1
Q
Common components of a theory
A
- definitions: what are key terms?
- descriptions: characteristics?
- relational statements; deterministic (variables are related), probabilistic (the relationship isn’t inevitable)
2
Q
middle range theories
A
- limited in scope
- testable
- ex. Durkheim’s theory of suicide
3
Q
Grand theories
A
- general and abstract
- provide ways to look at the world
- ex. structural functionalism
4
Q
deductive method
A
- most common approach to social research
- begins with a theory
- understand specific phenomenon through background research
- develop hypothesis
- test with empirical data
- revise if necessary
5
Q
inductive method
A
- theories and interpretations are the outcomes of the theory
- gather and examine data first
- create theory from observations
6
Q
grounded theory
A
- deriving theory from observations
- used by qualitative researchers
7
Q
epistemology
A
- how do we know the world?
- how does knowledge become acquired?
8
Q
positivism
A
- follows the natural sciences
- uses the principle of empiricism
- generate hypothesis to test (deduction)
- can provide foundation for induction too
- science is value-free (intersubjectivity)
- scientific statements are of key importance
9
Q
interpretivism
A
- critique on positivism
- goal is to grasp the subjective meaning of people’s lives
- people interpret the reality of their own lives
- views the social world from the POV of the social actor
10
Q
symbolic interactionism
A
- major perspective in soc that uses interpretism
- ex. George herbert mead, Irving Goffman
11
Q
critical approaches/theory
A
- critiques of positivism
- use both inductive and deductive methods
- reject “value-free” science
- anti-oppressive in practice and political in nature (Karl Marx and conflict theory)
- involves praxis: putting one’s theoretical positions into practice
12
Q
ontological considerations
A
- what is considered real
- ontological assumptions about reality effect: research question formulation; the way research is carried out
13
Q
ontological debate 1: objectivist perspective
A
- social phenomena have an objective reality, independent of our perceptions
14
Q
ontological debate 2: constructionist perspective
A
- constructionist hard: reality is merely a set of mental constructions; Nietzsche: there at no facts, only interpretations
- constructionist soft: more middle ground; there is an objective social reality that is marred by human interpretation
15
Q
quantitative research
A
- uses numbers and stats in the collection and analysis of data
- surveys, demographics
16
Q
qualitative research
A
- uses mainly words and other non-numeric symbols in the collection and analysis of data
- concerned with things that can’t be quantified
- motivations, reasonings, beliefs, understandings and how that shapes what they do
17
Q
values: a researcher’s values can contribute to bias in research
A
- choice of topic
- formulation of research question
- choice of method
- formulation of research design and data collection methods
- actual data collection
- analysis of data
- interpretation of data
- conclucions
18
Q
values: reflexivity
A
- researchers’ awareness that their values and decisions have an impact on the research
- personal biases are made explicit
19
Q
values: 3 different positions on values in social research
A
- research should be value-free
- research cannot be value-free, but researchers should be open and explicit about their values
- researchers should use their values to direct and interpret their investigations; value commitment is a good thing for researchers to have
20
Q
politics in social researchers
A
- researchers sometimes “take sides”
- funding: who gets it? are there strings attached? govt funding as strategic
- research subjects/participants: gatekeepers; who gets access?
- research findings: what sorts of findings are “acceptable” to those who fund or publish research?
21
Q
research questions: key characteristics
A
- as clear as possible so it is understandable to others
- be researchable
- relate in some ways to other research
- neither too broad nor too narrow
22
Q
research questions: choice of research orientations, design and method must match question
A
- is it a brand new phenomenon?
- world views?
- measuring impact
- hypothesis testing?
