• Create BookmarkCreate Bookmark
  • Create Note or TagCreate Note or Tag
  • PrintPrint

Lab 4.3 Exercises

4.3.1. Understand Usability and Usability Measures

a) Discuss what factors are used when performing a user analysis.




Answer: The data gathered about a user includes the target user group and its gender, age, and cultural characteristics. Also take into account the job characteristics such as job description, the main activities and responsibilities, the management reporting structure, and any system of rewards. We also need to take into account a user's background such as education, experience, knowledge, skills, and training. The work environment may also introduce usage constraints such as a factory floor, mobile position, and other nonoffice locations. Finally, any user preferences should be collected.

b) Discuss how usability differs from utility.




Answer: Utility is represented by the feature list of a software package—data manipulation functions. Usability is the measure of how well human users understand the user interface and how effective they are at using the features.

For many people this difference is not intuitive. The utility of a jumbo jet is pretty straightforward: moving a large number of people from place to place through the air. The same jet could be converted to cargo to transport goods or converted to a spy plane for military purposes. In every case, the utility of the plane (what functions it provides) is fairly simple. However, for the vast majority of the people the plane is too complex for us to use. Although we may eventually be able to learn how to fly it, a jumbo jet has limited usability for us.

Let's take two software products as another example: MS-Word and Wordpad, which is provided as a free editor with recent versions of MS-Windows. MS-Word has far more features than Wordpad. We could say that it has greater utility. However, the sheer number of menus and buttons can be intimidating. People will try to do fancy things and get lost because there is simply so much there. As a result, it has low usability.

Wordpad has a fewer features and therefore less utility. However, for most people, it is relatively more intuitive to learn. It is easier for most people to use, therefore it has greater usability.

c) Discuss likability and why it is important.




Answer: Likability refers to the concept of pleasurable engagements user have with a product. A system may be usable, but may leave users without a sense of achievement or pleasure in carrying out a task. For example, users who are accustomed to a GUI do not “like” the command line and get no pleasure out of using it, although it may be the only way of carrying out certain functions.

Consider Shackel's usability measure “attitude”. If a person is using a product only because someone else made the decision or he or she has no other choice, that user will typically have a bad attitude about the product. If the product is difficult to use (perceived or real), users are not as effective with the product. I know people who are almost intentionally careless with the command line because they say it's hard to use, so why bother making any effort (despite the fact this attitude is counterproductive).
d) Discuss the relationship of usability and effectiveness.




Answer: Effectiveness is essentially a measure of how well the desired goals have been achieved. An application is considered effective if it reaches or exceeds a particular goal. The easier a product is to use, the greater the likelihood that the goals will be reached and the greater the effectiveness of the product.

One key aspect is that effectiveness is not absolute. First, you have the obvious factor of human preferences. You may be familiar with the expression “Close enough for government work.” Some people seem to have the attitude that someone working for the government has (or needs) lower goals. Although this is meant as a jab at government employees, the implication is that the goal for government employees can be less than that for private businesses.

Consider a hotline. If you are talking about a mail-order company, then having as many us 10% of the people getting busy signals is acceptable. For a software support company, this may need to be as low as 5%. For an emergency service like the fire department or police, busy signals are not acceptable at all. Whether you need a new sweater or need to get your document to print is an emergency is a matter of opinion. Therefore, what constitutes an acceptable “drop” rate on callers differs for each situation. If the mail order only drops 7%, it is very effective, at least in the opinion of the manager who wants only a 10% drop rate. However, the software company is of the opinion that 7% is too high. Both have the same absolute value (7%), but their perception of the effectiveness is different.

As with a hotline, the effectiveness of software often depends on the user. As we discussed in Exercise 4.3.1c, how much the person likes the software helps determine the usability and therefore the effectiveness. If the interface is not usable (for whatever reason), the user cannot be effective. However, the product may still be considered effective as a whole. For example, your company with 500 employees might use MS-Word, but you prefer Corel's WordPerfect. Although there are plenty of reasons why WordPerfect is better than MS-Word, your company standardizes on MS-Word, sacrificing your effectiveness for a higher aggregate effectiveness. This is addressed in Table 4.2 in that a product is effective for “some required percentage of the specified target range of users.”

