CS/183/23/SS08 SOLUTIONS

UNIVERSITY OF SURREY ©

BSc Programmes in Computing

Level 1 Examination

CS183: Systems Analysis and Design

Time allowed: 2 hours Spring Semester 2008

********************************************************
ANSWERS
********************************************************

*IMPORTANT*
WRITE YOUR URN HERE …………………………………………………

THIS QUESTION PAPER MUST BE HANDED IN WITH YOUR

ANSWERBOOK AT THE END OF THE EXAMINATION

SEE NEXT PAGE

CS/183/23/SS08 - Page 2 - SOLUTIONS

SECTION A - Answer ALL questions in this section [one mark each].

*(answers are in bold)*

1. The systems development life cycle (SDLC) is the process of _____.

A. building an information system

B. delivering an information system to a client
C. designing an information system
D. understanding how an information system can support business needs
E. all of the above

2. In which phase of the SDLC is the system proposal developed?

A. analysis
B. design
C. implementation
D. planning
E. system delivery

3. The systems development methodology that emphasizes simple, iterative,

application development that provides rapid feedback to users on an ongoing
basis, with continuous development and testing is ________.

A. modular development
B. rapid application development
C. prototyping
D. agile development
E. throwaway prototyping

SEE NEXT PAGE

CS/183/23/SS08 - Page 3 - SOLUTIONS

4. The primary advantage of the Waterfall Development methodology is that _____.

A. a version of the system is quickly delivered into the users' hands

B. requirements evolve through users' feedback about the system
C. features and functionality of the system are explored through simple models
D. requirements are completely specified and held relatively constant prior
to programming
E. there is a long time lapse prior to completion

5. The principal disadvantage(s) with the waterfall development methodology is

(are) _____.

A. a long time elapses between completion of the system proposal and the
delivery of the system
B. it identifies system requirements long before programming begins
C. it minimizes changes to the requirements as the project proceeds
D. the design must be completely specified on paper before programming begins
E. a long time elapses between completion of the system proposal and the
delivery of the system and the design must be completely specified
(usually on paper) before programming begins

6. The analysis phase of the SDLC answers _____.

A. who will build the system and when will it be used

B. who will the system be for, what the system will do, when will it be used,
and where will it be used
C. why build the system, what the system will be, and how the system will work
D. why build the system, who will the system be for, when will it be used, and
how the system will work
E. why build the system, who will the system be for, when will it be used, and
where will it be used

SEE NEXT PAGE

CS/183/23/SS08 - Page 4 - SOLUTIONS

7. If the system under development will employ technology with which the analysts
and programmers are unfamiliar, the most appropriate methodology to employ is:

A. Waterfall Development
B. Throw-Away Prototyping
C. Parallel Development
D. Phased Development
E. Prototyping

8. The _____ development methodology is never recommended when system

reliability is critical.

A. parallel
B. phased
C. prototyping
D. throwaway prototyping
E. waterfall

9. Which of the following systems development methodologies involves developing

the system in a series of sequential versions?

A. Parallel Development
B. Phased Development
C. Prototyping
D. Throw-Away Prototyping
E. Waterfall Development

10. _____ development is a structured design methodology that divides the project
into a series of distinct subprojects that can be designed and implemented
together.

A. Parallel
B. Phased
C. Prototyping
D. Rapid Application
E. Throwaway Prototyping
SEE NEXT PAGE
CS/183/23/SS08 - Page 5 - SOLUTIONS

11. The _____ is generated by the department or person that has an idea for a new
information system.

A. feasibility analysis
B. gradual refinement
C. project sponsor
D. system request
E. work plan

12. The information gathering technique that is most effective in combining

information from a variety of perspectives, building consensus, and resolving
discrepancies is ________

A. interviews.
B. JAD (Joint Application Design) sessions.
C. questionnaires.
D. document analysis.
E. observation.

13. In a JAD (Joint Application Design) Session, which of the following is NOT a
proper role for the facilitator?

A. Assist the group in understanding the analysis techniques.

B. Document group output on a public display.
C. Ensure that the JAD session agenda is followed.
D. Cast the deciding vote on controversial issues.
E. Guide the discussion.

14. During document analysis the systems analyst learns that users in the purchasing
department have created their own forms. This is a clear sign that the _____.

