Data: From Raw to Ready
Ask anyone who analyses data for a living where their time goes, and the answer is cleaning. This course is about the unglamorous majority of real analysis: the export that arrives with three header rows, the age recorded as 512, the survey stored so awkwardly that no software will touch it.
It starts with the idea that saves the most grief later, tidy data: one row per observation, one column per variable. From there it builds the habits that keep a project trustworthy, a codebook so future-you knows what "grp" meant, validation rules that catch the impossible value at entry, and a cleaning workflow that never edits the raw file by hand. The middle lessons handle the recurring puzzles (outliers, recoding, reshaping, merging), and the course closes on reproducibility and privacy.
No coding background needed, though the principles map straight onto R or Python when you're ready. Pairs well with Methods before and Writing after.