4 Learning to use R statistical software for data mining — Putting it all together

By Dail Midgette

Provided by Kenan Fellows Program.

In Algebra I, students study data sets with one predictor variable and one response variable. However, in the real world, most response variables have numerous predictor variables, many of which may have a significant impact on the data. These different variables may also have differing effects on the situation at hand, so it is important to identify their effects and then use them appropriately to make sound, more valid predictions.

In this lesson, students will create their own data set and then use the free R statistical software to mine their data in an attempt to identify the variables that most significantly impact the selling price of a home. Students will utilize multiple methods of variable selection — forward selection, backward selection, and stepwise selection — in an attempt to determine which variables are most influential.

Learning outcomes

At the end of this lesson, students should be able to use R to:

• Identify influential variables in a multivariate data set using forward selection, backward selection, and stepwise selection
• Develop linear models that can be used to make predictions

Students should also be able to:

• Explain the purpose of data mining
• Understand the basic principles behind the forward, backward, and stepwise selection processes
• Identify some real-world uses of data mining
• Use linear models to make predictions

Teacher planning

Time Required

To complete this lesson, one block period or two traditional periods (for a total of approximately ninety minutes) would be necessary. This is the third lesson in a set of three.

Student handouts

Technology resources

• One computer per student with internet access
• One computer with projector for teacher
• R statistical software
• Microsoft Excel
• MLS Search Engine (I recommend the Fonville-Morisey website for homes in the Triangle of North Carolina.)
• Printing capability

Pre-activities

Prior to beginning this lesson, students should be able to:

• Distinguish between independent and dependent variables
• Create scatter plots and describe correlation
• Use a graphing calculator to find linear regression models for data sets
• Use R statistical software to create scatter plots, to calculate linear models, to determine correlation of variables, and to complete basic data mining tasks

Prior to implementing this lesson, teachers should:

• Become familiar with basic R commands
• Develop data sets that may be used for demonstration purposes, if necessary
• Ensure that R (a free software) and Microsoft Excel are installed on student computers
• Understand the basics of variable selection

Activities

Distribute the Putting it All Together worksheet to students. Lead students through the overall outline of their assignment. It may be helpful to go through the initial search engine setup with students so they can easily find the information they will need to create their own set of data. At this point in the instructional unit, students should be able to use their previous lesson worksheets to guide them through this assignment independently. At the end of the lesson, students should print a copy of their R script (including all commands used) and attach it to their worksheets.

Assessment

To assess students’ understanding of the material, you could:

• Collect the lesson packets and check students’ data tables, short answer responses, and R scripts with code
• Ask students to develop their own ideas of other situations in which it would be helpful to identify the most influential variables (other times when data mining could be useful)
• Ask students to describe the usefulness of R

Modifications

This activity can be used in almost any type of classroom setting. It may be helpful to pair students with learning disabilities or English language learners with higher-achieving students. Also, identify helpers — those students who quickly catch on to how the R environment works and can help you keep other students on track. One teacher trying to help thirty students write programming code can be a frustrating experience for everyone involved.

Critical vocabulary

data mining

a method commonly used to extract useful information from large sets of data

forward selection

a method of variable selection in which no variables are initially considered, but then are added if found to be relevant to a given situation

backward selection

a method of variable selection in which all variables are initially considered, but then are eliminated if found to be irrelevant to a given situation

stepwise selection

a method of variable selection similar to forward selection, but
variables that may initially be added may be later deleted in an attempt to identify the most relevant variables

analysis of variance

technique used to identify the effect of variables on a response

Comments

This instructional unit was created as part of the Kenan Fellows Program as part of a goal to create innovative curricula for K–12 teachers to use in classrooms across the United States. My work in data mining was supported by my mentors, Dr. Hao Helen Zhang and Dr. Yufeng Liu. Dr. Zhang is a professor of statistics at North Carolina State University in Raleigh, North Carolina. Dr. Liu is a professor of statistics at the University of North Carolina in Chapel Hill, North Carolina.

• Next:

Contents: Understanding data mining

• Next:
• Previous: Learning to use R statistical software for data mining — An extension of linear regression to multiple variables

Learn more

• The R Project for Statistical Computing This is where you can download the free software required for these lessons. However, there is much more to be found about statistical computing on this website.

Related topics

• Search LEARN NC for more resources on algebra, graphing, linear regression, mathematics, and statistics.

• Print this page

• Share: