Michael W (Michaelw714) · YouTube · 19:08
Use repeated polling simulations to visualize a sampling distribution and the reason estimates vary.
Use the learning inspector for key ideas and moments, or open the reading tabs for the complete notes.
Generated from the video's visuals and explanation; not verbatim speech.
The historical polling example reports a percentage with a margin of sampling error. A 43% estimate with a three-percentage-point margin gives 40% to 46%; this is an illustration of the reported interval, not a current poll.
Draw many random samples of the same size, calculate the support proportion in each, and plot the results. This distribution concerns a statistic across samples, not individual voters within one sample.
With a suitable random-sampling model and sufficiently many expected successes and failures, a sample proportion has an approximately normal sampling distribution. More simulation repetitions reveal that distribution; they do not change the sample size used in each poll.
The historical polling example reports a percentage with a margin of sampling error. A 43% estimate with a three-percentage-point margin gives 40% to 46%; this is an illustration of the reported interval, not a current poll.
Draw many random samples of the same size, calculate the support proportion in each, and plot the results. This distribution concerns a statistic across samples, not individual voters within one sample.
With a suitable random-sampling model and sufficiently many expected successes and failures, a sample proportion has an approximately normal sampling distribution. More simulation repetitions reveal that distribution; they do not change the sample size used in each poll.
The historical polling example reports a percentage with a margin of sampling error. A 43% estimate with a three-percentage-point margin gives 40% to 46%; this is an illustration of the reported interval, not a current poll. Draw many random samples of the same size, calculate the support proportion in each, and plot the results. This distribution concerns a statistic across samples, not individual voters within one sample. With a suitable random-sampling model and sufficiently many expected successes and failures, a sample proportion has an approximately normal sampling distribution. More simulation repetitions reveal that distribution; they do not change the sample size used in each poll.
YouTube11:26Another explanation of Sampling.
Sampling
YouTube11:26Next in Probability and statistics: Sampling → Hypothesis testing.
Hypothesis testing · View learning sequence
YouTube5:31Prerequisite in Probability and statistics: Random variables → Sampling.
Random variables · View learning sequence
YouTube12:03Prerequisite in Probability and statistics: Random variables → Sampling.
Random variables · View learning sequenceYouTube11:26Whiteboard for Sampling, compared with the current lecture.
Sampling