Bartlett’s Test for Homogeneity of Variances

  Bartlett’s test evaluates whether a dataset composed of k groups exhibits homogeneous variance among its elements. The logic of the test involves computing the T statistic and comparing it with the predicted value from the \(\chi^2\) distribution.


1 Equation:

\[ T = \frac{(N-k)\ln(s_p^2) - \sum_{i=1}^k (N_i-1)\ln(s_i^2)} {1 + \frac{1}{3(k-1)}\left(\sum_{i=1}^k \frac{1}{N_i-1}\right) - \frac{1}{N-k}} \]

  With:

\[ s_p^2 = \sum_{i=1}^k (N_i-1)\frac{s_i^2}{N-k} \]

Where:

  • \(s_i^2\) = variance of group i;
  • \(N_i\) = size of group i;
  • \(N\) = total sample size;
  • k = number of groups;
  • \(s_p^2\) = pooled variance.

2 Files:

  1. Bart1matrix program and example
  2. Program source code

3 Bart1matrix

  Based on a simple data matrix (column vectors), the program outputs the critical \(\chi^2\) value at 5% significance level (\(\alpha\) = 0.05), the T statistic, and the associated p-value.

4 Usage and example

1. Enter the data matrix (column vectors);
2. Run "Bart1matrix".


  The compressed file includes an example comparing the dry weight of control plant groups with two treatment groups. The data were obtained from the datasets library of the statistical computing environment RPlantGrowth. The figures below illustrate the execution of the program using this example dataset.
(a) Insertion of data matrix.
(b) Results obtained from Bartlett’s test.
Figure 1: Bart1matrix program running on the Android version of the HP50G calculator (Go49gp), showing data input and results using the example dataset.


  When compared with results obtained using R, the p-value was 0.2371 (\(\chi^2\) = 2.966).

5 References:

  1. NIST. Engineering Statistics Handbook. Bartlett’s Test. Available at: https://www.itl.nist.gov/div898/handbook/eda/section3/eda357.htm. Accessed on Feb 9, 2026.
Back to top