17.3.6.1 The Two-Sample Variance Test Dialog Box


Contents

Supporting Information

To open the Two-Sample Variance Test dialog box:

  1. With the worksheet active, click Statistics: Hypothesis Testing: Two-Sample Variance Test...

See Also:


Recalculate

Recalculate

Controls recalculation of analysis results

  • None
  • Auto
  • Manual

For more information, see: Recalculating Analysis Results


Input Data Form

Two-Sample Variance Test supports Indexed Raw, Summarized Data(SD) and Summarized Data(Variance) three input data form.

For help with range controls, see: Specifying Your Input Data

Indexed The input data consists of one sample data column and one grouping column. Factor variable and response data are stored in separate columns.
  • Group Range
The grouping column. Note that the grouping column will be set as categorical if Text column.
  • Data Range
The sample data column.
Raw The input data consists of two sample data columns. Each column contains response data from a level of the factor variable.
  • 1st Data Range
The first sample column.
  • 2nd Data Range
The second sample column.
Summarized Data (SD) Summarized Data is standard derivation and sample size prepared for samples.

Enter following Summarized Data by selecting columns from worksheet or entering into the dialog box manually.

  • Sample Name
Specify name of the samples. You can specify a column as sample name, or enter space separated strings, or leave it empty.
  • SD
Specify the standard derivation of the data. You can specify a column dataset, or enter space separated number. (SD should be greater than 0.)
  • Sample Size
Specify the total number of data. You can specify a column dataset, or enter space separated number. (Sample size should be greater than 0.)
Summarized Data (Variance) Summarized Data is Variance and sample size prepared for samples.

Enter following Summarized Data by selecting columns from worksheet or entering into the dialog box manually.

  • Sample Name
Specify name of the samples. You can specify a column as sample name, or enter space separated strings, or leave it empty.
  • Variance
Specify the Variance of the data. You can specify a column dataset, or enter space separated number.
  • Sample Size
Specify the total number of data. You can specify a column dataset, or enter space separated number. (Sample size should be greater than 0.)

Test

Test for Options:
  • Variance Ratio
  • Standard Deviation Ratio (Default option)
Test Ratio Specify the hypothetical ratio to be tested.
Null Hypothesis The null hypothesis to be tested (i.e. Variance1 = Variance2).
Alternate Hypothesis Options:
  • Variance 1/Variance 2 <> Test Ratio or SD 1/SD 2 <> Test Ratio
Test whether the two sample variance/SD are equal.
  • Variance 1/Variance 2 >Test Ratio or SD 1/SD 2 > Test Ratio
Test whether the variance/SD from sample one is larger than the variance/SD from sample two.
  • Variance 1/Variance 2 < Test Ratio or SD 1/SD 2 < Test Ratio
Test whether the variance/SD from sample one is less than the variance/SD from sample two.
Significance Level Specify the significance level.
Confidence Interval(s) Compute confidence intervals for the differences of the means.
Confidence Level(s) in % Specifies space-separated confidence levels. This option is only available when Confidence Interval(s) is chosen.
Test Method Select test method:
  • F Test
Computes the direct ratio of the two sample variances. Extremely sensitive to departures from normality. Even slight non-normality causes inflated Type I error rates. If the data conform to the normal distribution, then the F-test is typically more powerful than other method.
  • Bonett
If is for any continuous distribution. It will output the Bonett test result with P-Value, and the Confidence Intervals for Variance/Standard Deviation. Highly resistant to moderate non-normality and maintains higher statistical power. It can used for most continuous distributions with sample sizes of n > 20 per group. (For small samples with extreme skewness or heavy tails.)
When input data form is Summarized Data(SD)/(Variance), Bonett option is disable.


  • Brown-Forsythe
A modification of Levene's test. It calculates the absolute deviations of each observation from its group median and performs a standard 1-way ANOVA on those transformed values. Non-parametric level robustness. It can used for strongly skewed data, heavy-tailed distributions, or datasets containing notable outliers.
When input data form is Summarized Data(SD)/(Variance), Brown-Forsythe option is disable.


Power Analysis

Unset Ratio of Interest

Difference(s) for Specified Power When Ratio of Interest is not set, this option is available.

Output the differences for the specified power levels. When this option is selected, the specified power levels setting will affect the power level in power curve. If it is selected, it will use the default power levels.

Power Level(s) When check the Difference(s) for Specified Power option, you can enter the power levels separated by space in this edit box.

Set Ratio of Interest

Ratio of Interest Specify the minimum ratio of variances (or standard deviations) between two independent populations .


Note: When specify Ratio of Interest:

  • If Alternate Hypothesis <> Test Ratio, Ratio of Interest should be positive value and not be 1
  • If Alternate Hypothesis >= Test Ratio, Ratio of Interest should be > 1
  • If Alternate Hypothesis < Test Ratio, Ratio of Interest should be < 1
Sample Size(s) for Specified Power When specify the Ratio of Interest, this option is available.

Output the Sample Sizes for the specified power levels. When this option is selected, the specified power levels setting will affect the power level in power curve. If it is selected, it will use the default power levels.

Power Level(s) When check the Sample Size(s) for Specified Power option, you can enter the target power levels separated by space in this edit box.
Power Based on Test Data When specify the Ratio of Interest, this option is available.

Compute the power for the sample size of the test data. (the probability of rejecting the null hypothesis when it is false).

Hypothetical Power When specify the Ratio of Interest, this option is available.

Compute the power of the test at various sample sizes. When check this option, it will add corresponding curve to power curve.

Hypothetical Sample Size(s) When specify the Ratio of Interest and Hypothetical Power is checked, this option is available.

Specify hypothetical sample size(s) for power computation.

Plots

Histogram Create a histogram.
Box Chart Create a box chart.

Note: When outlier(s) detected in the datasets, a footnote will show under the Box chart to note you to review their case. If you want to exclude the outliers, you can click the triangle button at the end of the footnote to mark outliers and recalculate.

Arrange Plots of Same Type in One Graph Note that for the histograms, layers will be arranged vertically. For box chart, layers will be arranged horizontally.
Comparison Interval for SD Create a scatter plot with error bar for comparison interval for SD between two groups.
Note: When input data form is Summarized Data(SD)/(Variance), it will not output above plots.


Output

For help with the range controls, see: Output Results

Results Log Output

Select to output results to the Results Log.

Graphical Summary Report Create a summary graph report, including P value bar, effect size bar and data distribution graph. The result tables will also been attached to the graph to show the statistical values.

The footnote after the summary graph will diagnose these statistical results.

Graphical Power Analysis Report Create a graphical power analysis report with power curve(s).

A footnote will be added after the power curve graph to show you the diagnosis results of power analysis.

Output Plot Data Specify a worksheet for output of plotted data (Origin generates tabular data when plots are created).
Output Results Specify the worksheet for output of test results.
Note:
  • When input data form is Summarized Data(SD)/(Variance), Graphical Summary report will not be output.
  • When input data form is Raw or Indexed, Masked Data and Bad Data(missing value) table will be in the report sheet.