What is the fundamental difference between absolute risk and relative risk?

Absolute risk is the probability of an event occurring in a single group (ranging from 0 to 1). Relative risk is a comparison of two absolute risks, showing how many times more likely an event is in one group versus another.

When comparing Group A (risk 0.1) to Group B (risk 0.05), what does a relative risk of 2.0 signify?

It signifies that individuals in Group A are twice as likely to experience the event as those in Group B. It is calculated by dividing 0.1 by 0.05.

How does the range of possible values differ between absolute and relative risk?

Absolute risk must be between 0 and 1 because it is a probability. Relative risk can be any non-negative number (including values much greater than 1) because it is a ratio.

What is a common error when interpreting a relative risk of 0.5?

A common error is thinking the risk increased; actually, a relative risk below 1 means the event is less likely in the first group. Specifically, 0.5 means the group is half as likely (or 50% less likely) to experience the event.

Why is it a mistake to use the number of 'non-events' as the denominator in a risk calculation?

Risk is a probability, which requires the total number of trials (events + non-events) in the denominator. Using only non-events would calculate 'odds' rather than 'risk'.

What error occurs if you assume a small sample size result is 'biased'?

Small samples naturally have high variance; a result might deviate from the mean due to random chance (noise) rather than systematic bias. Bias requires a consistent, non-random trend.

Define 'Absolute Risk' in the context of probability.

Absolute risk is the estimated probability of a negative event occurring, calculated as the frequency of the event divided by the total number of trials.

What is 'Statistical Bias'?

Statistical bias is a systematic distortion in a statistic or measurement that prevents it from accurately representing the true population parameter.

What is a 'Fair' trial in probability?

A fair trial is one where every possible outcome has the theoretical probability it is supposed to have (e.g., 1/6 for a die), and the experimental results converge to these values over time.

Why do insurance companies rely more on relative risk than just absolute risk?

Relative risk allows them to identify specific risk factors (like age or location) that make one group more expensive to insure than the average, helping them set proportional premiums.

Library Podcasts

Courses

Referral & Rewards

Risk & Bias

Summary

Risk and bias are fundamental concepts in probability and statistics used to quantify the likelihood of negative outcomes and evaluate the fairness or accuracy of experimental data. Absolute risk measures the probability of an event occurring, while relative risk compares these probabilities across different groups to identify disproportionate impacts. Bias identifies systematic errors that cause experimental results to deviate from theoretical expectations.

1. Absolute Risk: The Probability of Failure

Absolute Risk is defined as the experimental probability that a specific negative or undesirable event will occur based on observed data.
It is calculated by dividing the number of times the 'risky' event occurs by the total number of trials or observations in the sample.
Mathematically, absolute risk is expressed as a value between $0$ and $1$ , where $0$ indicates no risk and $1$ indicates certainty of the event.

Formula: $\text{Absolute Risk} = \frac{\text{Number of occurrences of the event}}{\text{Total number of trials}}$

2. Relative Risk: Comparative Analysis

A bar chart comparing the absolute risk of two groups, illustrating how relative risk is the ratio between the two heights.

3. Understanding Statistical Bias

4. Key Distinctions: Absolute vs. Relative Risk

5. Exam Strategy & Common Pitfalls