This study investigates how blending of different instructional approaches with technology affects students' engagement. A computer networks course was designed and implemented for the first eight weeks of the semester as a lecture-based blended learning environment and for the second eight weeks of the semester as a problem-based blended learning environment. A single group repeated measures research design was carried out to understand if there are significant differences in measures of student engagement between these two blended learning approaches. Repeated measure ANOVA analysis on the data collected from 89 students revealed that Active Learning and Total Time on Task indicators of student engagement were significantly higher in the problem-based part of the course. Interaction and Level of Academic Challenge components and course satisfaction did not show any significant differences between the two parts. Regression analysis showed that the difference in Active Learning is not due to student individual differences but rather the learning environment provided in the problem-based blended learning. (Contains 11 tables and 1 figure.)