23
Q
research question
A
- states the purpose of the study in the form of a question
- qualitative: less specific research question; inductive; no hypothesis
- quantitative: can test causal models; deductive; narrowed research question to make a testable hypothesis
24
Q
research design
A
- a framework for the collection and analysis of data
- ask: what do I want to learn? what is the nature of research question? what kind of explanation will I want? (typically nomothetic and idiographic)
25
nomothetic explanations
- involve attributions of cause and effect, expressed in terms of general laws and principles
- typically quantitative
- 3 criteria of causation: correlation, time order, non-spuriousness
26
idiographic explanations
- involve a rich description of a person or group and seek to explain the particular
- typically quite limited
- typically qualitative
- it is not meant to apply to persons or groups who were not part of the study
27
select method
- questionnaire
- structured interview and semi-structured
- participant observation
- ethnography
- experiments
28
experimental design
- true experiments are common in psych and organizational studies but rare in soc and pols
- many variables of interest are not subject to experimental manipulation
- ethical concerns preclude performing experiments
- many phenomena of interest have long-term, complex causes that cannot be simulated in experiments
- even where applicable, experimental models do not get at the perceptions and feelings of research subjects
29
field experiments
- conducted in real-life surroundings
| - difficult to set up due to ethical concerns
30
lab experiments
- take place in artificial environments
- controls research experiment
- easier to randomly assign research subjects; enhanced internal validity
- easier to replicate
- weak external validity
31
key concepts to relevant experiments
- experimental or treatment group
- control group
- random assignment
- pre-test
- post-test
- causality underpins different types of research design
- expressed by variables
- IV are manipulated to see if they have impact on DV
32
classic experimental design
1. IV and DV are identified
2. the DV is observed/measured in each of the control and treatment groups and recorded at T1
3. treatment group receives while control group is left alone
4. DV measured in post-test and recoded as occurring at T2
5. any changes in each group are noted (ideally will only occur in treatment group)
33
internal validity
- validity in social research assesses the extent to which a research study addresses the issue that the research is intended to explore
- concerned with extent to which any given research design is a good test of the hypothesis under consideration
- third variable problem
34
Cook and Campbell (1979) threats to internal validity when lacking random assignment or control group
1. history: some event occurring after the treatment was given may have influenced the DV
2. testing: the pre-test may have influenced the DV
3. instrumentation: changes in the way a test is administered may account for pre/post-test differences
4. mortality: participants leave the experiment before it is over
5. maturation: participants change over time
6. selection: post-test differences between the control and experimental groups may have been caused by pre-existing differences
35
measurement validity (construct validity)
- "are you measuring what you want to measure?"
36
external validity: 2 primary causes
1. are the findings applicable to situations outside the research environment
2. can the findings be generalized beyond the people or cases studied?
37
Cook and Campbell (1979) threats to external validity in experimental research
1. the representativeness of the study participants
2. effects of the setting
3. effect of pre-testing
4. reactive effects of experimental arrangements
38
replicability
- the results remain the same when others repeat all or part of a study
- the procedures used to conduct the research are sound and spelled out
- very important
- currently in applicability crisis in research
39
the laboratory experiment
- greater control over environment is an asset
- easier to assign participants randomly to conditions
- limitations: low external validity; life in a test tube?
40
quasi-experiments
- differ from true experiments in that internal validity is harder to establish
- less control over variables
- ex. natural experiments --> covid pandemic
41
cross-sectional design
- no before or after comparisons
- do not include manipulation of the IV
- ex. questionnaires, structured interviews/observations
- 2 or more variables are measured in order to detect patterns of associations
- issues with internal validity and establishing direction of causation
- issues with external validity: random method should be used
- can examine the effect of variables that cannot be manipulated in experiments
- tend to be quantitative
42
longitudinal design
- cases are examined at a particular time (T1) and again at a later time(s) T2, T3, etc
- provide info about the time order of changes in certain variables
- helps to establish the direction of causation
43
longitudinal design: 2 basic types
1. panel study: the same people, households, orgs. etc are studied at different times
2. cohort study: people sharing the same experience are studied at different times, but different people may be studied at each time
44
longitudinal design: drawbacks
- attrition over time
- may be difficult to determine when subsequent waves should be conducted
- panel conditioning: people's attitudes/behaviours may change as a result of participation in a panel
45
case study design
- basic one involves in-depth study of a single case
