Quantitative Dissertations

The Quantitative Dissertations part of Lærd Dissertation helps guide you through the process of doing a quantitative dissertation. When we use the word quantitative to describe quantitative dissertations, we do not simply mean that the dissertation will draw on quantitative research methods or statistical analysis techniques. Quantitative research takes a particular approach to theory, answering research questions and/or hypotheses, setting up a research strategy, making conclusions from results, and so forth. It is also a type of dissertation that is commonly used by undergraduates, master's and doctoral students across degrees, whether traditional science-based subjects, or in the social sciences, psychology, education and business studies, amongst others.

This introduction to the Quantitative Dissertations part of Lærd Dissertation has two goals: (a) to provide you with a sense of the broad characteristics of quantitative research, if you do not know about these characteristics already; and (b) to introduce you to the three main types (routes) of quantitative dissertation that we help you with in Lærd Dissertation: replication-based dissertations; data-driven dissertations; and theory-driven dissertations. When you have chosen which route you want to follow, we send you off to the relevant parts of Lærd Dissertation where you can find out more.

Characteristics of quantitative dissertations

If you have already read our article that briefly compares qualitative, quantitative and mixed methods dissertations [here], you may want to skip this section now. If not, we can say that quantitative dissertations have a number of core characteristics:

If you choose to take on a quantitative dissertation, you will learn more about these characteristics, not only in the Fundamentals section of Lærd Dissertation, but throughout the articles we have written to help guide you through the choices you need to make when doing a quantitative dissertation. For now, we recommend that you read the next section, Types of quantitative dissertation, which will help you choose the type of dissertation you may want to follow.

Types of quantitative dissertation

Replication, Data or Theory

When taking on a quantitative dissertation, there are many different routes that you can follow. We focus on three major routes that cover a good proportion of the types of quantitative dissertation that are carried out. We call them Route #1: Replication-based dissertations, Route #2: Data-driven dissertations and Route #3: Theory-driven dissertations. Each of these three routes reflects a very different type of quantitative dissertation that you can take on. In the sections that follow, we describe the main characteristics of these three routes. Rather than being exhaustive, the main goal is to highlight what these types of quantitative research are and what they involve. Whilst you read through each section, try and think about your own dissertation, and whether you think that one of these types of dissertation might be right for you.

Route #1: Replication-based dissertations

Most quantitative dissertations at the undergraduate, master's or doctoral level involve some form of replication, whether they are duplicating existing research, making generalisations from it, or extending the research in some way.

In most cases, replication is associated with duplication. In other words, you take a piece of published research and repeat it, typically in an identical way to see if the results that you obtain are the same as the original authors. In some cases, you don't even redo the previous study, but simply request the original data that was collected, and reanalyse it to check that the original authors were accurate in their analysis techniques. However, duplication is a very narrow view of replication, and is partly what has led some journal editors to shy away from accepting replication studies into their journals. The reality is that most research, whether completed by academics or dissertation students at the undergraduate, master's or doctoral level involves either generalisation or extension. This may simply be replicating a piece of research to determine whether the findings are generalizable within a different population or setting/context, or across treatment conditions; terms we explain in more detail later in our main article on replication-based dissertations [here]. Alternately, replication can involve extending existing research to take into account new research designs, methods and measurement procedures, and analysis techniques. As a result, we call these different types of replication study: Route A: Duplication, Route B: Generalisation and Route C: Extension.

In reality, it doesn't matter what you call them. We simply give them these names because (a) they reflect three different routes that you can follow when doing a replication-based dissertation (i.e., Route A: Duplication, Route B: Generalisation and Route C: Extension), and (b) the things you need to think about when doing your dissertation differ somewhat depending on which of these routes you choose to follow.

At this point, the Lærd Dissertation site focuses on helping guide you through Route #1: Replication-based dissertations. When taking on a Route #1: Replication-based dissertation, we guide you through these three possible routes: Route A: Duplication; Route B: Generalisation; and Route C: Extension. Each of these routes has different goals, requires different steps to be taken, and will be written up in its own way. To learn whether a Route #1: Replication-based dissertation is right for you, and if so, which of these routes you want to follow, start with our introductory guide: Route #1: Getting started.

Route #2: Data-driven dissertations

Sometimes the goal of quantitative research is not to build on or test theory, but to uncover the antecedents (i.e., the drivers or causes) of what are known as stylized facts (also known referred to as empirical regularities or empirical patterns). Whilst you may not have heard the term before, a stylized fact is simply a fact that is surprising, undocumented, forms a pattern rather than being one-off, and has an important outcome variable, amongst other characteristics. A classic stylized fact was the discovery of the many maladies (i.e., diseases or aliments) that resulted from smoking (e.g., cancers, cardiovascular diseases, etc.). Such a discovery, made during the 1930s, was surprising when you consider that smoking was being promoted by some doctors as having positive health benefits, as well as the fact that smoking was viewed as being stylish at the time (Hambrick, 2007). The challenge of discovering a potential stylized fact, as well as collecting suitable data to test that such a stylized fact exists, makes data-driven dissertations a worthy type of quantitative dissertation to pursue.

