The interpretation of data and construction and interpretation of graphs are central practices in science, which, according to recent reform documents, science and mathematics teachers are expected to foster in their classrooms. However, are (preservice) science teachers prepared to teach inquiry with the purpose of transforming and analyzing data, and interpreting graphical representations? That is, are preservice science teachers prepared to teach data analysis and graph interpretation practices that scientists use by default in their everyday work? The present study was designed to answer these and related questions. We investigated the responses of preservice elementary and secondary science teachers to data and graph interpretation tasks. Our investigation shows that, despite considerable preparation, and for many, despite bachelor of science degrees, preservice teachers do not enact the ("authentic") practices that scientists rountinely do when asked to interpret data or graphs. Detailed analysis are provided of what data and graph interpretation practices actually were enacted. We conclude that traditional schooling emphasizes particular beliefs in the mathematical nature of the universe that make it difficult for many individuals to deal with data processing the random variation found in measurements of natural phenomena. The results suggest that preservice teachers need more experience in engaging in data and graph interpretation practices originating in activities that provide the degree of variation in and complexity of data present in realistic investigations.