Syllabi
University of Ljubljana
The summary or overview of a course includes the course content and structure, the prerequirements, a brief description of the learning objectives, mandatory and/or voluntary course literature as well as important examination information.
Prerequisites (knowledge of topic)
Prior knowledge of quantitative research methods is required. Basic knowledge of SPSS, or willingness to learn SPSS individually.
Hardware
Please bring your own laptop (mac or windows).
Software
Make sure to install SPSS version 27, Jamovi (free software) 2.3.28 and Process version 4.2 or newer on your computer before the start of the course.
Learning objectives
After the successful completion of this course, participants will:
- be well-equipped to design, conduct, and analyze experiments in a rigorous and informed manner
- understand how to update and improve their experiments and develop their skills in critically evaluating experimental research
- be able to run the appropriate analysis depending on the chosen experimental design
- understand the principles and the logic of the following statistical techniques: Regression, Analysis of Variance, Analysis of Covariance, Contrast Analysis, Mediation Analysis, Moderation Analysis, Logit
- understand issues related to statistical power and effect size and ethical considerations in experimental research.
- interpret the data of a wide range of experiments and report the results in a scientific paper.
Course content
This dual-focused program is designed to equip participants from various disciplines (behavioral economics, management, medicine, health science, communication, marketing) with the necessary skills 1) to conduct rigorous experimental research and 2) to analyze data from various experimental designs. Against the backdrop of actual examples from practice and academia, each module of the course centers around a specific experimental design. Participants will learn how to improve these experimental designs and to conduct the appropriate data analysis techniques. The curriculum covers a diverse range of designs and corresponding analyses, including moderation, mediation, spotlight analyses, co-variate analyses, and logit. Participants will acquire a nuanced understanding of these techniques, their application in research, and learn how to report their findings in an academic article. A part of the course will be devoted to ethical considerations in experimental research, addressing contemporary concerns in research integrity.
Structure
Day 1:
- Improving experimental designs
- Experimental issues: Confounds, sampling, and demand effects
- Intuition behind ANOVA: Increasing systematic variance and reducing error variance
- Power and Effect size
- Research integrity issues: p-hacking, power-analyses, preregistration
Day 2:
- Experimental design with moderation
- ANOVA
- Simple effect analyses
- MANOVA
Day 3:
- Experimental design with repeated measures
- Repeated measures ANOVA
- Mixed design
- ANCOVA
Day 4:
- Experimental design with (moderated) mediation
- Simple and moderated mediation
- Process through moderation
- Causal chain model
Day 5:
- Experimental design with continuous moderator
- Regression
- Conditional process
- Spotlight and floodlight analyses
- Experimental design with binominal DV
- Logit regression
Literature
Mandatory
Book:
- Field, Andy P. (2023). Discovering statistics using SPSS, Sixth Edition, London: Sage Publications: Chapters: 3, 11, 13, 14, 15, 16, 17, 20.
Articles:
- Meyvis, T., & Van Osselaer, S. M. J. (2018), “Increasing the Power of Your Study by Increasing the Effect Size,”Journal of Consumer Research, 44 (5), 1157–1173.
- Pieters, R. (2017). Meaningful mediation analysis: Plausible causal inference and informative communication. Journal of Consumer Research, 44(3), 692-716.
- Simmons, Nelson, Simonsohn (2011) “False-Positive Psychology: Undisclosed Flexibility in Data Collection and Analysis Allow Presenting Anything as Significant”, Psychological Science, 22(11), 1359-1366.
- Spencer, S.J., Zanna, M.P., & Fong, G.T. (2005). Establishing a causal chain: Why experiments are often more effective than mediational analyses in examining psychological processes. Journal of Personality and Social Psychology, 89 (6), 845-851.
- Spiller, S. A., Fitzsimons, G. J., Lynch, J. G., Jr., & McClelland G. H. (2013), “Spotlights, Floodlights, and the Magic Number Zero: Simple Effects Tests in Moderated Regression,” Journal of Marketing Research 50, 277-88.
Supplementary / voluntary
- Abbey, J. D., & Meloy, M. G. (2017). Attention by design: Using attention checks to detect inattentive respondents and improve data quality. Journal of Operations Management, 53, 63-70.Giner-Sorolla, R.,
- McShane, B. B., Bradlow, E. T., Lynch Jr, J. G., & Meyer, R. J. (2024). “Statistical Significance” and Statistical Reporting: Moving Beyond Binary. Journal of Marketing, 00222429231216910.