Sometimes, the focus of data-driven dissertations is entirely on discovering whether the stylized fact exists (e.g., Do domestic firms receive smaller fines for wrongdoings compared with foreign firms?), and if so, uncovering the antecedents of the stylized fact (e.g., if it was found that domestic firms did receive smaller fines compared with foreign firms for wrongdoings, what was the relationship between the fines received and other factors you measured; e.g., factors such as industry type, firm size, financial performance, etc.?). These data-driven dissertations tend to be empirically-focused, and are often in fields where there is little theory to help ground or justify the research, but also where uncovering the stylized fact and its antecedents makes a significant contribution all by itself. On other occasions, the focus starts with discovering the stylized fact, as well as uncovering its antecedents (e.g., the reasons why the most popular brand of a soft drink is consistently ranked the worst in terms of flavour in a blind taste test). However, the goal is to go one step further and theoretically justify your findings. This can often be achieved when the field you are interested in is more theoretically developed (e.g., theories of decision-making, consumer behaviour, brand exposure, and so on, which may help to explain why the most popular brand of a soft drink is consistently ranked the worst in terms of flavour in a blind taste test). We call these different types of data-driven dissertation: Route A: Empirically-focused and Route B: Theoretically-justified.

In the part of Lærd Dissertation that deals exclusively with Route #2: Data-driven dissertations, which we will be launching shortly, we introduce you to these two routes (i.e., Route A: Empirically-focused and Route B: Theoretically-justified), before helping you choose between them. Once you have selected the route you plan to follow, we use extensive, step-by-step guides to help you carry out, and subsequently write up your chosen route. If you would like to be notified when this part of Lærd Dissertation becomes available, please leave feedback.

Route #3: Theory-driven dissertations

We have all come across theories during our studies. Well-known theories include social capital theory (Social Sciences), motivation theory (Psychology), agency theory (Business Studies), evolutionary theory (Biology), quantum theory (Physics), adaptation theory (Sports Science), and so forth. Irrespective of what we call these theories, and from which subjects they come, all dissertations involves theory to some extent. However, what makes theory-driven dissertations different from other types of quantitative dissertation (i.e., Route #1: Replication-based dissertations and Route #2: Data-driven dissertations) is that they place most importance on the theoretical contribution that you make.

By theoretical contribution, we mean that theory-driven dissertations aim to add to the literature through their originality and focus on testing, combining or building theory. We emphasize the words testing, combining and building because these reflect three routes that you can adopt when carrying out a theory-driven dissertation: Route A: Testing, Route B: Combining or Route C: Building. In reality, it doesn't matter what we call these three different routes. They are just there to help guide you through the dissertation process. The important point is that we can do different things with theory, which is reflected in the different routes that you can follow.

Sometimes we test theories (i.e., Route A: Testing). For example, a researcher may have proposed a new theory in a journal article, but not yet tested it in the field by collecting and analysing data to see if the theory makes sense. Sometimes we want to combine two or more well-established theories (i.e., Route B: Combining). This can provide a new insight into a problem or issue that we think it is important, but remains unexplained by existing theory. In such cases, the use of well-established theories helps when testing these theoretical combinations. On other occasions, we want to go a step further and build new theory from the ground up (i.e., Route C: Building). Whilst there are many similarities between Route B: Combining and Route C: Building, the building of new theory goes further because even if the theories you are building on are well-established, you are likely to have to create new constructs and measurement procedures in order to test these theories.

In the part of Lærd Dissertation that deals exclusively with Route #3: Theory-driven dissertations, which we will be launching shortly, we introduce you to these three routes (i.e., Route A: Testing, Route B: Combining and Route C: Building), before helping you choose between them. Once you have selected the route you plan to follow, we use extensive, step-by-step guides to help you carry out, and subsequently write up your chosen route. If you would like to be notified when this part of Lærd Dissertation becomes available, please leave feedback.

Choosing between routes

Which route should I choose?

A majority of students at the undergraduate, master's, and even doctoral level will take on a Route #1: Replication-based dissertation. At this point, it is also the only route that we cover in depth [NOTE: We will be launching Route #2: Data-driven dissertations and Route #3: Theory-driven dissertations at a later date]. To learn whether a Route #1: Replication-based dissertation is right for you, and if so, how to proceed, start with our introductory guide: Route #1: Getting started. If there is anything you find unclear about what you have just read, please leave feedback.

References

Hambrick, D. C. (2007). The field of management's devotion to theory: Too much of a good thing? Academy of Management Journal, 50(6), 1346-1352.