Abstract:

Lecturing plays an essential role in the learning process. The variations in lecture design profoundly affect the quality of learning. Researchers are proposing models and frameworks for designing quality lectures. Learning styles like Learning by Doing (LBD), Learning by Memorization (LBM), Learning by Imitation (LBI) and others forms the foundation of these proposed models and frameworks.

A simple lecture is composed of three lecture components, namely, lecture introduction, lecture body and lecture conclusion. The quality lecture designing models/frameworks proposed so far wrap-up lecture components in a single unit. But in practice, each lecture component is critical and must be designed independently. This research is the answer to this problem. This research proposes a model named "LEC-CONCD" for designing quality lecture-conclusions (LECON). We designed this model, keeping in mind the needs of the online education system.

This model uses pedagogical, linguistic and computer engineering related concepts like content complexity rigor (CCR), depth of knowledge (DOK), knowledge points (KP), learning styles, controlled natural languages (CNL), syntax parsing and others to design quality LECON.

The LECON's generated using this model can further be classified into four distinct types, namely:
(i) simple summary,
(ii) single-KP conclusions,
(iii) multi-KP conclusions and
(iv) visual graphs.

The quality of designed LECON's is measured using five quality attributes. After an extensive literature survey:
(i) simple,
(ii) understandable,
(iii) recall,
(iv) interesting and
(v) engaging are chosen as quality attributes.

We conducted an online computer science learning program for students. The main objective of this program was to teach basic computer science topics to 197 registered students. All lectures used in this program were having LECON's designed using our proposed model. We asked students to rate the LECON's at the end of every online class, followed by an online feedback form. Based on rating and data compiled using the feedback form, we observed exciting trends. Few trends (or results) are as follows.

1. The LECON's designed using "visual graphs" have very high recall rate.
2. For rhythmic lectures, single-KP conclusions are more effective concerning multi-KP conclusions.
3. For large/long lectures, the simple summary is found best.
4. It's better to design multi-KP conclusions by encapsulating 3-5 KP's only.

We believe that this study helps online educators, lecturers, course designers and education software designers to understand LECON's designing process better. This model also supports the education industry to design pedagogically effective education online systems.

Keywords:

Lecture Conclusions, Controlled Natural Languages, Knowledge Points, Depth of Knowledge, Content Complexity Rigor.