A. as-is system is meeting user needs

B. department was overlooked during the interview process
C. process does not need improvements
D. system needs to be changed
E. users in the department are innovative
SEE NEXT PAGE
CS/183/23/SS08 - Page 6 - SOLUTIONS

15. During an interview, the following question is asked; "How many times during a
typical week does a customer complain about inadequate service following a
sale?" This question is an example of a(n) _____________

A. probing question.
B. open-ended question.
C. closed-ended question.
D. inappropriate question.
E. opinion question.

16. A systems analyst has prepared an interview that begins with specific, detailed
questions, and then asks the interviewee to make general statements about policies
and procedure of the business process. This is a(n) _____ interview structure.

A. bottom up
B. closed
C. open ended
D. probing
E. top down

17. The examination of existing paperwork in order to better understand the as-is
system is an example of what information-gathering strategy?

A. document analysis
B. interviewing
C. joint application design (JAD) sessions
D. observation
E. questionnaires

SEE NEXT PAGE

CS/183/23/SS08 - Page 7 - SOLUTIONS

18. The information gathering technique that enables the analyst to collect facts and
opinions from a wide range of geographically dispersed people quickly and with
the least expense is the _____.

A. document analysis
B. interview
C. JAD session
D. observation
E. questionnaire

19. The systems development methodology(methodologies) that is(are) most effective

when the user requirements are particularly unclear is(are) _______.

A. waterfall development
B. parallel development
C. phased development
D. prototyping
E. prototyping and throwaway prototyping

20. Which of the following development strategies is preferred when the business
need is common and time is a limiting factor?

A. in-house, custom development

B. purchased software
C. end-user development
D. outsourcing
E. None of the above

21. The foundation of the Unified Modelling Language is the _____________

A. use case.
B. object.
C. activity diagram.
D. message.
E. method.
SEE NEXT PAGE
CS/183/23/SS08 - Page 8 - SOLUTIONS

22. The Unified Modeling Language diagramming technique that illustrates the
objects and messages involved in a use case is the _____________

A. use case diagram.

B. sequence diagram.
C. class diagram.
D. statechart diagram.
E. None of the above

23. Each use case describes how the system reacts to a(n) _____ that occurs to trigger
the system.

A. activity
B. event
C. model
D. session
E. transition

24. Which of the following is often left out by end users when identifying elements
within Use Case steps?

A. analytical analysis of the identity elements

B. JAD session classification
C. requirements definition
D. role play analysis
E. seldom-used activities or special cases

25. When developing use cases, the project team first identifies the _______.

A. managers that supervise the use case department

B. place where the use case occurs
C. time the use case begins
D. trigger that causes the use case to occur
E. users who perform the use case

SEE NEXT PAGE

CS/183/23/SS08 - Page 9 - SOLUTIONS

26. Validating that the use case is correct and complete is the major activity
performed during which step of the Writing Use Case Reports process?

A. identify elements within steps

B. identify the major steps within each use case
C. identify the use cases
D. confirm the use cases
E. write the use case report

27. Which of the following diagramming symbols represents a generalization

relationship in a use case diagram?

A.

B.

C.

D.

E.

28. Which of the following relationships represent an enhancement of the

functionality of the use case to cover optional behaviour?

A. association relationship
B. extend relationship
C. include relationship
D. generalization relationship
E. optional relationship
SEE NEXT PAGE
CS/183/23/SS08 - Page 10 - SOLUTIONS

29. Buying in packaged software as a design strategy is good if _____.

A. in-house functional and technical experience exists

B. the business need is common
C. there is a desire to build in-house skills
D. the timeframe is flexible
E. the business need is core to the business

30. A custom development design strategy is NOT good for _____.

A. building technical skills

B. creating functional knowledge in-house
C. improving creativity
D. increasing flexibility
E. lowering risk

31. A disadvantage of outsourcing the application development process is that the

organization may ______.

A. build technical skills and functional knowledge in-house

B. lose control over confidential information
C. create a unique business need
D. remove all risk from the project
E. make strategic changes during implementation

32. Manipulating certain parameters to refine the way features work in a software
package is called ______.

A. a workaround
B. adjustment
C. customization
D. feature adjustability
E. parameterizing
SEE NEXT PAGE
CS/183/23/SS08 - Page 11 - SOLUTIONS

33. An advantage of purchasing packaged software is that the organization may

______.

A. accept functionality that is not a perfect fit

B. build technical skills and functional knowledge
C. have developers climb the knowledge ladder
D. make strategic changes during implementation
E. save money on the purchase

34. “A-part-of” or “has-parts” relationship represents a(n) ______ relationship.

