Welcome to Risk in Statistics!
In everyday life, we often hear the word "risk" associated with danger, like the risk of getting an injury while playing sports. In Statistics, however, "risk" is just another way of talking about probability. It is the likelihood of a specific event happening within a group.
In this chapter, we will learn how to calculate the Absolute Risk (the chance of something happening) and the Relative Risk (comparing the chances between two different groups). This is a vital skill for understanding medical studies, insurance, and even weather reports!
1. Understanding Absolute Risk
Absolute Risk is simply the probability of an event occurring. It is calculated using relative frequency based on data we have collected.
To find the absolute risk, you use this simple formula:
\( \text{Absolute Risk} = \frac{\text{Number of people the event happened to}}{\text{Total number of people in the group}} \)
Example:
In a survey of \(200\) students, \(40\) of them said they caught a cold during the winter term.
The Absolute Risk of a student catching a cold is:
\( \text{Risk} = \frac{40}{200} = 0.2 \) (or \(20\%\))
Quick Review: Probability Basics
Remember that risk is a probability, so it will always be a number between \(0\) and \(1\):
- A risk of \(0\) means the event is impossible.
- A risk of \(1\) means the event is certain.
- You can express risk as a decimal (\(0.25\)), a fraction (\(\frac{1}{4}\)), or a percentage (\(25\%\)).
2. Expected Frequency
Once we know the absolute risk, we can predict how many times we expect the event to happen in a larger or different-sized group. This is called the Expected Frequency.
\( \text{Expected Frequency} = \text{Absolute Risk} \times \text{Total Frequency (New Group Size)} \)
Example:
If the risk of a rainy day in April is \(0.3\), how many rainy days would you expect in a town of \(1,000\) people if each person experiences the weather individually?
Wait—weather is a bad example for frequency! Let's try a better one:
If the risk of a computer being faulty is \(0.02\), how many faulty computers would you expect in a shipment of \(5,000\)?
\( \text{Expected Frequency} = 0.02 \times 5000 = 100 \text{ computers} \)
Key Takeaway: Absolute risk tells you the "chance" for one individual, while expected frequency tells you the "count" for a whole group.
3. Understanding Relative Risk
Relative Risk is used when we want to compare the risk of an event happening in two different groups (Group A and Group B). It tells us how much more (or less) likely an event is to happen in one group compared to another.
The formula is:
\( \text{Relative Risk} = \frac{\text{Risk for Group A}}{\text{Risk for Group B}} \)
How to Interpret Relative Risk (RR):
- If \( \text{RR} = 1 \): The risk is exactly the same in both groups.
- If \( \text{RR} > 1 \): The risk is higher in Group A.
- If \( \text{RR} < 1 \): The risk is lower in Group A.
Don't worry if this seems tricky! Just remember that "Relative" means "compared to." We are just dividing one probability by another.
4. Step-by-Step Example: Comparing Two Groups
Let's look at a study involving a new vitamin tablet and the flu.
Group A (Took Vitamin): \(10\) people out of \(100\) got the flu.
Group B (No Vitamin): \(30\) people out of \(150\) got the flu.
Step 1: Find the Absolute Risk for Group A.
\( \text{Risk}_A = \frac{10}{100} = 0.1 \)
Step 2: Find the Absolute Risk for Group B.
\( \text{Risk}_B = \frac{30}{150} = 0.2 \)
Step 3: Calculate the Relative Risk of getting the flu (A compared to B).
\( \text{Relative Risk} = \frac{0.1}{0.2} = 0.5 \)
Conclusion: The Relative Risk is \(0.5\). This means people who took the vitamin were half as likely to get the flu as those who didn't. (Because \(0.5\) is half of \(1\)).
5. Common Mistakes to Avoid
1. Mixing up the groups: Always read the question carefully to see which group should be the numerator (on top) and which should be the denominator (on the bottom). If the question asks for the risk of A relative to B, B goes on the bottom.
2. Absolute vs. Relative: A "Relative Risk" might sound scary (e.g., "Double the risk!"), but if the "Absolute Risk" is very small (e.g., \(0.00001\)), the actual danger is still very low. Always look at both numbers!
3. Forgetting that Risk is a Probability: Ensure your absolute risk values are never greater than \(1\). If you get a number like \(1.5\) for absolute risk, you’ve likely put the fraction upside down!
6. Summary Checklist
- Absolute Risk: The probability of an event happening (\( \frac{\text{event}}{\text{total}} \)).
- Expected Frequency: \(\text{Risk} \times \text{Total population}\).
- Relative Risk: Comparing two risks by dividing them (\( \frac{\text{Risk A}}{\text{Risk B}} \)).
- Comparison: Relative Risk is a powerful way to show how factors like medicine, habits, or environments affect outcomes.
Note: For more complex probability questions involving two-way tables or tree diagrams, check out the "Two-way tables, sample space, tree and Venn diagrams" chapter in this section.