Usability Testing Research Papers - Academia.edu (original) (raw)

Statistical analysis has meant to provide statistical methods for making tests of significance and trustworthy estimations of the magnitude of the effects indicated by the results for the reduction of data. Statistical method on the other hand involves the use of certain logical ideas appropriate to experimental procedure. As this research intends to review the different statistical methods for data reduction during the administration of usability evaluation as well as provides readers with statistical methods’ definitions and concepts, it will help researchers especially the one who conducts usability evaluation becomes familiar and find appropriate statistical methods to treat the data collected and these methods include: (1) frequency distribution, (2) percentage, (3) mean score, (4) standard deviation, (5) likert scale, (6) analysis of variance, (7) chi-square, and (8) pearson’s r (pearson product-moment correlation coefficient).

The main focus of this paper is to provide other researchers a review on the different statistical methods used to treat data obtained from usability evaluations for data reduction. In order to achieve this specified goal, the researcher will present various previous studies like the study of Thuseethan et al., Hammouche, Penha et al., Zhao, Deotale, Iqbal et al., Manlai, Sauro and Kindlund, and Joo and describes the process of how they statistically treated data collected from different usability evaluation techniques.

Although this paper’s primary focus is about statistical methods, an overview about usability evaluation and its methods were discussed to provide readers a brief introduction about usability concepts and familiarize them with most usability methods used in usability testing and evaluation.

Various evaluation methods exists that serve to measure system’s usability and these methods can be analytical or empirical. Analytical usability methods are conducted by usability experts who put themselves as the users of the application or the system while the empirical usability methods consist of various usability tests and questionnaires. The empirical usability methods can be done if a prototype of the system is already available and ready to use.

An example of analytical usability method is the Heuristic Evaluation that serves to measure a system’s usability. The recent Heuristic Evaluation – Usability Techniques used today was designed by Deniese Pierotti of Xerox Company that comprises of 13 heuristics: (1) Visibility of system status, (2) Match between system and the real world, (3) User control and freedom, (4) Consistency and standards, (5) Help users recognize, diagnose, and recover from errors, (6) Error prevention, (7) Recognition rather than recall, (8) Flexibility and minimalist design, (9) Aesthetic and minimalist design, (10) Help and documentation, (11) Skills, (12) Pleasurable and respectful interaction with the user, and (13) Privacy.

The empirical usability method on the other hand is an approach which consists of various usability tests and questionnaires. Examples for usability evaluation techniques that measures usability are: (1) System Usability Scale (SUS), (2) Software Usability Measurement Inventory, (3) Post-Study System Usability Questionnaire (PSSUQ) and Web-based Learning Environment Instrument (WLEI), and (4) post-tasks walkthrough.

In this study, various statistical methods will be explored and reviewed. The selection of these methods will focus primarily on studies where data obtained from usability evaluation techniques such as: (1) Frequency distribution, (2) Percentage, (3) Mean score, (4) Standard deviation, (5) Likert scale, (6) Analysis of variance, (7) Chi-square, and (8) Pearson product-moment correlation coefficient (or Pearson’s r).

Although statistical treatment for the reduction of data is preferably done by statistician to avoid uncertainty of data, listing various statistical methods in this study will help researchers to understand its importance and use.

The findings from each selected research show proofs and evidences that choosing appropriate methods for evaluation and statistical treatment is important and the information gathered from previous articles and studies support this research to help other researchers become familiar on choosing appropriate statistical methods used in usability evaluation and since the results of the statistical analysis process were recorded through this research it will help the future researchers with similar study to easily retrieve it for future use.