Methods in human_geography

Methods in Human Geography
A guide for students doing a research project
R. Flowerdew and David Martin
This summary is made possible by Ellen van Dalen
2. Philosophies underlying human geography research
*
3 important assumptions when choosing a research topic with sub-questions:
• That the questions themselves are intelligible
• That they can, in principle, be answered
• That the answers will add to our understanding of the topic
The purpose of your dissertation is to produce ‘warranted geographical knowledge’; if you
claim to know something, you must be able to justify that claim when it comes under attack.
*
You cannot avoid Philosophy because methodological choices are, at heart, philosophical
choices, and are crucial to any justification you will have to make when undertaking your own
research. The claims you make in your research must be defended against competitors.
*
Any research will have a subject matter and a method or approach. These two dimensions
define the basic methodological framework of the project, together with theories and
underlying philosophies they form the research design.
*
Deciding your research design is not easy, in order to make an informed choice, you need to
understand the philosophical basis of that choice. Exploring some of the major philosophical
debates that underlie the choice of research methodologies in human geography is necessary:
One of the most fundamental choices that any social scientist has to make is that between the
philosophical positions of naturalism and anti-naturalism.
Naturalism: The claim that research in the social sciences is essentially the same as that in
the natural sciences with their methods (empirical evidence, falsification,
natural laws).
Anti-naturalism: The claim that social science is not like natural science, anti-naturalism has
roots in hermeneutical or interpretative movements in philosophy with a key
role of mental qualities of human beings. Human geography is anti-naturalistic
and is unlike physics because it deals with human action and experience.
(Terms: experiential place, text, discourse.)
Another choice is the debate between the ontological positions of realism and anti-realism
(idealism).Which simply means that these positions involve opposing claims about the basis
of existence.

Realism: A common sense philosophy which maintains that there is a ‘real’ world out
there, even if we don’t see it and independent of our conceptions of it (material
existence, it is ‘really’ there).
Anti-realism: There is no possible justification for believing in a reality other than that
constituted by the human mind. We make our world in our minds. This view is
sometimes known as (metaphysical) idealism.
Realism/anti-realism does not neatly map the distinction of naturalism/anti-naturalism.
Mostly: idealism as a hermeneutical and hence anti-naturalistic philosophy (but leaving open
the possibility of naturalism). Realism can be positive and hence naturalistic (but leaving open
the possibility of anti-naturalism).
*
Judging between competing social theories can be problematic. Nevertheless, it is possible to
detect some dimensions of choice which provide grounds for accepting or rejecting certain
ways of conceptualising social interaction.
Marxism is a particular way of interpreting the world which is often called historical
materialism it takes as its starting point the material conditions of human existence, the
presence can only be understood in term of the past. A central concept of Marxian social
theory is the ‘mode of production’ which characterises a society, a mechanistic model of
society beyond individuals. Capitalism is seen as a functioning system with its own internal
logic, and individuals are reduced to passive agents in the production and reproduction of the
system, which ultimately depends on the economic base. ‘Contemporary’ capitalism is seen as
inherently unstable and thus inevitably bringing about its own demise. (Marxism is very
critical about idealism.)
A social theory that challenges Marxism in a certain way is Feminism. At the core of
Feminism is the concept of patriarchy, which sees an understanding of the historical
relationship between men and women as basic to any analysis of how societies are ordered. If
the Feminists are right then Marxist theory must be modified so that gender relations and their
interplay with economic relations are seen to form the substructure of society. As a social
theory, however, Feminism is not dissimilar to Marxism in that it is a holistic theory which
explains individual conduct by appealing to much larger scale social facts.
Another challenge concerns Marxism’s treatment of human agency. Marx conception of
capitalism is hotly contested by those whose view of the individual of society is informed by
the ideas of Humanism. Humanism is not itself a social theory but rather a diverse set of ideas
which have in common an emphasis on the humanity of individuals. People are capable of
being creative (or destructive), reflective (or not) and above all, they are moral beings, which
is to say that there is a moral dimension to their actions (the characteristics of any social
theory). Consciousness and intentionality are central concepts and human meanings and
values are emphasised. This is the active view of human agency which rejects any
deterministic interpretation of human behaviour.
Marx provided a grand theory describing the way in which societies are transformed by the
inner logic of capitalism. His mission was to uncover what he calls ‘the law of motion of
capitalist society’. Almost like the natural lows of ‘hard’ science. Postmodernism however,
rejects the notion of a single general theory of society and sees social order as much more
partial and even contradictory. Postmodernism presents a special challenge and opportunity

for human geographers. It embraces, just like post-structuralism, a complexity of ideas both
about the nature of contemporary Western society and appropriate ways of thinking about and
studying that society. Yet the very toleration of different approaches (where all interpretations
are equally valued) is also its greatest weakness.
*
Consistency and a deeper understanding of the basis of your research design demands that
you take philosophy seriously. Underlying your dissertation research will be certain
philosophical assumptions and, probably, a particular conception of the nature of relationships
in society. However, the practice of geography, its approaches and methods, does not map
neatly onto the philosophical choices you might make.
Any research project starts from a question or problem that defines the topic or what is to be
studied. The approaches and methods used to tackle the research problem, or how you are
going to study, should be appropriate to the question posed. First, the way in which the aim of
a piece of research is conceptualised and expressed contains within it assumptions that tend to
direct researchers towards certain methods. Secondly, it is the framing of the research
problem that links epistemology and social theory to method.
The term ‘warranted knowledge’ indicates that answers to research questions need
justification. It should now be evident that philosophical and theoretical choices form part of
that justification along with the details of empirical evidence collected.
Section A: Preparing for the research project
3. Choosing a topic
*
Where can you turn for sources of inspiration?
• Consider a course, or set of lectures, or maybe just one class that has interested you,
and was worth following up.
• Authors; will usually end with a paragraph or more that calls for further research.
Books; browsing the library shelves, see if any of the titles appeal.
• Browsing newspapers, magazines, or television and radio programmes.
• If you have settled on a broad topic area, why not try to ‘solicit’ a topic from an
organization, or individual working for such?
• Seek to develop one of your own outside interests into a research topic (football).
• You can, off course, always talk to one of your lecturers, a post graduate student or
research assistant.
What should you avoid when generating ideas?
• Previous projects (not university level).
• General geographical descriptions without any sense of research problem.
• Use a technique in search of a problem. (First find a research problem, then worry
about what kind of approach, method, or technique is appropriate.)
• That it has to be useful! (intellectual curiosity is enough)

