The search for a reliable measure of anxiety constitutes the basis for the research discussed in this report. Possible influences on responses to self-report questionnaires are discussed and "Unwitting Defensiveness" is stressed as being one of the main contaminants of anxiety measurement. The test procedure used on successful and unselected applicants for positions in commission sales is outlined. Two scores for each applicant were devised; the first as a measure of underlying anxiety. Scores were classified as secure, average or insecure. Discussion of the study's results indicates that the second measure displayed more insecure applicants than the first; many insecure cases can be missed by using the outward measure alone; and anxiety measurement serves to sharpen test accuracy and to predict the effects of stress on future performance. The correlation of applicant security with variables such as compliance, verbal reasoning, and ambition is also documented. In conclusion, the failure of anxiety tests to demonstrate that they evince stress reactions normally associated with tests is highlighted, and the need for further research is emphasized. (AEF)