chapter 4 (information gathering: unobstructive methods) Flashcards
Unobtrusive Methods
- Less disruptive
- Text analytics to analyze qualitative data
- Insufficient when used alone
- Multiple methods approach
- Used in conjunction with interactive methods
Sampling
A process of systematically selecting representative
elements of a population
Involves two key decisions:
– What to examine
– Which people to consider
The reasons systems analysts do sampling are to
– Contain costs
– Speed up data gathering
– Improve effectiveness
– Data gathering bias can be reduced by sampling
Too costly to
– Examine every scrap of paper
– Talk with everyone
– Read every Web page from the organization
Sampling
- Sampling helps accelerate the process by gathering
selected data rather than all data for the entire population - The systems analyst is spared the burden of analyzing
data from the entire population
Sampling Effectiveness
- Sampling can help improve effectiveness if information
that is more accurate can be obtained - This is accomplished by talking to fewer employees but
asking them questions that are more detailed - If fewer people are interviewed, the systems analyst has
more time to follow up on missing or incomplete data
Sampling Bias
- Data gathering bias can be reduced by sampling
- When the systems analyst asks for an opinion about a
permanent feature of the installed information system, the
executive interviewed may provide a biased evaluation
because there is little possibility of changing it
To design a good sample, a systems analyst must follow
four steps:
– Determining the data to be collected or described
– Determining the population to be sampled
– Choosing the type of sample
– Deciding on the sample size
Four Main Types of Samples
- Convenience
- Purposive
- Simple random
- Complex random
Convenience Samples
- Convenience samples are unrestricted, nonprobability
samples - This sample is the easiest to arrange
- The most unreliable
Purposive Sample
- A purposive sample is based on judgment
- Choose a group of individuals who appear
knowledgeable and are interested in the new information
system - A nonprobability sample
- Only moderately reliable
Complex Random Samples
- The complex random samples that are most appropriate
for a systems analyst are
– Systematic sampling
– Stratified sampling
– Cluster sampling
The Sample Size Decision
- Determine the attribute
- Locate the database or reports in which the attribute can
be found - Examine the attribute
- Make the subjective decision regarding the acceptable
interval estimate - Choose the confidence level
- Calculate the standard error
- Determine the sample size
Investigation
- The act of discovery and analysis of data
- Hard data
– Quantitative
– Qualitative
Analyzing Quantitative Documents
- Reports used for decision making
- Performance reports
- Records
- Data capture forms
- Ecommerce and other transactions
Reports Used for Decision Making
- Sales reports
- Production reports
- Summary reports
Records
Records provide periodic updates of what is occurring in
the business
There are several ways to inspect a record:
– Checking for errors in amounts and totals
– Looking for opportunities for improving the recording
form design
– Observing the number and type of transactions
– Watching for instances in which the computer can
simplify the work (calculations and other data
manipulation)
Data Capture Forms
- Collect examples of all the forms in use
- Note the type of form
- Document the intended distribution pattern
- Compare the intended distribution pattern with who
actually receives the form
Analyzing Qualitative Documents
- Key or guiding metaphors
- Insiders vs. outsiders mentality
- What is considered good vs. evil
- Graphics, logos, and icons in common areas or web
pages - A sense of humor
- Email messages
- Memos
- Signs or posters on bulletin boards
- Corporate Web sites (note the interactivity of Web sites)
- Manuals
- Policy handbooks
Text Analytics
Software that can analyze unstructured qualitative data
from any source including:
– Transcripts of interviews
– Written reports
– Customers’ communication collected through email,
wikis, blogs, chat rooms, and other social networking
sites
Text Analytics
Unstructured, qualitative, or “soft” data are generated
through:
– Blogs
– Chat rooms
– Questionnaires using open-ended questions
– Online discussions conducted on the Web
– Exchanges occurring on social media
Text analytics can realize valuable insights into
– What customers are thinking about the organization,
the values and actions of the company
– Customer or vendor motivations for beginning,
maintaining, improving, or discontinuing a relationship