- Montoya, A. K., Reifman, A., Carpenter, T., Lewis Jr, N. A., Aberson, C. L., … & Soderberg, C. (2019). Power to detect what? Considerations for planning and evaluating sample size. Personality and Social Psychology Review, 10888683241228328.
- Vosgerau, J., Simonsohn, U., Nelson, L. D., & Simmons, J. P. (2019). 99% impossible: A valid, or falsifiable, internal meta-analysis. Journal of Experimental Psychology: General, 148(9), 1628.
Examination part
- Homework delivered during week of the course (4 assignments, 40%)
- Homework delivered after the course (1 assignment, 60%)
Supplementary aids
Open book.
Examination content
Assignment 1 (Due Tuesday Morning):
- Improving an experimental design
- Detection of confounds
Assignment 2 (Due Wednesday Morning):
- ANOVA
- Simple effect tests
- Interpretation and report of results
Assignment 3 (Due Thursday Morning):
- Mixed design
- Assumptions
- Follow-up tests
- Interpretation and report of results
Assignment 4 (Due Friday Morning):
- Mediation
- Moderated mediaton
- Mediation with categorical IV
- Interpretation and report of results
Assignment 5 (Due within one month after end of course):
- Analyse data set using analyses covered during the course
- Report and interpret results as for an academic journal
- Critically reflect upon research design and propose improvements
Examination relevant literature
All mandatory literature indicated above.
Prerequisites (knowledge of topic)
Basic knowledge of Python programming and familiarity with Machine Learning
Hardware
A laptop
Software
A google account to access colab
Learning objectives
On successful completion of this course, students will be able to:
– Understand the fundamentals of Generative AI and Large Language Models.
– Understand how to Implement and fine-tune LLMs for various tasks.
– Evaluate the performance of LLMs using various metrics.
– Understand the ethical and societal implications of using LLMs.
– Develop their own projects utilizing Generative AI.
Structure
Day 1: Introduction to Generative AI
Generative vs. Discriminative Models
Overview of Generative AI
Applications of generative AI
Day 2: Basics of Language Models
Understanding Language Models
Unigram, Bigram, N-gram Models
Neural network based (seq2seq) Language Models
Day 3: Transformer Models and Attention Mechanism
Introduction to Transformer Models
Self-Attention and Multi-Head Attention
BERT
Day 4: Large Language Models (LLMs)
Introduction to LLMs
Architecture of GPT-4
Fine-tuning LLMs
Day 5: Evaluating and Improving LLMs
Evaluation metrics
Issues in LLMs
Prompt engineering
Examination
We will have a take-home exam.
Supplementary aids
Open Book
Examination Content
Students are expected to work on one individual project related to the application of LLMs on a specific domain (e.g., healthcare, finance, or social marketing).
Prerequisites (knowledge of topic)
This course assumes no prior experience with machine learning or R, though it may be helpful to be familiar with introductory statistics and programming.
Hardware
A laptop computer is required to complete the in-class exercises.
Software
R (https://www.r-project.org/) and R Studio (https://www.rstudio.com/products/rstudio/) are available at no cost and are needed for this course.
Learning objectives
Students will leave with an understanding of the foundational machine learning methods as well as the ability to apply machine learning to their own field of work or study using the R programming language.
Course content
Machine learning, put simply, involves teaching computers to learn from experience, typically for the purpose of identifying or responding to patterns or making predictions about what may happen in the future. This course is intended to be an introduction to machine learning methods through the exploration of real-world examples. We will cover the basic math and statistical theory needed to understand and apply many of the most common machine learning techniques, but no advanced math or programming skills are required. The target audience may include social scientists or practitioners who are interested in understanding more about these methods and their applications. Students with extensive programming or statistics experience may be better served by a more theoretical course on these methods.
Structure
The course will be designed to be interactive, with ample time for hands-on practice with the Machine Learning methods. Each day will include several lectures based on a Machine Learning topic, in addition to hands-on “lab” sections to apply the learnings to new datasets (or your own data, if desired).
