Comp. by: VPugazhenthi Stage : Revises1 ChapterID: 9781405132879_4_Q Date:31/
3/09 Time:19:14:09 Filepath:H:/00_Blackwell/00_3B2/Gregory-9781405132879/appln/
3B2/revises/9781405132879_4_Q.3d
notions ofsexualityandgender, their repre-
sentations or their effects (seeessentialism;
feminist geographies;performativity: see
also Jagose, 1996; Sullivan, 2003).
The dilemma with any dictionary definition
of this term is that it is meant to refuse fixing
or defining, or delimiting a research trajectory,
because such a move logically excludes facets
of sexualities, or strategies to rethink them.
This conundrum reflects the tendency of
dominant discourses to assimilate any moves
that resist or stand outside of them (seedis-
course). Thus there is a discernable anxiety
about whether the radical potential of this
term is being evacuated as its popularity
grows and it is stretched to cover any work
being done by or about gays and lesbians
(Sothern, 2004).
Queer theory’s relationship with geography
has been two-directional, if rather uneven.
Queer theorists demand that geographers rec-
ognize – and attend to – their own heteronor-
mativity, but it also introduces the problematic
ofhomonormativity, which is to say, the discip-
lining effects of privileging and imposing any
sexuality or desire to the exclusion or stigma-
tization of others (Sothern, 2004). It has en-
abled geographers interested in sexuality and
spaceto maintain a focus onplace, yet also
incorporate an appreciation and awareness of
movement, migration, placelessness and
multiplescales(e.g. Knopp, 2004). Geog-
raphers, historians and architects have tried
to point out that queer theory’s roots in the
humanities, often marked by a suspicion of
empiricism, and penchant for discourse analy-
sis of literarytexts, open it to charges of a
rather unqueergeographical imagination,
where the materiality of the world is some-
times lost in a sea of textuality. While
queer theorists have been willing to
appreciate the significance ofspatialityand
nature–society dualities (Halberstram,
2005), they have been rather slow to acknow-
ledge the work of geographers who themselves
are informed by queer theory. mbSuggested reading
Jagose (1996); Sullivan (2003).questionnaire An instrument for collecting
data forsurvey analysis. Every respondent is
asked the same questions, in the same way and
in the same sequence. This differs from the
more open-ended, conversational format of
interviews and otherqualitative methods.
Questionnaires can be used to collect informa-
tion to classify respondents (e.g. gender,income or age), to learn about their behaviour
(e.g. voting practices, number of hours of
television viewing per week or recycling behav-
iour) and to understand their attitudes.
Questionnaires may be administered face-
to-face, over the phone or via theinternet,
or they can be mailed and self-administered.
There are advantages and disadvantages to
each mode of administering questionnaires,
including cost, response rate, accuracy of re-
sponses and the level of detail that can be
obtained. One mode of delivery is not superior
to another: the advantages and disadvantages
are weighed in relation to the research question.
Because one of the strengths of survey re-
search is the capacity to establish patterns over
large populations, questionnaires must be very
carefully designed so as to be unambiguous to
a large number of people. They must be un-
biased and simply understood, with clear in-
structions of how to respond. There are
standard pitfalls to avoid, such as the inclusion
of double negatives and double-barrelled
questions. Nevertheless, there is no avoiding
the fact that questions are inevitably framed
from a certain perspective and, in the case of
face-to-face or telephone interviews, the pres-
ence of an interviewer will have effects that can
only be understood rather than removed (see
interviews and interviewing). A variety of
types of questions can be asked, including
open-ended ones. Closed-ended questions
standardize responses across a given set of
responses and are already structured for stat-
istical analysis. Open-ended questions require
coding before statistical analysis, but are suit-
able if the full range of possible responses is
unknown, or the topic is a sensitive or complex
one that requires some explanation or qualifi-
cation. Because questionnaire design is so dif-
ficult and so important,focus groupscan be
used to understand interpretations of particu-
lar questions and questionnaires are always
pre-tested.
Questionnaires are considered by some to
be best suited for discovering patterns and
less helpful for studying causal relationships.
Marsh (1982) assesses these criticisms and
provides a careful guide for questionnaire de-
sign so as to enhance the utility of question-
naires for statistical causal analysis. This
involves collecting data on many different vari-
ables, beyond ones identified as independent
(causal) and dependent (effect) variables, in-
cluding possible confounding, antecedent and
intervening variables. She also provides guide-
lines for asking people about their reasons for
doing things. This ‘reasons analysis’ involvesGregory / The Dictionary of Human Geography 9781405132879_4_Q Final Proof page 613 31.3.2009 7:14pm Compositor Name: ARajuQUESTIONNAIRE