Looking to the future
So, what does the future hold for Internetbased
survey sampling? At this point in
the Internet’s development, with its rapid
expansion and continued evolution, it’s truly
impossible to say. Yet we can hazard a few
guesses.
First, if the Internet continues to expand
but largely maintains its current structure,
then advances in sampling methods that willallow random sampling of, and inference
to, general populations will be at best slow
and difficult to develop. This follows from
the fact that Internet-wide sampling frames
are simply unavailable under the current
Internet structure/organization, and no general
frameless sampling strategies yet exist. Unless
the way the Internet is organized and operated
changes, it seems this will continue to be the
case into the foreseeable future.
That said, survey sampling methodologists
should endeavour to develop new sampling
paradigms for Internet-based surveys. The
fundamental requirement for a probabilitybased
sampling scheme is that every member
of the target population has a known, non-zero
probability of being sampled. While in traditional
surveys this can be achieved via various
frame and frameless sampling strategies, it
does not necessarily follow that Internet
surveys must use those same sampling
strategies. Rather, new sampling methods that
take advantage of the unique characteristics of
the Internet, such as the near-zero marginal
cost for contacting potential respondents,
should be explored and developed.
In addition, researchers considering conducting
an Internet-based survey should consider
whether the capabilities of the web can
be leveraged to collect the desired data in some
other innovative fashion. For example, Lockett
and Blackman (2004) present a case study
of Xenon Laboratories, an Internet-based
financial services firm that employed a novel
approach to market research. Xenon Laboratories
wanted to collect data on foreignexchange
charges by credit-card companies
on business travellers. They recognized
that neither the travellers nor the creditcard
companies were likely to respond to
a survey on this topic, whether fielded
over the web or otherwise. Instead Xenon
Laboratories developed the Travel Expenses
Calculator (www.xe.com/tec) and the Credit
Card Charges Calculator (www.xe.com/ccc)
and posted them on the web for anyone to
use for free. These tools help foreign business
travellers to accurately calculate the cost of a
business expense receipt in terms of their own
currency.
Lockett and Blackman (2004) say, ‘On
the basis of this information [input by those
using the calculators] it is possible to conduct
basic market research by aggregating the
inputted calculations. Xenon is now in the
unique position to analyse whether or not
the different card providers employ the same
charging levels and whether or not these
companies’ charge structures vary according
to geographical region.’ They go on to
conclude, ‘This value-added approach, which
is mutually beneficial to both parties, is
an important and novel approach to market
research.’
Second, it is also possible that technological
innovation will facilitate other means of
sampling for Internet-based surveys (and for
conducting the surveys themselves, for that
matter). For example, current trends seem to
point towards a merging of the Internet with
traditional technologies such as television and
telephone. It is quite possible that sometime in
the future all of these services will merge into
one common household device through which
a consumer could simultaneously watch
television, surf the web, send e-mail, and
place telephone calls. Depending on how this
evolves, various types of random sampling
methodologies, as well as new survey modes,
may become possible. For example, it may
become feasible, and perhaps even desirable,
to sample respondents via RDD (or something
very similar), have an interviewer call the
respondent, and then in real time present
the respondent with an interviewer-assisted,
web-based survey.
It is also possible that, as the technology
matures, these combined television–Internet–
telephone appliances will become as
ubiquitous as televisions are today, with
service offered only by a few large companies.
If so, then it may ultimately be possible to use
the companies’subscriber listings as sampling
frames (much as telephone directories were
used pre-RDD in the mid-1900s). Or it may
be that some other state emerges that lends
itself to some form of sampling that is not
possible today. The point is that the Internet
is still very much in its infancy, and the
current difficulties surrounding samplingfor Internet-based survey described in this
chapter may or may not continue into the
future.
To put this in a historical context, note that
while the telephone was invented in the late
1800s, and telephone systems developed and
expanded rapidly through the early 1900s, it
was not until the mid-1900s that telephone
coverage was sufficiently large and standards
for telephone numbers adopted that made
RDD possible. In fact, the foundational ideas
for an efficient RDD sampling methodology
were not proposed until the early 1960s
(Cooper, 1964), after which it took roughly
another decade of discussion and development
before RDD as we know it today became
commonplace.4 Thus, in total, it was roughly
a century after the invention of the telephone
before RDD became an accepted sampling
methodology.
In comparison, the web has been in
existence, in a commercial sense, only for
little more than a decade or two. As with the
telephone in the late 1800s and early 1900s,
we are in a period of technological innovation
and expansion with the Internet. However,
unlike the telephone, given today’s pace of
innovation, the Internet and how we use it
is likely to be quite different even just a few
years from now. How this affects sampling for
Internet-based surveys remains to be seen.