Abordagem híbrida para a avaliação da usabilidade de interfaces com o usuário.

Abstract

In this paper, three categories of software product usability assessment approaches are confronted, namely: user performance measurement during product use, product compliance inspection to an international standard, and product satisfaction user satisfaction survey. . The purpose of this study was to investigate the influence of satisfaction, performance and conformity to an international standard on product usability. The research consisted of investigating the relationship between pairs of usability evaluation factors: (i) performance and user satisfaction; (ii) user performance and product conformation to a standard; and (Hi) user satisfaction and product conformation to a standard. From a set of seven hypotheses, an analytical study was developed, aiming to test them. Product inspection sessions were conducted based on Parts 14, 15 and 76 of the international standard ISO 9241. Subsequently, a usability test was planned and conducted in a controlled environment (usability laboratory), with the participation of 40 users of test. Quantitative and qualitative data were collected: (i) profiles and opinions of test users on product usability; (ii) written and video recordings of user performance during product use; and (iii) evaluators' judgments, based on the ISO 9241 standard. Statistical and non-statistical analyzes were performed on the tabulated data. Finally, conclusions were formulated, based on the results of the statistical hypothesis tests. The results show that the evaluative approach adopted produced a more complete picture of the product usability flaws than the isolated application of any of the considered techniques. The research also resulted in a detailed description of the three approaches investigated, with recommendations on when and how to apply them, individually or integrated in a hybrid approach. The selection of usability evaluation methods is usually heuristic, based on the evaluator's previous experience. Additionally, the available literature usually does not sufficiently detail the proposed assessment methods. This paper discusses not only a number of criteria to consider when selecting methods, but also provides a solid foundation to support the selection process. The investigated approaches produced complementary results regarding the identification of flaws in the quality of the evaluated product. Although the results obtained from the different approaches adopted were consistent with each other, discrepancies were found that suggest the combination of evaluative approaches as a means of obtaining a broader view of product usability issues.