7.01
Module Overview
Module 07: Focuses on Mean and Slope in hypothesis testing, particularly using t-tests versus z-tests.
Hypothesis Testing
One-Sample Mean
z-Test vs. t-Test:
z-Test: Used when the population standard deviation is known.
t-Test: Used when the population standard deviation is unknown; relies on the t distribution.
Key Similarity: The process of conducting both tests is largely the same, with adjustments for the t-test.
Considerations for t-Test
Normality: Check if the original population is approximately Normally distributed.
Sample Size: The sample size influences the robustness of the t-test.
n ≥ 30: Assume Normality; the sample is considered sufficiently large.
15 ≤ n < 30: Verify approximate Normality without outliers; proceed cautiously.
n < 15: Verify approximate Normality without outliers or skewness; proceed with extreme caution.
Data Visualization: Use box plots, histograms, or Normal probability plots to assess Normality.
Properties of t Distribution
Critical Value: Unlike the standard Normal distribution, the t distribution calculates the area right of the critical value.
Visualization: Review how the t distribution differs in appearance from the standard Normal distribution to understand implications for hypothesis testing.
Practical Application: Inflatable Gorilla Factory Example
Scenario
Quality Assurance Team: Needs to determine if fans last longer than 35 days with a 95% confidence level.
Hypothesis:
Null Hypothesis (H0): µ = 35 days
Alternative Hypothesis (H1): µ > 35 days
Sample Data
Sample Size (n): 18
Sample Mean (x̄): 37.9 days
Sample Standard Deviation (s): 8.4 days
Test Statistic Calculation
Formula: ( t = \frac{x̄ - μ_0}{s/\sqrt{n}} )
Plugging in values: ( t = \frac{37.9 - 35}{8.4/\sqrt{18}} \approx 1.465 )
Finding the p-Value
Determine p-value for t = 1.465 with degrees of freedom (df) = n - 1 = 17.
Calculation Methods: Use a calculator or t distribution table to find ( P(t \geq 1.465) ).
Result: p-value ≈ 0.0806.
Conclusion
Interpretation: The p-value is greater than the 5% significance level, leading to the conclusion:
There is insufficient evidence to reject the null hypothesis (H0) that population mean is equal to 35 days.
Therefore, it cannot be concluded that the inflatable gorilla's fan lasts longer than 35 days.