Category: PROC CORR

Kline, P. (1999). Handbook of Psychological Testing(2^nd ed.). London: Routledge. P.13

Despite the dangers of boosting the reliability of a test by making the items highly similar to each other, in which case validity is reduced, reliabilities should ideally be high, around .9, especially for ability tests. Certainly alphas should never drop below .7, a value stressed by both Guilford (1956) and Nunnally(1978). The rationale and proof of these claims are bound up in psychometric theory and are given in Chapter 3.

 

For 0.7; Nunnally,J . (1978). Psychometric Theory. New York, McGraw-Hill.

For 0.8; Nunally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3^rd ed.) New Yort: McGraw-Hill.

 

Reference:

Abell, N., Springer, D. W., & Kamata, A. (2009). Developing and Validating Rapid Assessment Instruments. Oxford University Press.

 

SAS's proc cor procedure produces two types of cronbach coefficient alpha: raw value and standardized value.

proc corr alpha data=dataname_here;
var  item1 item2 item3 item4 item5 item6 item7;
run;

The result table includes two values:

Cronbach Coefficient Alpha

Variables Alpha
--------------------------------
Raw 0.74
Standardized 0.75

Standardized version is based on standardized values of all variables included in the analysis.  If you standardize the variables yourself by creating z-score version of items and apply the same procedure, you will get the same value for both raw and standardized values.

proc standardize data=dataname_here out=dataname_here_B mean=0 std=1;
var  item1 item2 item3 item4 item5 item6 item7;
run;
proc corr alpha data=dataname_here_B;
var  item1 item2 item3 item4 item5 item6 item7;
run;

Cronbach Coefficient Alpha

Variables Alpha
--------------------------------
Raw 0.75
Standardized 0.75