Lecture_7-mas
Introduction
Course Information: BB1719, Spring 2024, Introduction to Data Analysis
Instructor: Dr. Michelle Sahai
Lecture Focus: Probabilities, P-value, and Hypothesis Testing
Learning Outcomes
Understand key concepts, including:
Null hypothesis (H₀) and alternative hypothesis (H or H₁)
Type 1 error vs. Type 2 error
Power of a test
P-value and its interpretation
Hypothesis Testing Steps
Set the Hypothesis
Null Hypothesis (H₀): Assumes no effect or difference.
Alternative Hypothesis (H₁): Assumes some effect or difference exists.
Set the Significance Level (α)
Determines the criteria for a decision (commonly 0.05).
Compute the Test Statistics
Make a Decision
Determine if there is a statistically significant difference between the two groups.
Example Groups: Control group vs. Intervention group.
Null and Alternative Hypothesis
Null Hypothesis (H₀): States there is no difference between groups.
Example: Mean control = Mean intervention group.
Alternative Hypothesis (H₁): States there is a significant difference.
Example: There is a difference in blood sugar levels.
Type 1 and Type 2 Errors
Type 1 Error: Incorrectly rejecting H₀ when it is true (False Positive).
Probability of committing a Type 1 error is represented by α.
Type 2 Error: Failing to reject H₀ when it should be rejected (False Negative).
Probability of committing a Type 2 error is represented by β.
P-value and Interpretation
Definition: The probability of observing results as extreme as those in the data, assuming H₀ is true.
Interpreting P-values:
Low P-value (< α): Reject H₀ (suggests a statistically significant effect).
High P-value (> α): Fail to reject H₀ (suggests no significant effect).
Factors Impacting P-value
The significance level (α), typically set at 5%.
A p-value measures the evidence against the null hypothesis.
Assessing Statistical Significance
A comparison between two groups can help determine if a significant difference exists based on test results.
Types of Statistical Tests
t-Test: Compares the means of two groups.
ANOVA: Compares means across three or more groups.
Chi-Squared Test: Compares categorical variables and proportions.
Regression Analysis: Analyzes the relationship between two or more variables.
Power of a Test
Definition: The probability of correctly rejecting H₀ when it is false (i.e., detecting a true effect).
Calculation: Power = 1 - β (where β is the probability of Type 2 error).
Desirable Power Level: Studies should aim for a power of ≥ 0.8 (or 80%) for reliable results.
Influencers of power include sample size, effect size, and precision of results.
Summary of Key Concepts
Understand the relationship and differences between:
Null Hypothesis (H₀)
Alternative Hypothesis (H₁)
Errors (Type 1 and Type 2)
Interpretation of p-value
Concept of Power in hypothesis testing.