- a single case can be a person, family, org, event, country etc
- qualitative and/or quantitative methods
- tends to be inductive and start with qualitative approach
- achieving external validity is not main reason for case study
46
case study design: 3 types
1. critical case: illustrates the conditions under which a certain hypothesis holds or does not hold
2. extreme (or unique) case: illustrates unusual cases which help in understanding the more common ones
3. revelatory case: examines a case or context never before studied
47
research ethics
- need to be addressed in the initial stages of a study and kept in mind in every phase
- first priority of a researcher: ensure participants are not being harmed
- participant safety > knowledge
- researchers must constantly balance between potential gain and risk of harm
- SSHRC, CIHR, NSERC
48
research ethics board (REB)
- policy to due to inadvertent harm caused to participants
- all Canadian research requires REB approval FIRST; may require modifications, appeal process
- conflict of interests roles
- can be found in many institutions comprised of people from various interest groups
49
ethics approval in quantitative research
- considered by some to be easier to obtain
- stated hypothesis and specific plan for testing
- some REB favour quantitative work because it's considered to be more scientific
50
ethics approval in qualitative research
- flexibility for emerging themes means indeterminate methods
- may capture data on people that would not want their activities observed
- a cautious REB can: restrict project, prevent funding, prevent project
51
general ethics principles
- TCPS2 provides for variation in research methods
- 3 core overlapping principles: respect for persons, concern for welfare, justice
- consent form
- front end deception
- debriefing
- covert research is very intrusive
- duty to report
52
ethics information sheet outline
- research project
- methods
- potential risks/benefits
- names of those in charge
- assurances of confidentiality
- funding
- compensation
- how data will be stored
- institution and contact
- publishing info
53
quantitative research
- epistemological and ontological assumption
- the collection of numerical data
- a deductive relationship between theory and research
- a preference for the natural science approach to research (positivism)
- an objectivist conception of social reality
- major concern: measurement validity
54
quantitative research: the main steps
1. theory
2. hypothesis
3. research design
4. devise measures of concepts
5. select research site(s)
6. select research participants
7. administer research instruments/collect data
8. process data
9. analyze data
10. finding/conclusion
11. write up findings/conclusions
55
operationalization
- the process of converting concepts into indicators or into specific questions in a questionnaire or an interview
- you need questions that allow you to measure the phenomenon you are trying to study
56
concepts
- the ideas or mental representations of things
- may be independent (manipulated) or dependent (outcome) variables
- measurement allows a delineation of fine difference between people/issues we are interested in
- what is the correlation?
57
concepts: 2 types of definition
1. nominal: describes the concept in words, much like a dictionary definition
2. operational: describes how the concept is to be measured
58
indicators
- tell us that there may be a link and indicate how strong that link may be
- usually one indicator for each concept is adequate
- advantageous to use more than one indicator of a concept
- can be direct or indirect
59
benefits of using multiple indicators to measure a concept
- reduces likelihood of misclassifying some people because the language of question is misunderstood
- endures definition of underlying concept is understood correctly
- gets access to a wider range of issues related to concept
- allows for factor and cluster analysis
- helps to weed out response sets
60
Likert scale
- very common way of collecting data on peoples opinions
- asked to give their level of agreement with statements
- items must be statements
- must all relate to the same subject
- options should be interrelated
61
coding unstructured data
- derive codes: labels or titles given to the themes or categories
- assign numbers to codes
- basic principles to observe: categories must be exhaustive and not overlap; must be clear rules for how codes are applied
- very unreliable and little validity was of coding data
62
reliability: stability over time
- whether the results of a measure fluctuate as time progresses, assuming that what's being measured isn't changing
- can be measured using the test-retest method
- difficult to quantify because of factors in passage of time
63
reliability: internal reliability
- whether multiple measure that are administered in one sitting are consistent
- can be measured using Cronbach's alpha coefficient or the split-half method
64
reliability: inter-observer consistency
- all observers should classify behaviour or attitudes in the same way
65
face validity
- established if at first glance the measure appears to be valid
66
concurrent validity:
- established if the measure correlates with some criterion thought to be relevant to the concept
- a lack of correlation brings some doubt to the validity of the original measure
67
construct validity
- established if the concepts relate to each other in a way that is consistent with the researcher's theory
- confirmed by seeing that the results match what would be predicted given the theory
68
convergent validity
- established if a measure of a concept correlates with a second measure of the concept that uses a different measurement technique
