Hypothesis Testing Key Concepts and Errors
Hypothesis Testing Overview
Hypothesis testing is a statistical method used to determine if there is enough evidence to reject a null hypothesis based on sample data.
Confidence Levels
A common confidence level used in hypothesis testing is 95%. This means that if the procedure were repeated many times, approximately 95% of the confidence intervals constructed would contain the true population parameter.
Definitions
Null Hypothesis (H₀): A statement asserting that there is no change or difference, the status quo.
Alternative Hypothesis (H₁): The claim or statement for which we are trying to find evidence. This typically suggests a change or a difference.
Statistical Notation
X: Sample mean
S: Sample standard deviation
Z: Z-score for the confidence level used in hypothesis testing
Interval: The range within which we expect the population mean (μ) to fall with a certain level of confidence. If the sample mean (X̄) falls within this interval, H₀ cannot be rejected, supporting the idea that the sample mean is close to the population mean.
Method of Hypothesis Testing
Formulate hypotheses: decide on the null (H₀) and alternative (H₁) hypotheses.
Collect sample data to calculate the sample mean and standard deviation.
Determine whether the sample evidence is sufficient to reject the null hypothesis.
Example of Hypotheses
Null Hypothesis (H₀): μ = 25
Alternative Hypotheses:
H₁: μ < 25
H₁: μ > 25
H₁: μ ≠ 25
Type I and Type II Errors
Type I Error: Rejecting the null hypothesis when it is actually true. In a judicial context, this would mean declaring someone guilty when they are actually not guilty (i.e., the null hypothesis of innocence is rejected).
The probability of making a Type I error is denoted by α.
Example: Concluding that more than 2% of children have headaches from a new antibiotic when it is actually less than 2%.
Type II Error: Not rejecting the null hypothesis when the alternative hypothesis is true. This would mean concluding someone is not guilty when they actually are guilty.
The probability of making a Type II error is denoted by β.
Example: Concluding that only 2% of children have headaches from an antibiotic when in fact more than 2% do.
Practical Application Case Study
Medco Pharmaceutical Company Scenario
Context: Researchers from the FDA are testing whether the percentage of children experiencing headaches from a new antibiotic is greater than 2%.
Hypotheses formulated:
Null Hypothesis (H₀): p = 0.02 (2% experience headaches)
Alternative Hypothesis (H₁): p > 0.02 (more than 2% experience headaches)
Type I Error: Incorrectly concluding that more than 2% of children have headaches when the true percentage is actually less than 2%.
Type II Error: Failing to reject the null hypothesis when in fact more than 2% of children experience headaches from the antibiotic.
Conclusion Statement After Hypothesis Testing
If the null hypothesis is rejected: There is enough evidence to support the claim that the alternative hypothesis is true.
If the null hypothesis is not rejected: There is not enough evidence to support the claim for the alternative hypothesis.