Mat 318

Project 4 (25 pts) Due Friday 12/7

Linear Regression

Collect some real data on two quantitative variables that you think might be related. The data you obtain must be paired data and it must be a sample from a population. (It doesn’t have to be a random sample, because it is quite hard to obtain those.) You should have somewhere between 10 and 30 data points. You may not get the data from our textbook.

Your assignment is to WRITE A REPORT (2-3 pages) that discusses the relationship between the two variables. In this report you should provide the following:

1. A very brief discussion of why you thought the two variables might be related.
2. A brief description of how you collected the data.
3. A list of the raw data in a horizontal table.
4. A scatterplot with the regression line and the confidence/prediction interval.
5. An analysis of the data. There are many things that could be included in here, but you can’t put them all in. Decide what you think is important in your case. Discuss any problems you may see in the data, such as: outliers, clusters, non-linear relationship, other patterns in residuals.
6. Draw conclusions about whether the two variables are indeed related and whether the linear model is useful. Discuss also any restrictions of the model that arise from range of data and your sampling method.

You report should be self-contained, i.e. it should be understandable to someone who has NOT read this assignment. Do not just give answers to individual questions.

Finally, up to 5 points of extra credit may be given for special level of creativity or initiative in choosing the variables. This project does not have to be of serious nature. So

HAVE FUN!