Day 15: STA205 - Introduction to Statistical Methods

Last time

Next time

• Announcements:
  • Your exam is returned:
    • Your score is out of 100.
    • The Key
    • A curve: add 2.5 points to your raw score out of 100, top of the page.

• Today:
  • I wanted to start with a little demo that illustrates the importance of $\alpha$ in an Hypothesis test. Brandon and I were talking about the choice of $\alpha$, and this idea seemed to help him. So I decided to share it with everyone.

    I'll try to create a few more demos to illustrate topics, when it seems useful. For example, here's one about Confidence Intervals.

  • One more example, to show you why we do this stuff:

    • Headline: Fracking linked to increased hospitalizations for skin, genital and urinary issues in Pennsylvania: Rashes, urinary tract infections, and kidney stones requiring hospital stays are more common in areas with more drilling, according to a new study

      Researchers found that the more fracking wells were in a county, the more hospitalizations the county saw for genital and urinary problems like urinary tract infections, kidney infections, and kidney stones. Fracking, another name for hydraulic fracturing, is a process of extracting oil and gas from the Earth by drilling deep wells and injecting liquid at high pressure.

    • Now let's check out the research: Unconventional natural gas development and hospitalizations: evidence from Pennsylvania, United States, 2003-2014. I want to look specifically at Table 4. This is why we're studying the things we're studying....

  • Let's get into Packet 6, which has as its major objective determining whether two (categorical) variables are independent of each other or whether they are related to each other.

    I've mentioned this issue in other packets: for example, we're studying violent behavior in KY adults, but if our sampling frame excludes felons, say, we might be skewing our results (if there is indeed a relationship between violent behavior and felon status -- so we might want to study that!).

• Links:
  • The Standard Normal Table of probabilities.
  • Computing chi-squares in Statcrunch is carried out using the

    Stat -> Tables -> Contingency ->

    (then either data or summary, depending on what you have)

  • Here's a Mathematica demo about Chi-Squares.