View
219
Download
1
Tags:
Embed Size (px)
Citation preview
Lecture Objectives
to be aware of various methods for data gathering (“fact-finding”) in respect of information system development
to understand the usefulness and suitability of various data gathering methods for particular problem situations
Data gathering in systems development
Data gathering is a major task of systems analysis.
Systems analysis involves:• Understanding and describing how the current system
functions• Determining what users would like their new system to
do (user requirements)
Necessary for system analysts to collect information about current and future situations, problems,opportunities, constraints
What data to gather?
The business or organisation:• Data about the nature of the business and its
market and business environment
• Data about business goals and objectives that dictate what and how work is done
• Data about organisational structure: major functions, departments etc
• Data about major business subsystems and how they interact
• Data about business policies and guidelines
Users of the system:• Roles and responsibilities• Reporting structures• Job specifications and actual tasks performed• Information needed to do their jobs• Formal and informal communication and
workflow channels
What data to gather?
Users of the system:• Data about roles and responsibilities• Data about reporting structures• Job specifications and data about actual
tasks performed• Data about information needed to do their
jobs• Data about formal and informal
communication and workflow channels
What data to gather?
The existing system:• Data about tasks and workflow: functions,
processes, sequence of processes, methods and procedures, inputs, outputs
• Data about the data (definition, volumes, size etc.)• Data about interactions with other systems• Data about work volumes and processing cycles• Data about performance standards and criteria• Data about control mechanisms: e.g security,
accuracy• Data about problems: e.g. efficiency, information
What data to gather?
The new system:• Data about system requirement: a need or
desire to be met by a proposed system• Data about both functional requirements
(processes and functionality) and
non-functional requirements (security, performance, service etc.)
• Data about constraints e.g. existing technology• Data about interactions with other systems• Data about relationship to existing system/s
What data to gather?
Sources of data
Users and other stakeholders Documents about the system Documents about the organisation Documents and data used within the
existing system Transactions within existing system External sources
Users• System sponsor/owner: overall project objectives• Managers: high level, broad view of existing
system and requirements• End-users: detailed, operational level view of
existing system and requirements• Technical staff: technology capaabilities, limitations
etc. External stakeholders: e.g. customers
Sources of data
Documents about the system and organisation:• Organisation charts• Policy manuals• Business reports: financial, annual etc.• Jobs, procedure, operations manuals• Training manuals• Existing system documentation• Internal reports relating to the system
Sources of data
Documents and data used within the existing system:• Files, databases, programs, forms, reports• Informal: Memos, bulletin boards, files
External sources:• Other organisations’ systems• Hardware & software vendors• Business & industry publications
Sources of data
Interviews
Questionnaires
Observation
Sampling documents and transactions
Research and site visits
What data gathering methods?
Interviews
Generally the most important and widely-used method for data gathering
May be formal/structured (specific questions) or informal/unstructured (general goal or purpose)
Need an interview strategy for the entire interviewing process
Need an interview plan or guide for each interview
The interview strategy
Establish general objectives and guidelines for the entire interviewing process:
e.g. information to be obtained, sources, formats, documenting, analysis
Identify the users to interview:• Ensure all key people are included
Determine the sequence of interviews: Co-ordinate the interviewing process:
• Compare results, select follow ups etc. Need individual interview plans Need to consider:
Who has the information you need?
Where to conduct the interview?
When is the best time to interview?
How should the interview progress?
The interview strategy
The individual interview Before the interview:
• Arrange time and place, necessary materials, inform interviewee of interview purpose
Conduct the interview After the interview:
• Write an interview report• Review this with the interviewee at a follow
up interview
The interview structure Preliminaries:
• Introduction, purpose, environment and procedures e.g. permission to tape
“Body”:• Define what you already believe to be true and
confirm this, explore points & issues further, new areas (questions)
Conclusion:• Summarise and confirm your findings• Schedule a follow up interview
The interview plan Interview plans:
• Decide on interview structure• Determine content of questions• Decide on question types
See Whitten et al (2001),Figure 6.5 pp 233 for a sample interview guide.
Interviews: types of questions
Closed: how many transactions per day?
Limits available responses Open: tell me about …..
Leaves options open for interviewee Probe: tell me more about the problem
with the ….
To clarify and expand Mirror: From what you said, I understand
that….
To confirm what was said etc.
Interviews: types of questions
Avoid long, complex, or double-barrelled questions:
what decisions are made during this process and how do you make them?
Avoid leading questions;
you don’t need the customer number on this report, do you?
