This study explores the influence of framing and activity type on expectations of learning and enjoyment as well as performance in a paraphrase identification task. In the first experiment, 80 students played one of three activities framed as either a "play" or "learning" task. Students then completed one of three activities; learning only, an educational game, or a play only activity. Results showed that the play frame had an effect on learning expectations prior to completing the activity, but had no effect after completing the activity. Students who completed the educational game scored significantly higher on the posttest learning assessment than those in the play only activity. Pairwise comparisons also indicated that students who completed the educational game performed just as well as the learning only activity when given the posttest learning assessment. Performance in the paraphrase identification task was collected using data logged from student interactions, and it was established that although there was an interaction between performance and activity type, this interaction was due to a significant difference during the second round. These results suggest that framing can influence initial expectations, and educational games can teach a simple writing strategy without distracting from the educational task. A second experiment using 80 students was conducted to determine if a stronger frame would influence expectations and to replicate the effect of activity type on learning and enjoyment. The second study showed no effect of framing on expected or reported enjoyment and learning. The performance results showed a significant interaction between performance and activity type, with the interaction being driven by the first round that students completed. However, the effect of activity type was replicated, suggesting that game features can enhance student enjoyment and are not a detriment to learning simple strategy-based tasks.