MWF 1:10-2:00 Introduction to Statistics -- Fall 1997

Groups: You may choose to work alone (with no penalty), however groups of two or three people are encouraged. Be careful to pick people with whom you can work. Turn in a group roster with your project proposal.

Group Participation: Any member of your group must be able to present your results to the class (we will have some presentations at the end of the semester if there is time). Each member of the group will make a confidential recommendation on the distribution of the participation points. The instructor will take these recommendations into consideration when determining the distribution of points.

Presentation of results: You must present your results in an attractive pamphlet constructed of no more than six 8.5 x 11 inch pieces of paper (you may use both sides). Submit the original plus one copy for each member of your group in a standard 9 x 12 envelope. There is a collection of old projects that you can look at for ideas.

The Audience: The audience will be taken to be a PLNC graduate who satisfactorily completed this course more than five years ago. She or he will have forgotten all but the most basic formulas; all that remains is a sense of how one should approach a statistics problem.

Difficulty: The higher the difficulty level of the project the higher the grade. One option is to take a simple project and extend it in some way. For example, your group may choose to do more examples or treat a harder case.

Depth: Your group must decide how thoroughly to treat the project. A cursory treatment will result in a lower grade.

Digestibility: Clarity is essential. Statistical information that is poorly presented is of little use to the consumer. In keeping with this idea your pamphlet should be attractive, well organized and clear. The danger is that in the attempt to present too much information one can confuse the reader.

Format: You are expected to follow the project report format (see attached)

Basics: Proper grammar and spelling are expected. Correct computation is also expected.

Grading: The following list gives the number of points for each category:

Group Participation	15
Difficulty	15
Depth	30
Digestibility	20
Format	10
Basics	10

Your individual grade may differ from your group grade. 10 points will be deducted for each day that the project is late.

Project Description: In this section, you state the what (objective) and why (rationale) for your research. you should clearly state your objective (what are you investigating?) and your rationale (why are you doing this?).

Data Collection: In this section, you describe how you will collect your data. This section should include the following:

1. DESIGN: This should contain sufficient detail so that someone else could repeat your study.

• What is the population you are studying? What is the size of the population?
• How will you select your units from the population? If you are doing random sampling or randomization, be specific.
• What will be your sample size?
  

2. VARIABLES: This should include a complete description of all variables you are measuring and how each variable will be used to answer your research question.

• Describe each variable, in words.
• Specify whether each variable is quantitative or categorical.
• Describe, in detail, how you will measure each variable.
  

3. CODEBOOK: This tells how you plan to enter data into your spreadsheet.

• What variable names will you use?
• What are the expected range of values each variable can take on?
• Will you be using any codes? (for example, 0=male and 1=female)
• How will you enter the data in the spreadsheet? How many columns will you have, what will be their names?
  

Plans for Data Analysis: How will you analyze your data?

1. Each quantitative variable should be described by its distribution and appropriate summary statistics.
2. Each categorical variable should be described using appropriate tables and bar graphs.
3. Relationships between variables should be examined using appropriate graphs, tables, and statistics.
   
   
   
   

The project report must follow this format!

1. TITLE AND ABSTRACT

The first page of your report contains the title, centered horizontally, followed by the names of the team members, also centered horizontally, one name per line. The abstract follows the title, in about the middle of the page. What is an abstract? An abstract is a very brief summary of the contents of the report. It should not exceed a half page. The purpose of the abstract is to provide the reader with an overview of the report without having to read the details of the report. The abstract may contain a concise statement of the research question and the rationale for conducting the study. It also includes a short summary of the findings.

2. INTRODUCTION

In the introduction, you will discuss the objective and rationale for you research. You should again summarize your methods and findings, in more detail than in the abstract, but not in as much detail as in later sections. Any background information or pertinent literature should also be included.

3. METHODS

The methods section provides complete detail about your data gathering. In particular, you should address the design of your study and how the data was collected. This includes such details as blinding and randomization. In addition, you should clearly describe all variables that you measured and how you measured them. This includes a description of measuring instruments used (often, the brand and model is specified), how the instruments were used, and the unit of measure. This section also briefly describes the statistical methods used, along with a description of computer hardware and software used. For example:

We used histograms and boxplots to describe the distribution for all quantitative variables along with means and standard deviations (or medians and interquartile ranges) to describe the center and spread of the distribution. Categorical variables were displayed using bar graphs and tables. All data was analyzed using Excel on an IBM- compatible personal computer.

4. RESULTS

The results section provides any results of your statistical analyses, with appropriate reference to tables and graphs. you may include tables and graphs directly in the section, or may put them in an appendix. All tables and graphs should be clearly labeled. Tables should be numbered Table 1, Table 2, and so on, while graphs should be labeled Figure 1 etc. Each table or graph should also have a descriptive title. Graph axes should be clearly labeled with a descriptive variable name and unit of measure, if appropriate. Variables appearing in tables should also be descriptive. If using abbreviations, be sure to define them elsewhere in your report.

5. DISCUSSION

The discussion section is the place to interpret your results and to provide any explanations. you also should discuss any limitations of your study (to what population can you generalize your results?) you can also include any recommendations for future studies.

6. REFERENCES

Be sure to include any references, such as texts and computer programs.

7. APPENDIX

Include a listing of your raw data and any computer output. Be sure the appendix is well-organized. If necessary, "cut and paste" so that only relevant output is included.

