Monday, May 21, 2007

Project instructions

II. Data Report (20 points)
Due Wed, May 30
This section should be a thorough explanation of how you collected your data and be a beautiful example of how much you've learned this year about the difficulties getting a representative sample.  Examples of what you should include are:
  • How you collected your data
  • Why you are confident your sample matches your population
  • Biases avoided and not avoided
  • A copy of any survey that was filled out
  • Your data:  either in "Excel" format or in a table/matrix summary.  You need at least 75 people/subjects for each group you want to compare.  Your data should be broken up by any category you want to compare.

III. Exploring the Data (25 points)
Due Fri, June 1
  • This section should be an outstanding example of exploratory data analysis (the first unit in our text).  Graphs should show the comparisons between all relevant groups you are comparing.  You should state any preliminary conclusions that can be drawn by using your eyeballs.
  • Graphs of your data
  • Statistics from your data
  • Descriptions of the graphs and statistics.

IV. Analyzing the Data (25 points)
Due Fri, June 1
Analyze your data using whatever method(s) is appropriate for your data.  Your conclusion should be nicely written using all appropriate statistical support.  Remember that confidence intervals can be a powerful method for comparing different groups.
  • Hypothesis Test (with conditions checked) and/or
  • Confidence Intervals (with conditions checked) and/or
  • Regression
  • Your Grand Conclusion!

V. Presentation (20 points)
2nd and 5th: Tuesday, June 12th
4th:  Friday, June 1st
Must include:
  • Clearly communicate your question and how you collected your data. (5 pts.)
  • Visually communication of some sort—you choose the method. (5pts.)
  • Clearly communicate your conclusion (5 pts.).
  • Be interesting to listen to and give us some sort of "hook" to inspire us to listen (5 pts).
  • Do not read off your visual aid—use note cards or your report.  Be careful how you communicate numbers to the class—too many numbers at once is confusing, as is too many decimal places.

VI. Success!? (15 points)
I will evaluate the overall success and difficulty of your project.  More challenging data collection issues add to your score.  Small sample sizes or an overly simple question will lower your score.

Details You Should Know
  • Tardies during the project will cost you points.
  • NO NO NO NO NO late work is accepted without penalty—even with excused absences.  No late presentations will be scheduled
  • Make your own copies anything you turn in.  Once you turn in one part of your report, I need to keep it.
  • Don't get lazy—this is to be the summation of what you've learned all year.  
  • Please type your paper.  Hand-done work is acceptable for some graphs, etc if it is done very neatly.

Friday, May 04, 2007

Choosing a test

Do this!!!

http://www.ltcconline.net/greenl/java/Statistics/StatsMatch/StatsMatch.htm


Note: the wording on this website is more vague than the AP test. The AP test always asks for "a 95% confidence interval" if it wants an interval (or some other %). Also, the AP problems are longer, and thus have more information that will guide you to deciding what to do.

However, I think that after you 10 or so of these, you'll get used to the wording of this site and it will be good practice. If it is just always frustrating, switch to finding the tests and intervals in your notebook. Every released test (A, B, C,...) has at least one.

Note: this website includes prediction intervals, which you do not
need to know.

Friday

See you this afternoon!

CW: Inverse Normal: '97 MC #12, '02 MC #10
Experiment vs Randomization: G-3a, M-3a, I-4a

Thursday, May 03, 2007

Thursday

H-2

Wednesday, May 02, 2007

Wednesday

No official homework tonight. I trust you're motivated to do some
studying on your own. The link below this post would make for some
great practice tonight.

Choosing a test

Here is a website you can use to practice choosing which procedure to
use for a given problem:

http://www.ltcconline.net/greenl/java/Statistics/StatsMatch/StatsMatch.htm

Note: this website includes prediction intervals, which you do not
need to know.

Tuesday, May 01, 2007

Tuesday

Pick 3 of the previous night's HW and do them (green sheets)
Also do: MC '97 #2, 5, 6, 16, 29