Linear Regression Project

Does a linear relationship exists between two quantitative variables?

Maybe you are interested in knowing if

the height of a swimmer has an effect on their 50 meter sprint time
the grades one receives in high school can be a good predictor for how one does in college
the salary of an individual has a strong correlation with their level of happiness

Data Sources

Proposal

At the end of the linear regression unit you will put together a presentation similar to the one on the right (the Cricket Thermometers one). You haven't yet learned all of the components that you need to complete the presentation, however you can get started by finding some variables that you are interested in analyzing and add to the presentation as you learn more.

For this project:

You may work with one partner or by yourself.
Click through the Cricket Thermometers slides to see all of the components of the presentation, then watch the Math is Everywhere video down below to hear how former students tackled this project. If you want some more ideas, there is a folder with presentations from previous classes linked under the Cricket presentation that you can look through.
Create a question you are interested in answering and find the data to support answering that question. You may NOT use JMP sample data; it is old and probably not what you are actually interested in. Search the web for data sources or check out the Ever Growing List of Data Sources above.
Fill out this Google Form when you have decided on your research question and data set. Click here to see your classmate's responses if you want.
You can use this Presentation Slide Template to create your presentation. Please save your work here so I can show future classes.

8.2 Linear Regression Example Crickets

Need more examples? Have a look at these Presentations from previous classes.

Project Elements

Here's a list of all of the required elements for your presentation in a list: Presentation Grading Rubric, or if you prefer, below is a written description of these elements.

Introduction

Using your proposal, create the introduction. Explain your research question, the context of your question, and why it is an important one to study. Clearly state what your predictor (x) and response (y) variables are and where your data came from and how many samples are in your dataset. Include a link to your data if available.
If you are having trouble starting, try using this worksheet to organize your work. Fill this out first, then use it as an outline for your presentation.

Data

Describe the distributions of your two variables with some 1-variable statistics. Include graphs.
Calculate and interpret the correlation of your variables and create an appropriate scatterplot. Discuss the shape, strength, direction, and anything else extraordinary about your points. Include a graph.

Methods

Add a least squares line to your scatterplot (and optional residual plot), report the r² value and equation of the line. Include a graph.
Interpret the slope of the line, the y-intercept, and r² value.
Optional: Show and interpret the spread of the residuals.

Discussion & Conclusion

Determine if your linear model is a good model for the data you have selected (see notes on Regression Assumptions and Conditions).
Generalize your findings and summarize them for your audience.
Give at least three suggestions about how future studies could set out to answer the question you asked better next time.

Why do we do this project? Hear what students have to say.

Google Sites

Report abuse