STAT 200 - Inference

STAT 200 - Inference

1. One Categorical: A single proportion

2. Two Categorical: Difference in proportions

3. Two Categorical: Chi-square test for two-way tables

4. One Quantitative: A single Mean

5. One Quantitative, one categorical: Difference in means

6. Two quantitative Variables: T-tests for slope and correlation

   • Scatterplot
     • Scatterplot with Regression Line
       • 1. Check linearity 2. Consistent variability
   • Correlation
     • Measure of strength and direction of linear association
     • Testing for Correlation
       • H0: p = 0 Ha: p ≠ 0 t = (statistic - null) / SE = (r - 0) / sqrt((1 - r^2) / (n - 2))
   • Regression line
     • response-hat = a + b * explanatory
     • Regression formula
       • y = B0 + B1x + e y = response variableB0 = interceptB1 = slope x = explanatory variablee = random error
   • Inference for Slope
     • Confidence Interval
       • Statistic +- multiplier * SE b1 +- t* * SE
     • Test for Slope
       • H0: B1 = 0    No linear relationship Ha: B1 ≠ 0    Some linear relationship t = (b1 - 0) / SE