EXAM B with SOLUTIONS

Part I (worth 40%)

Attempt four questions from the six in this part. Each question is worth 10 marks.

Question 1

A designer of a user interface needs two kinds of knowledge. One kind is specific to the job in hand; the other is more general design knowledge.

(a) An example of the former kind is knowledge of the user. Name four user characteristics which a designer would want to find out about when designing a user interface. [2]

(b) Besides knowledge of the user, give two more areas of knowledge which are important to a designer of a user interface and specific to the job in hand. [2]

(c) Describe briefly the difference between a design principle and a design rule. Illustrate your answer with one example of a design principle and one of a design rule. [4]

A designer of a web site uses mostly black 12 point Roman font for her text. Occasionally she uses red 18 point Bold Roman. Describe an instance where this change to a larger font might be appropriate. State which design principle she is ignoring when she uses this occasional change of font. [2]

Answers to Question 1

(i) and (ii) are easy, to get them started

(i) 4 distinct characteristics from: age; gender; physical limitations; general educational background; computer/IT experience; motive; attitude; cultural background (or anything else reasonable and distinct from the others cited).

½ mark for each

(ii) 2 from: knowledge of tasks; knowledge of the context, and knowledge of the domain

1 mark for each: easy marks to get them started.

(ii) A design principle is a rule which is general over any design (the text says ‘abstract, high-level guides for design’).

A design rule is specific to a context, and can be followed with the minimum of interpretation/translation by the designer.

An example of a principle is: ‘a designer should use interface objects consistently’

An example of a rule is: ‘a MicroSoft window can be closed using the x button on the top right hand side’

4 marks: 1 each for the description, which must use words like abstract/general/high-level for principle, and concrete/specific/contextual for rule.

1 each for examples. If it isn’t absolutely clear which example is for principle, and which for rule, then no marks. Just naming the principle (e.g. ‘consistency’) - only ½ mark.

The rule need not be derived from the principle.

She might use 18 point bold to draw the user’s attention to something specially important or significant e.g. bargain offers on an e-shopping site. Consistency is the principle violated.

2 marks; 1 for drawing the user’s attention with concrete example (I can’t think of any other real reason – ‘providing variety’ is too vague) - ½ if no concrete example;

1 for naming consistency.

Question 2

(a) A designer is observing a video of a user performing typical tasks on an existing user interface. Give four distinct examples of user behavior which should alert the designer to problems with the interface. [4]

(b) Give one advantage and one disadvantage of the two techniques of think-aloud protocols and

retrospective protocols. [4]

Identify which of these two techniques would be more appropriate in each of the following situations.

(i) You want to investigate why it takes a user so long to do an apparently straightforward task.
Your most important need is to measure the time taken to do the task, but at the same time, you need to know the user’s thought processes as s/he performs the task. [2]

Answers to Question 2

(i) Four from:

- hesitancy

- demonstrating confusion

- making errors

- Seeking help (from on-line sources, or manuals, or colleagues, or external aide-memoires);

- Using work-arounds (complicated, improvised sequences of actions to achieve a function which isn’t provided).

- (or anything else reasonable and distinct from the others cited).

4 marks, 1 for each

(ii)

Protocol / Advantage / Disadvantage
Think-aloud / Should give access to the user’s conscious thoughts as she performs the task. / The act of vocalization disturbs the train of thought – not the natural way of doing the task.
Retrospective protocol / Doesn’t perturb the natural way of doing the task / The user may forget what s/he was thinking as she carried out actions. OR
The user might do some post-hoc rationalization (i.e. present her thinking processes as being much more rational than they actually were).

1 mark for each reasonable advantage/disadvantage – total of 4

(a) is verbal protocol; (b) retrospective (because of the importance of timing). If students do not give these answers, but give a compelling reason for their own answers, then full marks.

1 mark for each.

Question 3

(a) Consider a GUI help system based on keyword search. Describe in one sentence a disadvantage of this interaction style. [1]

(b) Identify a suitable interaction style for each of the following systems. The style you choose should be different for each system.

In each case,

· give a brief reason for your choice

· describe briefly a disadvantage normally associated with the chosen style

· explain why this disadvantage is not an issue with the system given.

(i) A chess-playing system, in which the user plays the system [3]

(ii) a system for clerical officers to enter assignment information (assignment number; date returned; grade obtained) into a university’s students’ database. [3]

(iii) a system for UNIX system administrators to keep track of a large network [3]

Answers to Question 3

(a) A disadvantage of a system searching on key words is that the user may not use the exact key words known to the system.

1 mark for anything reasonable along these lines

(b)

(i)

Direct manipulation

Reason This is consistent with the interaction style used with most graphically-based games programs

OR it uses the metaphor of how people play chess in real life.

Disadvantage. The user may not recognize the meaning of the icons OR the metaphor

In this context: there are well known and recognizable icons for chess pieces, as used in books etc.

(ii)

Form-fill.

Reason: uses a metaphor that the users are likely to be familiar with.

Disadvantage: lack of flexibility.

In this context: the task is very constrained, so lack of flexibility is not an issue

(iii)

Command line.

Reason Gives a lot of flexibility, by way of parameters and/or macros.

Disadvantage: people have to learn the syntax and semantics of the language (alternative: the syntax and semantics of the language are difficult to remember)

In this context: the users have UNIX system administration as the main focus of their work, so are likely to be willing to expend effort in learning the language

OR: the users are likely to be familiar with using programming languages and won’t find learning a command language either particularly difficult or unmotivating

(alternative: administrators are likely to be working full-time using the command language OR they are experienced computer users so used to finding help).

