Project Proposal by Martin Stacey
Spelling Tutor
|
Software |
Language with good GUI tools, or Java and XHTML for a web-based system |
|
Covers |
Programming, systems design, web interface development, psychology of learning |
|
Skills Required |
Programming, object-oriented design, some interest in how people learn |
|
Challenge |
Conceptual Technical Programming |
Brief Description
The aim of the project is to build a computer tutor to teach primary school children to spell, and/or test them on how well they've learnt. Part of the challenge is developing software that provides appropriate practice for poor spellers, such as dyslexics. Or, alternatively, people learning to recognise tricky differences in foreign languages. What do you really want to teach, and how? What range of learning experiences can be provided by different content sets?
The major technical challenge here is integrating sound output into your own software.
Relatively simple systems can be effective teaching tools. However any interactive system for primary school children needs to be very robust, and the interactions between the system and the children should keep them entertained. Designing an interface to be bright and lively can conflict with making it usable - but creating an interface that's both stimulating and usable shouldn't be an insoluble problem.
An important part of building an effective computer teaching system is having a teaching strategy that fits how people learn. You'll need to think about what activities and ways of providing correction and reinforcement will fit how people learn, and how to put activities into effective sequences. One way to add challenge and sophistication to the project is to research appropriate teaching techniques and HCI approaches for dyslexics.
Variants
Similar technical challenges are presented by systems for teaching vocabulary and spelling in foreign languages.
Many children, especially those who are dyslexic, might benefit from a phonetics tutor that teaches them the mappings between spellings and pronunciations.
An applet or website accessible through the web would be an especially valuable resource for learners and teachers.
Extensions
You should consider how you can design your system with a clear separation between the system architecture and the information display mechanisms, and the content - the sets of words and spelling patterns they teach. Can you design your system as a language-independent tutoring shell plus a replaceable set of content files that can be used to teach spelling within a specific language?
The system would be a lot more usable with an easy-to-use editor to create and modify teaching materials, questions and question sequences.
If you can include speech input within your system, you could ask the system's users to read words and see if they've got them right. Learning decoding rules and then applying them is a serious issue for dyslexics. Reading and understanding continuous text presents different challenges to learning readers and dyslexics - a computer tutor can be more patient, and provide unemotional non-judgemental negative feedback, so might be a useful tool for reading practice.
Cross-Reference
For some kinds of computer tutors, having a way of measuring how well the students are doing is important, but building a complex testing tool might be an elaborate task, especially if you want to take the results seriously for evaluating student performance. I've proposed an Internet Quiz Administration System as a separate project.
Back to