Data science with R, RMarkdown, RStudio, Git and GitHub 0800-OG-DATAR

Contact hours with teacher (obligatory: 30 hrs, voluntary: 10 hrs): - participation in laboratories – 30 hrs - consultations – 10 hrs Self-study hours (35 hrs): - preparation for the laboratory session (watching the provided video and reading the supplied material) – 15 hrs - writing and reviewing assignments (projects, reports, etc.) – 20 hrs Altogether: 75 hrs (3 ECTS)

Student W1: has comprehensive knowledge of the data analysis lifecycle, including preprocessing, exploration, and reporting. W2: understands advanced data transformation and visualization techniques and their impact on analysis outcomes. W3: has a solid understanding of building, validating, and interpreting statistical models. W4: understands how to write robust, efficient, and reproducible scripts using RMarkdown and integrated tools. W5: recognizes the importance of collaboration and communication in data science, leveraging version control tools such as Git and GitHub.

Student U1: is proficient in importing, cleaning, and transforming diverse datasets, including spatial and textual data. U2: can design, implement, and evaluate statistical models to solve real-world problems. U3: is capable of creating fully reproducible workflows using RMarkdown, ensuring transparency in analyses. U4: is skilled in using Git and GitHub for collaborative data science projects, including managing conflicts and maintaining code quality. U5: can critically assess and incorporate ethical considerations into data analysis and communication processes. U6: is capable of conducting peer reviews of assignments through GitHub Pull Requests, fostering critical thinking and collaboration.

Student K1: understands the significance of ethical considerations in data analysis. K2: recognizes the importance of reproducibility and transparency in scientific research. K3: is prepared to collaborate effectively using version control systems. K4: is aware of the impact of data visualization on data interpretation. K5: values continuous learning and staying updated with advancements in data science tools and methodologies.

Expository teaching methods: - online materials (e.g., video lectures, reading materials) Active teaching methods: - practical exercises - group projects - interactive tutorials - peer reviewing

- Basic understanding of statistics. - Familiarity with at least one programming language is advantageous but not mandatory.

Assessment methods:
- group project focusing on reproducible research and collaboration using Git and GitHub, culminating in a public presentation that is also evaluated. The presentation should demonstrate transparency and reproducibility
(W1, W2, W3, W4, W5, U1, U2, U3, U4, K2, K3, K4);
- completion of all assignments and peer reviews is mandatory. Students must submit all assignments and conduct reviews for their peers to pass the course. This is assessed on a binary scale: all tasks completed – pass, any task missing – fail (U6, K3, K5);
- individual report assessing cooperation, punctuality, and ethical considerations in completing and reviewing assignments. The report must include:
• Self-assessment of timely completion of assignments and reviews (K5).
• Evidence of peer reviews, including screenshots of comments from GitHub (W5, U6, K1, K2, K3, K4).
• Reflection on ethical considerations in data analysis (U5, K1).

Verification of learning outcomes:
Knowledge (W1-W5): Evaluated through project documentation and use of appropriate analysis techniques.
Skills (U1-U6): Verified via project implementation, RMarkdown reports, and peer review activities on GitHub.
Social competencies (K1-K5): Assessed through individual reports, teamwork in projects, and peer review contributions.

Fail – less then 50%
satisfactory – 50%-59%
satisfactory plus – 60%-69%
good – 70%-79%
good plus – 80%-89%
very good – 90%-100%

Additional information (registration calendar, class conductors, localization and schedules of classes), might be available in the USOSweb system:

• Description of 0800-OG-DATAR in USOSweb