Hypothesis Testing -- Two-Sample Test for Variance
 
  Skip Navigation Links
 

Two-Sample Test for Variance

There are two preliminary assumptions of t-test, which are normal distribution and homogeneity of variance.
This function performs F-test for testing whether the variances are equal, and creating a confidence interval for the sample variance ratio.

You can set the following features when you perform the two-sample F-test for variance:

  1. Specify your dataset mode, indexed or raw
  2. Determine which alternative hypothesis is performed, upper lower or two-tailed
    • Variance1/Variance2<>1: two-tailed F-test will be performed
    • Variance1/Variance2>1: upper-tailed F-test will be performed
    • Variance1/Variance2<1: lower-tailed F-test will be performed
  3. Compute the Confidence Intervals by your specified Confidence Level(s)
  4. Origin supplies two plots for you, Histogram and Box Chart

  Two Sample Test for Variance interface

You can also set where to output your results as well as plot data and whether to output to the Results Log dialog.

An illustration of Two Sample Test for Varaince results being displayed in Origin

The figure above represents a typical results sheet obtained while performing the two-sample test for variance in Origin. While observing 30 students, half between the ages of 6-10, the other half between 11-16, we are interested in whether the variance of the 6-10 year old students' eyesight is equal to that of 11-16 year old students' eyesight.

 

Skip Navigation Links.
Copyright © 2012 OriginLab Corporation. All rights reserved.
20+ years serving the scientific and engineering community