Lecture 1
- Social science and social scientists are interconnected with society
→ What is social science research?
- to search for an answer to a question
- to extend knowledge, unravel mysteries, build theories
- verify facts
→ based on a scientific method
science= collecting systemic knowledge that is capable of resulting in a correct prediction or
reliable outcome
scientific method = seek to explain events of nature in a reproducible way and is reliable
(repeatability) and valid (are we measuring what we were supposed to measure?)
1. The goal is inference → generating new knowledge about something we don’t know on
the basis of something we can observe/we know
2. The procedures are public / transmissible / cumulative
3. Conclusions are uncertain / Falsifiable
4. The method!
Pseudo = false
The importance of research methods:
Generally:
• Being able to conduct good scientific research to understand various (safety and security)
issues/problems and find solutions
• Developing a critical eye And
On a practical note:
• Developing skills to understand the (empirical) literature in other classes
• Developing skills for your bachelor/master thesis, but also beyond…
Doing research:
1. Research topic
2. Literature review
3. Research design
4. Data collection
5. Data analysis
6. reporting
Research design
1. Research philosophy. 4 worldviews:
- Positivism: believe that there is a single reality which can be measured and known. A
theory/observation is always true. Use quantitative methods Eg. if u drop a pen it will
fall
- Interpretivism: believe there is no single reality/truth. It needs to be interpreted. Use
qualitative methods to find multiple realities.
, - Pragmatism: believe reality is constantly renegotiated/debated and interpreted and
therefore the best method to use is the one that solves the problem. Combi of two
above
- Critical: paradigm situates research in social justice issues to address political, social
and economic issues. Hte seek to be the voice of the marginalised ones.
● ontology= what constitutes reality and how can we understand its existence ?
● epistemology= what constitutes valid knowledge and how can we obtain it?
● methodology= how do we go about finding it out? {insert table from powerpoint slide}
2. Research approach
Deduction → from theory to confirmation of the theory
Induction → from observations to theory
3. Mono/mixed methods
→ qualitative, quantitative, mixed methods
4. Time horizon
How much time to do research?
5. Data collections
How to collect data? And how to analyse? → Eg. surveys, case study, ethnography
Lecture 2
Methods
Interviews & focus groups: strengths and challenges
Interviewing = a conversation that has a structure… the purpose is to obtain knowledge
- to obtain knowledge: align your topic with the right respondents, the right questions
and the right responses!
→ gain trust first, not too heavy questions
→ strive to gain knowledge of their conditions and believes → could explain certain
actions
Types of interviews:
- the structured interview → close to survey
- the semi-structured interview → open-ended questions and no predefined options for
the respondent
- in-depth interviews → no predefined structure & provide insight on topics that need to
be investigated in a more structured manner & discover new points of view
,Interviews vs focus groups
→ identifying group norms & discovering variety within a group, observe group interaction
→ snowballing of ideas = enables students to initially think about their own response to a
question, then converse with just one other person, and then join together with other pairs to
enable a wider discussion
→ group think & dominant persons
→ more difficult to address sensitive topics
→ difficult to guarantee confidentiality & to organize & to generalize
Case study and comparative method: strengths and challenges
What is a case study & what is it good for?
- Case = a spatially delimited phenomenon observed at a single point in time or over
some period of time
- Case study = intensive study of a single case (for the purpose of understanding a
larger class of similar cases, i.e. a population)
- can be either inductive or deductive
Comparative study = the study of a larger number of cases
→ why do we compare?
- to explain an outcome identify broad patterns/regularities or relationships among
variables, test/refine theory and make predictions
Process tracing method
→ single case research method that can be used to make within case inferences about
presence/absence of causal mechanisms
→ enables strong inferences about causal relationship between X and Y
→ links dependent and independent variable
- one part will lead to action in the second one, which will lead to outcome
- connects causes and consequences
→ focus is on studying causal mechanisms using in-depth single case study
Process tracing tests for causal
inference
→ Straw-in-the-Wind = unusual piece of
evidence
→ Hoop = about the area and place of
the crime
→ Smoking-Gun =
→ Doubly Decisive =
Method of elimination = used by sherlock
holmes → eliminating possible variables
, Experiments
● Module: Sampling
→ a deliberate test of a causal proposition, typically with random assignment to conditions
→ investigators design experiments to evaluate the causal impacts of potentially informative
explanatory variables
What is a cause?
→ something is a cause if changing it - eliminating it, adding it, or varying its level - would
affect the outcome (keeping the rest of the system constant
→ causes are difference makers. A change in the causal factor (independent variable) leads
to a change in the dependent variable (see slide)
Causality and the logic of experiments
→ causality is inferred by comparing two scenarios in which only the
suspected/hypothesized cause changes while everything else remains the same
→ since we have no access to parallel realities, how can we get to scenarios in which (only)
the hypothesized cause changes?
→ in other words, how to have access to factual and counterfactual versions of the world?
What is a treatment/manipulation?
→ manipulations are deliberate interventions developed to create variation in a putative
causal variable
→ in experiments we manipulate the independent variable
How are experiments different from other research designs?
