|
Data Collection Methods
We've certainly seen many arguments favoring one form of data collection
over another. For example, one often hears that the advantage of having personal
interviews is that the interviewer can clarify questions, which are unclear to the
interviewee.
Balderdash! That issue always reminds us of an occasion, years ago, when
one of our principals, Tom Schori, was intercepted by an interviewer in a Phoenix shopping
mall and asked to take part in a marketing research study. (As a side bar, Tom feels
guilty to this day that he did not acknowledge that he was employed in a marketing
research capacity.)
To Tom, it was very apparent what the marketing research study was all
about: evaluating alternative packaging for a Coors® brand. The interviewer was using a
tachistoscope to briefly (maybe, 50 milliseconds) present an image of a store shelf which
contained various brands of beer, including the Coors® brand. The interviewee, after each
brief presentation, was to indicate what brands had been presented. According to Tom, the
interviewer did not know how to properly operate the tachistoscope. Furthermore, he had a
great deal of trouble explaining the nature of the project and asking the questions. So
much for the "clarification" value of personal interviews!
The critical factor doesn't revolve around how the data are
collected (be it via personal interviews, or any other mode of data collection). The
critical factor is the survey instrument itself, i.e., the questionnaire. If the questions
are not asked or phrased in an understandable way and aren't presented to each respondent
in exactly the same fashion, the findings risk being ambiguous. A poorly designed survey
instrument will result in data of limited value, regardless of how the survey instrument
is administrated.
Data collection methods
- Personal interviews
- Telephone interviews
- Mail surveys
- Internet surveys
We always recommend using the data collection method which, we believe, will work best
given the nature of the data to be collected, and the time-constraints which may be in
place, and of course using a well-designed survey instrument. Regardless of what data
collection method we employ, we generally charge about 50%-60% of what our nationally
recognized competitors charge. And, while they only provide "cross tabs" for that
money, we always undertake a detailed data analysis, draw well thought-out conclusions,
provide readily implementable recommendations, and make an "in-person" final
presentations to enable our clients to probe, question, and challenge our findings,
conclusions and recommendations in whatever detail they desire.
Personal Interviews.
There are many times that we use personal interviews. They work very well when it is
necessary to display special effects, such as "story boards," television or
radio advertisements," or graphic displays. For example, personal interviews may
involve "mall intercepts" or interviews in homes, in offices, or in restaurants.
While we often use subcontractors or hire interviewers to conduct interviews, we ensure
that they are well-trained and well-supervised.
Telephone Interviews.
For very simple surveys or when very quick response time is necessary, telephone
interviews work great. We've long worked with a telephone subcontractor who can have 300
interviewers on the phone simultaneously in one location. If we're doing a
substantial telephone interviewing effort, we'll ordinarily expect them to complete all
the interviews within a 2 or 3 week time frame. However, for an urgent
"down-and-dirty" effort, they've invariably come to our aid because they truly
enjoy working with Millennium Marketing Research®. They are kind enough to accept a job on a Friday
afternoon and provide us with "top line" results on Monday. All at no extra
cost!
Mail Surveys.
When the survey instrument is complex, we find that mail surveys work quite nicely.
We've long said that the design of the survey instrument is a learned skill, and over
the years we've learned many valuable lessons about survey design. Tom Schori, one of our
principals, had developed a procedure which we term the Optimal Brand Positioning
Model©.
It requires the gathering of some pretty complex information. Mike Garee, our other
principal, developed a seven-page survey instrument, which was then mailed to those within
the client's target market. When the completed questionnaires started arriving, it became
very apparent that respondents were having a great deal of trouble completing the
questionnaire. Tom was perplexed until it dawned on him that every other time he'd
employed the Optimal Brand Positioning Model©, he had gathered the necessary
information with a one-page questionnaire. Mike then redesigned the questionnaire, which
we re-fielded at our own expense. Not only did respondents now fill out the questionnaires
completely, the response rate for the re-fielded questionnaire was twice that which we had
expected. Many people have the false impression that mail surveys take a long time to
field and generate useful data. The fact of the matter is that 90% of the responses will
have been received after the instrument has been in the field just a week!
Internet Surveys.
Anything that can be done with a mail survey of course can also be accomplished with an
Internet survey, and much more. And it is really fast. Bringing that point home, Tom
Schori, one of our principals, who has taught marketing management and marketing
research for many years, recently conducted an Internet survey project with his marketing
research students. Individuals within the target market were contacted via personalized
e-mail and were invited to participate in a research effort by following a hyperlink to
the site where the survey instrument was stored. When a respondent finished the survey,
he/she simply clicked on "submit" and his/her responses were stored in a secure
site for later data analysis. Within four days after sending out the e-mail
invitations to participate, about 1,200 surveys had been fully completed. This represented
about a 70% response rate among those with active e-mail accounts. Outstanding! You can
bet that more and more survey efforts will be conducted over the Internet over the coming
years.
|