Given below is a list of suggested projects. You may design your own project or choose one of the projects listed. Note that you may not use human subjects. This means that you may not do a survey of any sort. If your project requires data from a survey, the math department has a large data set and I can help you get the information that you want. It is always a good idea to talk to a professor in your major department for project ideas.

H.I.V. Testing: The state of Illinois mandated H.I.V. testing before a marriage license was granted. Discuss sensitivity and specificity in this test. What are the positive and negative effects performing this test on a large group of people? Should you be worried if your H.I.V. test comes back positive? See the American Family Physician July 1988 (pages 117-122), and January 1989 issue (page 53) at the UCSD Medical Library.

Library Books: Determine an appropriate sample size, and then select a random sample of Library shelves (or columns, row, etc.). Count the number of books on each of the selected shelves. What can you infer from this sample? How accurate is your inference?

Real Estate: Is proximity to the coast correlated with median resale price on single family dwellings? What is the appropriate type of regression needed to model this data? See San Diego Tribune 14-Feb-1993.

Game Show: Examine the odds of winning on an event in a TV game show. The Price is Right is full of games. Compute the probability of winning.

Counting Calories: Discuss the statistical methods and findings of the following paper. What are the implications of the findings? "Counting Calories--Caveat Emptor" in the JAMA September 22/29, 1993--Vol. 270, No. 12. (page 1454-56) This is available at the UCSD Medical Library.

Churches: Examine the district minutes of the Church of the Nazarene for the same district in two different years. Compute the ratio of attendance to membership in the first year. Compare this with the gain or loss in church attendance between the two sets of minutes. Interpret you results. These are available through Ryan Library.

Census or NORC data: There are on-line survey data sets that can be analyzed. Talk to a sociology professor or come see me for more details.

Common Stocks: Are common stocks a hedge against inflation? Update the data attached and compare 1970-1986 regression results to the ones obtained from the post-W.W.II period through 1969.

Stock values and consumer prices, post-World War II period (through 1969)

Rate of change, % per year

	Country	Stock prices, Y	Consumer prices, X
1.	Australia	5.0	4.3
2.	Austria	11.1	4.6
3.	Belgium	3.2	2.4
4.	Canada	7.9	2.4
5.	Chile	25.5	26.4
6.	Denmark	3.8	4.2
7.	Finland	11.1	5.5
8.	France	9.9	4.7
9.	Germany	13.3	2.2
10.	India	1.5	4.0
11.	Ireland	6.4	4.0
12.	Israel	8.9	8.4
13.	Italy	8.1	3.3
14.	Japan	13.5	4.7
15.	Mexico	4.7	5.2
16.	Netherlands	7.5	3.6
17.	New Zealand	4.7	3.6
18.	Sweden	8.0	4.0
19.	United Kingdom	7.5	3.9
20.	United States	9.0	2.1

Source: Common Stock Values and Inflation: The Historical Record of Many Countries, National Bureau of Economic Research. Suppl. March, 1974. table 1. p. 4.

Retail Sales: Using 1990 Census data, and figures from the California State Board of Equalization and/or the Franchise Tax, can you say there is a "strong" correlation between the population of a California county and its retail sales? (Note: this may not be an obvious "yes" for many reasons.)

Brochure Mailing: Using 1990 Census data: if I have enough money to mail 10,000 brochures about my new clothing boutique to "older" women in San Diego county, what percent of the potential market ("older" women) will I be able to cover with the mailing? What percent of San Diego county zip codes does this represent?

Market Share: Using the Pacific Bell Yellow Pages, how many auto dealers in San Diego are there for each auto manufacturer? What percent of the total dealerships does each manufacturer represent, e.g. Toyota, Ford, etc. Can you infer San Diego market share of each manufacturer based on this information? Why?

Research Paper: Discuss the statistical methods and findings of the following paper. What are the implications of the findings? The paper is titled "Premature Rupture of the Membranes and Sepsis in Preterm neonates" and is in Nursing Research January/February 1991 Vol. 40, No. 1 pages 36-41.

Research Paper: Discuss the statistical methods and findings of the following paper. What are the implications of the findings? The paper is titled "Effect of the Position On Cardiovascular Response During the Vasalva Maneuver" and is in Nursing Research July/August 1990 Vol. 39, No. 4 pages 198-202.

Research Paper: Discuss the statistical methods and findings of he following paper. What are the implications of the findings? The paper is titled "Prevalence of Cigarette Smoking in Hispanic Women of Childbearing Age" and is in Nursing Research March/April 1991 Vol. 40, No. 2 pages 103-106.

Research Paper: Discuss the statistical methods and findings of the following paper. What are the implications of the findings? "AIDS-Related Risk Factors, Medical Diagnosis, Do-Not-Resuscitate Orders and Aggressiveness of Nursing Care" and is in Nursing Research November/December 1990 Vol. 39, No. 6 pages 350-354.

Research Editorial: Discuss the statistical comments and the thought processes involved in getting work published. The editorial is titled "Scientific Conversation" and is in Nursing Research May/June 1988 Vol. 37, No. 3 page 131.

Research Paper:  Discuss the statistical methods and findings of the following paper. What are the implications of the findings? The paper is titled "Shared Values: Impact On Staff Nurse Job Satisfaction and Perceived Productivity" and is in Nursing Research May/June 1989 Vol. 38, No. 3 pages 172-177.

NORC data: There are on-line survey data sets that can be analyzed. Select three or more subgroups from the american population and run an analysis of variance on some characteristic. Interpret the results. For more help, talk to a psychology professor or come see me for more details.