- Bootstrap methods assume that our sample is an SRS and will give trustworthy conclusions only if this condition is met.
-
These methods make no assumptions about the distribution your data comes from.
- The Bootstrap Test for Two Population Means is an alternative to the two-sample $t$-test when the guidelines for its use are not met (such as when the data is strongly skewed or has outliers with low sample size).
- The Bootstrap Test for Two Population Proportions is an alternative to the two-large-sample $z$-test when the guidelines for its use are not met (such as an insufficient number of successes and failures).
| Sample 1 | Sample 2 | |
| Sample data goes here (enter numbers in columns): | ||
| Number of Successes | Sample Size | |
| Sample 1: | ||
| Sample 2: | ||
| Test for a Difference of Two: | ||
| $H_0: \mu$$\mu_0$ | ||
| $H_a:\mu$ $\mu_0$ | ||
| Level of Significance: $\alpha=$ | ||
| Number of Bootstrap Samples: |
| Sample Sizes: | $n_1=$ | $n_2=$ |
| Sample Means: | $\overline{x}_1=$ | $\overline{x}_2=$ |
| Sample Difference: | $\overline{x}_1-\overline{x}_2=$ | |
| Sample Proportions: | $\hat{p}_1=$ | $\hat{p}_2=$ |
| Sample Difference: | $\hat{p}_1-\hat{p}_2=$ | |
| Lower Critical Value: | $\delta_{L}^*=$ | |
| Upper Critical Value: | $\delta_{U}^*=$ | |
| $p\mbox{-value}$: |
| Frequency | ||
| Sample 1 and Sample 2 Data | Bootstrap Differences $\delta^*$ |