Geography 237 Geographic Research: Methods and Issues

1
Geography 237Geographic Research Methods and
Issues

• Research Design
• (Babbie, Chapter 4)

• Problems and Questions
• Purpose(s) of Research
• Causal Explanations
• Units of Analysis

2
Research Problem

• "The formulation of a problem is often more
  essential than its solution"! (Einstein)
• quite general
• form of a statement
• e.g., Worldwide, millions of children die each
  year from diarrhea.
• e.g., Some people recycle more than others.
• e.g., People smoke, despite general scientific
  agreement that smoking has serious negative
  health effects.
• e.g., The trucking industry has deregulated.

3
Research Questions

• more specific
• some conceptual distinction
• form of a question
• e.g., Why do children in Grenada not receive
  simple, relatively cheap, treatments for
  diarrhea?
• e.g., Do smokers know about the negative health
  effects of smoking? Do they believe what most
  scientists believe about negative health effects?
  If so, why do they still smoke?
• e.g., What are the impacts of trucking
  deregulation on large, medium and small trucking
  companies?

4
Exercise

• Out line a social research problem and some
  research questions to accompany it.
• State some specific hypotheses.
• e.g. the internet
• e.g. hurricanes
• e.g. other

5
Three Purposes of Research

• 1) Exploration
• topic new to researcher
• new topic entirely
• often idiographic explanations
• often qualitative
• often lack representativeness (see sampling)
• e.g., How do people perceive the risks from
  genetically modified foods?

6
Three Purposes of Research

• 2) Description
• patterns of phenomena
• precise sampling and observation
• often quantitative
• little or no explanation of patterns
• what?, where?, when? and how?
• e.g., census
• e.g., How widespread are genetically modified
  foods?

7
Three Purposes of Research

• 3) Explanation
• why?
• nomothetic or idiographic explanations
• e.g., Why do experts and laypeople seem to
  disagree on the risks posed by genetically
  modified foods?

8
Nomothetic Causal Explanation(in the
probabilistic sense)

• Does A cause B?
• E.g., Does drug addiction cause crime?

9
Three Conditions forNomothetic Causal Explanation

• Correlation
• A is associated with B
• Drug users are often convicted of theft.
• Time Order
• A precedes B
• The thieves were drug users before they became
  thieves.
• Nonspuriousness
• C does not cause A and B
• The drug user thieves are also very poor i.e.,
  being poor correlates with both.

10
Spurious Relationship
11
Unfair Criticisms of Nomothetic Causality

• Complete causation
• A is not the only cause of B
• Exceptional Cases
• B does not always occur when A happens
• Majority of Cases
• majority of A are not B
• more likely is all that matters

12
Types of Causal Variables

• Necessary
• A must be present for B to happen
• e.g., Geog 237 needed for Geog degree
• e.g., female to get pregnant

13
Necessary Cause
14
Types of Causal Variables

• Sufficient
• B happens in presence of A
• other As may also be sufficient
• e.g., 82 gets you into grad school
• best explanatory variables are both
• often explanatory variables are neither (e.g.,
  drug use and crime)

15
Sufficient Cause
16
Units of Analysis

• The level at which you want to make
  generalizations.
• individuals
• e.g., students, Canadians, hispanics
• groups
• e.g., families, households, census tracts
• organizations
• e.g., universities, banks
• social artifacts
• e.g., music, grades
• usually unit individuals

17
Units of Analysis
18
Units of Analysis
10 households 12 families 4/12 33 2/10 20
19
Units of Analysis Exercise

• What is the unit of analysis (see review chapter
  4)?
• This paper examines variation in job title
  structures across work roles. Analyzing 3,173
  job titles in the California civil service system
  in 1985 we investigate how and why lines of work
  vary in the proliferation of job categories that
  differentiate ranks, functions, or particular
  organizational locations (Strang and Baron 1990
  479)

20
Units of Analysis Exercise

• What is the unit of analysis (see review chapter
  4)?
• Women watch tv more than men because they are
  likely to work fewer hours outside the home than
  men....Black people watch an average of
  approximately three-quarters of an hour more
  television per day than white people (Hughes
  1980 290)

21
Units of Analysis Exercise

• What is the unit of analysis (see review chapter
  4)?
• Of the 130 incorporated UW cities with more than
  100,000 inhabitants in 1960, 126 had at least two
  short-term nonproprietary general hospitals
  accredited by the American Hospital Association.
  (Turk 1980 317)

22
Dangers With Units of Analysis

• Ecological Fallacy
• Drawing conclusions about individuals from
  studies done on groups
• E.g., census tracts
• Reductionism
• Explaining complex phenomena using a single,
  narrow set of concepts
• E.g., sociologists vs geographers
• But what about parsimony?

23
Temporal Considerations in Research Design

• Cross-sectional
• snapshot in time
• e.g., surveys
• Longitudinal
• observation at different points in time
• e.g., analysis of multiple censuses
• Longitudinal generally produce more convincing
  findings!

24
Longitudinal Designs

• Trend
• Changes within a population over time
• e.g., national censuses over time
• Cohort
• Follow specific subpopulations over time
• e.g. baby boomers.
• Panel
• Follow the exact same group over time
• e.g., 50 farmers from Oxford county

25
Something seem amiss?

• The COMMIT study incorporated both cohort
• and cross-sectional designs. In the case of
  attitude
• measurement, cohorts of smokers and non-smokers
• identified in the 1988 baseline survey were
  resurveyed
• in 1989, 1991 and 1993 to determine their
• attitudes towards smoking before, during and
  after
• the intervention. In addition, a subset of
  respondents
• to the 1993 final prevalence survey was asked
• the same attitude questions. As a result, two
  approaches to examining community-level changes
• in attitudes were possible (1) the cohort
  approach
• involving the comparison of attitudes across the
• three points in time and (2) the cross-sectional
• approach comparing the 1989 cohort survey data
• with the 1993 final prevalence survey.
• (Taylor et al. 1998 111)

26
Research Process
(Nomothetic) Research Process