*
What criteria makes a project to be research?
• Originality: main criteria. (Does not necessarily mean the collection of data first
hand.)
• Topicality: is the subject of contemporary interest, eighter to academics or the
public or both? (Anything in the current papers is, by definition, topical!)
• Interesting: select a topic that will keep your interest over several months.
*
You should be able to find a geographical angle to most topics, but you should be prepared to
defend its geographical character: which has a fundamental concern with place, landscape,
space and location.
*
When looking for a topic you will need to consider some practical constraints, revolving
around time, cost, personnel etc. Have you obtained any permission or clearance necessary to
collect data? Are ethical issues involved? Do never pick a topic that puts you in any danger!
4. Finding previous work on the topic
*
It is important to find out what other work has been done on the topic:
• You are a university student and acquiring an education involves increasing your
knowledge by finding out about your chosen subject.
• Other people working in the area will have thought about it and come up with some
interesting ideas and findings.
• People expect a literature review as a component of a dissertation.
• A bad reason: somebody else might already have had your idea; you should not worry
about this!
*
Approach to finding relevant literature:
• Talk to your supervisor and other people in your department.
• An up-to-date textbook: may provide some general context or contain references to
highly relevant studies.
• Local authority departments.
• Subject catalogue of your library.
• Bibliographies (but may contain lots of material which is not easy available).
• Abstracts: summary, almost all journal articles have them at the top of the article or
on-line. (GeoAbstracts/GeoBase)
*
Once you have found one article it is easier to find more:
• Reference lists: which are no good in finding up to date literature on a topic.
• Citation index: if you have found a relevant article you can use it to look up other
articles that have cited it. This allows you to work forward from a relevant article,
to find recent work on your topic.

Always take note of recent developments in your subject.
*
Be pragmatic, do your best in the time available.
5. Secondary data
*
Secondary data: information that has already been collected by someone else and which is
available for you, the researcher, to use.
Primary data: which you gather for yourself.
The secondary data sources relevant for dissertations comprise existing information which is
publicly available even if not already published like major government surveys, statistical
resources, records of all kind, newspapers and publications, maps, photographs and video and
sound recordings.
*
The strengths of secondary data:
• It exists already (and so is cheaper and quicker to obtain than primary data).
• It provides the researcher with contextual material for his/her primary research.
• It is usually of proven quality and reliability.
• A very wide range of secondary data material is available.
The weaknesses of secondary data:
• Its inflexibility (you cannot customise it to your needs).
• Its quality is unverifiable since it is not replicable (you have to trust it).
• It may cost you money and takes you time to obtain.
• Secondary data is a cultural artefact, produces for administrators with priorities and
ways of seeing the world that may be very different from those underpinning your
dissertation.
Why you should take the best use you can of the secondary data:
• They sketch out the issue or question you will pursue in your own primary research.
• To provide a context for the primary data you will subsequently collect (provides three
overlapping types: geographical, historical and socio-economical data).
• It helps you show potential employers that you can be relied on to find out what is
already known about any topic.
Warning: secondary data are not static; you have to be careful for representativeness
(validity).
*
For human geographers perhaps the single most widely used source of secondary data is the
population census. The best starting points are websites of national and international statistical
organisations and directories of official publications (Eurostat).
*

Online: single sites, maps, catalogues, compendia, portals (websites that does not
provide information itself, but is a list of hyperlinks to other websites usually
having something in common), search engines (narrow it down, be precise, if
‘tourist’ / ‘tourism’ etc. Search for ‘tour-‘ / ‘touris-‘, keep trying all the
synonyms you can think of, use AND/NOT/OR).
Websites are usually unregulated and need to be treated with some caution.
*
Finding comparable data may be difficult, u need the same spatiality, date, methodology, and
with cartographic issues need the same scale.
The principle functions of secondary data are to provide a context for more detailed case-
study work which you might conduct yourself.
Section B: Collecting Primary Data
6. Questionnaire design and sampling
*
Main focus: Analytical surveys, they are more concerned with explanations and causality
and are therefore more frequently adopted by academic researchers.
The questionnaire survey is an indispensable tool when primary data are required about
people, but should be avoided when secondary data can answer your questions. The need for a
survey must arise from a carefully considered set of research objectives and not from a
general desire to ‘go out and do a survey on my topic’. Conducting questionnaire surveys
entail a lot of work.
*
Three types of survey data:
• Data that classify people, their circumstances and their environment (respondent
variables).
• Data that relate to the behaviour of people.
• Data that relate to attitudes, opinions and beliefs.
Four types of variables considered in analytical surveys:
• Experimental or independent variables: these include all variables that may be
possible predictors of the main effects that are being studied.
• Dependent variables.
• Controlled variables: factors that need to be held constant to compare results.
• Uncontrolled variables: factors that have different answers (distance).
*
The stages of a questionnaire survey:
• Initial research idea: refine and develop analytical design, research objectives, and
literature.
• Design of research: hypothesis, consider variables, choice of survey methodology.

• Further refinement of the research instrument and sampling: pilot survey, sampling
frame (representive, sampling size: each cell at least 30 respondents).
• Main fieldwork.
• Processing / analysis of data: Data control, coding of data, data transcription,
statistical analysis, production of tabulations.
• Results: test hypothesis, research report and conclusion.
*
Fundamental: researchers must address the twin issues of reliability (can the results be
replicated?) and validity (does the survey measure what it was intended to?).
Sampling errors: associated with how respondents have been selected (sample size).
Non sampling errors: by questionnaire design biases.
Response errors: by the process of interviewing (understanding opinions, expected answers,
inappropriate questions).
Non response errors: arise through biases in who did and did not respond.
*
The content of the questionnaire needs to be firmly rooted in the research question of
hypothesis under investigation. A pilot survey is very useful. The survey should not be too
long, the content must be practicable and relevant to respondents (understandable language,
do not jump from topic to topic, use an introduction).
Frequent mistakes: Double-barrelled questions (2 in 1), Negatives, recurrent of habitual
behaviour (not ‘recently’ but ‘in the last month’), leading or lauded questions,
potentially embarrassing questions, classifying questions (age).
Use a number of different question types. Basic distinction is between open and closed
questionnaires.
Closed: when most/all of the alternative answers are provided.
Open: do not force respondents into giving particular answers, but are more work.
Attitudes and opinions are the most difficult category of social survey data to collect and are
best put on scale: attitude battery (not interesting – very interesting and in between, or so).
*
Surveys are usually conducted by sampling. Ask yourself: What are the appropriate units of
study? (A geographical boundary, a tempotal boundary (time), a boundary defined by
population characteristics.) What is the target population?
Sampling terminology see Box 6.3 on page 96!
*
The choice of survey technique is of great importance to the overall success of the research:
Observational studies: do not require interviews.
Unstructured interviews: informal approach, need preparation.
Questionnaire surveys: face-to-face, internet surveys (low costs but a high non-response),
self-completion, postal surveys (cheap for large geographical coverage),
telephone (good compromise between postal and face-to-face surveys,
responses are high and call-backs easy).

