Select the analysis options for Factor Analysis

Enter matrices or loadings to use for the initial extraction, and specify options for maximum likelihood estimation.

Matrix to Factor

• Correlation: Select to calculate the factors using the correlation matrix. Use the correlation matrix to standardize variables when the variables are measured using different scales.
• Covariance: Select to calculate the factors using the covariance matrix. Use the covariance matrix if you do not want to standardize variables. The covariance matrix cannot be used if you select Maximum likelihood as the extraction method on the main dialog box.

Source of Matrix

• Compute from variables: Select to use the correlation matrix or the covariance matrix that is calculated from the measurement data.
• Use matrix: Select to use a stored matrix for calculating the loadings and coefficients. If you use a stored matrix, Minitab ignores any raw data you enter in Variables on the main tab.

  Note

  If you choose this option, Minitab cannot calculate the scores.

Loadings for Initial Solution

• Compute from variables: Select to calculate the loadings from the raw data.
• Use loadings: Select to use loadings which were previously calculated, then specify the columns containing the loadings. You must specify one column for each factor.

  Tip

  You can use stored loadings to examine the effect of different rotations or to predict factor scores with new data. Use the main dialog box to enter different types of rotations or to enter new columns of data for the variables.

Maximum Likelihood Extraction

Use initial communality estimates in

Usually, the default values lead to a solution that converges. However, you can enter a column of initial communality estimates to obtain the following:

• More precise estimates, which are typically larger than the default values
• Less precise estimates, which are typically smaller than the default values, to determine whether the final factor loadings are sensitive to the initial communalities

Enter the column that contains the initial values for the communalities. The column should contain one value for each variable.

Max iterations

Enter the maximum number of iterations allowed for a solution. The default is 25. Usually, 25 is enough iterations for the likelihood function to converge. If the function does not converge, you can enter a larger number to increase the maximum number of iterations.

Convergence

Enter the criterion for convergence (occurs when the uniqueness values do not change very much). This number is the size of the smallest change. The default is 0.005.

Usually, the likelihood function converges to a point where it changes by less than 0.005 in an iteration. If the function does not converge, you can enter a larger number, but a larger number can make the analysis more sensitive to the choice of initial communalities.