A. generalization
B. association
C. aggregation
D. subsetting
E. vague

35. “A-kind-of” relationship represents a(n) ______ relationship.

A. generalization
B. association
C. aggregation
D. subsetting
E. vague

36. Which of the following would not be an appropriate class name?

A. student
B. patient
C. James
D. customer
E. doctor

SEE NEXT PAGE

CS/183/23/SS08 - Page 12 - SOLUTIONS

37. A(n) _____ describes information about an object.

A. attribute
B. behaviour
C. operation
D. message
E. instance

38. Which of the following is NOT one of the four basic functions of a computer
application?

A. application logic
B. data access logic
C. data storage
D. networking logic
E. presentation logic

39. Assume that your network has a server and three clients, the network is an
example of a(n) _____.

A. 4-tiered architecture
B. single-tiered architecture
C. three-tiered architecture
D. tierless architecture
E. two-tiered architecture

40. A server in a typical client-server based system performs which of the following
application functions?

A. application logic
B. data access logic and presentation logic
C. data storage and data access logic
D. data storage and presentation logic
E. presentation logic and application logic
SEE NEXT PAGE
CS/183/23/SS08 - Page 13 - SOLUTIONS

41. What is one primary problem with server-based computing architectures?

A. Servers are no longer made by hardware vendors.

B. Client terminals are no longer made by hardware vendors.
C. As users place heavier demands on the server, it is costly to increase
server capacity.
D. It is difficult to maintain security in the server-based environment.
E. Servers are difficult to connect to client terminals.

42. Which of the following is NOT one of the fundamental computing architectures?

A. Client-based computing
B. System-based computing
C. Server-based computing
D. Client-server based computing

43. An analyst depicts the static view of an information system with ______.

A. use-case models
B. structural models
C. behavioural models
D. interaction diagrams
E. statechart diagrams

44. On a sequence diagram, an object name of aStudent:List would indicate that

_____.

A. aStudent is an instance of the List class

B. List is an instance of the Student class
C. List is a method of the Student class
D. the Student and List objects are combined for that step in the sequence
diagram
E. a message is being passed from the Student class to the List class

SEE NEXT PAGE

CS/183/23/SS08 - Page 14 - SOLUTIONS

45. In a sequence diagram, an analyst draws a lifeline with an X at the end. This
lifeline represents _____.

A. a message that cannot be delivered

B. an object that is destroyed at a point in time
C. a message that is delivered at a point in time
D. an object that is complete at a point in time
E. an object that arrives at its final destination

46. The UML diagramming technique that shows the different state that a single
object passes through in reponse to events is a(n) _____ diagram.

A. class
B. sequence
C. behavioural state machine
D. use case
E. component

47. In a sequence diagram, an analyst draws a long box shape on top of a lifeline. This
box represents _____.

A. the time in which the object is exchanging messages

B. the time in which the object does not exist
C. a transition from one state to another
D. the encapsulation of the object
E. the time in which the object exists

48. The order of messages on a sequence diagram goes from ________ .

A. right to left
B. bottom to top
C. left to right
D. top to bottom
E. left to right, top to bottom

SEE NEXT PAGE

CS/183/23/SS08 - Page 15 - SOLUTIONS

49. The process of building new systems by combining packaged software, existing
legacy systems, and new software written to integrate everything together is called
______.

A. customization
B. formal methodology
C. outsourcing
D. systems integration
E. workaround

50. Which of the following is not a fundamental way to approach the creation of a
new information system?

A. develop a custom application in-house

B. rely on an external vendor to build the system
C. purchase a software package and customize it
D. rely on end-users to develop it themselves
E. all of the above are ways to create new information systems

51. An advantage of custom development is that the organization can _____.

A. accept functionality that is not a perfect fit

B. build technical skills and functional knowledge
C. install in a short period of time
D. remove all risk from the project
E. save money on the purchase

52. A(n) _____ file stores past transactions that may no longer be needed, is usually
stored off-line, and can be accessed on an as-needed basis.

A. audit
B. history
C. look-up
D. master
E. transaction
SEE NEXT PAGE
CS/183/23/SS08 - Page 16 - SOLUTIONS

53. The process of adding redundancy back into a physical data model is known as
_____.

A. balancing
B. clustering
C. denormalization
D. indexing
E. normalization

54. The most efficient tables in a relational database in terms of storage space have
_____.

A. no redundant data and few null values

B. no redundant data and plenty of null values
C. redundant data and few null values
D. redundant data and plenty of null values
E. repeat customer information

55. ______ is the general template used to define and create specific instances or
objects.

