In Chapters 9 and 10, we look at hypothesis testing in different contexts. It’s important to understand that hypothesis testing is just how statisticians (by “statisticians” I mean people who are working with data, e.g. analysts, auditors, economists, whoever) answer quanitative questions. We intuitively know that not much is truly certain, and that it’s really hard (if not impossible) to collect data on everything we are curious about. So we use samples and try to extrapolate to a larger population. For this discussion, I would like each of you to come up with your own interesting hypothesis test (extra credit if you do something similar to those in Chapter 10). So, you will do the following:
1) Explain the context or story for which you have a problem stats might be able to help with.
2) Come up with the null and alternative hypotheses. This can be in words, or numbers, or a combination of both (I’ll give an example below).
3) Explain where/how you might collect data to reach a conclusion for the test.
Example:
I do a fair amount of competitive downhill mountain biking. Naturally, I would like to win more races. One way to win more races is by going faster. So, I would like to figure out what would improve my speed. One theory I might have is that I’m a more confident and faster rider if I have caffeine in my system. Here’s what this test might look like:
Null: Caffeine doesn’t affect my performance.
Alternative: Caffeine makes me a faster rider.
To test this, it would probably be best to ride the same trail (one I’m familiar with) many times to get a pretty big sample. The only thing I’ll vary between each ride is how much coffee I drink. I’ll keep everything else the same (time of day, weather, etc). I’ll keep detailed notes of how much coffee I had and my riding time for that session. Then I can find my average riding time for each situation (lots of coffee or a little) and perform the differences in means hypothesis test above.
Last Completed Projects
| topic title | academic level | Writer | delivered |
|---|