Survey research design
Pages | 419-421 |
DOI | https://doi.org/10.1108/EUM0000000006543 |
Date | 01 December 2001 |
Published date | 01 December 2001 |
Author | Joseph Janes |
Survey research design
Joseph Janes
Introduction
In a previous column, we looked at the design
of a survey or questionnaire form, paying
special attention to things like the wording of
questions, but focusing primarily on the form
itself. Here I want to expand a bit and look at a
somewhat bigger picture ± the design of the
overall survey as a research enterprise. Here we
are concerned with how the survey will be
administered, as well as potential problems with
each administration method and the survey
method overall.
A survey or questionnaire will be at its best
when getting a snapshot of the current state of
affairs in a given group or population, what
researchers call descriptive work. Among your
biggest concerns when doing a survey is the
representativeness of your sample ± in other
words, that the group you are asking looks
enough like the entire group you are interested
in so that the answers you get can be
confidently ascribed to the larger group. In the
column on sampling, we talked about ways in
which drawing a good sample helps to give you
that confidence, but there are methodological
issues discussed here that are important as well.
First of all, as is the case with any research
enterprise, you have to know why you are doing
the study. If you are going to ask for people's
time and energy in filling out your form or
answering your questions, you ought to have a
pretty good reason. There is increasing
evidence that more and more people are
declining to answer telephone surveys, for
example, because of all the stupid telemarketing
questioners masquerading as survey
researchers, which is just serving to honk people
off. This raises real concerns about the
representativeness of the respondents, and thus
of the validity of the responses. So do not do a
survey unless you have to. In addition, every
question should be there for a reason. Surveys
and questionnaires should be as brief as they
can possibly be, and each one should be
important for the research you are undertaking.
Do not ask questions just because you are
curious, or because you can. If you cannot think
in advance of how you are going to use the
results, you probably do not need to ask the
question.
The author
Joseph Janes is Assistant Professor at The Information
School of the University of Washington and is Founding
Director of the Internet Public Library.
E-mail: jwj@u.washington.edu
Keywords
Statistics, Research surveys
Abstract
This column continues a series on topics in research
methodology, statistics and data analysis techniques for the
library and information sciences. It discusses issues related
to the design of survey research enterprises.
Electronic access
The research register for this journal is available at
http://www.mcbup.com/research_registers
The current issue and full text archive of this journal is
available at
http://www.emerald-library.com/ft
On research
419
Library Hi Tech
Volume 19 .Number 4 .2001 .pp. 419±421
#MCB University Press .ISSN 0737-8831
To continue reading
Request your trial