ELSE Command (DO IF command)

ELSE executes one or more transformations when none of the logical expressions on DO IF or any ELSE IF commands is true.

  • Only one ELSE command is allowed within a DO IF—END IF structure.
  • ELSE must follow all ELSE IF commands (if any) in the structure.
  • If the logical expression on DO IF or any ELSE IF command is true, the program ignores the commands following ELSE.

Example

DO IF (X EQ 0).
COMPUTE Y=1.
ELSE.
COMPUTE Y=2.
END IF.
  • Y is set to 1 for all cases with value 0 for X, and Y is 2 for all cases with any other valid value for X.
  • The value of Y is not changed by this structure if X is missing.

Example 

DO IF (YearHired GT 87).
COMPUTE            Bonus = 0.
ELSE.
IF (Dept87 EQ 1) Bonus = .12*Salary87.
IF (Dept87 EQ 2) Bonus = .14*Salary87.
IF (Dept87 EQ 3) Bonus = .1*Salary87.
IF (Dept87 EQ 4) Bonus = .08*Salary87.
END IF.
  • If an individual was hired after 1987 (YearHired is greater than 87), Bonus is set to 0 and control passes out of the structure. Otherwise, control passes to the IF commands following ELSE.
  • Each IF command evaluates every case. The value of Bonus is transformed only when the case meets the criteria specified on IF. Compare this structure with the second example for the ELSE IF command, which performs the same task more efficiently.

Example

* Test for listwise deletion of missing values.
 
DATA LIST / V1 TO V6 1-6.
BEGIN DATA
123456
    56
1 3456
123456
123456
END DATA.
 
DO IF NMISS(V1 TO V6)=0.
+  COMPUTE SELECT='V'.
ELSE
+  COMPUTE SELECT='M'.
END IF.
 
FREQUENCIES VAR=SELECT.
  • If there are no missing values for any of the variables V1 to V6, COMPUTE sets the value of SELECT equal to V (for valid). Otherwise, COMPUTE sets the value of SELECT equal to M (for missing).
  • FREQUENCIES generates a frequency table for SELECT. The table gives a count of how many cases have missing values for one or more variables, and how many cases have valid values for all variables. Commands in this example can be used to determine how many cases are dropped from an analysis that uses listwise deletion of missing values.