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.

© 2012 Samuel L. Baker

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.