A. a class
B. inheritance
C. a message
D. polymorphism
E. a method

56. ______ is essentially a function or procedure call from one object to another, and
is the information sent to objects to trigger methods.

A. encapsulation
B. information hiding
C. inheritance
D. a message
E. polymorphism

SEE NEXT PAGE

CS/183/23/SS08 - Page 17 - SOLUTIONS

57. The most common threat to a computer system is posed by _________.

A. internal hackers.
B. industrial espionage.
C. device failure.
D. viruses.
E. natural disasters.

58. When all files are regularly tested for worms, bugs, and illicit programs; this is an
example of _____ requirements.

A. access control
B. encryption and authentication
C. portability
D. technical environment
E. virus control

59. A file that stores information on who, when, and how data was altered is a(n)
______ file.

A. audit
B. history
C. look-up
D. master
E. transaction

60. The normalization process is performed primarily to increase the database's

___________

A. storage efficiency.
B. indexing scheme.
C. access speed.
D. multiple dimensions.
E. none of the above.

SEE NEXT PAGE

CS/183/23/SS08 - Page 18 - SOLUTIONS

61. The ability of the same message to be interpreted differently by different classes
of objects is referred to as _____________

A. encapsulation.
B. polymorphism.
C. inheritance.
D. association.
E. None of the above.

62. When the analyst is evaluating a data model to ensure that all fields in a record
depend fully on the entire primary key, which step of normalization is being
performed?

A. Base normal form

B. First normal form
C. Second normal form
D. Third normal form
E. Cannot tell from the above information.

63. If the logical data model contains fields that depend on another non-primary key
field, then it is in violation of the rules of _____.

A. base normal form

B. first normal form
C. non-normal form
D. second normal form
E. third normal form

64. The process of ensuring that values linking tables together through the primary
and foreign keys are valid and correctly synchronized is _____.

A. hierarchical integrity
B. primary integrity
C. table integrity
D. referential unity
E. referential integrity
SEE NEXT PAGE
CS/183/23/SS08 - Page 19 - SOLUTIONS

65. In order to reduce the number of joins that must be performed in a query and to
increase the speed of data access, the data analyst can _____ the physical model.

A. cluster
B. denormalize
C. index
D. normalize
E. optimize

66. The first step in the Design Phase is to _____.

A. convert logical process and data models to physical

B. create the deliverable
C. design the architecture
D. design the inputs and outputs
E. present design alternatives

67. The plan for the creation of the hardware and software infrastructure is commonly
called _____.

A. architecture design
B. hardware specifications
C. software specifications
D. network model
E. presentation logic

68. The number of errors found is likely to be highest in the _____ stage of testing.

A. unit test
B. integration test
C. system test
D. acceptance test (alpha)
E. acceptance test (beta)

SEE NEXT PAGE

CS/183/23/SS08 - Page 20 - SOLUTIONS

69. _____ is the development of all parts of the new system including the software,
documentation, and initiation of new operating procedures.

A. Construction
B. Design
C. Documentation
D. Management
E. Testing

70. Training for a business system should focus on _____.

A. all the capabilities of the new system

B. complex computerized modules and code
C. helping the users to accomplish their jobs
D. how to use the system
E. not using the system

[Total for Section A: 70 marks]

SEE NEXT PAGE FOR SECTION B

CS/183/23/SS08 - Page 21 - SOLUTIONS

SECTION B - Answer ONE question from this section.

1. This question is about Unified Modeling Language (UML) Functional Models

(a)
i) An overview use case diagram is a use case diagram that does not go into great
detail, but instead attempts to show only the high level overview of the
requirements. Typically an overview use case diagram is created early in the
process of understanding the system requirements.
[2 marks]

ii) The Doctor, Patient and Management are called ‘actors’ in this kind of diagram.
This term is particularly appropriate because they refer to ‘roles’ and not
necessarily particular people. In a movie, a movie actor plays a part, or role. For
example Patient refers to the role of being a patient, and not any particular one
patient. Indeed, here, there are many such possible patients, shown on the diagram
as one Patient actor (i.e. role).
[4 marks]
(b)
CS/183/23/SS08 - Page 22 - SOLUTIONS

