International Studies - Post Midterm Flashcards
(104 cards)
1
Q
Qualitative Research (4 main points)
A
concerned primarily with words and images Usually inductive Tends to be interpretivist Constructionist Takes naturalist perspective
2
Q
naturalist perspective
A
When doing research, the social world should be left as undisturbed as possible
3
Q
Constructionist
A
Social life not seen as fixed, but constructed from interactions and negotiations
4
Q
interpretivist
A
Concerned with finding out what an action or event means to those involved
5
Q
inductive
A
Starts with field research ⇒ concepts and theories
6
Q
Steps in Qualitative Research (eight)
A
- Establish a general research question
- Select a relevant site and subjects: Where is the research being conducted and who are the research subjects?
- Collect the data–determine which methods to use
- Interpret the data–determine meanings that research subject put to activities
- Conceptual and theoretical work–evaluate the data related to your research question
- Tighter specification of the research question and
- Collection of further data
- Writing up and findings/conclusions–the researcher must prove the credibility of the research and why it matters
7
Q
Theories used in Qualitative research
A
often grounded theory
- Use of data to develop theories
- May involve an iterative process: going back and forth from data to theory
Begin with abroad definition of a concept and narrow it down through the research process
8
Q
Definitive concepts
A
defined with nominal and operational definitions, as in quantitative research
9
Q
Sensitizing concepts
A
provide only general sense of reference and guidance as to the content of the concept
10
Q
Difficulties in qualitative research
A
achieving external reliability due to ever changing circumstances
achieving external validity due to small sample sizes
11
Q
Methods in qualitative research
A
Ethnography
12
Q
External Validity
A
Degree to which a study can be replicated
Can the findings be generalized across social settings?
13
Q
Internal Validity
A
Do different observers see the same things?
Is there a good match between what is observed and the resulting theoretical ideas?
14
Q
Trustworthiness (Lincoln and Guba)
A
Credibility
Transferability
Dependability
Confirmability
15
Q
Credibility
A
- Do the people studied agree with the interpretation of their thoughts and actions offered by the researcher?
- Conducted through respondent (member) validation
16
Q
Transferability
A
- Can findings be applied to contexts/people not studied?
- ’Thick’ description tells us if transferability is possible
17
Q
Dependability
A
- Were proper procedures followed?
- Can the study’s theoretical inferences can be justified?
18
Q
Confirmability
A
Was the researcher objective and unbiased?
19
Q
Authenticity
A
speaks to integrity and quality of qualitative research process (constructivist)
- Fairness
- Ontological authenticity
- Educative authenticity
- Catalytic authenticity
- Tactical authenticity
20
Q
Fairness
A
Evenhanded representation of all viewpoints
21
Q
Ontological authenticity
A
- Construction and reconstruction of a person’s perspective as it becomes more sophisticated
- Consciousness-raising of sociopolitical, economic, cultural contexts
22
Q
Educative authenticity
A
Increased understanding of and respect for the values of others & how these values frame their perspectives
23
Q
Catalytic authenticity
A
Research process my facilitate, stimulate, or evoke action
24
Q
Tactical authenticity
A
Holds inquirer (constructivist word for ’researcher’) and the research process to standard of effective (from points-of-view of stakeholders) change
25
Goals of Qualitative Research
1. Empathy / empathetic understanding
2. Comprehensive description and emphasis on context
- Behavior can ’make sense’ when context is described
- Behavioral observation in its own environment 3. Emphasis on process
- How do events and patterns form over time
- Long time ’in the field’ allows researcher to understand change and its context
4. Flexibility and limited structure
- Questions tend to be quite general
- Early on little theory driving research
- Topics explored may change during research process
26
Critiques of Qualitative Research
Can be too impressionistic and subjective
Bias can result from personal relationships built during research
Difficult to replicate - Reactive effect can be expected
Problems of generalization: is that always a bad thing?
Often a lack of transparency
27
Participant observation
refers to the observational
| component of ethnographic work
28
How do we observe behaviour?
in an unstructured way, and often in-depth, unstructured discussions and interviews are held with the people studied.