Advantages must be weighed carefully against the disadvantages compared with face-to-face
interviewing (see box 6.4 on page 102).
*
The design of a pilot survey should take in consideration: question design and format, length,
output, classification questions (do you really need income?), serialisation (put a number on
each survey) and other information.
Response error can be avoided or minimised by careful consideration of interviewing
technique (safety, time and location, your appearance, identification).
*
The main editing and data processing tasks begin once the survey is complete. Think good
about how to deal with low frequently answered question options, non-response or multiple
answered questions.
When analysing: refer back to the original research plan: What were the main dependent and
independent variables? Which types of statistical tests might be required?
Final report:
Fieldwork locations and dates should be included in the methodology section as well as
details of sampling. Provide a justification of the survey technique, mention failings and
include a copy of the questionnaire.
Be selective in the data tables you include in the text. Do not get frustrated with a few results,
that is normal. Acknowledge institutions which gave assistance and send them a summary of
your main findings.
7. Tell me about...: using interviews as a research methodology
*
Questionnaire surveys might push respondents’ answers into particular categories. Interviews
are an alternative and are unstructured or semi structured, they are a dialogue rather than an
interrogation. Unlike a questionnaire, the aim of an interview is not to be representative but to
understand how individual people experience and make sense of their own lives.
Critique on in-depth interviews: interviewers bias the respondents’ answers; interviewers are
not or cannot be objective or detached.
*
Recruiting informants for interview is to select an illustrative sample; it has to be a
theoretically motivated decision. Who you want to interview depends on your own identity,
recognise your personality and be reflexive. Watch out for power relations between you and
the informant but also in the informants surrounding.
Different ways to approach potential informants:
• Via a questionnaire: to get information and ask for willingness.
• Cold Calling: trace telephone numbers of possible informants.
• Gate keepers: individuals in an organisation that have the power to grant or
withhold access to people or situations for the purposes of research.
• Snowballing: using one contact to help you recruit another contact, who in turn can
put you in touch with someone else.

Where to hold interviews: At their own territory (relaxed and you learn more about the
person). Avoid busy places where you can get disturbed. Always take safety precautions.
*
Brief yourself on the topic carefully. Interviews do not rely on a rigid set of questions worded
beforehand but work out a list of themes or some key questions on which you may then hang
a series of follow-up ideas. If they start talking about your last theme first, follow their train of
thought, otherwise the spontaneity of the interview is lost.
Encourage interviewees to talk by: ‘Tell me about...’, use silences, ‘uhuh’, or repeat back
statements as questions, use gestures, facial expressions and body language.
Asking the questions is only half the skill of interviewing, the other half is learning to listen
and respond to your participants.
*
Always tell interviewees that they do not have to answer any questions which they do not
want to and remind them of this during the course of the interview. Let them stay anonymous
and confidential, and always be objective (especially with unequal power relations / cultural
differences).
*
Advantages of taping: you can concentrate better, you can have a proper conversation, more
detailed, and you can listen to it over and over again.
Always check if it works and think about the location.
Try and transcribe the tape as soon as possible (some hard parts you might still remember and
it takes a lot of work).
Write down your experiences after the interview.
Write a thank-you letter with a summary of your findings.
8. Focus groups
*
Focus group: a group of individuals selected and assembled by researchers to discuss and
comment on, from personal experience, the topic that is the subject of research.
Group size: 4-10 participants as an ideal, with a researcher (or two) to facilitate and
moderate the conversation.
Start with an introductions followed by an issue-based discussion, with participants
contributing in the ways and at the times they wish to. The conversation is usually audio
recorded (or video recorded), and transcribed as a written text. The text may be supplemented
by moderators’ notes regarding group dynamic, mood, gesture etc. The result is a rich
qualitative record of a focused group conversation.
Focus groups are useful in two main regards:
• Gaining insight into views that individuals hold regarding a particular issue.
• The nature of their interaction and dialogue over that issue.
Focus groups lie somewhere between individual interviews and participant observation.
*
History:

The origins of focus groups are generally held to lie in work undertaken at Columbia
University in the 1940’s (experimental procedure of 12 people together in a studio with
yes/no buttons). A series of subsequent studies with alternative procedures were developed. It
was not until the 1980’s that focus groups began to re-emerge as a prominent method in the
social sciences (marketing, group therapy).
People construct environmental and social issues, share their knowledge, experiences and
prejudices, and argue their different points of view.
*
Usefulness of focus groups as a core research method to:
• Gain insight into participants’ understandings and views of an issue.
• Gain insight into how these views relate to each other (debates and arguments).
• Look at how such views differ by social grouping.
• Probe the gap between what people say and do.
Also useful for:
• Preliminary brainstorming: generating potential lines of enquiry.
• Developing survey or interview questions.
• Disseminating research findings and receiving feedback.
*
Central issues: determining a focus, recruiting participants, formulation of question schedules
or other forms of discussion prompts, nature of the role of the moderator, and
how to analyse the qualitative materials.
Recruiting: Two options are known as ‘natural’ (individuals are likely to be familiar) and
‘assembled’ (strangers) focus groups. Always over-recruit. How many groups? That depends
on the range of subgroups you are interested in and the number that could make ‘theoretical
saturation’ possible (having heard all the main points of view on the issue).
*
Running a focus group: normally seated in a circle, room must be well lit, heated and
ventilated. Practice run the audio recording. Start with introducing yourself and the others.
To help initiate a focused conversation: use participants statements, images, distribute
preliminary questionnaires.
If the conversations dry up: prompt, guide and intervene. The length of a focus group should
last somewhere between 60 and 120 minutes. If the moderator senses a point of saturation,
draw the focus group to a close.
The moderator’s involvement should centre around:
• Ensuring the conversation does not significantly diverge from the predetermined
focus.
• Intervening when there is an unproductive conversational dynamic emerging between
participants, such as when the volume or intensity of one contributor is notably
inhibiting others from speaking freely.
*
Transcribing focus group material can be difficult when people talking over each other. Tone
is very significant and could mean different things.

