Datasets
Dataset strategy
The course uses a small set of connected synthetic datasets. Reusing contexts helps students focus on the analytical idea rather than relearning a new domain every lesson.
The datasets do not contain real employee, customer, hotel, company, or student information.
Core datasets
1. Employee and organizational-practices survey
File: data/employee_survey.csv
Rows: 480 synthetic employees
Uses: cleaning, description, correlations, reliability, PCA/EFA/CFA, moderation, mediation, and categorical analysis.
2. Management-training experiment
File: data/training_experiment.csv
Rows: 220 synthetic participants
Uses: independent and paired t-tests, including the distinction between one-sided and two-sided hypotheses.
3. Marketing mix
File: data/marketing_mix.csv
Rows: 156 weekly observations
Uses: linear regression, multicollinearity, residual diagnostics, and influential observations.
4. Hotel upgrades
File: data/hotel_upgrades.csv
Rows: 900 synthetic stays
Uses: logistic regression, odds ratios, predicted probabilities, and responsible operational interpretation.
5. Replication sample
File: data/employee_replication.csv
Rows: 360 synthetic employees
Uses: reproducing analyses in another sample and comparing estimates.
New applied-extension datasets
6. Employees nested within teams
File: data/multilevel_teams.csv
Rows: 600 employees in 50 teams
Uses: nested data, team clustering, ICC, random intercepts, Level-1 and Level-2 predictors.
7. Stagnant retail sales
File: data/stagnant_retail.csv
Rows: 288 branch-month observations (24 branches × 12 months)
Uses: management diagnosis, scenario-based analysis, regression-based what-if comparisons, and the link between repeated observations and multilevel data.
Dictionaries
- Original dictionary:
data/data_dictionary.csv - v0.3 extension dictionary:
data/v0_3_data_dictionary.csv
Why synthetic data?
Synthetic teaching data allow:
- public sharing without privacy concerns;
- controlled examples with recognizable patterns;
- deliberate inclusion of methodological problems;
- repeatable classroom demonstrations.
Synthetic results are not evidence about real organizations. Do not present estimates from these files as empirical findings about actual employees, firms, customers, or Luiss students.