Functional Requirements for Creating Reliable Self-Screening Tests by Non-Developers

Abstract

Self-screening test is a questionnaire created for public users to guide them to determine his/her health condition and its severity. Computer-based online self-screening tests are easily accessible and also allow the creation of more inclusive questions. However, there are two prevailing issues: 1) The need to depend on the application developers to create the tests and, 2) There are doubts about the reliability of the respondents' answers to the comprehensive questions. Existing self-screening tests do not provide adequate supports to tackle the dependency and reliability issues, which our research addresses. To deal with the dependency on the application developers, we propose to equip computer-based test setters with an editing template. It can facilitate test authors (authorised experts without programming skills) to create the self-screening tests on their own. To deal with the reliability concern that originates from inconsistent users responses, we propose to facilitate the test authors to specify definite rules to detect and handle inconsistent responses in the screening test. Following the initial stages of the Engineering research approach, we gathered and analysed a number of health screening questionnaires and reviewed various methods for detecting inconsistencies in users' feedback. Based on our review and analysis, we identified four key components of the test editing template: Question setting, Answer scoring, Consistency checking and Decision-making. Consequently, we specify the functional requirements of each of these components in this paper, which form the basis for the design and development of the screening test editing template.