What type of digital content can help students learn?

Below is an email response by Sridhar Rajagopalan (co-founder Educational Initiatives / ASSET / Mindspark) to someone who asked on how to use digital content to aid self-learning by students.

The goal is to make the learning as independent as possible for the average student. (It is much easier to achieve it for a few motivated students or a few intelligent students who breeze through the material)

At the outset, I would say this is a very difficult problem to solve as the final solution involves a very high level of quality in a number of different dimensions (like pedagogy, design, technology, etc.). 

I will also add that in EI's case, we are not really trying to create independent learning for now, and the role of the teacher or parent is pretty critical in setting the goals and pace and even monitoring the students.

Overall, I would say that all these factors are important in achieving reasonably independent learning by students using technology

1. A commonality of objectives:

a. The student needs to know what he or she is doing so that the process does not become mechanical. This is obviously not easy where the student is learning something she does not know. It may be easier to understand that there is a story and the questions at the end have to be answered, or that the 10 problems in the set are to be solved. But where such objectives are not there (for example, if we want the child to go through certain exercises to understand the concept of area, it is more difficult to define this in a way that the student can completely do it herself. (This is related to the point 'non-mechanical / repetitive nature of content' in point 3b below)

b. Visibility of progress: This is an important corollary. If a student knows that they have 10 problems to solve and they have solved 4, they are much more likely to complete the remaining 6, than if the number of questions and time are indeterminate. 

c. Where the larger goal may be difficult for the child to understand, certain goals like solve 10 questions daily or work for 15 minutes daily can be quite effective too.

2. Interactiveness of the content: We find that it is more difficult to retain children's attention with non-interactive content than interactive content. Though TV programmes like Sesame Street (Gali Gali Sim Sim) and Blue's Clues are exceptions (and were backed by tons of research, aimed at smaller children and not trying to make kids understand specific conceptual ideas like area in Maths or develop reading comprehension skills), for computer-based learning programs, students seem to quickly move away from one-way communication. In our experience, the solutions are either 

- regular responses (where the student has to respond regularly, typical more than once a minute) OR

- short videos (allowing interaction between the videos)

Getting regular responses from the students also allows the system to understand what the child is following and what they are not.

3. Quality of the content: We believe that this is the MOST important factor. This is particularly true in a situation now where there is no shortage of questionable quality material all around. The points are:

a. does the content really help the child to understand and learn? If not, the child will not continue to do (and teachers / parents will not encourage or push for the same)

b. non-mechanical / repetitive nature of content: Very often programs take a certain question type and simply repeat them especially in Maths. Except when students are forced by, say, a school timetable, children get bored and stop doing such content. 

c. pacing including moving to the next level at the appropriate stage - again needed to ensure the child is not bored (content too easy) or too challenged (content too difficult)

d. ability to detect student deviant behaviour (if students click at random, or are struggling too much, the system may need to be able to detect and change what it is giving.) A corollary of this point is that a fixed content of questions in a module for all children is unlikely to work very effectively.

4. Appropriate reward mechanisms for motivation: Though for some people, the rewards of learning seem to be the learning itself, this may happen only within topics of interest (eg. student A may love learning language but not Maths) even for those students. Other kinds of rewards or 'acknowledgments of progress or effort' like badges, stars, etc. help in some cases. In other cases, social recognition helps. Marks and grades in the school system - apart from praise by teachers - are examples of these and some kind of digital equivalent may be needed. It is important to remember that the rewards that are valued, change and lose their attraction over time, so they may need to be modified every few years.

5. Some periodic testing / monitoring of learning (to check that learning is actually happening): It is useful both in order to tweak and improve the content that has been developed, and to show progress that an independent measure of the learning improvement is available. For example, when the content is created, we may assume that doing a set of questions improves students' understanding of the related concept. [In fact, learning in schools works on the same principle - it seems extremely self-evident.] But all content is not created equal, and it is important to verify that learning is happening / the product is leading to learning. This is strongly related to point 3a above.

A related but different aspect of this is for the system itself to have tests (sometimes called A-B tests) to determine what aspect of the system is working or working better. This is important for the system to be improved continuously.

6. A balance of system-direction, student direction and mentor direction: Who decides what content the child gets next? At one extreme, the choice could be entirely with the child. At the other extreme, everything is controlled by the system which decides what next to serve up. And the third option is for the teacher or mentor to drive this (which is the typical school model). We find that each of these is important. Child control should be higher for students at a higher level of motivation or ability. [Choice of how to proceed or what to do is also a reward which is valued by such students.] System direction is important as the system is able to access the student's past performance better than the child or even a teacher. And mentor direction is key because few systems today are able to provide effective direction completely by themselves. (We find medical examples of highly effictive and flexible machines being used to guide and help doctors a very close analogy)

7. Provision for an adult (like a teacher or parent) to get a dashboard of the student's progress: Due to points 5 and 6 above, the visibility of progress being available to a parent and even more importantly, a teacher who is monitoring a number of students, becomes very important. This also means, unfortunately, that we are still some time away from a goal of completely independent, technology-mediated learning for students...