The pitfalls of serial administration of personality measures as a research selection device were explored. Three samples of subjects were administered the Self-Consciousness Scale (Fenigstein, Scheier, & Buss, 1975) either first or last in a series of questionnaires. As predicted, those who responded to the measure last obtained higher private self-consciousness scores. However, the strength of the effect was moderated by the number of other measures included in the testing. Unexpectedly, public self-consciousness scores also were influenced by order of administration of the measure. Subjects who were administered the scale last evidenced lower public self-consciousness scores although this effect reached significance in only one of three samples of subjects. The broader implications of these findings for researchers who employ mass test administration procedures to select subjects for personality research are considered.