quality control is called peer review (or refereeing). (Peer review heet in het Nederlands collegiale toetsing (koppeling).)
You can use papers from the internet or working papers, provided that you document your sources (check APA Style on how to document internet sources and working papers). However, working papers are work in progress that hasn’t been refereed and accepted for publication in a scholarly journal (yet) (see: What is a refereed (or peer-reviewed) scholarly article?). As a result, working papers may be of lower quality than papers published in scholarly journals: they may contain errors or may be flawed otherwise. If you find a working paper that is interesting, check in EconLit whether a more recent version of the paper has been published in a scholarly journal. If that is the case, use the refereed & published version, not the working paper. Most of your references should be articles from refereed scholarly journals. Many refereed scholarly articles are available electronically via the VUB library web site (via V-link and Article Database ADB): such electronic versions are usually PDF-files of the articles as they appeared in the print version of the journal.
A call number is the code used to physically locate a publication on a shelve in a library. Take for example the following publication: Friedman, M. (1970). Price theory (7th ed.). Aldine.
Log in to the VUBIS on-line catalogue of the VUB library (link) and use the search window to look up the call number for the publication above. Verify that the call number is 338.5 G FRIE 70. In this call number:
While doing your research in a library catalog, add each reference (in APA Style, as shown above) to your working bibliography and add in square brackets the library and call number:
Check wether you can find Friedman (1970), but leave the book in place. Print copies of periodicals (with code “E” rather than “G”) are stacked in a separate section of the library, but according to the same principles. Rather than the first four characters of the author’s name, the call code has an abbreviation for the title. For example, the call number for American Economic Review is 330 E AERE. Locate the most recent issue of American Economic Review in the VUB library.
If you use one of the electronic databases accessible via the VUB library to locate a journal article, the record for the reference will usually display a V-Link icon. Click on the V-Link icon. If an electronic copy is available, this will be shown. Click on the link to download the pdf (portable document format) file. (In you use a bibliography manager, link the pdf on your hard disk to the reference in your .bib file). If no electronic copy is available, locate the journal in another Belgian scientific library using the Union Catalogue of Belgian Libraries (UniCat) or click the link “Make an inter-library loan (ILL) request” to request the journal article at another scientific library. The Union Catalogue of Belgian Libraries (UniCat) is available via the VUB library (link: databases > A-Z: U). See “How do I obtain books or articles that are not in the university library?” for more on inter-library loans.
The university library increasingly subscribes to the electronic versions of scholarly journals, so make sure whether an article you are looking for is available electronically. You do so by clicking the V-Link next to the citation in the university library catalogue. If not, order the book or article via the university library’s Interlibrary Loan Service. This is possible on-line or in the office of the Interlibrary Loan Service, located in the library. The Interlibrary Loan Service will try to find what you need by contacting participating libraries, which range from other university libraries (in Belgium and elsewhere) to the British Library in London. Place your order via de library web site [insert link] or contact the librarian in charge of the Interlibrary Loan Service for help (she’s in the office next to the statistics section of the library). Usually it takes a couple of weeks to get what you want. Articles are usually e-mailed (if a pdf version exists); books can only be consulted in the librarian’s office.
It is essential that you take notes while you are reading. In the past, researchers did this on index cards, but nowadays most researchers (including me) use a text editor or a bibliographic reference manager to maintain a bibliography and take notes. The reading notes are an essential tool when you write your paper.
If the paper has an abstract, copy it (verbatim and in quotation marks) to the text file for this source (or, if you use bibliography manager, in the abstract field). If the paper has no abstract, write one yourself after you finished reading the paper; clearly indicate that you wrote the abstract yourself. When you start reading, take notes as you go along—don’t wait until you finished reading the paper. Distinguish between verbatim citations (enclosed in quotation marks), paraphrases (not enclosed in quotation marks), and your personal comments (enclosed in square brackets). Here’s an example of reading notes: link. I usually reduce a 30-page paper to 3 to 5 pages of notes.
In a literature review you give a structured and critical overview of past research on your topic. Your annotated bibliography and your notes (one set of notes per source, plus thematical notes that tie together sources) are the raw materials for your literature review. A literature review is not a sequence of summaries (“X said this … Y said that …”); in a literature review you compare and contrast ideas, approaches and results (see here and “Comparing and contrasting” in The Little Seagull Handbook).
Because science should be open and reproducible, share your data and code with your promotor and other readers of your paper. You can do so easily if you create a GitHub account (link).