29
Two dimensions of interview/ observation access
Nature of Disclosure, and
| Setting
30
Two Natures of Disclosure
’Overt’: the people being studied know they are being observed by a researcher
’Covert’: the people being studied do not know they are being observed by a researcher
31
Two types of settings in interviews/research
’Open’/public settings: e.g., public parks, downtown sidewalks, etc.
- May be difficult to make observations and talk to people
’Closed’: i.e., private or restricted settings: meetings of private clubs, social movement organizational centres, etc.
32
Drawbacks to having key informants
- Researcher may ignore other group members
- Key informant’s view may not be representative of the group
as a whole
33
4 roles for Ethnographers
Complete participation
Participant-as-observer
Observer-as-participant
Complete observer
34
Complete participation (ethnography)
- Covert operations
- Researcher adopts a secret role in the group
- Method gets closest to participants and their activities
* * Risk of over-identification or developing a strong dislike of the participants
* * Involvement
35
Participant-as-observer (ethnography)
- Researcher adopts a role in the group
- Participants aware who the researcher is
* * Risk of reactivity–subjects’ behaviour changes b/c they know they are being studied
36
Observer-as-participant (ethnography)
- Researcher observes and interviews from the periphery
| * * Risk of reactivity & incorrect interpretation of activity
37
Complete observer (ethnography)
- Researcher does not engage the participants at all
- No risk of reactivity
- Limited info for understanding actions of participants
* * Detachment
38
Types of Field Notes
Mental notes
* Remember and write later
Jotted notes (scratch or rough notes)
* Brief notes made at time to jog memory when writing detailed notes later
Full field notes
* Detailed notes of what was seen, heard and reflection on situations
* Include comments on possible reactivity Analytic memos
* Link observations to concepts & theories that may apply
* Notes on data not data notes
39
Visual Ethnography (two approaches)
Realist
- the material depicted treated as ’factual’, objective reality
Reflective
- consideration of how the researcher influenced what the materials reveal
e.g., the researcher’s ideological views may have affected what was depicted and how it was depicted
- this approach recognizes that the visuals may be ’collaborative’, and may be subject to multiple interpretations by different people
40
Difficulties with Using Visual Ethnography
Stem from interpretation
- Context of when, where, how, and by whom the visual material was taken
- Different meanings may be ascribed to the visual material by the researcher and by different participants
- Potential for researcher to influence the perception of the subject
41
How do qualitative researchers decide when to bring their research to a close?
Because there are no theories to be tested, research and development of theory can go on indefinitely
- One place to stop is when categories are saturated
- New data does not result in new concepts or categories, or in an elaboration of existing concepts or categories
42
Types of Probability Sample
Simple Random Sample (SRS)
Stratified Random Sampling
Multi-stage Cluster Sampling
43
Element or unit
a single case in the population
44
Population
all cases of interest to researcher
45
Sampling frame
the list of elements from which sample
| will be selected
46
Sample
the elements (subset of a population) selected for investigation
47
Representative sample
sample that contains same essential characteristics as the population
48
Probability sample
a sample selected using a random process so that each element in the population has a known likelihood of being selected
49
Non-probability sample
a sample selected using a non-random method
50
Sampling error
estimation error that occurs because of differences between the characteristics of the sample and those of the population
51
Non-response
when an element selected for the sample does not supply the required data
52
Census
data that comes from information from all elements in the population
53
Sampling Bias is the result of ..
1. Not using a random method to pick the sample
2. Sampling frame is inadequate:
3. Non-response:
54
Inadequate sampling frame
If human judgement selects one group over another
| ** Example: IRC selection of sites in Democratic Republic of Congo (DRC)
55
Non-response sample frame
Some people in the sample fail to participate which skews the data.
