Mathematics
Discrete vs. Continuous Distributions
Quick fact
A continuous distribution assigns zero probability to any single exact value — only intervals have non-zero probability. That's why we use integration instead of summation.
Why this is interesting
If you roll a die, you can only get 1, 2, 3, 4, 5, or 6 — but when you measure someone's height, there are infinitely many possibilities. Why does that difference shape how we calculate probabilities?
Read the full explanation
Understanding Discrete vs. Continuous Distributions
Imagine you have a bag of numbered marbles from 1 to 10. When you pick one, the outcome is one of those specific numbers. This is a discrete scenario: the outcomes are countable and separate. Now think of measuring the time it takes for a cup of coffee to cool from 80°C to room temperature. The exact time could be any number in a continuous range — 10.5 minutes, 10.53 minutes, 10.534 minutes, and so on. There's no gap between possible values. In probability, discrete distributions (like the binomial) assign probabilities to each individual outcome using a probability mass function (PMF). Continuous distributions (like the normal) describe probabilities over intervals using a probability density function (PDF). The key is that for continuous cases, the probability of any exact value is zero; only ranges have positive probability.
A deeper explanation
The mathematical distinction stems from whether the sample space is countable or uncountable. For discrete distributions, we sum probabilities: Pr(X=x) = PMF(x). For continuous, we integrate the PDF over an interval: Pr(a ≤ X ≤ b) = ∫ₐᵇ PDF(x) dx. This difference matters because applying the wrong type leads to incorrect calculations (e.g., summing densities gives nonsense). It also influences which statistical tests are appropriate: many tests assume normality (continuous), while others like chi-square tests handle counts (discrete). Understanding this dichotomy is crucial for modeling real-world phenomena — from counting defects in manufacturing (discrete) to measuring voltage fluctuations (continuous).