The ability to assess the quality of a service (QoS) is important to the emerging cloud computing paradigm. When many cloud service providers exist offering many functionally identical services, the prospective users of these services will wish to use one that offers the best quality. Many techniques and tools have been proposed to assess QoS, and the ability to deal with uncertainty surrounding the QoS verdicts given by any such techniques or tools is essential. In this paper, we present a probabilistic model to quantify confidence in QoS assessment. More specifically, we take the number of QoS data items used in assessment and the variation of data in the dataset into account in our measure of assessment reliability. Our experiments show that our confidence model can help consumers to select services based on their requirements effectively.