Statistics for Data Science - Study
Guide
Descriptive stats, probability, distributions, hypothesis testing and regression -
with intuition
Complete Study Guide | Key Concepts | Examples | Practice Questions
, 1. Descriptive Statistics
Descriptive statistics summarise data. Central tendency tells you the middle; spread tells you the variability - both
are needed to understand any dataset.
Measure Meaning Sensitive to outliers?
Mean Arithmetic average Yes
Median Middle value when sorted No
Mode Most frequent value No
Variance / Std dev Spread around the mean Yes
IQR Q3 - Q1 (middle 50%) No
- Right-skewed data (income): mean > median; use the median for the 'typical' value.
- Standard deviation is in the data's units; variance is in squared units.
- Box plots show median, quartiles and outliers in one picture.
- Correlation (between -1 and 1) measures linear association - not causation.
TIP: Always plot your data first. Anscombe's quartet: four datasets with identical stats but wildly
different plots.
Page 1
Guide
Descriptive stats, probability, distributions, hypothesis testing and regression -
with intuition
Complete Study Guide | Key Concepts | Examples | Practice Questions
, 1. Descriptive Statistics
Descriptive statistics summarise data. Central tendency tells you the middle; spread tells you the variability - both
are needed to understand any dataset.
Measure Meaning Sensitive to outliers?
Mean Arithmetic average Yes
Median Middle value when sorted No
Mode Most frequent value No
Variance / Std dev Spread around the mean Yes
IQR Q3 - Q1 (middle 50%) No
- Right-skewed data (income): mean > median; use the median for the 'typical' value.
- Standard deviation is in the data's units; variance is in squared units.
- Box plots show median, quartiles and outliers in one picture.
- Correlation (between -1 and 1) measures linear association - not causation.
TIP: Always plot your data first. Anscombe's quartet: four datasets with identical stats but wildly
different plots.
Page 1