- a measure that is not reliable will not be valid
- a measure may be invalid but still reliable
69
main goals of quantitative research
- measurement
- establish causality (internal validity)
- generalization of findings to those not studied
- replication
70
critiques of quantitative research
- researchers fail to distinguish people and social institutions from "the world of nature"
- the measurement process produces an artificial and false sense of precision and accuracy
- the reliance on instruments and procedure produces a disjuncture between research and everyday life
- the analysis of relationships between variables ignores peoples everyday experiences and how they are defined and interpreted
- explanations for findings may not address the perceptions of the people to whom the findings purportedly pertain
- quantitative researchers tend to assume an objectivist ontology - social order is fixed
71
open questions: advantages
- allow for replies that researcher may not have contemplated
- make it possible to tap the participants' unprompted knowledge
- salience of particular issues for respondents can be examined
- can generate fixed choice format answers
72
open questions: disadvantages
- more time consuming
- answers must be coded
- less convenient to compose an answer
- may require transcribing
- face inter-interviewer variability
73
closed questions: advantages
- minimize intra and inter-interviewer variability
- may make it easier to understand question because answers are provided
- can be answered quickly and easily, reduces response rate issues
74
closed questions: disadvantages
- loss of spontaneity and authenticity because relevant answers may be excluded from choices provided
- respondents may differ in the interpretation of the wording of fixed responses
- respondents may not find a fixed response that they feel applies to them
75
prominent sources of error in survey research
- poorly worded questions
- interviewer error in asking a question
- misunderstanding on part of the interviewee
- interviewee lapses in memory
- interviewer error in recording info
- mistakes in entering data into computer
- baises related to characteristics of interviewers and/or interviewee
76
types of questions
- personal factual questions - demographic
- factual questions about others
- factual questions about entity or event
- questions about attitudes, beliefs, knowledge
77
rules for designing questions
- keep the research question in mind
- focus on exactly what you want to know
- put yourself in the position of the respondent
- consistent wording and balanced questions
- don't overstretch respondents memories
- question order
- run a pilot study
- using existing questions can be beneficial
78
what to avoid when designing questions
- avoid: ambiguous terms, long questions, double-barreled questions, very general questions, leading questions, 2 in one questions, negative/double-negatives
- minimize technical terms
- carefully consider "don't know" answers
- avoid provoking response set
79
vignette question
- presenting people with one or more scenarios and asking them how they would respond
- anchor the choices in realistic situations
- weakness: how people say they would act in a situation can be different from how they will actually act
- can provide useful info or at least a starting point for further research
80
telephone interviews: strengths
- good for national or govt research
- cheaper and quicker to administer
- easier to supervise and therefore reduce interviewer errors upfront
- reduced bias arising from "interviewer effect"
81
telephone interviews: disadvantages
- may exclude people without phones, unlisted numbers, cell phones and hearing impaired
- hard to sustain for long period of time
- cannot collect additional information
- difficult to be sure targetted respondent is actually one answering
- visual aids cannot be used to assist
82
online interviews: advantages
- quality of face-to-face interviewing with efficiency and economy of the internet
- opportunity for more considered, fuller responses
- ease of contact for follow up
- no need to transcribe
83
online interviews: disadvantages
- fairly high drop out rate, but may be overcome by developing mutual trust
- answers may be short on detail
84
conducting interviews
- know the schedule
- prepare the introduction
- establish the rapport
- introductory statement
85
questionnaires
- essentially structured interviews without an interviewer
- involve filling out a form
- has to be very clear and easy to follow
- have fewer open questions, have short simple designs
- online surveys popularity is growing
- email surveys
- web surveys
86
questionnaires: advantages
- cheaper, quicker and more efficient to administer
- no interviewer effects
- social desirability bias seems to be reduced
87
questionnaires: disadvantages
- researcher cannot explain the question
- greater risk of missing data because of a lack of probing or supervision
- difficult to ask a lot of questions and open questions
- cannot track whether the questions were answered in original order = order effects may occur
- not appropriate for certain respondents
- designated respondent may not have completed it
88
researcher driven diaries
- form of questionnaire
- participants record their feelings, perceptions, actions etc on a form shortly after they occur
- participants should be given explicit instructions on how to complete, the time periods for recording responses, types of experiences to be recorded
- used for qualitative and quantitative
89
secondary analysis of survey data
- large amounts of quantitative data already exist
- raw stat data available on many topics
- survey data is needed as a check and counterpoint to official stats
