7. What if respondents have problems with the survey?
Problems almost always
come up even with well-designed surveys. For targeted surveys you’ll need a way
for respondents to get help for problems such as ID numbers that don’t work, or
computers that crash in the middle of a survey, or other obstacles. It’s a good
idea to provide a phone number for respondents so they can report problems or
complete the survey by phone if necessary. This will increase your response
rate. Sometimes, respondents have concerns they would like to express that were
not in the survey itself, which could be very valuable information.
8. Can we view real-time results from the survey?
The ability to view data
very soon after a survey is “live” is an incredible asset. Viewing responses as
they come in can help validate or troubleshoot a survey. You can determine
quickly if the questions you are asking are “working,” and whether the
respondents understand the questions and are providing the answers that you
need.
If 15 out of the first
20 respondents unexpectedly answer “Don’t Know” to a question, it probably needs
to be reworded or removed. For open-ended questions, respondents might not
always answer in the way you expect them to. Viewing those answers early in the
fielding process allows you to flag potential problems and correct them. Many
survey software packages offer viewable real-time results, but they vary in
their detail and quality, and in the ability to make changes when needed.
9.
How will we direct respondents to our survey?
For most Web surveys,
respondents can be invited to the survey via e-mail. Compared to other
invitation methods, e-mail is fast, easy, and cheap. However, with the
proliferation of spam and other unwanted e-mail, survey response rates from
e-mail invitations can be very low if they are not constructed properly. It’s
extremely important to have an e-mail subject line that will be accurate and
enticing. A direct subject line such as
“Take [company name]’s survey and receive a
$25 honorarium”
will have less risk of
getting sent to the spam file.
While the subject line
is important in order to get recipients to read the e-mail, the body of the
e-mail is even more important. A message with too many graphics may be sent
directly to a quarantine folder or may be rejected outright. Simple and
easy-to-read text that tells the respondent how long the survey will take, and
why they should take it, has the best chance of success.
10.
Are we taking advantage of all the valuable options we have with a Web survey?
One of the biggest
advantages of an Internet survey is flexibility. Queries can be added to an
existing instrument and then tested in the field much more quickly than with
other modes. Also, the order of questions and the response choices can be
easily randomized, which will improve the reliability of the results.
Another advantage of Web
surveys is the ability to include audio and visual images among the actual
survey questions. This can be very helpful in focusing the research on exactly
what you want. Another feature that we like to include in surveys is a
progress bar, a graphical display that shows the respondent how much of the
survey is remaining. Skip patterns, branches, and dynamic content are other
features that improve both the quality of information obtained and improve
response rates.
We hope the responses to these questions have helped shed light on the specific
pros and cons of using an Internet survey. In general, online surveys can help
you get more accurate information more quickly from more of your key customers.
But getting to that point will require either investment in internal resources
or a relationship with the right vendor. If you have additional questions about
this approach (or other survey methodology), please feel free to contact The HSM
Group and we’ll be happy to share our thoughts and experience.
Read #1-3 in this top 10
list
Read #4-6 in this top 10 list