The approximate schedule will be as follows:
Day 1: Introducing Machine Learning with R
- How machines learn
- Using R and R Studio
- Data exploration
- k-Nearest Neighbors
- Lab sections – installing R, choosing and exploring own dataset (if desired)
Day 2: Intermediate ML Methods – Classification Models
- Quiz on Day 1 material
- Naïve Bayes
- Lab sections – practicing with kNN and Naïve Bayes
Day 3: Intermediate ML Methods – Numeric Prediction
- Quiz on Day 2 material
- Decision Trees and Rule Learners
- Linear Regression
- Regression trees
- Logistic regression
- Lab sections – practicing with classification and regression methods
Day 4: Advanced Classification Models
- Quiz on Day 3 material
- Neural Networks
- Support Vector Machines
- Random Forests
- Lab section – practice with neural networks, SVMs, and random forests
Day 5: Other ML Methods
- Quiz on Day 4 material
- Association Rules
- Hierarchical clustering
- K-Means clustering
- Model evaluation
- Lab section – practice with these methods, work on final report
Literature
Mandatory
Machine Learning with R (4th ed.) by Brett Lantz (2023). Packt Publishing
Supplementary / voluntary
None required.
Mandatory readings before course start
Please install R and R Studio on your laptop prior to the 1st class. Be sure that these are working correctly and that external packages can be installed. Instructions for doing this are in the first chapter of Machine Learning with R.
Examination part
100% of the course grade will be based on a project and final report (approximately 10 pages), to be delivered within 2-3 weeks after the course. The project is intended to demonstrate your ability to apply the course materials to a dataset of your own choosing. Students should feel free to use a project related to their career or field of study. For example, it is possible to use this opportunity to advance dissertation research or complete a task for one’s job. The exact scoring criteria for this assignment will be provided on the first day of class. This will be graded based on its use of the methods covered in class as well as making appropriate conclusions from the data.
There will also be brief quizzes at the start of each lecture, which cover the previous day’s materials. These are ungraded and are designed to provoke thought and discussion.
Supplementary aids
Students may reference literature and class materials as needed when writing the final project report.
Examination content
The final project report should illustrate an ability to apply machine learning methods to a new dataset, which may be on a topic of the student’s choosing. The student should explore the data and explain the methods applied. Detailed instructions will be provided on the first day of class.
Examination relevant literature
The written report is only expected to reflect knowledge of the material presented in class, which covers chapters 1 to 10 in the Machine Learning with R (4th ed.) textbook noted above.
Prerequisites
Basic knowledge of statistics, data analysis, and programming.
Hardware
Participants need to bring laptops.
Software
Google Colaboratory environment (https://colab.research.google.com/). Participants will receive setup instructions before the course.
Learning objective
Large language models like ChatGPT or Llama-3 are incredibly useful tools for research in the social and behavioral sciences. In this course, you will (1) learn about the fundamental principles of large language models, (2) learn how to employ state-of-the-art open-access large language models using the Hugging Face ecosystem, (3) learn about the rich opportunities that large language models offer for behavioral and social science research, and (4) gain experience in applying large language models to answer personal research questions.
Course content
Open-access large language models (LLMs) like Llama-3 or Phi-3 are powerful alternatives to proprietary LLMs like ChatGPT that offer state-of-the-art performance and crucial advantages in terms of reproducibility and scale. This course introduces the use of open-access LLMs from the Hugging Face ecosystem for research in the behavioral and social sciences. In short lectures, participants will learn about key concepts (e.g., embeddings, causal attention, feature extraction, classification, prediction, fine-tuning, and token generation) and practical examples from social and behavioral science. In hands-on exercises, participants will apply language models to answer research questions from psychology, political science, decision-making research, and other fields. During and after the course, participants will engage in a personal research project applying large language models to a personal research question. Two lecturers will hold this course, implying a high level of support during the exercises and research project design.
Structure
Day 1:
Morning: Welcome and intro to large language models
Afternoon: Applying the Hugging Face ecosystem for open-access large language models
Day 2:
Morning: Intro to feature extraction and embedding models
Afternoon: Applying large language models to predict the relationships between survey items and questionnaires in personality psychology
Day 3:
Morning: Intro to fine-tuning for classification and regression
Afternoon: Applying large language models to evaluate and classify texts in political science
Day 4:
Morning: Intro to token and text generation
Afternoon: Using large language models to predict people’s responses in decision-making situations
Day 5:
Morning: Overview of additional applications of large language models for qualitative data analysis
Afternoon: Project pitches
Literature
Mandatory:
Hussain, Z., Binz, M., Mata, R., & Wulff, D. U. (2023). A tutorial on open-source large language models for behavioral science. PsyArXiv (available in December)
Supplementary / voluntary:
Tunstall, L., von Werra, L., & Wolf T. (2022). Natural Language Processing with Transformers. John Wiley & Sons.