When analysing focus group material: The researcher is centrally interested in the themes that
are evident within the discussion of the research issue. It is also important to be attentive to
the unexpected issues and ideas that emerge within the conversation, whether these fit with
the a priori research framework interests or not. Unforeseen responses may be of particular
significance.
The overall aim of the analysis should be an understanding, as noted earlier, of the spectrum
of different views a group of individuals has on an issue.
*
Focus groups are able to generate rich conversational data that provides insights into complex
social issues. This material is relatively unstructured, however, is not statistical and is not
readily generalisable to a wider population. It is thus important to consider whether focus
groups are the most appropriate method for addressing your research aims.
9. ‘Participatory’ approaches and diagramming techniques
*
Participatory approaches: working with rather than on people; generating data and working in
ways that increase participants’ ability to bring about positive change in their own lives (not
about simply describing or analysing social realities). There is a distinction between
alternative research techniques such as participatory diagramming, and a participatory
approach to research; one does not necessarily imply the other.
*
Conventional research: a process that begins with an externally developed research design,
proceeds with the extraction of data from ‘the field’ and ends in the presentation of results.
Participatory diagramming techniques: participants, rather than the researcher, produce the
maps, transects, charts and matrices that describe the research results. After raising a question
and suggesting a visual means of addressing it, he or she takes a ‘back seat’ while participants
generate the diagram. The task of diagramming is tactical as well as verbal; it facilitates the
contributions of less eloquent individuals in the group (ethical reason for pursuing a
participatory approach). All diagramming techniques are highly visual and flexible, with the
use of loose materials to put on the diagram.
Gold standard: integrating participation into all aspects of a research project (most projects
fall short of this goal).
Methodology:
• The initial diagram should not be viewed as an end product but you should see it as a
mechanism through which to gather further qualitative interview data and as a means
to stimulate further discussion and self-analysis among the participants.
• Plan your methodology so you have mapped out in advance the particular
diagramming session and also its place within the project as a whole.
• Aim to achieve two goals: obtain the best and fullest impression possible of the
subject you are investigating; facilitate participants’ own learning, self-reflection and
analysis.
• If alternative techniques and participatory approaches are to run smoothly,
considerable forward planning is required.

• You can cover a wide range of issues relatively quickly and get on to difficult subjects
more readily than if you use more established methods.
*
Suggestions about how you can integrate participation into one or more stages of your
research project are summarised in box 9.2 on page 160. These set out an approach within
which you might decide to use a particular methodology.
Comments when you are thinking about participatory research:
• You will need to make some decisions about the scale at which you wish to operate.
• Think about the level at which you attempt to enter the power/decision-making
structures of society.
• The presence, or absence, of pre-existing structures through which to work and gain
access to participants (snowballing).
• Should the group be composed of strangers or naturally occurring groups of people.
*
The scientific validity of ‘participatory’ approaches and techniques:
• It can legitimise the researcher’s presence, and position the participants as
stakeholders with a reason to help the project succeed.
• If participants are involved in the data analysis, the method is likely to produce more
‘credible accounts’ that participants themselves recognise as accurate. Triangulation
and ‘member checking’ (respondents certifying the validity of data) are built in.
• Because process is such a central concern and is meticulously recorded, participatory
research is well placed to provide a full and open audit of how and under what
conditions ideas and themes arose.
• The final written report incorporates multiple accounts, including the participants’
own analyses of the data; it provides a narrative ‘between’ the perspectives of
participants and those of the researcher.
• The epistemological foundation that underlies all qualitative research is that
researchers need to build trusting and reciprocal relationships with ‘respondents’ id
they hope to gain really deep insights into their life-worlds.
10. Participant observation
*
Participant observation: a method which involves living and/or working within particular
communities in order to understand how they work ‘from the inside’. By participating in a
community or observing a community. It is a three-stage process in which the researcher
somehow, first, gains access to a particular community, second, lives and/or works among the
people under study in order to take on their world views and ways of life and, third, travels
back to the academy to make sense of this through writing up an account of that community’s
‘culture’.
*
Access: It does not have to be an isolated community in some faraway part of the world; it can
legitimately involve anything like several months of sustained work during a summer holiday.
Access has to be negotiated through various ‘gatekeepers’. If a community is not open to you

and gives you doubt and frustration than this is where the fieldwork starts. You should not
expect things to proceed according to any pre-planned schedule.
*
Roles: it is not only who you contact during the gatekeeping and subsequent stages of your
research that is important, but also the way that you present yourself when doing this: what
role you take on in the course of your work. Take in consideration whether you should
provide a full explanation of your role, or conceal your purpose and identity and to what
degree your research can or should be participatory and/or observational. Think about any
impacts it might have on the loves or the participants in the future.
*
Constructing information: the most important information that you construct from your
participant observation you put in your field diary. How you were able to access the
community, power relations, expectations and motives, what you have observed, day by day,
and what you have come to (mis)understand as a result.
*
Key questions which you will continually have to ask yourself along the way if you choose to
take on this method:
• How can I vividly convey what I have learned during my fieldwork to those who will
read the final dissertation? (Creative writing!)
• How can I integrate this kind of account with the kinds of theoretical, methodological
and substantive discussions which are usually required in a dissertation? (Set out the
relationships between theory and practice.)
Section C: Analysis of data
11. Analysing numerical spatial data
*
Data analysis is the part of a research project with the greatest trepidation, it is most closely
scrutinised and most easily criticized. In order to increase the positive aspects of data analysis
and reduce the negative ones; some simple questions to ask yourself:
• When should I start thinking about data analysis?
• Why should I undertake data analysis?
• How do I decide what technique to use?
• How geographical is my analysis?
• What problems might I encounter?
The appropriate time to start thinking about data analysis: before the data are collected rather
than after!
*
Why should you undertake data analysis?

