GW#: GW# International Break II – What can Fantasy Football show us about Research Approach?

This week, I thought it might be interesting to take a look at how our Fantasy Football game can help us understand some concepts related to Research Methodology and – specifically – two terms related to Research Approach which often confuse us.  

This post is, hopefully, useful for our final year students, but also for all students ahead of taking the Research Methods module in Yr2 Semester 2 before undertaking the final year dissertation.  

The terms we’ll explore here are deductive reasoning and inductive reasoning.  When we design a research project, we will usually lean towards either a deductive or inductive approach.  So.  what are these approaches?  How do they relate to research?  What’s the difference between these approaches?  And how can Fantasy Football help us understand these concepts in relation to our academic work.   

First of all, its vitally important to recognise that we use both deductive and inductive reasoning every day.  You already know how to do this, but are perhaps not aware (a) when you’re doing this or (b) the distinction between the two.

Transfers and Decision Making in Fantasy Football

Each week, we look at our team and we try to figure out the best way to adapt our team for the upcoming fixtures.  When we do this, we use reasoning to help us formulate arguments about the best strategy to adopt.  

Whenever we do research, we are really asking a single Major Research Question.  This Major RQ frames our project.  So, in this blog post, the question we’ll focus on is: 

Major Research Question:  “Which player should I transfer in?”

Deductive Reasoning

Deductive reasoning is a logical process that begins with a theory and then uses data to either prove or disprove that theory. 

To do this, deductive reasoning often begins with the formulation of a hypothesis (theory) – not an open-ended question.  That hypothesis is based on two premises: a major premise and a minor premise.

For example: 

Major Premise: Norwich have the best attack in the Premierleague

Minor Premise: Teemu Pukki is the best attacking player for Norwich

Hypothesis: The best player to transfer in is Teemu Pukki

We would then look at a range of statistical data from sources such as the Official FPL website/app, OPTA stats, or sites like Fantasy Football Scout or Fantasy Football Hub.  

Now, you’re probably thinking “why the hell would we even consider Teemu Pukki?  Norwich are terrible”.  Well, yes.  So in all likelihood, we would examine the data and the hypothesis would be rejected.  We would not transfer Pukki in, because his stats show the Major Premise to be false.  And this is how deductive reasoning works.  Only if both the Major and Minor premise are true will we be able to confirm the hypothesis.  

So let’s try a different example.  

Major Premise: Liverpool have the best attack in the Premierleague.

Minor Premise: Mo Salah is the best attacking player for Liverpool.

Hypothesis: The best player to transfer in is Mo Salah

This is the same deductive approach, but in this case we can see that both the Major and Minor premises are (arguably) true.  We can then look at the available data and either confirm or reject the hypothesis.  

So the key feature of deductive reasoning – and way to distinguish it from inductive reasoning – is that it starts with a theory (hypothesis) and then tests that theory (hypothesis) using empirical data, with the aim of confirming or rejecting the theory (hypothesis).  Research that uses a deductive approach is usually identifiable through the presence of hypotheses.  If you see a Hypothesis being tested in research, you can be pretty sure that research is using deductive reasoning.  

So a project using deductive reasoning to answer the Major RQ “Which player should I transfer in?” would likely set Hypotheses for all Premierleague players (or at least a list of 5-10 players we’re thinking about transferring in) and then analyze the data to confirm or reject those hypotheses. 

So what do you think?  Is this a good way to make a decision? Maybe it is.  And it might work. But, it would take a significant amount of time to do this every week, and would only really tell you about past performance – which isn’t always a reliable indicator of future performance.   So maybe we can think about a different approach…..

Inductive Reasoning 

The key distinction to keep in mind here is this:  the starting point.  Deductive reasoning begins with a theory and moves to confirm or reject that theory depending on what your data tells you.  

Inductive reasoning takes a different approach.  It starts with gathering the data, then moves to analyze the data, and then builds a theory.  

So we don’t generally see hypotheses in research that adopts an inductive approach.  What we see instead are open-ended questions.  All projects have a Major Research Question (Rq), but deductive approaches lead us to formulate and test hypotheses, while inductive approach leads us to ask more open-ended “minor” research questions. 

