PROC CALIS Documentation Example

The SAS^© program first creates an internal 'sales' dataset containing observations for four variables (q1 to q4). Then, it executes PROC CALIS multiple times with different configurations to analyze the data's covariance structure. The options `covpattern=diag` and `covpattern=sigsqi` are explored, as well as the effect of `chicorrect=0` on the chi-square test results. This demonstrates how to specify covariance matrix structures in the analysis.

Data Analysis

Type : CREATION_INTERNE

The data is created directly within the SAS script using a DATA STEP statement with `datalines`.

1 Code Block

DATA STEP Data

Explanation :
This DATA STEP block creates a dataset named 'sales' with four variables (q1, q2, q3, q4) using raw data provided directly in the script (datalines). This data will serve as input for subsequent CALIS analyses.

Copied!

1	DATA sales;
2	INPUT q1 q2 q3 q4;
3	DATALINES;
4	1.03 1.54 1.11 2.22
5	1.23 1.43 1.65 2.12
6	3.24 2.21 2.31 5.15
7	1.23 2.35 2.21 7.17
8	.98 2.13 1.76 2.38
9	1.02 2.05 3.15 4.28
10	1.54 1.99 1.77 2.00
11	1.76 1.79 2.28 3.18
12	1.11 3.41 2.20 3.21
13	1.32 2.32 4.32 4.78
14	1.22 1.81 1.51 3.15
15	1.11 2.15 2.45 6.17
16	1.01 2.12 1.96 2.08
17	1.34 1.74 2.16 3.28
18	;

2 Code Block

PROC CALIS

Explanation :
This PROC CALIS call performs a factor analysis or structural equation modeling on the 'sales' dataset. The `covpattern=diag` option specifies that error variances are modeled as independent and of the same value (diagonal covariance matrix for errors).

Copied!

1	PROC CALIS DATA=sales covpattern=diag;
2	RUN;

3 Code Block

PROC CALIS

Explanation :
Similar to the previous block, this PROC CALIS execution also uses `covpattern=diag`. The `chicorrect=0` option disables the Satorra-Bentler correction for the chi-square test, which may be relevant when data do not follow a multivariate normal distribution.

Copied!

1
2	PROC CALIS
3	DATA=sales covpattern=diag chicorrect=0;
4	RUN;
5

4 Code Block

PROC CALIS

Explanation :
This block executes PROC CALIS with `covpattern=sigsqi`. This option models error variances as independent and different for each variable (diagonal covariance matrix with different elements on the diagonal).

Copied!

1	PROC CALIS DATA=sales covpattern=sigsqi;
2	RUN;

5 Code Block

PROC CALIS

Explanation :
This final PROC CALIS execution combines `covpattern=sigsqi` with `chicorrect=0`. It models independent and heterogeneous error variances, while disabling the Satorra-Bentler correction for the chi-square test, as explained previously.

Copied!

1
2	PROC CALIS
3	DATA=sales covpattern=sigsqi chicorrect=0;
4	RUN;
5

This material is provided "as is" by We Are Cas. There are no warranties, expressed or implied, as to merchantability or fitness for a particular purpose regarding the materials or code contained herein. We Are Cas is not responsible for errors in this material as it now exists or will exist, nor does We Are Cas provide technical support for it.

Retour à la liste