Avoid loaded questions:
when did you first discover the mistake?
i.e. how long have you known and done nothing?
Interviews: advantages obtain extensive, complex detailed information
get insights and opinions
discover informal procedures
flexible e.g. explore issues further or new issues
establish rapport with interviewee and understand their attitudes
reveal the ‘politics’ of the system environment
information is revealed both by the spoken word and by the interviewee’s body language
guaranteed response
Interviews: Disadvantages
Time-consuming
Costly
Danger of bias
More difficult to tabulate and analyse results e.g. to obtain an overall picture
Success in interviewing depends on the inter-personal skills of the interviewer
Questionnaires
A structured method of data gathering in which written questions/comments are provided for the participants to respond to in written form
The questionnaire can take many forms - write comments/ select from a list of possible responses/ mark on a scale
May permit either quantitative or qualitative data (mark out of 10/grade from good to bad)
Usually involves no direct contact between data gatherer and respondents
Questionnaires Useful when small amounts of data are required
from a large number of people For geographically dispersed respondents Types of questions:
• Open-ended (free format)• Fill-in-the-blank• Multiple choice• Rating• Ranking
Designing questionnaires What facts and opinions to be collected Who to sample and sample size Types of questions and wording (precise,
accurate, unambiguous) How to administer e.g. paper, online, mail out etc. Format and layout (grouping, crosschecks etc.) Test on small sample of respondents How completed questionnaires will be returned
and collated How analysis of the data will be carried out
Questionnaires Useful for:
• Obtaining simple opinions, facts• Quantifying what was found in interviews• Identifying issues before interviewing• Determining extent of problems
Not useful for detailed or complex information or exploring issues in depth
Can supplement other methods
Questionnaires: advantages
most economical method for gathering data from large numbers of people
quick and easy to administer results can be tabulated rapidly and analysed
readily allow respondents to be anonymous gives respondents time to reflect on answers respondents complete in their own time
Questionnaires: disadvantages
difficult to construct effective questionnaires specific and limited amounts of information possible low return rates possible bias and misinterpretation cannot probe issues further (inflexible) cannot clarify vague or incomplete answers lack non-verbal communication
Observation
observing the actual processes of a system need to prepare beforehand, and report on
data collected gain first hand knowledge of current system’s
operations clarify other information collected understand complex procedures inexpensive behaviour distortions may affect reliability unrepresentative samples affect reliability
Sampling of documents and transactions
Sampling: collecting a representative sample of documents, forms, transactions
Useful for specific information e.g. transaction volumes and types, file sizes
Useful where large volumes exist Information about existing system operations Representative samples must be selected:
determine sample size, appropriate range, avoid bias
Research and site visits
Most problems not unique: learn from experiences of other organisations
Professional societies can provide contacts for site visits
Computer trade journals and magazines and the internet can be sources for research into the problem/s e.g. do appropriate software packages exist?
Other data gathering methods
Other “modern” methods used:
Discovery prototyping JAD (Joint Application Development)
sessions Focus groups
A data gathering strategy Data gathering must be carefully planned
in order to make the most of the time and resources available:• Information sources• Data gathering methods• Recording and documentation methods• Data analysis methods• Procedures for reviewing results with
management and users
A data gathering strategy E.g. a “top down” approach:
• Initial interviews with management to determine major system activities and data
• Document and verify this• Expand major system component descriptions
into detailed descriptions:
Interview operational users, sampling, questionnaires, observation etc
• Document and verify this• Repeat these last two steps as necessary• Review findings with management
A data gathering strategy
Consider costs: allow for time and resources required for initial and ongoing information gathering
Use the least expensive methods first Plan how to check the validity of data:
• Cross checking between groups, methods• Evaluate data for inconsistencies• Ask further questions
Plan documentation of data e.g. records of interviews etc. data dictionary, system models
Data gathering in practice
Completeness? Accuracy? Objectivity? Biases? Stability? Representative? Finished?
References
WHITTEN, J.L., BENTLEY, L.D. and DITTMAN, K.C. (2001) 5th ed., Systems Analysis and Design Methods, Irwin/McGraw-HilI, New York, NY. Chapter 6
HOFFER, J.A., GEORGE, J.F. and VALACICH (1999) 2nd ed., Modern Systems Analysis and Design, Benjamin/Cummings, Massachusetts. Chapter 7
DWYER, J. (1997) The Business Communication Handbook (4th
edition) Prentice-Hall, New York, N.Y. Chapter 5