Numerical Variables: Discrete and Continuous
You'll learn how to distinguish between two fundamental types of numerical data: discrete (counts) and continuous (measurements).
The Core Difference
Numerical variables represent quantities you can measure or count. They come in two flavors:
Discrete numerical data consists of whole numbers that represent counts of distinct items. You can't have fractional values between them. Think: number of customers (1, 2, 3...), number of products sold, or number of clicks on a website. You can't have 2.7 customers!
Continuous numerical data represents measurements that can take any value within a range, including decimals. Think: height (5.8 feet, 5.81 feet, 5.812 feet...), temperature, time, or weight. Between any two values, infinitely more values can exist.
Practical Analogies
Imagine counting apples in a basket versus measuring water in a glass:
- Discrete (apples): You have exactly 5 apples or 6 apples. Never 5.3 apples.
- Continuous (water): You might have 250ml, or 250.5ml, or 250.52ml — precision is limited only by your measuring tool.
Why This Matters in Data Science
Understanding this distinction affects:
- Statistical methods you choose (some tests only work for continuous data)
- Visualization decisions (bar charts vs. histograms)
- Model selection (certain algorithms handle discrete vs continuous differently)
A common example: age can be recorded as discrete (23 years old) but is technically continuous (23.4567... years). The context and precision matter!
Key Takeaway: Discrete numerical variables are countable whole numbers, while continuous numerical variables are measurable values that can take any number within a range—knowing the difference guides your entire analysis approach.