In this example, we might see several minor research questions being asked:  

• Who has the easiest fixtures this week?  

• Which player is playing against a team where they have a good attacking record?  

• Which attacking player is currently in-form?

• When we consider all of these, which player is the strongest candidate?  

The theory we’re trying to build here is a theory on which is the best player to transfer in.  The way we do this is we (i) gather data, (ii) identify questions to ask, (iii) see what the data says, (iv) develop a theory.  

When we see open-ended research questions, this is *usually* an indicator that the research we’re looking at uses inductive reasoning.  The research doesn’t aim to test theory, but generate theory through an exploration of data.  

So what do you think?  Is this a good way to make a decision? Again, maybe.  But it also has weaknesses – we might only consider analyzing information we think is relevant, and miss other more important factors. There are also so many variables that impact player performance – some measurable, some not – so 

Reasoning in Research Projects

The question of research approach can confuse us.  This is often because we think we need to “choose” a research approach before we are clear on what our Major Research Question is.  

In most cases, a well-designed Major Research Question will dictate – to a certain extent – what type of research you need to conduct, and whether that research will primarily utilise deductive or inductive reasoning.  I say “primarily” because, in many cases, we can and do use both.  Yet when it comes to research methodology, it is important to be able to explain which type of reasoning characterizes your project and your approach.  

A mistake we often see is that students try to decide their research approach before they’ve either (a) developed a well-designed research question, or (b) mapped out a basic design for their research project.  This is putting the cart before the horse and is best avoided.  

The “Process” of Writing a Dissertation

It’s also important here to highlight that, while dissertations are read in order – Intro, lit. review, methodology, findings & analysis, conclusion – they are rarely written in such a linear order.  We have to revise, refine, adapt and change consistently.  Writing a dissertation is about confronting a problem and trying to understand that problem.  Writing a dissertation isn’t like a race with a start and finish, its more like a fight with an opponent you know little about. You’re the aggressor, and you’ve got an opponent who has superb defence and counter.  Your role with methodology is like choosing how to breach their defences – boxing, kickboxing, throws, grappling, direct attack or evasively stalking your opponent - and reveal the nature of their strengths and weaknesses.  If you get your methodology wrong, you won’t get close to understanding the problem they present.  Its therefore vital that you know the difference in approaches to your opponent and how they work to answer the questions you have about them. 

We often see students insisting that they are using deductive reasoning, but they are actually asking an open-ended question and exploring statistical data to theorize about their topic: an inductive approach. For example, the student might ask “To what extent has the time we spend streaming online content on Netflix and Amazon Prime been impacted during the Covid lockdown?”.  The student looks at average viewing times and compares and contrasts this statistical information.  But they start with the data; they look at what the data says, and then develop a theory about how our viewing habits have changed.  This is inductive reasoning.  

Similarly, we sometimes see students testing theories using qualitative research, but insisting that they are using inductive reasoning.  An example of this might be a student comparing advertising posters for different smartphone brands.  They develop a hypothesis (theory): “Apple use more emotional marketing messages in their posters, while Samsung use more informational messaging”.  The student then tests their hypothesis by undertaking textual analysis of their dataset: Apple and Samsung’s advertising materials – with the aim of confirming or rejecting their hypothesis. This is not inductive reasoning.  The approach tests this theory through the systematic analysis of data.  This is deductive reasoning.  

The Key “Lesson”

Your dissertation is the crown jewel of your undergraduate studies.  All honours degrees require the completion of an independent research project we know as “the final year dissertation”.  This is what makes your degree and an “honours” degree.  

Research methodology isn’t some mystical, complicated dark art.  Its part art, part science – like craftwork or artisan work.  What it reveals about you is your ability to think about a problem, develop a strategy and design a project to improve our understanding of that problem.  

Each and every problem can be investigated, evaluated and analyzed from myriad perspectives.  Being able to articulate your “perspective” is important, as it helps you strengthen your argument and the contribution you make to understanding.  

And that should always be your goal: not to be “right” rather than “wrong”, but to be “good” rather than “bad”.  Not to be “correct” but to “make a contribution” to understanding. 

Understanding the reasoning and approach that characterizes your perspective – and showing this in your dissertation’s methodology section – is, therefore, pretty important.