This article describes how to perform a Barlett Test of Sphericity. This test is used to make sure that the correlation matrix of the variables in your dataset diverges significantly from the identity matrix. An identity matrix is a matrix in which all of the values along the diagonal are 1 and all of the other values are 0. If the p-value from Bartlett’s Test of Sphericity is lower than your chosen significance level (common choices are 0.10, 0.05, and 0.01), then we can assume that the variables in the correlation matrix are not orthogonal.
Requirements
A Displayr document containing a data set which contains two or more variables which have a Structure set as Numeric or Numeric-Multi. Note that binary variables can also be used as inputs but will be converted to numeric when calculating the correlation coefficients. If you use categorical or ordinal variables, they will be coerced to numeric based on their values for the purposes of running the test.
Method
- Select Anything > Advanced Analysis > Test > Barlett Test of Sphericity.
- In the object inspector, specify the variables to use under Inputs > Input Variables.
- OPTIONAL: Select Missing data to change the way missing values are handled
- Select Use partial data (pairwise correlations) to treat each pair of variables separately and only include observations that have valid values for each pair in the data set.
- Select Error if missing data if you do not want the test performed if there is any missing data
- Select Exclude cases with missing data if you only want to use cases with valid values on all the variables
- Select Imputation (replace missing values with estimate) to request that imputation techniques be used to replace missing values with estimates
- OPTIONAL: Select Variable names to display variable names in the output instead of the variable labels
- OPTIONAL: Select More decimal places to display numeric values with 8 decimal places.
Next
How to do a principal components analysis in Displayr