Examination
The course grade will be determined based on the quality of a project pitch at the end of the course and a two-page research paper submitted after the course. The paper communicates an analysis applying large language models to a personal research question, including all parts of a traditional research paper (introduction, method, results, and discussion). The research paper can be based on the examples during the course.
Content
Statistical mediation and moderation analyses are among the most widely used data analysis techniques in social science, health, and business fields. Mediation analysis is used to test hypotheses about various intervening mechanisms by which causal effects operate. Moderation analysis is used to examine and explore questions about the contingencies or conditions of an effect, also called “interaction”. Increasingly, moderation and mediation are being integrated analytically in the form of what has become known as “conditional process analysis,” used when the goal is to understand the contingencies or conditions under which mechanisms operate. An understanding of the fundamentals of mediation and moderation analysis is in the job description of almost any empirical scholar. In this course, you will learn about the underlying principles and the practical applications of these methods using ordinary least squares (OLS) regression analysis and the PROCESS macro for SPSS, SAS and R invented by the course instructor.
Topics covered in this five-day course include:
- Path analysis: Direct, indirect, and total effects in mediation models.
- Estimation and inference about indirect effects in single mediator models.
- Models with multiple mediators
- Mediation analysis in the two-condition within-subject design.
- Estimation of moderation and conditional effects.
- Probing and visualizing interactions.
- Conditional Process Analysis (also known as “moderated mediation”)
- Quantification of and inference about conditional indirect effects.
- Testing a moderated mediation hypothesis and comparing conditional indirect effects
As an introductory-level course, we focus primarily on research designs that are experimental or cross-sectional in nature with continuous outcomes. We do not cover complex models involving dichotomous outcomes, latent variables, models with more than two repeated measures, nested data (i.e., multilevel models), or the use of structural equation modeling.
This course will be helpful for researchers in any field—including psychology, sociology, education, business, human development, political science, public health, communication—and others who want to learn how to apply the latest methods in moderation and mediation analysis using readily-available software packages such as SPSS, SAS and R.
Prerequisites (knowledge of topic)
Participants should have a basic working knowledge of the principles and practice of multiple regression and elementary statistical inference. No knowledge of matrix algebra is required or assumed, nor is matrix algebra ever used in the course.
Hardware and Software
Computer applications will focus on the use of OLS regression and the PROCESS macro for SPSS, SAS and R developed by the instructor that makes the analyses described in this class much easier than they otherwise would be.
Because this is a hands-on course, participants are strongly encouraged to bring their own laptops (Mac or Windows) with a recent version of SPSS Statistics (version 23 or later), SAS (release 9.2 or later), or R (version 3.6 or later) installed. (Only one statistical package is required, but participants can use more than one if desired) SPSS users should ensure their installed copy is patched to its latest release. SAS users should ensure that the IML product is part of the installation. You should have good familiarity with the basics of ordinary least squares regression, as well as the use of SPSS, SAS, or R. You are also encouraged to bring your own data to apply what you’ve learned.
STATA users can benefit from the course content, but PROCESS makes these analyses much easier and is not available for STATA.
Literature
This course is a companion to the second edition of the instructor’s book Introduction to Mediation, Moderation, and Conditional Process Analysis, published by The Guilford Press. The content of the course overlaps the book to some extent, but many of the examples are different, and this course includes some material not in the book. A copy of the book is not required to benefit from the course, but it could be helpful to reinforce understanding.
Examination
100% of assessment will be based on a written final examination at the end of the course. The exam will be a combination of multiple choice questions and short-answer/fill in the blank questions, along with some interpretation of computer output. Students will take the examination home on the last day of class and return it to the instructor within one week.
During the examination students will be allowed to use all course materials, such as PDFs of PowerPoint slides, student notes taken during class, and any other materials distributed or student-generated during class. Although the book mentioned in “Literature” is not a requirement of the course nor is it necessary to complete the exam, students may use the book if desired during the exam.
A computer is not required during the exam, though students may use a computer if desired, for example as a storage and display device for class notes provided to them during class.
Among the topics of the exam may include how to quantify and interpret path analysis models, calculate direct, indirect, and total effects, and determine whether evidence of a mediation effect exists in a data set based on computer output provided or other information. Also covered will be the testing moderation of an effect, interpreting evidence of interaction, and probing interactions. Students will be asked to generate or interpret conditional indirect effects from computer output given to them and/or determine whether an indirect effect is moderated. Students may be asked to construct computer commands that will conduct certain analyses. All questions will come from the content listed in “Course Content” above.