3 marks for each answer: 1 for each of reason, generic disadvantage and consideration of context. Actual choice of interaction style doesn’t matter, providing the 3 interaction styles are different and that the reason, generic disadvantage and consideration of context are convincing. The consideration of context must refer to the identified generic disadvantage.

Question 4

(a) Describe how card sorts might be used to inform

(i) the structure of menus

(ii) how web-pages within a site are linked. [4]

(b) Whether or not a feature of a design works well depends on the context of system use.

For each of the following design features, describe a context in which it would work poorly, and explain why.

(i) Hypertext links in the home page of a web-site are represented by diagrams..

(ii) Alarm conditions are signaled by a loud noise.

(iii) A single button press accomplishes a single critical user goal. [6]

Answers to Question 4

(a)

- A set of relevant concepts (e.g. actions) is written on a set of cards and users are asked to divide the cards into groups. The subsequent groupings can be used to inform how the concepts are divided between menus and how they are grouped within menus.

- For web-pages, the main theme of each page can be written on a card and sorted into groups as above. From the groups, it can be established how closely linked web-pages should be. For example, web-pages in the same group might be linked directly.

4 marks: in each case, 1 for a description of the technique, and 1 for how the results might be used.

(b)

(i) This is difficult for visually impaired people using screen readers OR it may take an unfeasible time to download

(ii) Wouldn’t be useful in a noisy environment OR one where there are many alarm conditions OR where the users might be hearing impaired.

(iii) This feature would be poor on a device where it is very easy to press the button accidentally. For example, a single button press on a mobile phone to ring 999 might lead to many accidental calls [this is an example given in Chapter 1]. OR might lead to a significant and unwanted result.

6 marks: 2 for any reasonable context and explanation.

Question 5

(a) Describe the evaluation elements of observe, listen, measure and compare as applied to a heuristic evaluation undertaken by an expert. [4]

(b) Suppose you are designing a brand-new system, for which, of course, there are no existing users.

How would you choose participants for your user observations? [2]

How would your choice influence your interpretation of the results? [2]

Answer

(a) evaluation elements for an heuristic evaluation:-

observation: of interface OR of expert observing herself [I prefer the former but the text gives the latter]

Listening: to the expert (who may be oneself)

Comparing: against a list of heuristics

Measuring: how many heuristics are violated and the frequency of each such violation.

1 mark for each of the above – ½ marks for each of the points made in measurement.

(b) Choice of participants. A new product should be designed with target customers in mind (e.g. as established by market research). The participants should be chosen from the target market.

If this isn’t possible, then one can choose colleagues, friends, family, anyone else.

1 mark for target market; 1 for anyone else

Interpretation of data. Data from the target customers are more to be trusted. Data from colleagues, friends, family, anyone else should be regarded as being possibly biased: they may just want to please you, and they may not meet the profile of the intended users of the system (indeed, if they do, then they’re part of the target market). In the latter case, they may not react to the system in the same way as the target customers.

1 mark for data from target customers being more trustworthy; 1 for some description of how data from other sources might be biased.

(c) Satisfaction can be quantified by use of an ordinal scale in questionnaires and structured interviews, that is, a scale that is, or can be converted to, numbers. For example, dislike very much…… like very much. SUMI questionnaires can also be used.

2 marks for a reasonably complete answer, as above.

Question 6

(a) It is proposed to redesign a user interface. For each of the following groups, suggest two distinct concerns which might cause them to be reluctant to support the idea and/or execution of a redesign.

(i) the users who work with the existing user interface

(ii) their line managers. [4]

(b) You are working as a usability expert in a Management Information Systems (MIS) department of an organization which develops systems and provides IT support to the rest of the organization. Briefly describe three arguments to convince the Finance Director of the benefits of designing usable systems. [6]

Answers to Question 6

(a) Users might be reluctant to support any redesign because

- they are reluctant to learn to use a redesigned system

- because they fear the effect a redesigned system may have on their job security (it might make them redundant)

Managers might be reluctant, because of time pressures, to spare users for usability observations and evaluations

They may take the proposed redesign as an implicit criticism of the work they do currently

4 marks: 1 for each distinct reason (accept anything supported by a reasonable argument)..

(b)

- a more usable system will cut down on training costs

- a more usable system will cut down on requests for help, both from the user to MIS departmental members & from the user to his/her colleagues. Thus time – and hence money- will be saved.

- a more usable system should cut down on errors and hence rework. Again, time, and hence costs will be saved.

Marks: 2 for each argument: 1 if it isn’t couched in terms of cost. Accept any reasonable argument – again, only 1 mark if not couched in terms of financial attributes

Part II (worth 60%)

Question 7

In the context of the KitchenMade scenario, identify six specific considerations that should be taken into account when designing the user interface. Your answers must be drawn from any or all of the following areas.

· users

· tasks

· the environment

Justify why each consideration is important, and briefly outline how it could be addressed by the user interface design.

[12 marks. 2 marks for each of 6 different issues]

Answers to Question 7

The following are some of the issues. Each issue must be linked to the scenario.

Users

Age. The scenario specifically says that the system is targeted at elderly people. This might affect their physical abilities and attitude as below. It should not be too ‘flashy’ in overall appearance, in a way which would appeal to a younger age-group and put off an older.

Physical abilities. Given the age of the target user group, the system developers should support users who have problems with eyesight, hearing and motor control. There should be a variety of input and output options, with care taken especially about visual elements of the interface.