To perform statistical analysis I recommmend Python (in Jupyter Notebooks) or R (I use both but mostly Python). Menu-based statistical software such as SPSS is acceptable but I’m not a fan, among other things because running your analysis from a menu makes it hard to replicate your work, both for yourself and for others (such as me). Wolfram Alpha is useful to do quick statistical computations on-line. More on Python, R, and Wolfram Alpha below. Don’t use a spreadsheet program such as Microsoft Excel, LibreOffice Calc, or Numbers: spreadsheet programs are poor tools for graphing data and for statistical analysis. Burns (2005) gives some reasons why (Burns, P. (2005). Spreadsheet Addiction (link). For some examples, see Harford, T. (2023, February 11), Spreadsheet disasters: What happens when spreadsheets go wrong? More or Less: Behind the Stats (BBC Radio 4, podcast) (link). For a more comprehensive and technical discussion, see the papers by Yu (2008); Yalta (2008); and McCullough & Heiser in Computational Statistics and Data Analysis, 52(10)).
Python. Python is a powerful general-purpose programming language with great data analysis tools. If you have no preference between R and Python, I recommend that you use Python in Jupyter Notebook:
Notebook documents (or “notebooks”, all lower case) are documents produced by the Jupyter Notebook App, which contain both computer code (e.g. python) and rich text elements (paragraph, equations, figures, links, etc…). Notebook documents are both human-readable documents containing the analysis description and the results (figures, tables, etc..) as well as executable documents which can be run to perform data analysis. (link)
First download and install Python (link). Start Python and (from within Python) install Jupyter Notebook (link). Here’s a brief explanation on the basics of using notebooks: link (starting at section 2.4). (If you took the Business Information Systems / Bedrijfsinformatica course, you already used Spyder as a front end to Python. Spyder is fine but I encourage you give Jupyter Notebook a try.) I made a Jupyter Notebook that illustrates how to use Python to draw the diagrams and do the computations from an introductory applied statistics course. You can view the notebook in your browser here. If you use Spyder, copy the Python code from the grey code boxes in the notebook viewer. If you use Jupyter Notebook, download the notebook here.
Resources on using Python to do statistical analysis:
R. “R is (…) a freely available language and environment for statistical computing and graphics which provides a wide variety of statistical and graphical techniques.” R runs on Windows, macOS, and Linux. R can be downloaded for free on the R Project homepage. This document explains how to install and run R and RStudio on your computer (macOS or Windows): link. This short document shows how to use R to do the statistical computations from an introductory course in statistics: link. RStudio is a graphical user interface that makes using R easier. Resources on using R:
Wolfram Alpha. Wolfram Alpha (as a web page or a smartphone app) is useful to do quick statistical computations on-line. Here is a document that shows how to do the basic computations from an introductory applied statistics course.
SPSS. “SPSS Statistics is a set of programs used for interactive, or batched, statistical analysis.” VUB students can download SPSS on softweb.vub.be. A useful book is: Field, A. (2013). Discovering Statistics Using IBM SPSS Statistics. London etc.: Sage. Available in the VUB library, call number: 311.17 G FIEL 2018.
To report statistical results, follow the guidelines of APA Style: link.
You can use qualitative research methods (such as methods based on data from interviews) when such methods are appropriate and provided that you apply the methods with rigor. Your paper should contain a section giving a rationale for the method you selected. For a primer on the use of qualitative methods in the social sciences see: Brett Davies, M. (2007). Doing a Successful Research Project: Using Qualitative or Quantitative Methods. Palgrave Macmillan (ISBN: 978-1403993793) (VUB library call number: 371.1 G DAVI 2007). For a primer on using interviews, see Knott, E., Rao, A. H., Summers, K., and Teeger, C. (2022). Interviews in the social sciences. Nature Reviews Methods Primers, 2(732022) (link). VUB students can download the qualitative data analysis computer software package NVivo on softweb.vub.be.
Yes, if a survey is an appropriate method to gather data: “Surveys are an essential approach for eliciting otherwise invisible factors such as perceptions, knowledge and beliefs, attitudes, and reasoning.” (Stantcheva, 2022). You should also design and administer the survey well. But to get a sampling frame and a random sample is much harder than most students think. If your sample is not random, you should test whether the sample is representative of the population for characteristics that may matter. The samples I see in many master’s theses are not representative). And remember what Dr House says. If you want to find out how people behave and why they behave that way, it is better to look at what people do rather than at what they say they do (revealed preferences vs stated preferences). Listen to the section on surveys (from 18’28” to 20’19” here in Episode 495: “Why Are There So Many Bad Bosses?” of the Freakonomics podcast. Essential reading is: Stantcheva, S. (2023). How to run surveys: A guide to creating your own identifying variation and revealing the invisible. Annual Review of Economics, 15, pp> 205-234 (link). VUB students can download the Qualtrics software “for surveys, feedback and polls” on softweb.vub.be.
