This correlation applet allows you to see plots with a specified correlation; to compare plots, varying the correlation and standard deviations of the variables; and to play games to test your ability to estimate the correlation coefficient for a given plot.
Type of Material:
Recommended uses for this applet would be for an instructor or student to visually display scatterplots with known correlation, as well as allowing students to distinguish different plots and compare correlation between them.
The technical requirements for this applet is an up-to-date java plug-in and an internet connection with a web browser. Does not run on the iPad.
Identify Major Learning Goals:
The learning goal for this applet is for students to estimate the correlation between two variables by examining a scatterplot.
Target Student Population:
introductory statistics students
Prerequisite Knowledge or Skills:
This applet does not actually teach what correlation is, so a pre-requisite would be to have an understanding of correlation or additional source explaining what correlation is.
Evaluation and Observation
The applet allows students or instructors a variety of ways to explore and compare correlation using graphs. It is very visual and easily interpreted to an audience that understands what correlation is.
The applet does not actually teach what correlation is, how it is calculated, or what it means in any context. It does not explain its uses for only linear relationships, the range that r can have, or the relationship between slope of the line and r. The correlation is to be estimated to 2 decimal places. It is doubtful a student could get that close without running the data through a calculator or computer.
Potential Effectiveness as a Teaching Tool
This applet would be easy to use in a classroom setting and allow students a fun experience of playing games guessing correlation. There are scores immediately available for students and instructors to assess the understanding of the concepts.
The applet does not aid in the understanding of how correlation is calculated, its meaning, nor its application to statistics. The assessment score, "scorr" is not clearly defined, nor is it explained how it is calculated. Particularly with the interface that allows a user to guess the r-value of a given scatterplot, "scorr" is not easily interpreted.
Ease of Use for Both Students and Faculty
The applet is very easy to use and has clearly defined instructions. The applet also has several different interfaces with varying levels of difficulty and speed depending on the targeted audience. It is very easy to interpret and use.
There are typographical errors in some of the instructions. It may also be helpful to explain a little more about X and Y as random variables; for example, that for all scatterplots they both have a mean of 50. It is not immediately obvious that the interface menu expands when clicked upon, but it is fully functional. Also, again, it is difficult to get 2 decimal point accuracy.