56
Non-responsive sampling issue
this is a concern if those not responding are systematically different from those responding
- That is, non-responses being non-random is a problem and will bias our results
57
Population heterogeneity sampling issue
the more varied the population, the bigger the sample size you need
58
Simple Random Sample (SRS)
- Each element has the same probability of being selected - Each combination of elements has the same probability of being selected
59
How do you select an SRS (4 steps)
1. Devise a sampling frame: a list of all elements in the population
2. Number all the elements consecutively, starting at some number
3. Pick a sample of size (n) from the total population (N)
4. Use the R function, sample() to generate your random sample
60
Sampling Ratio
n/N
n = sample size
N = population size
61
Stratified Random Sampling
Ensures that subgroups in the population are proportionally represented in the sample - Not a guarantee using SRS
** reduces sampling error
62
Why use Stratified Random Sampling? (3 reasons)
- To ensure low sampling error for each stratum:
- Using this procedure ensures that we can say something conclusive about each country.
* * This will not give us a sample that is representative of the population as a whole.
63
How do you ensure low sampling error by using Stratified Random Sampling?
1 .Stratify the population, i.e., divide it into strata (countries, regions, etc)
2. Select a simple random sample from each stratum of an appropriately large size (say 400 for each country).
64
Multi-stage Cluster Sampling
- Used for large populations
- No adequate sampling frame
- Elements are geographically dispersed
- It involves two or more stages
- Selecting clusters (groups of elements), then selecting subunits within clusters
65
Types of Non-probability Sampling
1. Convenience sampling
2. Snowball sampling
3. Quota sampling
66
Convenience (non-probability) sampling
- Cases are included because they are readily available.
- Useful for pilot studies, for testing the reliability of measures to be used in a larger study, for developing ideas, etc.
* * Problem: can’t confidently generalize to larger population
67
Snowball (non-probability) sampling
- Form of convenience sampling
- Researcher makes contact with some individuals, who in turn provide contacts for other participants, and so on
* * no two snowballs alike
68
Quota (non-probability) sampling
- Collecting a specified number of cases in particular categories to match the proportion of cases in that category in the population
- Quotas for people in groups: age, gender, ethnicity, etc.
69
Four sources of error in research
1. Sampling error
2. Sampling-related error
3. Data collection error
4. Data processing error
70
What is done prior to data collection in quantitative analysis?
- Data analysis decisions are made
| - Must be fully aware of what analysis techniques will be used
71
Why should the data analysis decisions/techniques be known before designing questionnaires, observation schedules, & coding frames?
- Statistical techniques used depend on how a variable is measured
- Size and nature of the sample may make some techniques unsuitable suitable
- But, more importantly, the types of variables matter for analysis
72
Types of Variables in quantitative analysis (categorizing variables)
1. Nominal
2. Ordinal
3. Interval/Ratio
73
Nominal variables in quantitative analysis
observations are different b/c they are in one or another category
- Categories can not be ordered by rank
- Cannot do arithmetic or mathematical operations with the categories
ex: "mean(region)" would not work, & "table(region)" would work with a nominal variable.
74
Ordinal variables in quantitative analysis
categories of the variable can be rank ordered
- Distance or amount of difference between categories may not be equal
- Cannot do arithmetic or mathematical operations with the categories
75
Interval/Ratio variables in quantitative analysis
Distance or amount of difference between categories is uniform
- Can do arithmetic and mathematical operations with the categories
- Ratio variables have a ‘0’ start position
76
Univariate Analysis
Analysis of one variable at a time
77
Nominal variables in quantitative analysis
observations are different b/c they are in one or another category
- Categories can not be ordered by rank
- Cannot do arithmetic or mathematical operations with the categories
* * ex: "mean(region)" would not work, & "table(region)" would work with a nominal variable.
