|M.Sc Student||Alperin Regina|
|Subject||Statistic Analysis of Hebrew Word Software Users' Complaints|
and Preparing Measures for Estimation of these
|Department||Department of Quality Assurance and Reliability||Supervisor||Dr. Lev Reznikof|
The quality of a software, like that of any other product, is measured by it’s compatibility with the users’ expectations and requirements as well as by his/her satisfaction level. In the case of a software like Microsoft Word intended for a wide range of users rather than for a specific category, the user’s impression is the dominant factor in the evaluation in the absence of formal requirements. Accordingly, the relevant information must be collected from the potential users and translated into quality solutions. The big question is how to do that.
The classic model of McCall comprises 11 software quality factors, defined from the software engineer’s point of view, rather than from the user’s. My own approach is based on the latter, whereby the number of factors was reduced to six and their definitions were modified accordingly, namely efficiency, reliability, usability, correctness, flexibility and interoperability.
Microsoft Word was the example used in illustrating the process of data collection and analysis.
The materials used in the study were answers to a questionnaire and users’ complaints collected from the Internet and from the questionnaire. The statistical techniques used were factor and distribution analysis, stepwise regression, one- and two-way ANOVAs, Pareto analysis, etc.