The conventions on the use of generative AI in scholarly work (including assignments, bachelor papers, and master’s theses) are evolving. Consult the university policy and the policy set out in the guidelines for the master’s thesis on the SharePoint page of the faculty of Social Sciences and Solvay Business School [add link]. In any case, cite the tool you used. This blog post explains how to do that: McAdoo, Timothy (2024, February 23). How to cite ChatGPT. APA Style Blog (link). I also expect you to follow the APA policy on the use of generative artificial intelligence (AI) in scholarly materials (link):
For this policy, AI refers to generative LLM AI tools and does not include grammar-checking software, citation software, or plagiarism detectors.
- When a generative artificial intelligence (AI) model is used in the drafting of a manuscript for an APA publication, the use of AI must be disclosed in the methods section and cited.
- AI cannot be named as an author on an APA scholarly publication.
- When AI is cited in an APA scholarly publication, the author must employ the software citation template, which includes specifying in the methods section how, when, and to what extent AI was used. Authors in APA publications are required to upload the full output of the AI as supplemental material.
Carefully read the section on abstracts in Bullock, R., Brody, M., and Weinberg, F. (2014). The Little Seagull Handbook (2nd edition). W.W. Norton & Company; or in American Psychological Association. (2001). Publication Manual of the American Psychological Association (5th ed.). American Psychological Association. Following the title page, the master’s thesis should have an abstract which is “a brief, comprehensive summary (…)” (American Psychological Association, 2001, p. 12). A good abstract is accurate, self-contained, concise and specific, nonevaluative, coherent and readable (see American Psychological Association, 2001, pp. 12-14 for more explanation). The reader should get the main points (problem, method, and results, including key numerical results) from reading the abstract. A common mistake is that the abstract reports the main questions but fails to provide the answers. An abstract is also no substitute for the introduction of the paper. “Abstracts should not exceed 120 words” (American Psychological Association, 2001, p. 12). Abstracts never refer to tables or figures. Abstracts usually don’t cite sources. Write the abstract after you have written the main text (after all, it’s a summary of the main text), but place the abstract before the body of the report, following the title. Many students erroneously write the abstract before having written the main text, and as a result their main text lacks an introductory section with the context and the data of the problem (they seem to think that the abstract also serves as the introduction; it doesn’t). Abstracts are double spaced, not indented, in a single paragraph. In business or consultancy reports abstracts are sometimes called executive summaries; in that case, the language is usually non-technical, and the abstract emphasizes the practical meaning of the results for the decision maker.
Collect all the relevant materials for your paper in a research file (create a folder/directory on your computer). Keep a paper copy of the research file, and bring the complete research file (on paper) when you visit your paper/thesis advisor. Typically, a research file contains the following items (I explain elsewhere what each item represents):
Make sure you keep backups of all your work (several, stored in different physical places). Buy an external hard disk to make back-ups and store a copy in the cloud using Microsoft OneDrive, Dropbox, iCloud, or a similar service.
To protect citizens’ privacy, the General Data Protection Regulation (GDPR) of the European Union requires anyone who processes personal data to follow certain principles and obey certain rules. The regulation also applies to students who process personal data such as responses to a survey or interviews. “Processing” includes: collecting personal data using surveys, interviews, focus groups and so on (but also collecting personal data from existing databases); consulting personal data from existing databases; and organizing, structuring, and storing personal data. Inform participants of the purpose of the survey or interview, and explain how you will code and process the data and which safeguards you will use. Ask the consent of each participant. To do so include an information and consent section and the beginning of the survey or interview key [templates for informing participants and get their consent will be available on the VUB web site]. Encrypt personal data using the free encryption program Veracrypt (link). Make a register of processing activities in which you describe who processed the data, which data were collected, how the data are coded, what the purpose is, where the data are stored, which safeguard measures are taken, and how long the data will be stored. Include the register of processing activities in the methodology section of your paper. Keep the data for one year after you got the grade for the course of master’s thesis; then destroy the data.
The VUB Internet Security and privacy web page gives more information on the General Data Protection Regulation (English version; Dutch version).
I grade the paper on the following criteria:
Consult the syllabus for the weigths of the criteria. Other thesis advisors or readers may use different criteria and different weights.
