Question
Bootstrapping can approximate a statistic’s unknown population distribution named after this process. For 10 points each:
[10e] Name this process, whose varieties include “stratified” and “simple random.” This process can be done with or without replacement to select a subgroup of a population.
ANSWER: sampling [or word forms such as sample; accept simple random sampling or stratified sampling; accept sampling distribution; prompt on surveying or polling and equivalents with “What process is used to choose the participants?”]
[10h] Bootstrapping can estimate the standard deviation of the sampling distribution, which is used to approximate the margins for these constructs. The analog to these constructs in Bayesian statistics are described as “credible.”
ANSWER: confidence intervals [prompt on intervals] (Frequentist statistics uses confidence intervals, while Bayesian statistics uses credible intervals.)
[10m] If a statistic falls outside a confidence interval, then this value for the statistic must be less than the significance level. “Hacking” this value can lead to erroneous statistically significant results.
ANSWER: p-value [accept p-hacking]
<Other Science>
Summary
| 2024 ACF Fall at Ohio State | fall | Y | 2 | 15.00 | 100% | 50% | 0% |
| 2024 ACF Fall at Washington | fall | Y | 1 | 30.00 | 100% | 100% | 100% |
| 2024 ACF Fall at Georgia | fall | Y | 1 | 20.00 | 100% | 100% | 0% |
| 2024 ACF Fall at North Carolina | fall | Y | 3 | 13.33 | 67% | 33% | 33% |
| 2024 ACF Fall at Illinois | fall | Y | 9 | 20.00 | 100% | 89% | 11% |
Data
| UBC A | UBC B | 10 | 10 | 10 | 30 |