• To summarise large amounts of data through descriptive statistics (arithmetic mean,
median, interquartile range, coefficient of variation, centre of gravity, standard
distance etc.).
• To explore data through visualisation: this is called exploratory analysis (suggest
aspects of data or a modelling technique that might be worth investigating in more
detail, presented in visual format such as pie charts or box plots).
• To infer aspects of the population from samples or processes from observations: the
answers to the question are then provided by hypothesis testing and statistical
inference which fall into the category of the more usual deductive-based confirmatory
analysis.
• To develop models of the real world.
There are a bewildering number of statistical techniques available; there are two properties of
data that limit the range of appropriate numerical techniques: the number of samples and the
type or level of data. How to determine an appropriate statistical technique is shown in Table
11.1 on page 198.
There are three levels of data that are commonly encountered:
• Nominal: discrete (data can only take certain values, categories have no ranking.
• Ordinal, weakly ordered: discrete, data are in categories which are ranked.
• Ordinal, strongly ordered: discrete, data are ranked.
• Ratio: continuous.
Non-parametric statistics: statistical techniques that can be applied to nominal and ordinal
data.
Parametric statistics: data that provide the most information since they are continuous and
have metric properties (ratio).
*
Weakly geographical: uses data for different locations but effectively throws the interesting
spatial component away and utilises the data as if they are simply lists of
numbers that could be reshuffled without affecting the outcome of the analysis
(ordinary least-squares regression analysis).
Strongly geographical: explicitly uses the information on the spatial location of the data points
(calculation of spatial means, standard distance, spatial variation).
What problems you have to consider:
• The modifiable areal unit problem: in the analysis of aggregate spatial data, the
conclusions reached might be dependent upon the definition of the spatial units for
which data are reported.
• Spatial non-stationarity: relationships might vary over space making ‘global’ models
less accurate. Global models will also hide interesting geographical variations.
• Spatial dependence: to what extent is the value of a variable in one zone a function of
the values of the variable in neighbouring zones? Strong spatial dependence can
affect statistical inference. Data should be independent of each other.
• Non-standard distributions: examine the nature of the data being analysed. Not all data
are normally distributed! Consider experimental methods of statistical inference if
data have an unusual distribution. Experimental distribution methods may be used
to confirm or reject hypotheses.

• Spurious relationships: make sure that any relationships identified are meaningful and
are not caused by a variable or variables omitted from the analysis. Each parameter
indicates the relationship between the dependent variable and one independent
variable, everything else being equal.
12. Analysing categorical data
*
Categorical data is encountered mostly with questionnaire surveys. Both nominal and ordinal
variables are discrete (i.e. can take only certain values) and are types of categorical data. It is
always possible to convert data from a higher level (ratio) of measurement to a lower one, but
not in the reverse direction. It is consequently important that results are not blindly accepted
as correct just because the computer software produced them. Ensure that relationships
between variables are examined in as thorough a manner as possible. It is therefore a good
idea to undertake exploratory analysis of the data before using confirmatory methods to test
hypotheses.
Interrelationships: Multiple causes cause misleading conclusions, this is particularly likely
when interactions are present (i.e. the association between two categorical
variables changes according to the value of a third factor). Linear models test
the significance of particular associations (interrelationships) or interactions
between variables.
*
Methods of exploratory data analysis use descriptive summary and graphical display variables
(SPSS, MiniTab, Excel). They can be used to examine trends in data and are also effective at
detecting outliers (check for any outliers early in an analysis).
Associations between nominal variables can be investigated by cross-tabulating the data to
form contingency tables. It is important to think about the question you wish to answer before
deciding the direction in which percentages should run. They should sum to 100 within the
categories of the explanatory variable.
Appropriate exploratory methods for ordinal variables differ according to whether the data are
weakly (treat the data as if they were nominal) or strongly ordered (scatter plots).
*
You will need to undertake some form of confirmatory analysis to test the hypotheses. A
statistically significant result in this context is one where the trend in the sample data is
sufficiently strong that there is only a small probability (conventionally under 5%) that it
could have occurred by chance. Non-parametric methods have lower statistical efficiency than
parametric methods, but they also make fewer assumptions about the parameters of
populations from which samples are drawn.
Irrespective of the type of data and tests involved, you should pay some attention to the
presentation of results from confirmatory analyses. Including the output from every test
conducted is usually unnecessary and conveys the impression of a lack of thought. A suitable
approach might be to discuss one or two in detail and then summarise the remainder in a list.
*

Modelling more complex relationships like with several different possible explanatory
variables, it is necessary to use types of generalised linear models. These have the advantage
that they can be used to test explicitly for the presence of particular combinations of
associations and interactions between variables.
It is important that results are presented in a succinct and interesting manner.
13. Analysing qualitative materials
*
Analysing qualitative material deals with verbal or textual material. Analysis of the material
should not be an afterthought, but needs to be included in early research plans, because
getting to grips with the materials properly can take as long as creating them in the first place.
Large quantities of material generated by qualitative research, so called field materials are
interviews or open-ended questionnaire responses, observations and notes made by the
researcher, sketches, maps, itineraries, diaries, letters, memoirs etc. They all need different
approaches when analysing them.
*
The first step is to get them into a presentable, readable form. The amount of detail in the
transcription really depends on what sort of study you are doing, you can afford to miss out
irrelevant portions but sometimes you may need more detailed transcripts. Styles and
transcripts vary between researchers and approaches, notations can be used for pitch, pause
length, pronunciation and even body language.
What tends to be done, while sifting and sorting the material, is that the researcher goes back
over all those notes and transcripts and starts to formalise them into categories or codes. Each
time people refer to a particular event, it is given the same code. After that you cut up a copy
of the materials and put the coded sections into piles, so each code forms a pile.
*
The notes to yourself are often termed theoretical memos and are worth keeping, and need to
be labelled so they can be linked back to the materials they refer to. One of the important
reasons for keeping theoretical memos is so you can trace back through them as well as see
what made you change your ideas. This in the end becomes one way of justifying your
conclusions, by being able to show how they were created.
*
You should be clear whether you see the relationships of codes to materials as your own
interpretation, as structures underlying the issues you studied, or as relationships your
informants used more consciously. There are obviously grey areas between all these and your
position will depend on a variety of theoretical and ethical considerations, which will vary for
every project and researcher.
It is important to note he different epistemology (how you can claim to know something) from
many quantitative projects. What is not so much of interest are the codes itself rather the text
they denote, not how often they occur but what is in them. The codes have emerged from an
iterative process, going through the material, what is called analytic induction.

