The effects of web-mediated instructional strategies and cognitive preferences in the acquisition of introductory programming concepts: a Rasch model approach

Mohamad, M 2012, The effects of web-mediated instructional strategies and cognitive preferences in the acquisition of introductory programming concepts: a Rasch model approach, Doctor of Philosophy (PhD), Business Information Technology and Logistics, RMIT University.


Document type: Thesis
Collection: Theses

Attached Files
Name Description MIMEType Size
Mohamad.pdf Thesis Click to show the corresponding preview/stream application/pdf;... 1.79MB
Title The effects of web-mediated instructional strategies and cognitive preferences in the acquisition of introductory programming concepts: a Rasch model approach
Author(s) Mohamad, M
Year 2012
Abstract This is a quasi-experimental study that aims to examine the interactive effects of web-mediated instructional strategies and learners’ cognitive preferences in their acquisition of introductory programming concepts. The thesis knowledge domain is an introduction to computer programming knowledge, designed for non-computer science students. It is the intention of this study to extend the McKay (2000a) research to examine the interactive effects of web-mediated instructional strategies and cognitive preferences in the acquisition of introductory programming concepts in Malaysian universities. Whereas the instructional strategies investigated by McKay (2000a) were paper-based, they involved graphical metaphors.

Instead, this thesis interrogates whether or not information communications technology (ICT) elements such as signals (or cues), interactive animation, navigational tools, words and graphics influence students’ cognitive performance, and whether there are interactive effects of their cognitive preferences that contribute to the results. This study recognises the complex nature of the web-mediated learning environment and the difficulties experienced by students due to their lack of prior programming knowledge, which were first analysed by Bagley (1990) and revised by McKay (2000a).

This empirical study examined the performance of novice-learners (or -programmers) with different cognitive preferences using two web-mediated instructional strategies: 1) text-plus-textual format and 2) text-plus-graphical format. The participants were primarily second year undergraduate students, in Malaysia, who are required to enrol in an Introductory Programming course as part of their program requirement for the first time.

They were classified for the purpose of this research as novice-learners. The subsequent data collected included scores from three instruments: 1) pre-test; 2) post-test and 3) cognitive styles analysis. The validity and reliability of the cognitive performance measurements (pre- and post-test) were confirmed according to Rasch model using the Quest interactive test analysis system software (Adam & Khoo, 1996). These cognitive performance measurements were found to have strong validity with the item fit statistics produced by the Quest estimate and ranged from 0.81 to 1.22, strongly upholding that the test-items sampled the one construct. Further analyses were conducted, based on effect size (Cohen’s d) to compare the magnitude of difference between groups (Cohen, 1977). The results suggest that there is no clear evidence that a cognitive preference plays an important role in cognitive performance when learning from web-mediated instructional modules. However, it has been observed that the analytic-verbalisers performed better when the instructional format they received suit their cognitive preferences with a medium effect size. As the participants were novice-learners, the influence of prior domain knowledge cannot be ignored.

The findings of this doctoral study contribute new knowledge to the existing literature in the related disciplines of human–computer interaction, web-mediated instruction, human cognitive processing and research methodology. Firstly, the use of high-quality measurement tools to assess the effectiveness of web-mediated instructional strategies is important. Secondly, the findings can provide guidelines on how to design web-mediated instruction for high-element interactivity knowledge domains such as the acquisition of programming concepts. Finally, the learner’s cognitive learning preference profile should be considered when designing web-mediated instruction and in particular their level of prior domain knowledge.
Degree Doctor of Philosophy (PhD)
Institution RMIT University
School, Department or Centre Business Information Technology and Logistics
Keyword(s) Cognitive preferences
web-mediated instruction
instructional strategies
Rasch model
Versions
Version Filter Type
Access Statistics: 374 Abstract Views, 519 File Downloads  -  Detailed Statistics
Created: Mon, 24 Sep 2012, 11:41:14 EST by Brett Fenton
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us