AS A RESULT OF THESE CONDITIONS, PROGRAM PEOPLE FEEL THAT THEY MUST LIE TO FUNDING PEOPLE IN ORDER TO MEET THE CRITERIA REQUIRED FOR A PROPOSAL TO BE FUNDED. THE RESULT OF SUCH PHONY PROPOSALS IS PHANTOM PROGRAMS. THE CRUX OF EVALUATIVE RESEARCH IS THE ASSESSMENT OF THE DEGREE TO WHICH THE PROGRAM HAS ACHIEVED ITS SPECIFIED GOALS. FREQUENTLY, THEN, THE FIRST TASK ATTEMPTED BY THE EVALUATOR IS TO SPECIFY MEASURABLE PROGRAM GOALS. THIS MAY NOT BE UNREASONABLE EXCEPT THAT JUST AS FREQUENTLY PROGRAM PEOPLE DO NOT THINK IN TERMS OF SPECIFIC MEASURABLE GOALS. IF THE RESULTS OF PROGRAM EVALUATORS ARE TO BE TAKEN INTO ACCOUNT BY PRACTITIONERS, ADMINISTRATORS, OR POLICYMAKERS, THESE RESULTS CANNOT BE DERIVED FROM PHANTOM PRGRAMS OR SPURIOUS GOALS. IF EVALUATIONS ARE NOT CREDIBLE AND THUS NOT TAKEN INTO ACCOUNT IN DECISIONMAKING PROCESSES, THEN THERE IS NO SENSE IN DOING THEM. AVOIDING THE GOAL-TRAP INCREASES THE PROBABILITY OF CREDIBLE AND USEFUL EVALUATION RESEARCH. THERE ARE AT LEAST THREE WAYS IN WHICH THE TRAP MAY BE AVOIDED. THEY ARE (IN REVERSE ORDER FROM THE MOST DESIRABLE, AND PROBABLY IMMEDIATELY FEASIBLE, TO THE LEAST) AS FOLLOWS: 'PROCESS ANALYSIS' (VIEWING SUCCESS OR FAILURE IN TERMS OF PROCESS RATHER THAN OF INPUT AND OUTPUT); GREATER EMPHASIS ON UNINTENDED PROGRAM CONSEQUENCES; AND THE NEGOTIATION OF VARIOUS ALTERNATIVE SCENARIOS IN A SEARCH FOR MORE MEANINGFUL PROGRAM GOALS. A FOUR-PAGE REFERENCE LIST IS INCLUDED....ELW