Organise the materials so that interesting relationships can be seen, the status of codes needs a
little clarifying. It needs a little caution about what your codes do and do not mean. The
division between emic and etic codes should be considered.
Emic codes: Those used by informants themselves.
Etic codes: Those assigned by the analyst to describe events and attribute meanings and
theories. These are more theoretically oriented, responding to issues in the
research.
In thinking through the relationship of category and material, theory and evidence, researchers
have to consider what sort of account they want to give. (It might sometimes be useful to turn
to styles that pay more attention to the perhaps unsaid structures.)
*
Structures of meaning that may not be apparent to the informants themselves like that the
researcher is seen as knowing best or structures that relate to real-life situations, account for
regularities but they do not explain what actors think. This type of analysis (of underlying
structures) will be developed from semiotics.
Semiotics: the study of signs and the construction of meaning. It is concerned with the
way words, things, pictures and actions come to be signs, that is to convey
meanings in particular times and at particular places. A few tactics:
• The first principle of semiotics is that meanings are relational rather than fixed. Signs
derive their meaning from other signs and from the wider system of signs not from
their actual form or content.
• Charles Peirce divided communicative signs into three types: iconic, indexial and
symbolic. Iconic: depicts what it means (picture of woman with child = woman with
child). Indexial: part of a wider complex (picture = motherhood). Symbolic:
relationship between sign and meaning is purely conventional.
• A third tactic is to look for oppositions between categories; we see meaning created by
the relationships between things or semiotic clustering.
• Complex cases, simple oppositions developing a semiotic square of rules. The square
organises incidents into the 4 categories or especially outside the categories. The
interesting parts are the more shifting relations between categories.
These tactics should make clear how organising the relationships between categories might
help you develop ideas. It is not necessary to use these particular devices; they are examples
of developing arguments from the relationships between your categories. You can also use a
common diagram which is not discussed.
*
Narrative analysis: people tend to make sense of things by seeing them as stories, comprising
events, imputing motives, agency and roles, rather than in terms of static characteristics.
You can think about the story on several levels. At the grand level, you can think of whether it
talks of progress or decline. On a more detailed level, you can look at how events themselves
are defined as significant or not. The level of detail tells us something about the importance of
an element. The position of the teller, the actors and roles can be thought about. On a final
level, the ideas of drama and scripts can be seen, what is really meant with a certain message,
why a certain role is adopted.
*

There is no single mandatory way that materials have to be understood or used. They
suggestions in this chapter are ways of thinking about your materials that might help you
develop your own approach.
14. Textual analysis: reading culture and context
*
What are texts: data that you need for an original research project, through art, films, videos
and books.
How do you approach texts: through theoretical windows and interpretations of text through
semiotic, Marxist, feminist, psychoanalytic, post-colonial and post-structural critiques.
*
Text: written material that occupies anything from a newspaper article to a volume sitting on
shelf of a library. But also other communication systems such as fashion, etiquette and urban
design. Diverse subjects as zoology, medicine, musical codes. Today geographers even read
images such as paintings, maps and landscapes with social, cultural, economic and political
institutions as texts. These productions exhibit text-like qualities and are interpretable using
textual methods.
Literary texts: geographers turn to literary texts so that the concept of place may be reinserted
into discussions of societal transformations through modern times.
Travel writing, diaries and other anthropological texts: Research on travel writing attempts to
elaborate imaginary geographies in the context of othering and masculine
heroic projects. It is possible to glean how sexuality, race and power are
geographically elaborated through culture and globalisation.
Visual texts: geographers use textual methods to help read paintings, photographs, movies
etc. There certainly is some methodological tension between the use of textual
and visual metaphors to understand the contexts of the geographic world.
Musical texts and performances: understudied topic in geography, although music is clearly
an important cultural text for regional differentiation, cultural diffusion,
perception of place rather than anything specifically textual.
Reading maps: Maps may be described as visual texts and analysis in geography found great
potential in cartography.
Landscape as text: describing the morphology of contemporary landscapes.
*
How we come to understand texts, simple how-to steps: See list on page 242, in short: find
something interesting, decide if textual analysis is appropriate, read the text, ask why the
author created the text, read everything you can find on the topic, read the text, look for
structure, determine how your text relate to other texts, look for subtexts/missing texts, read
text, make connections to larger geographies and social processes, create dissertation.
The treatment of any kind of geography as a text that can be read implies that, at least at the
moment of reading, the text has some meaning which may be grasped. Some of the theories
perspectives and ways of knowing that make these bodies of knowledge more accessible.
Semiology: a science of signs: uncovering the meaning of texts by dealing with what signs are
and how they function. Any sign could be divided into two components: the

signifier and the signified. The point that makes the analysis of texts interesting
and problematic is that there is no logical connection between them.
Intertextuality, metaphors and metonym: Important concepts of semiology. Intertextuality:
refers to the conscious of unconscious use of previously created texts. Readers
must be familiar with the original text to fully appreciate the new text.
Metaphor: the relationship between two things is suggested through analogy,
are used to give objects and events different identities. Metonym: relationship is
based upon association of names. The use of metaphors and metonyms implies
the existence of codes in people’s minds that enable them to make connections.
Hermeneutics: a method of interpretation: much in common with semiotics but its base is in
literary criticism rather than linguistics and its scope is broader, embracing
methodological principles of interpretation which incorporate the context of the
reader as well as the context of what is written.
Marxist analysis: the production and consumption of texts and ‘the political unconscious’:
Powerfully suggestive way of interpreting texts. Assumes that beneath the
superficial randomness of things there is a kind of inner logic at work.
Everything is shaped by the economic system of society, which affects ideas
and consciousness. A secondary concern is how the economic base can be
defined without having recourse to the categories which are themselves
superstructural.
Feminist and psychoanalytic criticism: Feminism addresses aspects of woman’s geographies
that have heretofore remained silent. It’s about woman/environment reciprocity
and renewal rather than male stoicism; they talk about personal vulnerability
rather than dominance and power. Another aspect is the focus upon the
gendered power relations inherent within texts.
Psychoanalytic theory; we form an image or text of ourselves that takes the
place of the real us, but we are always trying to rediscover the real.
Post-structural?: Treating texts as social products that are dynamic, and culturally mediated by
discursive practices. The student brings something of themselves to the text.
The readers’ knowledge is used and is interconnected. A post-structural
question is not only what is known, but also how it comes to be known.
*
All forms of text are produced in large part through discursive practices, which structure
flows of understanding. Discursive practices imply intertextuality rather than rambling chaos,
but these practices are as always embedded in social relations of power and ideology that give
authority to some texts while subverting others.
15. Visual methodologies: what you see is not always what you get
*
Of all our senses, the visual is our most useful and, by extension, our most acute, what we see
is important and there is a lot of confusion about what actually constitutes visual imagery.
Work images on our senses as narrative-like texts?

