A knowledge survey (KS) is a series of content-based questions sequenced in order of presentation during a course. Students do not answer the questions; rather, they rank their confidence in their ability to answer each question. A 304-question KS was designed and implemented for a multisection, multi-instructor introductory biology course to determine whether this tool could be used to assess student learning. The KS was administered during the first 2 wk and the last 2 wk of the semester online via WebCT. Results were scored using one point for each “not confident” response (level 1), two points for each “possibly confident” response (level 2), and three points for each “confident” response (level 3). We found that scores increased significantly between the pre- and post-KS, indicating that student confidence in their knowledge of the course material increased over the semester. However, the correlation between student confidence and final grades was negligible or low, and chi-square tests show that KS scores and matched exam questions were not significantly related. We conclude that under the conditions implemented in our study, the KS does not reliably measure student learning as measured by final grades or exam questions.