E0430 Data Analysis in Biomedical and Environmental Sciences II

Přírodovědecká fakulta
jaro 2025
Rozsah
1/1/0. 2 kr. (plus ukončení). Ukončení: zk.
Vyučováno kontaktně
Vyučující
Mgr. Albert Kšiňan, Ph.D. (přednášející)
Mgr. Gabriela Kšiňanová, Ph.D. (přednášející)
Mgr. Hynek Pikhart, Ph.D., M.Sc. (přednášející)
Garance
prof. RNDr. Jana Klánová, Ph.D.
RECETOX – Přírodovědecká fakulta
Kontaktní osoba: Mgr. Albert Kšiňan, Ph.D.
Dodavatelské pracoviště: RECETOX – Přírodovědecká fakulta
Předpoklady
The course is open to students from all levels. The only prerequisite is that students should have working knowledge of OLS regression and data handling in SPSS (or related statistical package).
Omezení zápisu do předmětu
Předmět je nabízen i studentům mimo mateřské obory.
Mateřské obory/plány
Cíle předmětu
The goal of this course is to teach students advanced methods of statistical analyses. These include multilevel/hierarchical modeling (MLM/HLM), path analysis, and structural equation modeling (SEM). Students will learn the basic framework for understanding these analyses, will be able to carry them using specialized software (MLWin for MLM/HLM; AMOS for path analysis+SEM). Special attention will be given to longitudinal data analysis. The syntax for all the topics will be also provided in R.
Výstupy z učení
After this course, the student will be able to:
1. Carry out analysis employing clustered data, understand basic concepts of MLM/HLM (nested structure of data, random effects, longitudinal analysis with data in a long format)
2. Carry out basic analyses in SEM framework, understand basic concepts of SEM (latent variables, model fit, multigroup analysis, longitudinal analysis with data in a wide format)
3. Work with datasets – transform data (wide/long) based on the type of analyses, make analytical subsets
4. Conduct analyses using the MLWin and IBM AMOS sofware
5. Report the results in a proper format
Osnova
  • Lectures
  • 1. Introduction and class overview
  • 2. Introduction to multilevel models vis-à-vis OLS regression
  • 3. Fixed effects for two-level nested data
  • 4. Random effects, three-level models
  • 5. Multilevel models for binary outcomes
  • 6. Repeated measures ANOVA and longitudinal multilevel models
  • 7. Basics of SEM, model diagram symbols, causal inference, recursive and nonrecursive models
  • 8. Model estimates, path analysis
  • 9. Model identification, Model fit, alternative fit indices
  • 10. Confirmatory factor analysis (CFA)
  • 11. Structural regression models
  • 12. Hierarchical models, multigroup models
  • 13. Latent growth model (LGM)
  • 14. TBD
  • Practical sessions
  • 1. Correlation, OLS regression
  • 2. Overview of of MLWin software
  • 3. Estimating two-level multilevel model
  • 4. Estimating three-level multilevel model
  • 5. Estimating logistic multilevel model
  • 6. Estimating longitudinal multilevel model
  • 7. Different methods of missing data handling, overview of AMOS
  • 8. Estimating path analysis model
  • 9. Interpreting path analysis model fit and estimates
  • 10. Performing CFA
  • 11. Estimating structural regression model
  • 12. Estimating multigroup SEM model
  • 13. Estimating LGM model
  • 14. TBD
Literatura
    povinná literatura
  • KLINE, Rex B. Principles and practice of structural equation modeling. 3rd ed. New York: Guilford Press, 2010, xvi, 427. ISBN 9781606238769. info
  • RAUDENBUSH, Stephen W. a Anthony S. BRYK. Hierarchical linear models : applications and data analysis methods. 2nd ed. Thousand Oaks: Sage Publications, 2002, xxiv, 485. ISBN 076191904X. info
    doporučená literatura
  • SINGER, Judith D. a John B. WILLETT. Applied longitudinal data analysis : modeling change and event occurrence. Oxford: Oxford University Press, 2003, xx, 644. ISBN 0195152964. URL info
Výukové metody
The teaching format is an in-person lecture supported by PowerPoint presentations followed by a practicum in the computer lab. Students will use MlWin or IBM AMOS software to perform statistical analyses discussed in the lectures.
Metody hodnocení
Students will complete a practical data analysis exercise during the practical session. Additionally, there will be three quizzes throughout the semester (multiple choice format). The quizzes will not be cumulative. Lastly, there will be no final exam. Instead, students will submit an assignment, which will consist of statistical analysis and its write-up using their own data or data provided by the instructors if necessary. There will be opportunities for earning extra credit throughout the semester.
The grading is broken down as follows:


Quiz 1 (20% of grade)
Quiz 2 (20% of grade)
Quiz 3 (20% of grade)
Final assignment (40% of grade)
This corresponds to the following grades: A (100%-92%), B (91%-84%), C (83%-76%), D (75%-68%), E (67%-60%), F (< 60%).
Vyučovací jazyk
Angličtina
Další komentáře
Předmět je vyučován každoročně.
Výuka probíhá každý týden.
Předmět je zařazen také v obdobích jaro 2022.
  • Statistika zápisu (nejnovější)
  • Permalink: https://is.muni.cz/predmet/sci/jaro2025/E0430