STATS Univ of the Pacific Workshop, June 9-16, 1996

Project Assignment

By the end of the week, your team is to make both an oral and a written presentation of your project. I have given various guidelines in e-mail messages and will continue to advise you during the week. This handout shows how I "pace" my intro stat class during the semester to do a similar project.

--- Mary Parker

Sunday: Agree on the project. Begin on non-text quiz E here. (Third page)

By Tuesday: Complete non-text quiz E. This should include any revision of how you will collect the data.

By Thursday: Complete the data collection and begin the analysis. (It is OK to start thinking about and working on the analysis part before you complete the data collection.) Instructions for my class reports are on the last page. You need not follow these exactly if you have better ideas. Do incorporate the discussion of sampling, experimental design, and the inferential analysis in your report, since you already know that material.

By Saturday: Complete the project report.

*********************************************************************************

Math 1563 -- Parker's sections

Non-text quiz A: (Due date to be announced. First week.) Credit: up to 3 points

Example: I might want to measure the lengths of cats. The cases (also called individuals) are cats, and the variable is "length", in inches, measured to the nearest inch. Some questions which occur are (1) Should I include the tail? and (2) If I include the tail, do I measure to the tip of the fur or just to the tip of the tail without the fur? Other variables which might be interesting are "weight" and their "eye color".

Think of some variables which you might be interested in analyzing sometime.

  1. Describe what a single case is.
  2. Describe three different variables that could be measured on a case.
  3. Pick one of the three variables (underline it) and describe how you would measure it and what the units will be. Are there likely to be any problems or ambiguous situations?
  4. For that variable, give four observations that would be reasonable to see.
  5. Do you have a calculator? Answer "yes" or "no" for this quiz question, and, if the answer is "no", explain when you will get one. Bring your calculator and your manual for the calculator to class. If you already have a calculator, even if you don't have the manual, don't go buy another one yet. Maybe another student has a similar calculator and will let you copy the relevant pages of their manual. If you are buying a new calculator, be sure to get one that does two-variable statistics. They are not more expensive than those which only do one-variable statistics, but you will need to look carefully to make sure that is what you are getting.

Non-text quiz B: (Due date to be announced. Second week.) Credit: up to 3 points

Use this set of numbers: 4, 5, 8, 9, 9

  1. By hand, showing all work, find the mean.
  2. By hand, showing all work, find the standard deviation.
  3. On your calculator, find the mean.
  4. On your calculator, find the standard deviation.
  5. Write down notes about exactly how to do these two computations on your calculator. (I want to see them, but the point is for them to be useful to you. Write as much or as little as you will need two weeks from now to remember how to do this.)
  6. Does your calculator do two-variable statistics? Tell me how you know.

If you are having trouble getting started with your calculator, read through these questions and your manual and try to answer them. If you have trouble with any of them, see me before the next class.

Non-text quiz C: (Due date to be announced. Start looking for the example immediately.) Credit: up to 6 points

In some newspaper, magazine, or some other book besides a statistics textbook, find an example of two-variable data that you are (at least mildly) interested in. Both variables must be quantitative variables (not categorical). (You might want to read pages 92-96 first.)

  1. Give the reference.
  2. Describe what a case (individual) is.
  3. What are the two variables and in what units are they measured?
  4. Which is the response variable and which is the explanatory variable?
  5. Why would someone want to predict values of this response variable? Of what use is it?
  6. List the data.
  7. Sketch a scatterplot of the data.
  8. Guess the correlation coefficient. (Is it positive or negative? Close to 0 or far from 0?)
  9. Give the correlation coefficient of the data. (If you don't have a two-variable calculator, use MINITAB or else get the instructor or another student to find this for you on their calculator.)
  10. Give the equation of the regression line to predict the response variable from the explanatory variable.

Non-text Quiz D: (Due date to be announced -- probably during Week 4) Credit: up to 9 points.

This is the first part of a project and should be done individually, although you may discuss it with each other as much as you wish. Later parts of the project should be done in groups of 2 or 3 people.

1. Formulating the question.

Think of a question of interest to you for which you could collect some data.

Examples:

a. What percentage of cars don't completely stop at the stop sign at the north corner of Waterston and Hartford?

b. In a blind tast test, do people prefer Classic Coke to Pepsi?

c. How accurately could we predict hat size from forearm length?

Write the question and explain how this question is of interest to you or someone. How might accurate knowledge about the answer change someone's actions?

2. Identifying relevant variable(s).

Explain what the variable(s) is and how you expect to measure it. Discuss any definitions needed. (Remember the example about the length of a cat -- does that include the tail or not?)

3. How would you collect useful data?

Is this a question for which you just go collect data or do you have to set up an experiment? (Example b above requires you to set up an experiment.) Explain how you could get a reasonable sample or set up a reasonable experiment. Describe the population in which you are interested and explain how this method of collecting data should get representative data from this population. I don't expect a sophisticated answer here. I am deliberately assigning this before we talk about sampling and experimental design in detail.

Non-text quiz E: (Due date to be announced. Approximately week 6.) Credit: up to 3 points each.

Form a group of 2 or 3 people and choose a question to investigate. The question can be one that one of the people has suggested already or a new one. Use the feedback provided on Non-text quiz D as a guide. Go through all the steps of Non-text quiz D on this question until you agree on them. (At this stage, you know more about sampling and experimental design than you knew before, so use that new knowledge as much as you can.) Then measure a few data points and revise your ideas of what you will do as needed.

As one paper from your whole group, turn in

Non-text quiz F: (Due date to be announced. Approximately Week 9.) Credit: Up to 9 points each.

Using the same group and question as in Non-text quiz E, turn in a project report with your tentative answer to the question you originally asked, and supporting information. (A handout describing this report is included here.) This supporting information is the data you collected (with a description of how you collected it), an analysis based on descriptive statistics and exploratory data analysis, and a discussion of the limitations of your study. Later in the semester we will study inferential statistical techniques, which are more appropriate for actually drawing conclusions based on data.

Non-text quiz G: (Due date to be announced. Approximately Week 14.) Credit: Up to 9 points.

This quiz must be written and turned in by individuals, rather than a group. However, you may work with others to analyze and discuss this. This will be a further analysis of the data in your project from Non-text Quiz F, but with inferential techniques applied and an analysis of whether your experimental or sampling design has given you data for which inferential statistics are appropriate.

Descriptive Statistics Project Report (Group work)

Your report must be typed or written very neatly and readably. It should include:

  1. A half-page executive summary. This is a report to an intelligent reader who has not taken a statistics course, who will not read the full report, but who needs to know what you studied and what you found. In this summary, you should describe your research question, briefly describe the methods used, report your findings and what they mean, and how they compare to what you expected to find.
  2. A brief description of your question and why it is important.
  3. A description of the variables used to investigate the question, the question stated in terms of the population parameter(s), and a description of the population.
  4. The method of data collection, including a discussion of why you think your sample is representative of the population and any difficulties encountered in the collection of data.
  5. The raw data in some unobtrusive way. (In an appendix is OK.)
  6. A clear statement of conclusions with supporting graphs, data summaries, and explorations. (If the data is inconclusive about your question, that is OK. That often happens in real life. Be sure to state it clearly.)
  7. A discussion of what actions you would recommend on the basis of your study and/or what questions are suggested or left unanswered by your study.
  8. A description of the limitations of your study.
Extension to Inferential Statistics (Individual work)
  1. Is is reasonable to treat your data as if it is a simple random sample from the population?
  2. What type of inferential statistical technique would be useful to help you determine what information your data has about your question.
  3. Carry out that inferential statistical technique. Describe the result and discuss what this says about your question of interest.