StatPac has over 30 years
experience providing survey design and data analysis services to
thousands of businesses and institutions in the
private and public sectors. We offer the most
comprehensive tables and the fastest turnaround
in the industry. Our prices are extremely competitive.
Most of our work involves Internet surveys. A typical job would be: A
client provides us with a MS Word document of the survey questions. From
that we estimate the cost. When approved, we create a Web survey and host it
on one of our servers. Some clients ask us to email the survey invitations,
while others prefer to do it themselves. At the completion of the survey we
prepare topline and banner tables and supply the reports as MS Word
documents.
COST
The cost to prepare a Web survey depends mostly on the number of
questions in the survey, as well as the complexity of the survey. Things
like branching, piping, and question rotations take more time and increase
the cost.
In order to give an accurate estimate, we have to see the actual survey.
More often than not, we hear "there are only 25 questions", but when we
examine the actual survey, there are really parts a, b, c, etc. for many of
the questions. In terms of preparation time, these are separate items, even
though the numbering on the survey refers to them as the same question.
To a large degree, cost is affected by the number of changes required
after the Web survey has been prepared. Some things are very easy and
inexpensive (e.g., minor wording changes to questions or response choices).
Others, such as adding new questions or changing the order of questions, can
be time consuming and costly. The most efficient way to control cost is to
make sure that the survey has been given final approval by the appropriate
people before asking us to prepare the Web survey.
A typical Web survey with less than a 100 items will cost between $1,000
and $2,000 for us to turn it into a Web format. Unlike other survey methods,
the number of respondents does not affect the cost. Once a Web survey has
been prepared, the cost to collect data is the same whether there are 10 or
10,000 respondents.
SENDING EMAIL INVITATIONS
Often, we are asked to email invitations to participate in the survey. We
can "spoof" the return address so it appears to have come from you or your
client. If we send the invitations, we can also track who responded and who
didn't. This allows us to send follow-up invitations to only people who did
not respond to the first invitation. Asking us to email the invitations adds
a couple of hundred dollars to the cost. The actual list of email addresses
must be provided by you and you must have a prior relationship with the
people on the list (or they must have agreed to receive a mailing from you).
In other words, we will not participate in unsolicited email (spam). We do
not provide any email lists.
ONLINE HOSTING
We will host your survey on one of our four secure servers. About 90% of
all responses to an email invitation happen within three days. Follow-up
invitations to non-responders are typically sent a week after the original
invitation, and most online surveys are done in two weeks. Depending on your
situation, you may want us to keep the survey online for a longer time. If
so, it is important for you tell us your intentions before we begin so that
we do not schedule other clients on the same hosting server.
SECURITY
Web surveys are normally "anonymous". This means that we do not know who
completed a particular survey. There are three ways to make a survey
non-anonymous. The first is to send email invitations that contain a
tracking or ID number. When the respondent clicks on the link in the email,
their ID number will be recorded with their responses to the survey. The
second is to ask the respondent for identifying information as part of the
survey itself. The third is to provide potential respondents with a unique
password they can use to access the survey. All three methods allow us to
identify how a particular respondent answered the survey. If none of these
methods are used, the survey will be anonymous. A "confidential" survey is
one where you know who responded but you promise not to tell anyone. If you
use one of the above methods, your survey can still be considered
confidential.
Web surveys can also be secure by using SSL technology. This means that
the data is encrypted when it travels from the respondent's browser to the
hosting server. The only time you might need SSL is when the survey itself
contains questions that identify the respondent. If the survey questions do
not contain any identifying information then SSL is usually not necessary.
We can provide SSL on any of our servers.
When we host a survey on one of our servers, then the data is stored in a
folder that cannot be accessed by anyone except us. We keep the data on our
server until the survey is closed. After we have delivered the data and
final reports to you, the data is removed from our server and we keep no
copies of it. Upon your request, we can keep the data for up to one year.
FINAL REPORTS
While the survey is live, we will be preparing the commands to create the
topline and banner tables reports. So in most cases, we can provide final
reports the day after the survey closes.
HOW TO GET AN ESTIMATE OF THE COST
It's easy to get an estimate of the project costs. For a rough estimate,
all we need is a Microsoft Word document of the survey. If you want a firm
bid, then we will need additional information (e.g., your specific reporting
requirements)..
Let us
bid on your next Internet survey and data analysis project.
You'll be surprised at how little it costs!