78
Univariate Analysis
Analysis of one variable at a time
- frequency tables
** ex: "table(df)" or "prop.table(df)"
interval/ration variables may be combined but needs a theoretical or empirical reason to do so (as you lose info)
79
Measures of Central Tendency
average, or typical, value for a group
80
'Mode' measure of Central Tendency
score that shows up most in a particular category (“modal” response)
81
'Median' measure of Central Tendency
the exact middle score when all scores lined up in order
| - If even number of observations, then median is mean of the 2 middle scores
82
'Mean' measure of Central Tendency
sum of all scores, divided by total number of observations
- Is sensitive to ’outliers’ (see graph on following page)
- Can be ’information-poor’
83
Measures of Dispersion
The amount of variation in a sample
84
'Range' measure of Dispersion
Highest score minus lowest score
| - Shows the influence of outliers
85
'Standard deviation' measure of Dispersion
Measures the amount of variation around the mean
| - Influenced by outliers
86
Bivariate Analysis
Determines whether there is a relationship between two variables
Note: determination of a relationship is not proof of causality
87
Coefficient of Correlation
- Use with two interval/ratio variables
- Values from 0 (indicates no relationship)
- to +1 (indicates perfect positive relationship)
- or -1 (indicates perfect negative relationship)
88
Bivariate Associations of non-interval Variables
whenever at least one of the variables in a bivariate relationship is non-interval (i.e., nominal or ordinal), use a contingency table to determine association.
89
A Null hypotheses
Tests the significance of the bivariate association.
- If the null is correct there is no relationship.
- If the null is rejected and the statistical significance (p) of the findings are ≤ .05 there is indirect support for the research hypothesis
90
Chi-square (χ2)
- Used with contingency tables
- Measures the likelihood that a relationship between the two
variables exists in the population
- Calculated by comparing the observed frequency in each cell with what would be expected by chance (if there were no relationship between the variables)
- The chi-square value is affected by the sample size
91
Multivariate Analysis (elaboration)
- Examines the relationship between three or more variables.
- Can be used to test for spuriousness.
- Can be used to test for intervening variables
- Can be used to test for interactions
- Is used in multiple linear regression
92
What is spuriousness?
- Spuriousness exists if two variables are correlated but only through a third variable
- In a spurious relationship, an antecedent third variable is producing the variation in the two variables of interest.
93
What is an intervening variable?
X ⇒ Y (intervening variable?) ⇒ Z
- If the possible intervening variable is controlled, and the relationship between X and Z disappears, then Y is considered an intervening variable.
* * e.g., Education ⇒ Income ⇒ Happiness
94
What is an 'interaction' between variables?
An interaction exists if the effect of one independent
variable varies at different levels to that of a second
independent variable
- The two vary independent of each other
- There is mild connection
- One independent variable moderates (has some impact on) the relationship between the other independent variable and its dependent variable
** e.g., the effect of age on having another source of exercise is different for men and women
95
Multiple linear regression
- determines how much of the variation in the dependent variable is explained (predicted) by the independent variables
- determines which, if any, of the independent variables is a significant predictor of the dependent variable.
96
Open Coding
- Identifies initial concepts that will be categorized together later
- Close to the data
97
Axial Coding
- Data are reviewed for linkages and re-organized according to those connections
- New codes may be developed
98
Selective Coding
- Selection of the core category / categories
- Validating the relationships
- Identifying gaps that need to be filled in
- Conceptualizing the phenomenon (emerging theory)
99
Substantive Theory
Observed patterns are related to each other and a theory is developed to explain the connections in that setting
100
Formal Theory
- Theory applied at a higher level
- Requires data collection in different settings
- Applicable to a variety of settings
101
Multi-Strategy Research (two approaches)
Triangulation, Facilitation & Complementarity
102
Triangulation in Multi-Strategy Research
- The use of quantitative research to corroborate qualitative findings, and vice versa (e.g., several observers, theoretical principles, sources of data and methodologies)
- Triangulation can also take place within a research strategy
103
Facilitation in Multi-Strategy Research
The use of one research strategy to assist with research that uses a different strategy
- May occur in a number of ways:
* providing hypotheses (often by using qualitative data to generate hypotheses for quantitative research);
* aiding measurement (often by using qualitative data to design questionnaire items);
* providing research participants: Quantitative studies may identify people who could participate in qualitative research.
104
Complimentary in Multi-Strategy Research
- Using two or more approaches where a single approach would not be sufficient
* Can be used to illustrate static and process features of a
social phenomenon
* Can be used to test the generality of findings encountered in qualitative research by using quantitative data gathered from random samples
- Qualitative data can be used to interpret the relationship between variables found in quantitative research
- Different methods can be used to study different aspects of a phenomenon.
