Advanced workshops of analysis methods - Introduction to systemic thinking and network science

Teacher(s)

Heeren Alexandre;

Language

English

Content

Over the past decades, many scientific fields have profoundly transformed the way they approach the world. Rather than attempting to reduce the complexity of their objects of study to simple models, they have increasingly developed approaches that explicitly represent such complexity, notably through network models.
This development has been made possible both by increases in computational power, allowing increasingly large and complex datasets to be analysed, and by the development of network science.
Network science offers several advantages that have contributed to its adoption across fields as diverse as ecology, sociology, and epidemiology. In particular, network approaches can be applied to many types of systems, in which nodes represent the elements under study (e.g., people, symptoms, or brain regions) and edges represent relationships between these elements. Network science also builds upon graph theory, which provides mathematical tools for describing, analysing, and better understanding network structures.
Psychological science has also increasingly incorporated these approaches.
The aim of the course is to provide a general introduction to systemic thinking, graph theory, and network analysis in psychological science. Applications to real datasets will be presented throughout the course. To illustrate the diversity of possible applications, examples may notably range from social networks to brain networks and symptom networks.
By the end of the course, students will notably be able to:

• develop a more systemic approach to psychological problems and move beyond siloed perspectives;
  
• understand the general principles of graph theory and network analysis;
  
• model and analyse network data using R, apply tools derived from graph theory, and use contemporary methods for network estimation and visualisation;
  
• understand and critically evaluate scientific work using network analysis or graph theory in psychological science.
  

Prerequisites

No specific prerequisite is required beyond the basic statistical knowledge and skills acquired through the statistics courses of the psychology programme.
No prior knowledge of network analysis, graph theory, or their implementation in R is required.

Sustainability and Transition Elements (S&T):

In line with recent charter "For Teaching that Meets the Ecological Emergency" (https://charteenseignantsecologie.be/la-charte/), this course also emphasizes the interdependence of individual, economic, ecological, and social aspects; the dangers of siloed thinking; and the urgent need to break down boundaries and to approach the various themes of psychology in their full complexity and interconnectedness.

Teaching methods

Teaching will combine interactive theoretical presentations with practical applications and exercises. Throughout the course, students will work on concrete examples and datasets in order to apply the concepts and methods introduced, notably using R.
Emphasis will be placed on conceptual understanding, methodological reasoning, and application rather than on the memorisation of procedures or computer commands.
Course materials and content will be available in English, allowing students to become progressively familiar with the scientific and methodological terminology of the field, for which most resources are available in English. This exposure also provides useful preparation for scientific and professional contexts in which English is frequently used, notably in research and consultancy.
Particular attention will nevertheless be paid to ensuring that language does not constitute a barrier to learning. Students may ask questions and interact in French whenever this facilitates participation or understanding. When international students are present, the instructor will reformulate or translate into English the elements necessary for collective understanding. The course thus provides an environment in which students can progressively become familiar with scientific English without oral proficiency in English constituting a barrier to participation or learning.
Teaching and assessment will take place on the UCLouvain campus in Louvain-la-Neuve, Belgium.

Evaluation methods

First session

• The final grade is based on two components:
  
• In-class exercises (4 points / 20). Participation in and completion of the practical exercises and activities carried out during the course are mandatory. Completion of these activities is required in order to be eligible to sit the oral examination during the first session.
• OPEN-BOOK ORAL EXAMINATION (16 points / 20). The examination consists of four application-based exercises:
  • one exercise assessing the student’s understanding, reasoning, and ability to apply the concepts and methods covered during the course;
  • three practical network-analysis exercises using R, based on the types of analyses and procedures practised during the course.

The examination is open-book. Students are therefore not expected to memorise R procedures or commands. They are expected, however, to understand and master the underlying concepts and to be able to identify, implement, explain, and justify the appropriate procedures.
For each exercise, students are expected not only to solve the proposed task but also to explain orally their reasoning, the different steps followed, and the methodological choices made, and to demonstrate a clear understanding of the concepts and procedures involved. Where relevant, students will also be expected to interpret the results obtained.
The examination focuses primarily on understanding, reasoning, and application of the course material rather than on the mere reproduction of theoretical knowledge or the memorisation of R syntax.
Examination instructions and questions are provided in both French and English in order to ensure accessibility for both French-speaking and international students. Students may take the oral examination and provide their answers in either French or English, according to their preference.
The use of Generative Artificial Intelligence tools is not permitted during the examination.
Second examination session. Assessment consists solely of an open-book oral examination (20 points / 20). The examination assesses students’ understanding of the course content and their ability to apply the concepts and methods covered during the course. It includes the completion and discussion of exercises during the oral examination. For each exercise, students are expected to explain their reasoning, the steps followed and, where relevant, to interpret the results obtained. Assessment focuses on understanding and application of the course content rather than on the memorisation of procedures or R commands. The use of Generative Artificial Intelligence tools is not permitted during the examination.

Online resources

Course documents, including slides and articles used as examples, sample R code, and the datasets used for practical exercises will be made available via Moodle.

Bibliography

- Network Psychometrics with R: A Guide for Behavioral and Social Scientists, Edited By Adela-Maria Isvoranu, Sacha Epskamp, Lourens Waldorp, Denny Borsboom
- Network Science, by ALBERT-LÁSZLÓ BARABÁSI (freely and legally available at : https://networksciencebook.com) 

Faculty or entity

> EPSY