*
It may be argued that visualisation is at the heart of geographic practice with the study of
landscape, maps, paintings and photographs.
There are many ways to study power relations in images. Geographical shapes, meanings and
arts do many things. They are used to represent real world places and people and geographical
images create landscapes, the traditional domain of geographers.
• Representational spaces are shaped and created by life experiences.
• The opposite is also true: conditions within material society are most certainly shaped
through their visual representations.
• Geographical images are not objective representations of the material world but are
socially created under the influence of particular ideologies, meanings are social
constructions.
*
Art and photography: much research derives from humanistic tradition, and involves
highlighting the ways that landscapes are depicted. The images become places
that can be analysed to better understand lived experience.
Film and advertising: Interest of film in geography often revolves around the power of film to
shape contemporary culture with the compilation of power, place, situation and
spectacle. (Braveheart; how does his representation of a brave but failed hero
reflect on contemporary Scottish politics?) Advertising provides momentary
gratification for the alienated spirit because it stimulates desire.
*
Before choosing your method and beginning your analysis:
• Read through and around the context of your image, get knowledge about his times
and place.
• To avoid analytical incoherence, you need to situate yourself within the debates of
visual culture to interpret your chosen image.
• Understand the theoretical position that you bring to your analysis.
Visual methodologies:
Semiology: a method and theory of signs: take images apart and then focus on how they relate
to larger systems of meaning. Interpret images by identifying signs and then
analysing what meanings those signs construct. Sign: everything that, on the
grounds of previously established convention, can be seen as something
standing for something else. Signification: the social process whereby
something (the signifier) comes to stand for something else (the signified) and
by questioning this process, semiotics provides a popular way of getting behind
and through images.
Psychoanalysis and feminist analysis: visual ideology and the male gaze: in history woman
are never represented as self in these images, but rather as other, the dark
continent, the love inspired in the hero etc. Today there is a significance of
space as a site of gender differences: home, family and femininity or masculine
space of adventure and action. Mapping politics onto body space is affected by
how the body is seen and recognised.

Discourse analysis: the ways meanings are connected through representations, texts and
behaviours become forms of disciplining and so discourses are also about
power and knowledge production but reject classical analysis. The discourse
produces subjects and landscapes but it also produces technologies like maps
and GIS that enable particular ways to visualize the world.
*
Visualisation becomes a constructive process, an active and creative act that emphasizes the
viewer’s awareness of a greater need to understand their own cultural reality. By utilising
visual methodology, geographers can form uniquely valuable understandings of the geospatial
environment. Various visual methodologies help us to understand how particular spaces are
produced by manipulating images of the landscape and the built environment, but there is no
precise relationship between landscapes and images. The relationship of images to narratives
and texts may be metaphorical at best: we do not really know.
16. Geographical information systems and spatial analysis
*
Geographical information systems can be used to support many different research projects in
which there is a requirement to process geographically referenced data, particularly where a
wide range of data sources are involved (large volumes of spatially referenced information).
GIS: Alternative real world scenarios can be explored with systems for the input,
storage, manipulation and output of geographical information. You can use
GIS to handle pretty much all kinds of information for which you might think
of producing a map. GIS are better at data manipulation than analysis and it
will often be necessary to use other software as well. The golden rule is to
perform each task because it is a logical requirement of your research, and not
just because it happens to be available in the GIS software.
*
It is important to know if the chosen system will perform all the tasks required in addressing
your research question. It is not necessary to understand the details of how the entire system
works.
There is a very wide range of software marketed as GIS and the choice will depend on what is
available in your own institution.
A commonly encountered division among strategies for data organization in GIS systems is
between vector and raster models.
Vector: location is recorded by using map coordinates. Include topological information
and store geographical features individually and stored separated in tabular
form.
Raster: location is recorded by row and column position in a grid. Based on entire
gridded layers.
*
Data availability will be a very major consideration in any research project using GIS. Data
comprise the largest single component of the financial costs associated with commercial GIS,
and there is a whole industry concerned solely with digital data creation and supply. In many

situations the most appropriate approach will be to georeference (assign specific locations to)
research data by use of an existing locational data set. Important geographical data such as
census boundaries, postcode locations and general topographical information are created by
different organizations and many of these are available to the academic community and some
through internet.
Whatever your data sources, it is particular important that you make appropriate
acknowledgement in your final report and that you watch out for licensed conditions and
copyright.
Important issues relating to digital data are those concerning data quality. Checklist of
potential sources of error:
• Age of data (out of data, incompatibilities between different layers).
• Differing definitions of objects (e.g. coast defined as high of low water marks).
• Quality of source document (poor-quality source causing geometric distortion).
• Coverage of data (not all data layers cover the whole study area).
• Locational recording errors (e.g. poor digitizing, duplicating or missing lines, poor
matching at edges of source map sheets).
• Attribute recording errors (e.g. wrongly labeled zones, census under-enumeration,
missing values).
*
Visualization is a very important aid to exploratory data analysis and you should be sure to
allow yourself time to simply study maps of your data and to question what you see. Think
geographically.
Start with undertaking background research that falls in more than one domain, use secondary
data, it is actually very unusual to encounter a GIS research project that involves only self-
generated data. Analysis and manipulation should always relate back to the original research
question, and not seen as a methodology per se.
*
In considerating the role which GIS can play in more complex data analysis, it is helpful to
distinguish between the terms weakly and strongly geographical research using spatial data.
Weakly: uses spatially referenced data, but fails to make much use of the spatial component.
Strongly: utilizing explicitly geographical concepts such as clustering, dispersion, distance,
contiguity, etc.
The data manipulation and quary tools available within GIS are able to provide many of the
input measures required by more specialized spatial statistics.
Section D: Producing the report
17. Designing the report
*
The structure and presentation of the written product are vitally important to a successful
project. The report needs to follow a logical sequence and this requires a great deal of thought
from an early stage.

