Page 92 - Design for Six Sigma for Service (Six SIGMA Operational Methods)

Customer Survey Design, Administration, and Analysis 73

Interviewers should not introduce any bias in the interview process and should
not express any opinions in response to the answers from the respondents.
Many companies that conduct telephone interviews use computer-assisted
telephone systems where the interviewer sits at a computer that dials the
telephone number and puts the questions to be asked on the screen so the
interviewer can read them to the respondents. The software that manages
this can take care of the data entry and coding, as the interviewer uses the com-
puter’s keyboard or mouse to indicate the respondent’s answers to the questions.

4.3.3 Administering an In-Person Survey
In-person interviews are the most expensive to conduct, in terms of both time
and money, and the most intrusive method. A major strength of the in-person
interview is the ability to deal with complex topics. Because you can see how
respondents react to the questions as you ask them, you will have a better
idea how well they understand the questions and what confuses them. You
will also have opportunities to resolve any glitches in the interview process.
The selection and training of in-person interviewers is even more important
than that of telephone interviewers. The selection methods and training of
interviewers are almost the same as that of telephone interviewers.

4.4 Survey Sampling Method and Sample Size

The main goal of a customer survey is to produce an accurate picture about
the population based on the information drawn from a scientifically selected
subset from that population. Sampling is necessary because it is impractical
to seek information from every member of the population. In this sampling
process, first we need to determine our population. Then we need to define
a sampling frame that is a list of elements in the population that may be
selected in the sample. Thirdly, we need to identify a sampling method,
that is, a method to select a subset from the sampling frame. Before we do
the sampling, we need to identify what is the adequate sample size for this
survey in order to ensure the credibility of the survey data analysis. In this
section, we are going to discuss these issues.

4.4.1 Population and Sampling Frame
The first consideration in survey sampling is the specification of the unit of
analysis. The unit of analysis is the individual, object, institution, or group
that bears relevance to the survey study. For example, for a fast-food chain