Project A Analyze Your Own Data-Inference for a Difference in Means

This project asks you to collect and analyze your own data for a difference in means. You must use a statistical software package to perform the analysis and you should incorporate output from the package to verify your answers. All results should include both computer output and a discussion of the results. You should submit a well-written report addressing each of the following parts.

Part 1 Clearly state what you are investigating. What is your population and what two groups are you comparing? What is your quantitative variable of interest? Define the parameter(s) of interest. (Note: Several suggestions are given below or come up with your own idea.)

Part 2 Describe clearly how you collect your data.

Remember Chapter 1! Devise a way to efficiently collect data in a way that will be relatively unbiased. After you have described your sampling method, discuss any ways that your sampling method may bias the results. (Note: If collecting your own data is not feasible, perhaps you could sample your statistics class.)

Part 3 Collect your data!

Include a copy of the dataset in your report. If appropriate, include a reference for the source of your data.

Part 4 Describe your data.

Include a graph of the data in your quantitative variable and include summary statistics (mean, standard deviation, five number summary). Discuss the results. In addition, give summary statistics (mean, standard deviation, and sample size) for each of the two groups. Include a comparative plot to compare the two groups and discuss what you see in the graph and the comparative statistics.

Part 5 create a confidence interval for the difference in means between your two groups.

Include the output from the software package you use (choose your own confidence level) and be sure to interpret your result in the context of your data situation.

Part 6 Test whether your data show evidence for a difference in means between the two groups. Specify the appropriate null and alternative hypotheses. Use a two-tailed alternative, unless you have some reason in advance to suspect a difference in a particular direction. Include the output from the software package you use and use the p-value to make a conclusion (in context) about the evidence for a difference in means between your groups.

Topic Suggestions

Determine whether there is a difference in average: Number of hours spent watching television in a week, between males and females Life expectancy, between countries in the Eastern and Western hemispheres

Number of hours spent studying per week, between first-year students and upper-class Students.

Number of pitches thrown in baseball games, between the National and American leagues

Price of textbooks, between introductory and upper-level courses