Page 166 -

ChaPter 5 • InformatIon GatherInG: unobtrusIve methods 133

CONSULTING OPPORTUNITY 5.1

Trapping a Sample

“Real or fake? Fake or real? Who would have thought it, even Sylva glances at him furtively and begins to interrupt,
five years ago?” howls Sam Pelt, a furrier who owns stores in New “But that will take months, and purchasing may come apart at
York; Washington, D.C.; Beverly Hills; and Copenhagen. Sylva the seams unless they know soon what—.”
Foxx, a systems analyst with her own consulting firm, is talking Pelt interrupts, “I don’t care how long it takes, if we get
with Sam for the first time. Currently, P & P, Ltd. (which stands for the right answers. But they have to be right. Not knowing how
Pelt and Pelt’s son) is using a networked computer that supports to solve this dilemma about fake furs is making me feel like a
package software for a select customer mailing list, accounts pay- leopard without its spots.”
able and accounts receivable, and payroll. Sylva talks to Sam Pelt a bit longer and then ends the
Sam is interested in making some strategic decisions that will interview by saying, “I’ll talk it all over with the other analysts
ultimately affect the purchasing of goods for his four fur stores. at the office and let you know what we come up with. I think
He feels that although the computer might help, other approaches we can outfox the other furriers if we use software to help us
should also be considered. sample opinions, rather than trapping unsuspecting customers
Sam continues, “I think we should talk to all the customers when into giving an opinion. But I’ll let you know what they say.
they come in the door. Get their opinions. You know, some of them This much is for sure: If we can sample and not talk to every-
are getting very upset about wearing fur from endangered species. body before making a decision, every coat you sell will have
They’re very environmentally minded. They prefer fake to real, if a silver lining.”
they can save a baby animal. Some even like fakes better, calling them As one of the systems analysts who is part of Sylva Foxx’s
’fun furs.’ And I can charge almost the same for a good look-alike. firm, suggest some ways that Sam Pelt can use software to
“It’s a very fuzzy proposition, though. If I get too far away adequately sample the opinions of his customers, store man-
from my suppliers of pelts, I may not get what I want when I need agers, buyers, and any others you feel will be instrumental in
it. They see the fake fur people as worms, worse than moths! If I making the strategic decision regarding the stocking of fake
deal with them, the real fur men might not talk to me. They can furs in what has always been a real fur store. Suggest a type
be animals. On the other hand, I feel strange showing fakes in my of sample for each group and justify it. The constraints you
stores. All these years, we’ve prided ourselves on having only the are subject to include the need to act quickly so as to remain
genuine article.” competitive, the need to retain a low profile so that competing
Sam continues, in a nearly seamless monologue, “I want to furriers are unaware of your fact gathering, and the need to
talk to each and every employee, too.” keep costs of data gathering to a reasonable level.

4. Make a subjective decision that the acceptable interval estimate will be ±0.02.
5. Choose a confidence level of 95 percent. Look up the confidence coefficient (z value)
in Figure 5.2. The z value equals 1.96.
6. Calculate s as follows:
p
i 0.02
s 5 5 5 0.0102
p
z 1.96
7. Determine the necessary sample size, n, as follows:

p11 2 p2 0.0510.952
n 5 1 1 5 1 1 5 458
s 2 p 10.01022 10.01022

The conclusion, then, is to set the sample size at 458. Obviously, a greater confidence level
or a smaller acceptable interval estimate would require a larger sample size. If we keep the
acceptable interval estimate the same but increase the confidence level to 99 percent (with a z
value of 2.58), the necessary sample size becomes 1,827, a figure much larger than the 458 we
originally decided to sample.
DETERMINING SAMPLE SIZE WHEN INTERVIEWING. There are no magic formulas to help a
systems analyst set the sample size for interviewing. The overriding variable that determines how
many people a systems analyst should interview in depth is the time an interview takes. A true
in-depth interview and a follow-up interview are very time consuming for both the interviewer
and the participant.