90
secondary analysis of survey data: advantages
- cost and time
- high-quality data
- opportunity for longitudinal analysis
- more time for data analysis
- fulfill wider obligations of social research
- opportunities for cross-cultural studies
- subgroup analysis
- re-analysis
91
secondary analysis of survey data: disadvantages
- lack of familiarity with someone else's data
- data sets can be complex
- ecological fallacy: data gathered by region or neighbourhood are used to make statements about individuals
- no control over data quality
- absence of key variables
92
element or unit
- a single case in a population; in social science, its usually a person; can sample other things (nations, regions, schools etc)
93
population
- all the cases about which you are seeking knowledge or all the cases to which your conclusions might apply
94
sampling frame
- the list of elements from which the sample will be selected
95
sample
- the elements selected for investigation; a subset of the population - may involve probability or non-probability
96
representative sample
- a sample that is a microcosm of the population; one that represents its characteristics
97
calculating a representative sample
- sample size doesn't change much for pops > 2000
- margin of error between 1-5%
- confidence level 99-95%
98
probability sample
- selected using a random process, such that each unit in the sampling frame has a known chance of being selected; goal is to minimize sampling error
99
non-probability sample
- a sample selected using a non-random method
100
sampling error
- error of estimation that occurs if there is a difference between the characteristics of the sample and those of the pop from which it was selected
- virtually impossible to eliminate
101
non-response
- a situation that occurs whenever some unit selected for the sample refuses or doesn't participate
102
census
- data collected from all elements of the population rather than from a sample
103
3 sources of bias in sampling
1. not using a random method to pick sample
2. the sampling frame
3. non-response
104
process of simple random sample
1. devise a sampling frame
2. number all of the elements consecutively starting at 1
3. pick a sample size (n) from the total pop (N)
4. use a random number table or program to generate a list of random numbers
5. the sample will be comprised of the cases whose element numbers match the randomly generated ones
- sampling ratio = n/N
105
systematic sample
- selected directly from the sampling frame, without using random numbers
- i= size of sampling interval
- to begin choose a number at random from 1 to i
- potential issue is periodicity: when/if the cases in the frame are arranged in some systematic order
106
stratified random sampling
- ensures that subgroups in the population are proportionally represented in the sample
107
multi-stage cluster sampling
- used for large populations
- involves 2 or more stages
- issues: technical complications; cluster samples are usually stratified as well
108
sample size
- the absolute size of the sample matters
- as sample size increases, sample error tends to decrease
- common sample sizes: 100, 400, 900, 1600, 2500
- each size increase cuts the sampling error by 1/2, 1/3, 1/4, 1/5 respectively
- the biggest change occurs between 100 and 400
- often dictated by financial concerns
109
issues with sample size: heterogeneity of the population
- generally the greater the heterogeneity of the pop on the characteristics of interest, the larger the sample size should be
110
issues with sample size: kind of analysis
- the size needed may vary depending on what sort of analysis will be done
- if small groups in the pop are to be compared to larger groups, it may be necessary to oversample the smaller group in order to make meaningful comparisons
- certain stat procedures, such as some multivariate analyses, require large sample sizes to work
111
convenience sampling
- cases are included because they are readily available
- problem: one cannot generalize the results to some larger population with any confidence
- useful for: pilot studies, testing reliability of measures for larger studies, developing ideas, learning how to do research etc.
112
snowball sampling
- form of convenience sampling
| - the researcher makes contact with some individuals, who in turn provide contacts for other participants
113
quota sampling
- collecting a specified number of cases in particular categories to match the proportion of cases in that category in the pop
- random methods are NOT used to fill the quotas
- weaknesses: not likely to be representative, judgements about eligibility may be incorrect, not appropriate for calculating standard error term
114
structured observation and sampling
- often no sample frame
- may involve time and place sampling
- may include behaviour sampling
115
limits to generalization
- even when a sample is selected using probability sampling, any findings can be generalized only to the population from which the sample was taken
- do the findings from an earlier date still apply today?
116
error sources
- sampling error
- sampling-related error
- data collection error
- data processing error
117
virtual sampling issues
- not everyone is online and has the technical ability to handle these kinds of questionnaires
- internet users are a biased sample of the pop
- few sampling frames exist from the general online pop
118
covert research
- very intrusive
- no consent
- benefits must outweigh any harm to participants
- little to no concern over reactivity
- permission if over sought after the fact
- researchers must ensure anonymity of non-consenting participants
SOC 315
flashcards
Decks in class (2)
# Cards
