Acceptance and mindfulness methods that emphasise the acceptance rather than control of symptoms are becoming more central to behavioural and cognitive therapies. Acceptance and Commitment Therapy (ACT) is the most developed of these methods; recent applications of ACT to psychosis suggest it to be a promising therapeutic approach. However, investigation of the mechanisms of therapy within this domain is difficult because there are no acceptance-based measures available specifically for psychotic symptoms. This paper describes the preliminary evaluation of a self-report instrument designed to assess acceptance-based attitudes and actions in relation to auditory and command hallucinations. Following initial scale development, a 56-item version of the Voices Acceptance and Action Scale (VAAS) was administered to 43 participants with command hallucinations as part of their baseline assessment in a larger trial. Measures of symptoms, quality of life, and depression were also administered. The scale was examined for reliability using corrected item total statistics. Based on this method, 31 items were retained. Internal consistency and test-retest reliability for the 31-item VAAS were acceptable. Subsequent examination of construct validity showed the VAAS to correlate significantly in the expected directions with depression, quality of life, and coping with command hallucinations. It also discriminated compliance from non-compliance with harmful command hallucinations. Although these results are preliminary and subject to a number of limitations, the VAAS shows promise as a useful aid in the assessment of the psychological impact of voices.