What makes users happy?

Reviewer: Thomas C. Lowe

This paper reports on a controlled survey which asked the question “What makes users happy__?__” The responses are based on a mailing sent to subscribers of Computerworld magazine. The question is answered by relating 17 independent variables (satisfaction with response time, the extent to which systems are meeting users' expectations, etc.) to one dependent variable—overall user satisfaction. Some results are not surprising: users like short response time, for example. Some are puzzling. The authors ask, for example, “Why, for instance, do mainframes have a relatively large negative effect on the dependent variable__?__” Certainly, the paper serves to remind us that the old criteria of what is good, better, or worse should be reviewed. However, it is not always possible to agree with the authors' reasoning. For example, it appears from Table I that about 7.5 percent of the users are reporting on microcomputers, and it is reasonable to conclude that very few of these are portables. If Figure 1 gives a clear picture of the survey instrument(s), then one of the questions requires that the subject provide a numerical percentage of the time the following is a correct statement: “System is power/energy efficient.” The analysis provides a single relative satisfaction provided by the “power/energy” factor for all categories of systems, from microcomputers to mainframes. The authors then attribute the importance of that factor to arguments relating to power supply, including both portables with their own sources and large computers with their own generators. It is a far reach from a small positive influence of the “power/energy” factor on overall satisfaction, to the reasoning explaining that correlation. This reasoning includes the statement that “Computers with their own power supply can be safer and more easily protected from instable sources of energy, and are likely to be more economical since they will not overheat, and go down very frequently.”