Matrix File Format

STATISTICA's matrix files (e.g., Correlation, Covariances, Similarities, and Dissimilarities) can be used in the modules that support the matrix input file format (e.g., Multiple Regression, Canonical Correlation, Reliability and Item Analysis, Cluster Analysis, Multidimensional Scaling, Factor Analysis, etc.). By default, matrix spreadsheets will be saved with the default file extension .smx. Most STATISTICA modules will read both full and lower triangular matrices. However, in order for STATISTICA to recognize the file as a matrix file, the file must meet the following conditions:

  • The number of cases (rows) = the number of variables (columns) + 4.

  • The matrix must be a square matrix and the case names should be the same as the variable names.

  • The last four cases contain the following case names and information:

Means. The mean of each variable is given in this row; this case can be left empty (i.e., do not enter anything in this row) for Similarities and Dissimilarities matrices.

Std.Dev. The standard deviation of each variable is given in this row; this case can be left empty (i.e., do not enter anything in this row) for Similarities and Dissimilarities matrices.

No.Cases. This required number is the number of cases from which the matrix was produced, not the number of cases (rows of data) in this matrix file.

Matrix. This required number represents the type of matrix file; 1 = Correlation, 2 = Similarities, 3 = Dissimilarities, and 4 = Covariance.

Note: When entering these last four cases into the matrix file manually, be sure to spell the case names exactly as they appear above (i.e., Means, Std.Dev., No.Cases, and Matrix).

Examples of correlation matrix files:

 

Var 1

Var 2

Var 3

Var 1

1.00

.20

.30

Var 2

.20

1.00

.10

Var 3

.30

.10

1.00

Means

12

11

10

Std. Dev.

3

5

2

No. Cases

50

 

 

Matrix

1

 

 

Examples of lower triangular correlation matrix files:

 

Var 1

Var 2

Var 3

Var 1

1.00

 

 

Var 2

.20

1.00

 

Var 3

.30

.10

1.00

Means

12

11

10

Std. Dev.

3

5

2

No. Cases

50

 

 

Matrix

1

 

 

For further information, see: Types of Matrices and Multiple Correlation Matrices.