Structured exploration
Exploratory data analysis, or EDA, is a way to learn what is in the data and check possible explanations. It is more than producing many charts.
EDA is a cycle: ask a question, check row meaning and data quality, describe distributions, compare relevant groups, interpret and revisit assumptions. A gallery of unrelated plots does not substitute for that reasoning.
For every figure, identify its population, axes and units, denominator, pattern, possible alternative explanation and limitation. Explain what would change the conclusion. If EDA is being used to choose predictive features, develop them inside training data and protect final test outcomes.
Source: W5 PDF pp. 3-51, 53. The lecture uses both a constructed 200-student support example and a synthetic 4,665-registration course dataset. They are different tables with different units.
Practice problems
Supplementary study questions on the lecture concepts — not verbatim exam questions. Each answer counts toward spaced review.
1. Conceptual understanding
Definitions and distinctions.
What is the purpose of structured EDA?
If EDA selects predictive features, where should that development occur?
2. Application
Apply the concept to a case or calculation.
Which is the better start to EDA?
Before comparing pass rates, what should be checked?
3. Interpretation and limitations
Evaluate assumptions and justify conclusions.
Which report separates observation from explanation?
EDA reveals many unmatched join keys. What should happen?
Explain it in writing
Self-checked, not marked. Write a few sentences, then compare with the example answer. Your writing is saved as you type.
Show an example answer
Name the population, axes and units, denominator, possible explanations and limitations. Explain what evidence could change the interpretation.
Check: did you answer the question, use evidence or working, and avoid claiming more than the evidence supports?
Show an example answer
Use training and appropriate validation data to develop features and choices. Keep final test outcomes separate until the fixed procedure is evaluated.
Check: did you answer the question, use evidence or working, and avoid claiming more than the evidence supports?