Math 216: Statistical Thinking
Small Sample Statistical Challenges
Normality Assumption: With samples smaller than 30, the Central Limit Theorem may not apply effectively, requiring the population distribution to be approximately normal for valid inference.
Standard Deviation Uncertainty: Using sample standard deviation \(s\) instead of population \(\sigma\) introduces additional variability that must be accounted for in our calculations.
Key Statistical Issues:
Statistical Significance: These challenges necessitate specialized methods like the t-distribution for valid small-sample inference!
For small samples where the population standard deviation is unknown, we use the t-distribution to construct confidence intervals: \[ \bar{x} \pm t_{\alpha/2, df} \cdot \frac{s}{\sqrt{n}} \] This formula accounts for the additional uncertainty inherent in small samples.






R Functions for Confidence Interval Calculations
Critical Value Calculations for 95% Confidence Intervals:
qnorm(0.975) = 1.960qt(0.975, df=10) = 2.228qt(0.975, df=5) = 2.571qt(0.975, df=2) = 4.303Confidence Interval Formulas:
Key Insight: As degrees of freedom increase, t-distribution approaches normal distribution, and confidence intervals become narrower for the same confidence level!
Practical Interpretation:
Practical Exercises Using R
Exercise 1: t-Distribution Critical Values for Confidence Intervals
qt(0.975, df=8) = 2.306qt(0.995, df=15) = 2.947qt(0.95, df=20) = 1.725Exercise 2: Small Sample Confidence Interval Calculation
50 + c(-1,1) * qt(0.975, df=9) * 8/sqrt(10)Exercise 3: Confidence Interval Width Comparison
Practical Applications Using t-Distribution
Application 1: Medical Research Interpretation
Application 2: Environmental Monitoring
Application 3: Quality Control