MATH 1342 – Day 1 handout.

Agenda for Week 1:

1. Learn about producing data. Chs 8 & 9 and Data Ethics Commentary.

2. Learn to use StatsPortal and your text.

3. Learn what is expected in this course.

4. Do something in each of the software packages:

a. Minitab in our classroom. Make a dotplot (Graph > Dotplot) for the data in Chapter 1, exercise 1.31. (No interpretations yet!)

b. CrunchIt at home. Make a dotplot (Graph > Dotplot) for the data in Chapter 1, exercise 1.31 (No interpretations yet.)

c. StatsPortal Statistical Applet “One Variable Statistical Calculator” and pick the first dataset in the list of available datasets. Then explore what happens as you change the number of intervals – between about 6 intervals and 20 intervals. Can you see that the same general shape of the distribution is visible at all of these? We’ll talk about this in Week 2. This week, I mainly want you to find out if you’re going to be able to use the applets at home or if you need to come to the computer lab.

Day 1:

Goals for today’s class

1. Help students understand the expectations in the course.

2. Elicit some ideas students have about designing a useful statistical study.

3. Introduce the basic ideas of experimental design. (Chapter 9)

4. Show how to use a Random Digits table to do random assignment of subjects to treatment groups. (Chapter 8)

What we do in class

1. Go over first-day handout. Pay careful attention to

what you are graded on

when things are due

what you have to do to remain enrolled in the course

2. Fill out info sheet and turn it in.

3. Sign attendance sheet. (VERY important that you do this every day you are here.)

4. Introduce yourself to two people and tell them why you’re taking the course.

Now let’s think about this week’s topic – Producing Data.

1. (2 minutes) Suppose that you need to plan a study to test whether Vitamin E, given orally, reduces the time it takes a burn patient to heal. How would you design a useful study? Don’t write sentences now, just a few words to remember what you think of.

2. (3 minutes) Pick one other person and talk to them. Do you have any new ideas for how you might design a useful study? Add some more words (possibly mark out some of your previous ones.)

3. Here are three principles of experimental design:

a. Control the effects of lurking variables on the response variable, most simply by comparing two or more treatments.

b. Randomize – use chance to assign subjects to treatments.

c. Use enough subjects in each group to reduce chance variation in the results.

4. Talk to your partner some more. Add to your few words here to remind yourself of what you think might be a good design.

5. (With partner) Do you think there is only one correct design for a study like this?

6. (With partner) Do you think it would be possible to design a study like this badly? Discuss.

7. As a class, we’ll talk through one way to do a study like this. It’s not necessary to write down this design. Pay attention and make a note of any words we use to describe parts of it if you hadn’t used those words yet. (If you forget what we say about them, you can look them up in the Vocabulary Resource in StatsPortal.)

8. What if we had wanted to study the effects of BOTH Vitamin E and Vitamin C on burn patients? Would the best thing be to use two separate studies – one for each Vitamin? If not, why not?

9. Let’s suppose we want to include both vitamins in the same study, and that we have four levels of Vit C (0 mg, 1000 mg, 2000 mg, and 5000 mg) and three levels of Vit E (0 mg, 400 mg, 800 mg). Make a list of all the combinations of those. Each combination is a treatment. How many treatments are there? If this were a different study with two factors (two different variables you are testing) will you always have to make the list to know how many treatments? How could you find the number of treatments more quickly than making a list?

10. Now, today’s material was all about Chapter 9 because it’s a little harder to get into and to read than Chapter 8. Most of Chapter 8 is best learned out of class by reading on your own. But you can skip over the part about using the Random Digits Table to find a Simple Random Sample. We’ll talk about that in class next time. Most students find that is easier when we discuss it in class first before you try to read about it

Quiz 1. Due Wed. Jan. 23 at the beginning of class.

8.38

8.40

9.30

9.36. As part of (a) identify the response variable, factors, levels for each factor, treatments, and how many subjects are needed. For (b), draw a picture like EITHER Fig. 9.4 (p. 231) or Fig 9.5 (p. 237).

Software

a. Data from exercise 1.31. Make a dotplot. Print out EITHER the dotplot from Minitab (use the IRT computer lab or the Learning Lab) or take a screenshot of your output from CrunchIt and print that. This should take no more than 5 minutes. If it is taking longer, answer this question by writing a note about what you still need to do and whether you need help.

b. Using the Statistical Applet “One-variable statistical calculator.” Find a computer on which you can use the applets, and follow the directions at the top of this handout, and then write down whether you see the same general pattern for all the values from 6 to 20 bins.