The structure that you decide upon will not be set in stone, flexibility is very important.
Various formal aspects of the structure can be identified but most important is that the story
(dissertation) is both informative and enjoyable to read.
*
Chapters: Findings should be written-up in a clear and logical manner.
Title: Clear and concise and should reflect the work that you have undertaken.
Clarity is probably a safer bet than originality.
Title page: title, name, year and institution putted in the golden section of the paper: the
eye naturally positions at about one-third of the page down.
Acknowledgements: thank those who have aided something to your research. Keep it brief
and not too personal.
Contents page: Make sure it summarizes the dissertation structure adequately. Each of the
figures and tables that you use in the dissertation are summed up in a different
contents page and require titles that neatly summarizes what it shows.
The abstract: should provide a clear idea in reader-friendly language of what the dissertation
is about, a brief summary of the methods you used and a description of the
major findings.
The introduction: an essential part of the logical argument you are putting forward. Clearly
state the aims of the dissertation by briefly describing what the main findings
on the subject have been previously, highlight why more work is needed and
how you want to archieve this. The introduction must be interesting, so that it
encourages the reader to continue.
Literature review: ‘if you steal from one author, it’s plagiarism; if you steal from many, it’s
research’ (W.Mizner, 1953). Cite those which are the most pertinent and those
who have been most influential in the field, put the rest in bibliographies. The
ideal literature review begins with a discussion of the topic in general,
progresses through the development of ideas on the topic, show what questions
have partial answers and how new questions arise. Contradictory findings
should be acknowledged. A literary review is longer than you think.
Methodology: Describe the procedures that you used in your research. This might include a
number of sections, depending on the methodological complexity of the study.
It is vital to include a section that discusses the philosophical approach and the
methods you decided upon. It is possible to identify problems that may have
influenced your study. Convince the reader of your methodology.
Results: It is tempting to include too much material. The results should be logically
structured, the results chapter can concentrate just on results. If there are a lot,
concentrate on the most interesting and relevant.
Discussion: To bridge the gap between the results section and the conclusion and provide a
personal interpretation of the results. Identify the limitations of the study and it
is inevitable that the research will not be perfect.
Conclusion: It means a logical outcome implied from the previous work. Do not report new
findings. Discuss how the results may be of interest to policy-makers and
avenues for further work. It does not have to be long.
References: Only contain those references that you have cited or were particularly relevant.
You do not have to read everything from cover to cover to include it.
(Bibliographies include lists of publications that are relevant to the subject.)
Appendices: for information that is important to the argument, or that would interrupt the
flow of the text unnecessarily, detailed eduations, example of the
questionnaire, copies of important correspondence etc.

18. Writing the report
*
It is important that you pitch the project at the correct level when writing up. Do not assume
readers know a great deal about the topic already. Show that you have done the preliminary
work and include all important steps (Chapter 17).
Putting the first words on paper is hard and therefore it is essential to begin writing early and
continue writing as the research develops.
*
Practical considerations:
• Follow the guidelines laid down by your institution
• There is no single correct style that you must adopt, usually write-up is in the third
person, with a formal structure but there are 3 narrative conventions: realist (third
person), confessional (highly personalized, detail on fieldwork) and impressionist
(reader experiences aspects like the researcher did).
• Language should be understandable and readable. Avoid long centences,
nominalization (substitution of verbs with nouns), Words like ‘very’ and ‘much’,
question marks and exclamations.
• Acknowledge your sources as a matter of etiquette, ascribing to others the credit for
their efforts. It is only plagiarism if it is unacknowledged but watch out for copyright
policies.
• Editing and presentation, use word processors and check logical page classification,
use diagrams and tables to replace words.
*
Your work will never be perfect in your own eyes!
17. Illustrating the report
*
Illustrations in a geographical dissertation could include any or all of: maps, diagrams, graphs,
charts, photographs, satellite images or aerial photographs. Whatever their specific purpose,
illustrations are normally used in a general sense to communicate ideas and to summarise in a
small amount of space on the page what would otherwise be difficult to explain by using text.
Use text only to refer and interpret the information.
The interpretation of map features are very critical when using maps in your dissertation.
Maps can serve to generate ideas and hypotheses, and they allow us to explore spatial
variations and relationships.
*
Do not leave finding maps to the very end! You should give some time and thought to the
type and number of illustrations from both primary and secondary sources and try to assemble

these as you go along. Cartography greatly facilitates the process of illustrating a piece of
work, helps to provide a professionally finished end-product and can also save time.
*
Two main caregories for sources of maps: existing maps, and wholly new maps created by the
author.
Existing maps can give a professional result if you redraw or digitize them in the features that
are most relevant to your application, to generalize and add information and to keep all your
illustrations in the same ‘house style’. Acknowledge clearly the sources of all illustrative
material.
*
Creating illustrations take numerous detailed functions of individual cartographic or mapping
software packages. It is important to consider at the outset which type of software might best
meet your needs. An attractive finished product has a good scale (whole field of work), all the
needed features and a good map design. The best maps and illustrations are often the simplest.
Principles of map and illustration design:
• The subject should always be placed on the same page close to the illustration.
• The frame should comprise a bold line delimiting the map from surrounding text. The
title, scale and north arrow can be fitted in, around the subject, where space is
available within the frame.
• Shading can be used to emphasize the idea of contrast. ‘Natural’ breaks in the
distribution can also be used to create the mapped categories. (Choropleth maps)
• If a photograph: make sure the date on which the image was captured is also included.
• The text should always include the figure number.
• You should have a separate list of figures and tables at the start of the dissertation.
• Illustrations should be positioned as close to, and after, their first mention in the text.
20. Conclusion
*
Your institution requires a copy, just like people that have helped. If your results have policy
implications, and if you feel your work is good enough to withstand critical scrutiny, it may
be worth passing on to a community group, local planners or the press. The abstract or
executive summary might then be the most important part (except for the title).
Think about what you have learned:
• You will have learnt a lot about whatever you have been studying.
• You will have learned much about methods for research.
• You will have learnt about conducting a research project.
• You will have learned something about yourself.
Clearly the best project is one which comes up with interesting findings and fulfils your
original goals. If it does not, however, it is not a complete failure. You may learn as much
from failure as from success.

Potential employers see more relevance in successful completion of research projects than in
examination questions so it can be relevant to the job market or is a good topic for discussion
at job interviews.

Methods in human_geography

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Methods in human_geography

Similar to Methods in human_geography (20)

More from Lisa Schmidt

More from Lisa Schmidt (20)

Recently uploaded

Recently uploaded (20)

Methods in human_geography