Data Visualisation and Data-driven Decision Making (Spring 2023)

Abstract

The course is focused on forming the basis of a data communicator. The students should recognize the various features of the human perception system, and operate around those limitations. We expect a student to be able to create effective visual encodings (charts), data-driven stories, dashboards and storyboards. The course will empower students with the underlying assumptions, guidelines and trappings of visualizing quantitative information. In doing so, the curriculum will put the field of data visualization in context; touching on its history from the first historic examples, right up until the propensity today of misleading people.

Description

The clear communication of analytic results is the primary end goal of a scientific endeavor. The origin of a data science project is the question “What do we want to communicate and how?” The “how” heavily depends on data visualization, a core activity needed from the very early exploratory phases to the eventual explanatory phases. Moreover, creating visualizations is not simply an activity to wrap up a project at the end: it is rather an objective that informs all phases of data analysis, including the early stages of interrogating datasets to better understand them. 

One cannot rely on the unprincipled application of graphical capabilities (and defaults of standard software) such as bar/line/pie charts, because each visual choice has repercussions in perception, interpretation, and internalization of information. In addition, the accumulation of smaller design decisions can result in a wide range of effectiveness. In light of these challenges, this course aims to substantially equip Data Science students with methods, tools and tactics for critical reasoning along their future paths of working with quantitative data.

The following subjects will be covered in the course:

1. Introduction– Why We Visualize and a History of Information Graphics
2. Visual Vocabulary: A Tour through the Dataviz Zoo
3. Human Perception and the Visual System
4. The Details of Information Design
5. Geospatial Visualization
6. Design Thinking, Human Computer Interaction (HCI) & Decision Support 
7. Aesthetics and Visual Appeal for Attraction, Clarity & Memorability
8. Network Visualization
9. Storytelling & Data-Driven Journalism
10. Visual Literacy (Graphicacy) and Xenographics
11. Miscommunication and How to Spot Lies | Ethical Encoding on a Truth Continuum
12. Uncertainty: How to Visualize it
13. Visual analytics / dashboarding

Formal prerequisites

The course mandatory for BSc in Data Science fourth semester.

The course requires basic programming skills and basic statistical knowledge. Specifically, the students should have basic skills in Python programming (loop structures, data types), and have knowledge of: probability spaces, random variables, conditional and joint probability, independence, expectation, variance, correlation and covariance, law of large numbers, confidence intervals.

Intended learning outcomes

After the course, the student should be able to:

• Define data visualization (versus information graphics, or information visualization) and explain the differences in visualization purposes, from exhibitroy to exploratory and explanatory.
• Identify the most appropriate visualization strategy given the result of an data analysis process and an intended communication objective.
• Follow a checklist of established design guidelines when building a visualization to account for human perception, communicating uncertainty, and reducing potential miscommunication.
• Conversely, students should be able to readily identify instances where general guidelines should be broken, and provide practical and theoretically informed arguments supporting their design decisions.
• Produce alternative visualizations for the same message, compare their relative strengths and weaknesses, and make a motivated choice for the preferred one given the intended communication outcome.
• Identify deceptive usage of visual communication, or unintentionally misleading graphical representations of data (evidence). Explain ethical considerations, specifically the consequences of misinformation, and the motivations why data should not be visualized in specific ways.
• Generalize standard visualization techniques and customize them to better fit the visual literacy and/or intended communication outcome for a particular audience with domain-specific problems.
• Likewise, participants will explore methods for relentlessly expanding the truthfulness of their models when abstracting data from the real world.

Learning activities

Lectures, feedback, peer-to-peer feedback, project work, active students’ participation, reflection, practical exercises & case studies.

Mandatory activities

Exercises are mandatory to be able to access the exam. Each exercise will result in a visualization made by the student. Students must have at least 5 approved exercises by the project delivery date. Students fulfilling the mandatory activities in the previous years must notify the course manager about their completion status.

The course aims at providing the students the ability to create informative and not misleading visual data presentations. It is thus necessary to develop not only the theoretical knowledge of the tools, but also the skills to put such theory in practice. Therefore, the exercises, which are mandatory activities, will require the students to practice data wrangling, manipulation, and projection, to serve the purpose of visualizing data, testing the concepts presented during the previous lecture.

The initial and simplest mandatory activities will only receive an "approved" / "not approved" score and feedback is only provided in the form of instructions for the not approved activities -- students can solicit feedback at any time. In the second half of the course, the teacher will select some remarkable visualizations and discuss them in class.

Upon receiving a "not approved" evaluation, the students will be able to use the teacher's feedback and re-submit their amended visualization. There is no limit for the number of attempts the students can make.

The student will receive the grade NA (not approved) at the ordinary exam, if the mandatory activities are not approved and the student will use an exam attempt.

Course literature

Cairo, A. (2016) The truthful art: Data, charts, and maps for communication (available at Academic Books på Søndre Campus (KU)). Additional readings will be provided through LearnIT.

Student Activity Budget

Estimated distribution of learning activities for the typical student

• Preparation for lectures and exercises: 14%
• Lectures: 33%
• Exercises: 33%
• Project work, supervision included: 14%
• Exam with preparation: 6%

Ordinary exam

Exam type:
D: Submission of written work with following oral, External (7-point scale)
Exam variation:
D2G: Submission for groups with following oral exam supplemented by the submission. Shared responsibility for the report.

reexam

Exam type:
D: Submission of written work with following oral, External (7-point scale)
Exam variation:
D22: Submission with following oral exam supplemented by the submission.

Time and date