This applet demonstrates Type I error. A Type I error is rejecting a hypothesis that is true.

When the applet starts, you see a set of points generated by a line with 0 slope and 30 intercept, with random error. This is the same as the data for assignment 1A's last part.
The New button generates a new set of points.
The True button shows the true line that generated the points. Press again to hide the true line.
The Errs button shows or hides the random errors that generated the points. Each point is the true line, plus or minus the random error.
The LS button shows or hides the least squares line through the points.
The Res button shows orhides the residuals, the vertical differences between the points and the least squares line.
The Squ button shows or hides the squares of the residuals. It also shows or hides the sum of the squares' areas and the t-statistic for the hypothesis that the true slope is 0. The critical t value at the 0.05 level for a two-tailed test is 2.57, so a t-statistic higher than 2.57 or lower than -2.57 leads to a Type I error. Click the New button repeatedly to see this.

Other HSPM J716 demos


© 2012 Samuel L. Baker
Creative Commons License
Type I error in linear regression by Samuel L. Baker is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
The author is solely responsible for this page. Its contents have not been reviewed or approved by the University of South Carolina.
. Its contents have not been reviewed or approved by the University of South Carolina.