Datalab 00: SDG indicators

We start with a short introduction to formulating a problem statement based on the given domain(s) of your SDG indicator(s). Subsequently, we will focus on putting our new knowledge into practice by forming a research question. Then, we will use both to create an introduction component for your data-lab project: the conference poster.

0) Learning Objectives

Enable you to:

  1. research your domain;
  2. write a problem statement;
  3. write a research question;
  4. write an introduction.

Today we'll learn how to create the following components of any data science project. If you exceed the estimated duration for a given component and feel like you are squeezed for time, feel free to ask us for help!

Table of contents:

  1. Mock-Assessment Review: 0.5 Hours
  2. Problem Statement: 2 hours
  3. Research Question: 1 hour
  4. Introduction: 2 hours
  5. Daily Reflection: From 16:00 to 17:00

Good luck!

0) Mock-Assessment review: Quantifying our World into Data

Now, let's all find a partner (as in a classmate, not a date) and get our mock-assessments out (assuming you made them) and discuss them. Specifically, analyse and evaluate one another work by:

  • Simply telling one another about what you identified and deconstructed as an object and phenomena and try to justify the decisions you made.
  • Listening to one another critically and giving feedback as your partner goes along explaining.

1) Forming a Problem Statement

We start our project by focusing on creating value with our data science solution; that's what data science is about, after all. That begins with learning about how to form a problem statement. Next, we are going to apply the techniques below to fill in our problem statement. Part of that is exploring the domain(s) we are working in because a problem cannot exist without its context.

Note that you need to log in to Linkedin LEARNING using your BUas account!

How to develop a problem statement from Academic Research Foundations: Quantitative by Rolin Moe

Now that we have acquired the basics of forming a problem statement, it's essential to use it methodologically. Listing the different ways in which you can approach

Create a file to make notes for yourself: a Word file, .txt file, a limestone tablet, or google documents file; whatever you prefer to write in.

Excercise 1: Write down the different ways you can get started writing your problem statement.

Hint: the answers are in the video above.

Excercise 2: Form a draft of your problem Statement

Step 1. Download and inspect the conference poster template. Save it to your GitHub repository and name it aptly: "SDGIndicatorConferencePoster_YourNameHere".

Step 2. Research the domain, the SDG Indicators. Before diving in the online documentation, I would recommend watching all videos below. Feel free to skip one of the first two introductions if you prefer reading; don't skip the technical one.

Watch the SDG Introduction

An easy introduction into the topic. If you cringe too much on TED talks, feel free to skip to the next video: it's your research process, after all.

SDG introduction to data scientists

Data scientiest on a conference in canada are introduced to SDG's.

After 12 minutes in he starts to talk about national collaboration and development examples which you can skip. It may be interesting later though as you make recommendations in your conference poster.

A more technical introduction

Specifically, watch until the 5th minute; the part about using it for a business is not vital but may inspire your problem statement.

Full documentation

Now that we have gotten a good impression of the SDG indicators and what they are for. I recommend you choose one of the 17 goals and start narrowing it down to an SDG indicator you want to investigate. You can view the complete documentation of the project here.

SDG Indicators by Goal (Source: Our World In Data)


Step 3. Choose your SDG Indicator (e.g. Extreme Poverty) to study.

If you need help choosing your topic, watch one of these video's to narrow:

Narrow your idea to a research topic from Academic Research Foundations: Quantitative by Rolin Moe

Note: You have to log in to Linkedin LEARNING using your BUas credentials!.

Or broaden your search:

Broaden your idea to a research topic from Academic Research Foundations: Quantitative by Rolin Moe

Step 4. Form your problem statement and write it down in your conference poster.

Watch this if you feel like you need more information on how to form a problem statement.

Problem statement from Apache Spark Essential Training: Big Data Engineering by Kumaran Ponnambalam

Note that we're not going to do extensive data engineering just yet :p, we are just conceptualising this block. However, later you may want to recommend a data processing pipeline and dashboard to monitor some real-world later in the project, depending on your problem statement.

2) Forming a Research question

Now, it's time for writing the research questions!

Step 1. Please watch the following videos to collect the knowledge you'll need to apply:

Elements of a research question from Academic Research Foundations: Quantitative by Rolin Moe

How to write a research question from Academic Research Foundations: Quantitative by Rolin Moe

Step 2. Form your research question and write it down in your conference posters. Note down the type of research question you're asking. Make sure it is concise, specific and simple!

3) Writing an introduction

Step 1. Watch the following video and note down which elements may be found in the introduction. You already define two fundamental elements: 1) the problem you're going to address and 2) you have your research question: your purpose is to find an answer to it.

Writing the introduction from Writing a Business Report by Judy Steiner-Williams

Step 2. Decide what information the reader needs to know and write the introduction in the poster template.

Step 3. Save your poster and notes to your Github repository.

4) In-Class discussion

At 16:00, we'll all get together in Datalab to discuss our progress and reflect on today activities.

Tomorrow, we're going to cover variables and data frames in R.

Questions or issues?

If you have any questions, please first ask your peers or send us a message on teams instead!

Further reading & other relevant information:

  1. A slow in-depth dive from a UK statistics institute: Royal Statistical Society: If you want more information, repetition or a deeper understanding of the SDG indicators; at multiple levels.