This paper extends our previous work (Kim, Weitz, Heffernan & Krach, 2009) which compared a classic worked examples (WE) condition with a tutored problem solving (TPS) condition. By classic we mean the WE condition does not include tutoring, a self-explanation component, or fading. The aim of the current study was to compare the WE and TPS conditions with a mixed condition, which presents students with WE-TPS pairs. More specifically, for conceptual problems a pure WE condition was compared with a WE-TPS condition and for procedural problems a pure TPS condition was compared with a WE-TPS condition. While overall learning occurred in all conditions no significant differences were found between conditions. Further, our findings echo the results of earlier studies, that students who receive worked examples learn more efficiently that is, they need significantly less time to complete the same learning material. This is an important finding for educators because building classic worked examples is considerably easier than building tutoring.