In Sections 8.1 - 8.3, we learned how to use Hypothesis Testing to find a "statistically significant" result.
For instance, we found out in a previous example that Iceland's data, used as a sample, showed that the proportion of COVID-positive people who are asymptomatic is significantly higher than 25%, the current number put out by the CDC. So the question remains, HOW much higher? What does the sample data tell us about that? Is it possible to be more precise than just saying, "Oh, yeah, btw, 25% is too low..."?
That's what we're going to explore! Along the way, we'll learn about Margin of Error and what it means to be in a "statistical tie" in, say, a political race. We'll also see how Hypothesis Testing and Confidence Intervals relate to each other.
By the way, the National Institutes of Health (NIH) is conducting a study to address the question of how prevalent coronavirus infections are in the general population for people who haven't already been tested. They are gathering participants right now so you can contact them to put your name of the list and be part of important (and historical) research: https://www.nih.gov/news-events/news-releases/nih-begins-study-quantify-undetected-cases-coronavirus-infection
Tip: Review the 3.2 notes on z-scores and the Empirical Rule before going through this lesson.
Notes and Videos:
Confidence Intervals
Hypothesis Testing and Confidence Intervals
References: