University of Twente Student Theses

Login

Testing time invariance of process parameters in serious games : measured and applied in the application game Leo’s Pad

Steenbeek, C.D. van (2015) Testing time invariance of process parameters in serious games : measured and applied in the application game Leo’s Pad.

[img] PDF
1MB
Abstract:The company ‘Kidaptive’ is a commercial business which develops apps and websites for children to give them a learning experience while they are having fun. One of their most popular products is Leo’s Pad, an adaptive-learning series for preschoolers which they can experience on a smart electronic device like a tablet or smartphone (Kidaptive, n.d.). In this application, children learn concepts like problem solving, creative thinking, social emotional awareness and other. Gamers who get immediate feedback by intelligent and adaptive tutoring, get more involved and motivated to play the game. For example, when a child completes a level far more quickly than he normally does or than his peers do, there should be given a different type of feedback than when a learner is much slower. This requires a statistical model, which is able to detect the variation in performances across players. Changes in the performance measurement can be used to inform a player and to provide stimulated feedback. The parameters of the statistical model are often assumed to be invariant over time. Then, scores of a gamer are not time-dependent and can be used to assess the performance and give informed feedback. However, the time invariance assumption of the model parameters needs to be tested. This leads to the research question: how can variations in item characteristics over time be identified. The developed tests will be applied to data of the learning game Leo’s Pad to test the time-invariant properties of the game’s items. This research has focused on the characteristics of the items studied over time. This is to find out if the items are consistent over time and are not sensitive to time-correlated variables like knowledge or impact of technology. Several statistical tests will be developed and applied using the response times as observations and the program ‘R’. The conclusion is that the likelihood ratio test is the best method to answer the research question.
Item Type:Essay (Bachelor)
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:77 psychology
Programme:Psychology BSc (56604)
Link to this item:http://purl.utwente.nl/essays/68544
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page