Use cases are shown in this diagram as ovals. The ‘New Patient’ and ‘Old Patient’ are
specialised actors, which are generalised (shown by an arrow) to the ‘Patient’ actor. The
‘Update Patient Information’ use case extends the ‘Make Old Patient Appointment’ use
case because sometimes the patient information is updated (the New Patient doesn’t have
this because a record cannot be updated before it is created). Similarly, the ‘Make Payment
Arrangements’ use case is an extension of the generalised (for both old and new patients)
Make Appointment use case, because sometimes new payment arrangements are made.
The Make ‘Old’ and ‘New’ Patient Appointment use cases are generalised to the ‘Make
Appointment’ use case. A new patient record is always created by the ‘Create New
Patient’ use case, when an appointment is made for a New patient, using the ‘Make New
Patient Appointment’ use case, so <<includes>> is used here. Similarly, when the
Management ‘Produce Schedule Information’, and a Doctor ‘Records Availability’, the
Manage Schedule use case is always invoked, so <<includes>> is also used here. (Note:
many use case diagrams do not use the * multiplicity syntax, so no marks will be lost if
they are not included here.)
Use Cases 1 mark each+1 mark for each actor+1 mark for each generalisation+ 1 mark for
each extend or include =[21 marks]
(c)
CS/183/23/SS08 - Page 23 - SOLUTIONS

Specialised actors for GP(General Practitioner) and Specialist can be added as shown,
generalised to Doctor, using the generalisation arrow – see diagram above [3 marks].

2. This question is about Requirements Determination (Requirements Gathering).

(a)
scribe (or scribes): assists the facilitator by recording notes, making copies
(sometimes using CASE tools for this). Basically a scribe acts as a secretary and does
not actually take part in the discussion.
facilitator: a person who sets the meeting agenda and guides the discussion, but does
not join in the discussion as a participant. He/she does not provide ideas or opinions
on the topics under discussion, remaining neutral throughout the session. However,
the facilitator does write group points and conclusions on a board/flipchart for all to
see. The facilitator must be an expert in both group process techniques and IS analysis
and design techniques.
agenda merry-go-round: this is when a group member keeps returning to the same
issue every few minutes and won't let go. One solution is to let the person have 5
minutes to ramble on, while every point is carefully written on a flip-chart. Then at
future occasions, the flip chart can be referred to whenever the person raises the point
again.
side discussions: this is a problem in a JAD session, when some participants persist in
private discussions with the person sitting next to them rather than taking part openly
in the group session. This can often be resolved by walking over to where they are
sitting.
violent agreement: is when participants really agree on the issues, but don't realise
they are agreeing because they are using different terms. For example, one might
argue that a glass is half full, whereas the other might be arguing that it is half empty.
To solve this, the facilitator has to translate the terms into different words to find
common ground so that the parties realise that they really agree.
[2 marks for each = 10 marks]
(b)
Interviews:
(1) Selecting interviewees: make schedule or list of all people to be interviewed.
(2) Designing interview questions: designing closed-ended, open-ended, or probing
questions.
(3) Preparing for the interview: create interview plan, anticipating possible answers
and providing follow-up questions.
(4) Conducting the interview: build up rapport and trust, explain why the questions
are being asked, present an unbiased view, take careful notes, and make sure every
answer is understood.
(5) Post-interview follow-up: prepare interview report, often also sending this to the
interviewees for approval/clarification (with thanks).
[one mark for each listed plus one for each explanation = 10 marks]

(c) List the other THREE main information-gathering techniques used in the analysis
phase.
(1) Document analysis
(2) Questionnaires
(3) Observation
CS/183/23/SS08 - Page 24 - SOLUTIONS

[one mark for each = 3 marks]

(d) JAD Sessions and Interviews provide the greatest amount of detail (greatest depth).
They both also help the analyst to understand the reasons behind the information
provided. They are useful for obtaining not only facts and figures, but also an
understanding of why those facts and opinions exist. (In contrast, document analysis
and observation really only are useful for obtaining facts. Questionnaires can provide
medium depth, regarding both facts and opinions, but obtain little understanding of
why.)
[4 marks]

(e) Interviews and JAD Sessions are the most costly; both having what might be called
‘moderate’ costs. JAD Sessions are much more expensive initially, because many
users and highly-paid managers and consultants need to be away from their offices for
a long period of time, BUT since JAD Sessions can significantly reduce the time
spent in design and information integration, that cost is offset in the long term.
Interviews have moderate costs because although personnel are removed from their
offices, it is generally one at a time, and therefore has less impact on productivity.

[3 marks]

EXAMINER: Dr J. Y. Clark