·
Two types of software: 1. Software packages to be installed and hosted on
the customer’s own server
2.
Application service providers (ASP) solutions – software and surveys are hosted
on the ASP company’s servers E.g., WebSurveyor.
·
Sampling: Respondents either
1. Intentionally navigate to the Web site which
contains a questionnaire.
In some cases, individuals anticipate a
survey at the web site and in other cases it is totally unexpected. (When the computer user does not expect a
survey on a web site and cooperation is voluntary, response rates are
low.) This information can be used to
build a database and profile the target market.
2.
Are alerted via e-mail to visit the Web site. Often they are members of consumer panels who
have previously indicated their willingness to cooperate. Or, they can be names of people who have
previously visited the Web site. Such
people already have an interest in the company/brand and so will be more
responsive. When individuals receive e-mail
invitations to participate, they are given passwords/PIN numbers and a link to
a certain web site that contains a welcome
screen—the first page in an Internet survey; it serves as a cover letter
(incenticizes motivation, plus requests the respondent enters a password or PIN
to move into the survey questions). (The
PIN restricts access to individuals who are not part of the scientifically
selected sample. Assigning an unique password code
also allows the researchers to track the responses of each respondent, thereby
identifying any respondent who makes an effort to answer the questionnaire more
than once.)
·
Questionnaire Design:
. Questionnaires should be personalized/customized to increase the response rate.
. Mix closed- and open-ended
questions to keep it interesting.
. Avoid open-ended questions
at the beginning since they require more work, especially if respondents can’t
skip them.
. Use progress bars to show how much of the survey remains (since this can’t be visually seen as with a paper questionnaire).
. Keep the survey to under 20 minutes in length.
. Use graphicswhere possible or appropriate o make the experience visually engaging.
·
Follow-up: Use e-mail to follow up the original request to participate (only use
one reminder—more than one reminder is considered pestering).
Ask for permission, online, to contact respondents offline for further information, if needed (richer info can be obtained in a phone interview).
Be prepared for some angry
responses since some people think it’s bad “netiquette"
to conduct MR on the Internet.
·
Advantages and Disadvantages;
-
Advantages:
1. Speed – You can send and
receive responses in 1 to 3 days (vs. four+ weeks for a mail survey). (75% to
80% of responses will return within 48 hours.) Inexpensive – Most servers allow
you to post a survey on the Web for free.
No costs of paper, postage, interviewers, or data editing, coding, and
entry. Marginal costs per response are
very low. (Overall, costs are about
one-half of phone or mall interviews).
2. Cost effective: no interviewer,
no postage, consumers enter the data for tabulation.
3. Large sample size – Since
marginal costs are almost zero. You can
conduct and analyze research with a asmple of over 100,000 reapondents. Also, tiny segments of the population can be
fully covered with samples in the hundreds and even thousands.
Reach difficult to reach
people - Because Internet surveys can be accessed anytime (24/7) from anywhere,
they can reach certain hard-to-reach respondents, such as doctors.
4. Easier – Just write the
questionnaire, post, let the answers come in, and analyze the data (no printing
questionnaires, mailing, interviewing, data entry, etc.).. Free e-mail survey services are available at www.surveypro.com, www.SurveyMonkey.com,
and at www.zoomerang.com; you must register to use them. However, such services do not include more
sophisticated research techniques like conjoint analysis.
5. Worldwide geographic reach.
.5. Personalized, flexible individual
questioning - Software
is available to allow questioning to branch off into two or more different
lines depending on each respondent’s answer to filtered questions. This can reduce time for respondents and
increase their involvement.
6. Secure confidential answers.
Interactive/flexibility of
questioning – Questions can be a function of respondents’ prior answers. The computer can sequence questions based on
previous responses.
7. Visual appeal - Color,
graphics, sound, and animation. These
help to increase participant cooperation and willingness to spend more time
answering the questionnaires.
8. Accurate real-time data
capture - Each respondent’s answers are directly entered into the researcher’s
computer as soon as the questionnaire is submitted. This eliminates interviewer bias and data
entry errors You can immediately view responses as collected in a
password-protected environment, allowing for real-time data analysis.
The questionnaire
software may be programmed to reject improper data entry.
9. Easy callbacks - Computer
software can automatically sending e-mail reminder notifications to those
solicited via e-mail who did not visit the welcome page. Because computer
software can identify the passwords of respondents who only completed a portion
of the questionnaire, customized messages can be sent to individuals who
terminated the questionnaire with only a few additional questions to answer.
10. Respondent anonymity -
Respondents are more likely to provide sensitive or embarrassing information
when they can remain anonymous.
11. Fewer errors – There are no interviewer errors
(e.g., accidentally skipping questions, inaccurate recording of answers,
etc.) Also, computers don’t vecome fatigued or cheat.
12. Fun for respondents. Therefore, they are more likely to add thoughtful
free-form comments to express their opinions than they would with paper
surveys.
13. More convenient for
respondents due to being able to complete at their leisure and easy to complete
with a few mouse clicks
14. Higher response rates – Due
to many of the above reasons
-
Disadvantages:
-
1.
Nonrepresentive samples are somewhat likely (although this is not the big
problem it was in the early days of the Internet when Internet penetration
rates were low) due to
-
(a) Lack of universal
Internet access (a bit over 2/3rds of Americans had Internet access at home,
work, school, or elsewhere in 2006. They
tend to be somewhat upscale, urban, and younger. This is an especially large issue in global
research and research among ethnic segments.
-
(b) Low response rates to
e-mail notification unless panel members (lower than phone or mail). (It’s okay to send one follow-up erminder, ,but beyond that it is
considered harassment.)
Self-selected
sample if the survey is a sample of those who visit a web page and
voluntarily fill out a questionnaire.
Such opt-in participants tend to be more involved with the subject of
the research. However, scientifically
drawn samples from a consumer panel or samples randomly generated in other ways
can yield representative samples. Even
here, there is concern that, as online research proliferates, there are more and more
“professional respondents”—people who sit at home and take a lot of surveys
2. Varying respondent capabilities. People have:
(a) different levels of technological expertise - they may not know how
to navigate through and provide answers to an Internet questionnaire.
(b) Computers
with different capabilities for receiving graphics, animation, sound, etc. Because of low speed Internet connections
(low bandwidth), many individuals cannot quickly download high-resolution
graphic files.
3. Security concerns - Hackers
or competitors may access web sites researching new product concepts, new
advertising campaigns, and other top-secret ideas. Many research suppliers specializing in
Internet surveying have developed password-protected systems that are very
secure. One important feature of these systems is to restrict individuals from
filling out a questionnaire over and over again.
4. Privacy concerns -
Respondents may worry if personal information will remain private. Secure locket layer (SSL) encryption
technology can be used to secure sensitive data
5. High setup costs – Programming,
debugging, etc. can be expensive