Despite the features, MS-Word is not very good at drawing. Within the “range of usage environments,” which includes graphics, one cannot use MS-Word very well to create graphics and it is therefore not effective.

4.3.2. Apply the Basics of User Interface Evaluation

a) Discuss why usability evaluation is important.




Answer: One reason evaluation is important is that quality (regardless of the context) is often arbitrarily defined and there is inadequate intuition about quality. One user calls a product “good,” but there needs to be some subjective measure of this goodness. In addition, people making the decisions about products often make mistakes about user needs and their reactions, particularly if they aren't the ones who are going to use the product, but are basing their choice on other sources. These same people often under/overestimate users' skills and limitations. When developing something internally, many people forget that developers are not ordinary users and cannot imagine users' confusion or lack of understanding.

We saw in Chapter 2 that evaluation lies at the heart of the modern software development life cycle, and it is involved in several iterations as implementation progresses. Therefore, it is essential that evaluation methods should be fast to apply and minimize work by evaluators. The outcomes of evaluation quite naturally should be the identification of problem areas and suggestions for possible solutions.

These objectives should be independent of the type of user interface type (character-based or GUI), the hardware and software, the design stage, the evaluation method used, and the type of data gathered.

Figure 4.5 shows some figures calculated by Hewlett-Packard for the costs of software production. The vertical axis is logarithmic and shows the cost of fixing faults at different stages in the software life cycle. Each black bar also shows the number of hours needed to perform usability evaluation at different stages of the life cycle of a software product. It is assumed that the user interface is improved after each evaluation, and the high cost during maintenance is caused by the deployment costs when a new version has to be distributed to end users.

Figure 4.5. Time and Costs for Evaluations in the Development Cycle (Hewlett-Packard)

Figure 4.5 shows that usability evaluation is a major software development cost factor. In fact, it forms 20 to 60% of the initial code development cost. Early detection of usability problems is needed as the cost during the specification and design stages is much lower. Problems found in later stages cost much more to fix. We saw in Chapter 2 that high usability measures can increase income, resulting in better reviews and more satisfied and productive users. Evaluation of usability shows tangible savings, from about 2:1 for small projects up to 100:1 for larger projects.

b) Discuss which of the two main types of evaluation testing is more practical.




Answer: The two types of evaluation technique are predictive models and experimental techniques. Predictive models rely on HCI theory, which is not perfected. In many cases, the evaluations are very subjective and many not apply in a specific circumstance. On the other hand, experimental techniques rely on controlled circumstances and so are more reliable. It is generally straightforward for you to define evaluation criteria that fit your particular situations and define specific results that you want to achieve.

By experiment, a number of useful measures have been used to determine the level of usability of systems under test. Table 4.3 shows the more important measures used.

Table 4.3. Usability Evaluation Measures
Task completion
  • Number of tasks correctly completed

  • Number of tasks completed in given time

  • Time taken per task

Action usage
  • Frequency of use of different commands

  • Use of command sequences

  • Use of special command (e.g., “help”)

  • Use of keyboard equivalents

Display perusal
  • Time spent looking at display

  • Comparative data for different screen designs

User errors
  • Classification of error types

  • Frequency of error types

  • Time spent in error situations

  • Time taken to correct errors

Input devices
  • Comparative time taken to execute tasks

Specialist tools are needed to set up, carry out, and analyze the results of usability evaluation experiments. Additional software development may be needed to instrument the systems under test so that they can record the measures in Table 4.3. Some of these measures can be extracted from direct observation or video recording. However it is done, teams of evaluators and locations in which to perform the evaluations are needed. We refer to the evaluation locations as usability laboratories.

All major software manufacturers now utilize usability laboratories in which to test prototypes of the products. The laboratories contain special observational equipment such as video cameras, hidden observation points, and specially instrumented computers and workstations.



Not a subscriber?

Start A Free Trial

  • Creative Edge
  • Create BookmarkCreate Bookmark
  • Create Note or TagCreate Note or Tag
  • PrintPrint