Check whether you have give an oral presentation of your paper and what the time constraint is (usually you get 10-15 minutes to present your research and 10-15 minutes to answer questions). Read appendix 18 “Guidelines for Giving an Oral Report” in Thomas L. Wyrick (1994), The Economist’s Handbook: A Research and Writing Guide, West Publishing Company; and Edward Tufte (2006). The Cognitive Style of PowerPoint, in: Beautiful Evidence, Graphic Press (link). Ignore most of what you have learned about PowerPoint presentations in your business courses— presenting research is not the same as giving a sales pitch and even a typical sales pitch with PowerPoint is usually lethally boring (“death by PowerPoint,” referring to a scene in the science-fiction novel The Jennifer Morgue by Charlie Stross involving a PowerPoint presentation that turns anyone who watches it into a murderous zombie). Tufte (2006) explains extensively what’s wrong with the typical slideware presentation: a typical presentation slide contains little information and often less information than what is needed to carefully develop an argument; slideware imposes a hierarchical, linear way of thinking that is not appropriate to explore many issues. If you understand Dutch, also watch this routine by a standup-comedian: link. In sum, I recommend not to use PowerPoint at all (nor similar presentation software like LibreOffice.org Impress or Apple’s Keynote)—Acrobat Reader and Preview have a full-screen view mode that does a fine job displaying text, tables, and graphs on a screen. Prepare a four-page handout for your audience, consisting of up to four A4 pages printed on A3-sized paper and folded once to obtain an A4-sized booklet. The handout is a mini-paper that contains your contact details, summarizes your key findings and includes the graphs and tables you discuss during your presentation. Here is an example produced using LaTeX: link. Use the format of a short paper, but omit the abstract (rules for written work). Include a list of references (including a reference to the the full paper). During the presentation, focus on the research hypothesis and the evidence. Don’t start by summing up what your going to say; you only have 15 minutes, so get to the point. In a thesis presentation, assume that your audience has read the paper in advance (if they haven’t, they are unprofessional and don’t deserve your time). Don’t oversell your point (again, a research presentation is not a sales pitch), but give a balanced assessment of the evidence. Stay strictly within the time allotted; the only way to time a presentation is to practice it at home with a test audience (room mates of family members are useful guinea pigs). Like a paper, a presentation has to be revised, revised, and revised again to obtain a satisfactory result. During the presentation face the audience, not the screen. Don’t read out what is shown on the screen (especially problematic with “bullet points”), nor read from a sheet. Use sequentially numbered 10 by 15 cm index cards with key words (not complete sentences). Leave you power suits in the closet: dress code is informal and you’ll have plenty of opportunities to dress up later.
The following criteria are relevant for evaluating an oral report:
For short student papers and most master’s theses a word processor is fine. For a master’s thesis with a lot of mathematical type or for doctoral dissertation I recommend that you use LaTeX (pronounced as “lah-tech”). LaTeX is a freely available program for typesetting research reports that runs under Windows, macOS, and Linux. It is widely used by scientists. LaTeX has several advantages over a word processor: the typesetting is beautiful, formatting (creating sections, subsections) is automatic, formulas can be correctly formatted, and referencing can be automated. You don’t have to worry about formatting details of APA Style (or any other style): use the default format of LaTeX with Harvard style author (year) citation. LaTeX uses a mark-up language that takes some effort to learn but is usually straightforward. For instance, to start a new section called “Review of the literature” type:
\section{Review of the literature}
Markups (\section) and text (Review of the literature) are in one plain text document. In sum, LaTeX is not “What You See Is What You Get” but “What You See Is What You Mean.” LaTeX editors such as TeXShop or TeXmaker are relatively easy to learn.
If you write in LaTeX, use a reference manager (such as (http://bibdesk.sourceforge.net/)BibDesk) for macOS or (http://jabref.sourceforge.net/)JabRef) for Windows) to maintain your bibliography. In LaTeX, the references list can be generated automatically from a BibDesk of JabRef bibliography file.
To run LaTeX on your computer, download and install the MacTex distribution (macOS) or the MikTeX distribution (Windows) form from the TeX users group page (link). The distributions contain a front-end (TeXShop for macOS, TeXMaker for Windows).
For more, see:
If you use LaTeX stick to the default LaTeX article format, but use parenthetical author-year citations by using the natbib package and the apalike bibliography style:
\documentclass[a4paper]{article}
\usepackage{natbib}
\begin{document}
\bibliographystyle{apalike}