University of Twente Student Theses

Login

Predicting persistency of usability problems based on error classification : a longitudnial study on improving mobility for the elderly

Zandbergen, R. (2015) Predicting persistency of usability problems based on error classification : a longitudnial study on improving mobility for the elderly.

[img] PDF
2MB
Abstract:Social isolation and loneliness are becoming increasingly serious problems among the elderly. MOBILE.OLD is a project which helps elderly stay independent, healthy and mobile by creating services for mobile devices. This study evaluated the usability of the designed prototype services for this project. Earlier studies have shown the need to look at persistency of usability problems for elderly users when you want to get a clear image of how well your new product is learned by them, as the elderly need a little bit more time to ‘get started’. For this reason longitudinal study designs are very appropriate for elderly users. However, due to deadlines and budget restraints, longitudinal designs are often not used. This study wanted to use the results from the MOBILE.OLD project to predict persistency of problems. If this would be possible, predictive measures could become a cheaper alternative to the longitudinal design. Prediction of persistency was attempted by using error classifications for the usability problems. An extended matching protocol was created to incorporate the error classifications in the matching steps. To help evaluators classify the incidents to error categories, a step-by-step classification guideline was constructed. Previous experience and technology enthusiasm or ‘geekism’ were used as predictors for persistency. A sample of twenty elderly users between the age of 61 and 82 tested ten different applications. Data was collected by capturing video, questionnaires and think aloud procedures. The longitudinal data was used to create persistency patterns for the problems. Three different groups of persistency patterns were further investigated: disappear early, appear late and persistent. Elderly encountered mostly knowledge-based problems and geekism was shown to influence the number of KBFR problems that users encountered. It proved to be difficult to predict the persistency of the problems that elderly encountered during their learning efforts, but some findings were promising in supplying further inspiration for new studies on persistency.
Item Type:Essay (Master)
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:77 psychology
Programme:Psychology MSc (66604)
Link to this item:http://purl.utwente.nl/essays/67069
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page