This applet from Statistical Java allows the user to generate bivariate data for analysis with simple linear regression. The page describes the equations used to generate the data and estimate the regression lines. Users manipulate data generated and create their own line of best fit to try and match the computer generated regression line. This page was formerly located at http://www.stat.vt.edu/~sundar/java/applets/Regression.html
Netscape 4.0 and up, Explorer 5 and up
Content Quality Concerns:
The notation used in the initial explanation may be a little too advanced for some lower level courses. Perhaps more importantly, the applet is rather limited in the types of data sets that are "created." For example, if datasets could be created which contained outliers; leverage values; or inherent non-linearity, then the instructor could demonstrate the impact of each of these on the residuals and the regression line.
Content Quality Strengths:
The item uses a very nice graphical display illustrates an important statistical concept, namely the reason why the least-squares line provides the "best" fit to data.
Ease of Use Concerns:
Ease of Use Strengths:
The applet is easy to use and has clear instructions.
Potential Effectiveness Concerns:
It is not possible to ask all students to work with the same generated data. Therefore it is not possible to have students to search for the same least squares line. This might make using the applet in a homework assignment difficult.
Potential Effectiveness Strengths:
The item is an effective and cleanly written illustration of the concept of least squares. It is easy to incorporate into a classroom presentation and would work with most introductory courses.
Potential Effectiveness Rating:
Source Code Available:
Source Code Available
Intended User Role:
Learner, Teacher, Author
Free for All