Study Notes on Confidence Intervals and Hypothesis Testing
Confidence Intervals and Population Parameters
- Confidence intervals provide a range of likely values for a population parameter (e.g., population mean, population proportion).
- Example: Given values create a confidence interval of (26.5, 34.5) with 95% confidence for bc (population mean).
Hypothesis Testing and Confidence Intervals
- If the null hypothesis bc_0 is within the confidence interval, we fail to reject the null hypothesis.
- If the null hypothesis bc_0 is outside the confidence interval, we reject the null hypothesis.
- This applies similarly to proportions (i.e., bphat).
Variable Types and Hypothesis Tests
- Determine the number of variables for the hypothesis test; only univariate tests are covered in exam three.
- Identify the type of variable:
- Categorical (binary or multiple categories).
- Quantitative (mean and standard deviation).
- Types of tests:
- T-tests are for quantitative variables.
- Z-tests are for proportions.
- Chi-squared tests for categorical variables with more than two categories.
Chi-Squared Goodness of Fit Test
- Used to determine if observed proportions fit expected proportions.
- Set hypotheses:
- Null: All expected proportions equal.
- Alternative: At least one proportion does not equal expected value.
- Formula: \text{Chi-squared} = \sum \frac{(O - E)^2}{E} (where O is observed count, E is expected count).
- Degrees of freedom: k-1 (where k = number of categories).
Statistical vs Practical Significance
- Statistical significance relates to probability (p-values); practical significance refers to the real-world relevance of a finding.
- A low p-value indicates the sample statistic is unlikely under the null hypothesis; however, it does not imply practical impact.
Exam Preparation and Expectations
- Familiarize with calculating confidence intervals and hypothesis tests.
- Understand the limit of reporting only statistically significant findings, leading to potential biases in interpretations.
- Recognize the distinction between significance and practical impact when analyzing research results.