Data Cleaning Practices (Medium)
A 7-question multiple-choice quiz grounded in a pandas hotel-bookings cleaning walkthrough. Mixes recall and applied reasoning about inspection, missingness, placeholders, types, and transformations.
Try this quiz
Play through the questions and see your score instantly
Ready to test your knowledge?
7 questions · Quick play · Instant results
Make your own quiz videos
Turn any topic into a polished video quiz — with AI-powered questions, voiceover, and animations. No video editing skills needed.
Unlimited quizzes, free to start
Create as many quizzes as you want. Describe your topic and AI builds the questions, answers, and explanations for you.
Customise everything
Pick from stunning templates, tweak colours and fonts, add your branding, and choose between vertical or landscape formats.
Export-ready videos
Download HD videos optimised for TikTok, YouTube Shorts, Instagram Reels, or full-length YouTube — one click, no editing.
No credit card required
Quiz Questions & Answers
Review every prompt, the correct responses, and helpful context to prep for your own run-through.
Question 1: Before changing any columns, which quick set of checks does the material recommend you run to surface structural issues and missingness?
head(), columns, dtypes, isnull().sum(), describe()
Immediately convert types to integers and booleans
Drop any column with more than 5 missing values
Run heavy imputations on all columns automatically
Question 2: If a dataset contains an 'Unnamed: 0' index-duplicate column, what action does the material recommend and why?
Keep it as a primary feature for modeling
Drop it, because index-duplicates are trivial and reduce readability
Fill it with zeros to preserve shape
Convert it to categorical without inspecting contents
Question 3: When a numeric identifier column has many missing entries but needs a placeholder, which sentinel approach does the walkthrough endorse?
Leave NaNs and convert to integer type immediately
Replace with the column mean
Fill with a distinguishable sentinel like -1 that is castable to numeric
Use the string 'unknown' in place of NaNs
Question 4: For categorical fields with missing values, what filling strategy preserves grouping semantics for joins and counts?
Delete the entire column immediately
Replace NaNs with explicit labels like 'unknown' or 'missing'
Randomly assign an existing category to each missing row
Encode missing categories as -1 numeric sentinel
Question 5: If only a handful of rows (e.g., 4 rows) lack a critical value, what trade-off driven action does the material suggest?
Duplicate similar rows to replace missing values
Always impute them with column median regardless of context
Drop those few rows to avoid complex imputation when information loss is minimal
Flag them and keep them unmodified until after analysis
Question 6: Why does the walkthrough recommend converting column data types only after handling NaNs?
So you can keep all columns as strings permanently
Because type conversion is unnecessary in most analyses
To ensure all values are randomized before modeling
Because conversions often fail or coerce types incorrectly when NaNs are present
Question 7: Which transformation technique does the material show to make continuous fields like lead time and compound date fields analysis-ready?
One-hot encode continuous lead-time directly without binning
Store transformed features only in a separate notebook and not the dataset
Normalize all continuous fields to 0–1 and keep compound date strings intact
Use pd.cut for binning continuous values and str.split(expand=True) to separate compound date parts