Does a large sample of voter preferences (proportions) provide evidence to support a campaign's hypothesis? Learn about the significance test for proportions.
- A candidate's campaign finds…that in a random sample of 500 eligible voters,…54% of those polled said they planned on voting…for this candidate.…The candidate needs over 50% of the vote…to win the election.…This candidate would like to test the hypothesis…that he will win this election.…Let's go through our four step process.…Step one, we're going to develop the hypotheses…and state the significance level.…H sub zero, our null hypothesis,…will be p is less than or equal to 0.50.…
This hypothesis states that the candidate…would get 50% or less of the votes,…and thus not have enough of the votes to win the election.…H sub a, our alternative hypothesis,…the candidate wins.…This would be the opposite of the null hypothesis.…This one would say that this candidate…would get a majority of the vote and thus win the election.…Our alternative hypothesis is p is greater than 0.5.…
Our significance level for this test will be 5%.…If this has less than a 5% chance of occurring,…then we reject our null hypothesis.…We're looking at a one-tailed test,…
Eddie Davila first provides a bridge from Part 1, reviewing introductory concepts such as data and probability, and then moves into the topics of sampling, random samples, sample sizes, sampling error and trustworthiness, the central unit theorem, t-distribution, confidence intervals (including explaining unexpected outcomes), and hypothesis testing. This course is a must for those working in data science, business, and business analytics—or anyone else who wants to go beyond means and medians and gain a deeper understanding of how statistics work in the real world.
- Data and distributions
- Sample size considerations
- Random sampling
- Confidence intervals
- Hypothesis testing