→ experiments work because random assignment to treatment and control conditions tends
to produce groups that are comparable in all respects but one - the variable manipulated by
the researcher
The logic of experiments
- importance of theory and deriving of hypotheses
- addressing the issue of causality: does change in X cause a change in Y
- standardization: the same intervention (X) applied to everybody in a treatment group
- randomization: random assignment to treatment and control condition
(between-subjects or within-subject design)
- replicability
Survey experiments (Sniderman 2011)
- basic idea: the method combines experiments and survey
- involves experimental intervention in the course of a survey
- this intervention is a measure of the influence of an independent variable of interest:
manipulated experimentally
- participants are assigned randomly to treatment condition(s) and control condition
→ typically involve the creation of experimental scenarios by verbal and/or graphic
narrations in which experimental stimuli are embedded and systematically manipulated
- Social science and social scientists are interconnected with society
→ What is social science research?
- to search for an answer to a question
- to extend knowledge, unravel mysteries, build theories
- verify facts
→ based on a scientific method
science= collecting systemic knowledge that is capable of resulting in a correct prediction or
reliable outcome
scientific method = seek to explain events of nature in a reproducible way and is reliable
(repeatability) and valid (are we measuring what we were supposed to measure?)
1. The goal is inference → generating new knowledge about something we don’t know on
the basis of something we can observe/we know
2. The procedures are public / transmissible / cumulative
3. Conclusions are uncertain / Falsifiable
4. The method!
Pseudo = false
The importance of research methods:
Generally:
• Being able to conduct good scientific research to understand various (safety and security)
issues/problems and find solutions
• Developing a critical eye And
On a practical note:
• Developing skills to understand the (empirical) literature in other classes
• Developing skills for your bachelor/master thesis, but also beyond…
Doing research:
1. Research topic
2. Literature review
3. Research design
4. Data collection
5. Data analysis
6. reporting
Research design
1. Research philosophy. 4 worldviews:
- Positivism: believe that there is a single reality which can be measured and known. A
theory/observation is always true. Use quantitative methods Eg. if u drop a pen it will
fall
- Interpretivism: believe there is no single reality/truth. It needs to be interpreted. Use
qualitative methods to find multiple realities.
, - Pragmatism: believe reality is constantly renegotiated/debated and interpreted and
therefore the best method to use is the one that solves the problem. Combi of two
above
- Critical: paradigm situates research in social justice issues to address political, social
and economic issues. Hte seek to be the voice of the marginalised ones.
● ontology= what constitutes reality and how can we understand its existence ?
● epistemology= what constitutes valid knowledge and how can we obtain it?
● methodology= how do we go about finding it out? {insert table from powerpoint slide}
2. Research approach
Deduction → from theory to confirmation of the theory
Induction → from observations to theory
3. Mono/mixed methods
→ qualitative, quantitative, mixed methods
4. Time horizon
How much time to do research?
5. Data collections
How to collect data? And how to analyse? → Eg. surveys, case study, ethnography
Lecture 2
Methods
Interviews & focus groups: strengths and challenges
Interviewing = a conversation that has a structure… the purpose is to obtain knowledge
- to obtain knowledge: align your topic with the right respondents, the right questions
and the right responses!
→ gain trust first, not too heavy questions
→ strive to gain knowledge of their conditions and believes → could explain certain
actions
Types of interviews:
- the structured interview → close to survey
- the semi-structured interview → open-ended questions and no predefined options for
the respondent
- in-depth interviews → no predefined structure & provide insight on topics that need to
be investigated in a more structured manner & discover new points of view
,Interviews vs focus groups
→ identifying group norms & discovering variety within a group, observe group interaction
→ snowballing of ideas = enables students to initially think about their own response to a
question, then converse with just one other person, and then join together with other pairs to
enable a wider discussion
→ group think & dominant persons
→ more difficult to address sensitive topics
→ difficult to guarantee confidentiality & to organize & to generalize
Case study and comparative method: strengths and challenges
What is a case study & what is it good for?
- Case = a spatially delimited phenomenon observed at a single point in time or over
some period of time
- Case study = intensive study of a single case (for the purpose of understanding a
larger class of similar cases, i.e. a population)
- can be either inductive or deductive
Comparative study = the study of a larger number of cases
→ why do we compare?
- to explain an outcome identify broad patterns/regularities or relationships among
variables, test/refine theory and make predictions
Process tracing method
→ single case research method that can be used to make within case inferences about
presence/absence of causal mechanisms
→ enables strong inferences about causal relationship between X and Y
→ links dependent and independent variable
- one part will lead to action in the second one, which will lead to outcome
- connects causes and consequences
→ focus is on studying causal mechanisms using in-depth single case study
Process tracing tests for causal
inference
→ Straw-in-the-Wind = unusual piece of
evidence
→ Hoop = about the area and place of
the crime
→ Smoking-Gun =
→ Doubly Decisive =
Method of elimination = used by sherlock
holmes → eliminating possible variables
, Experiments
● Module: Sampling
→ a deliberate test of a causal proposition, typically with random assignment to conditions
→ investigators design experiments to evaluate the causal impacts of potentially informative
explanatory variables
What is a cause?
→ something is a cause if changing it - eliminating it, adding it, or varying its level - would
affect the outcome (keeping the rest of the system constant
→ causes are difference makers. A change in the causal factor (independent variable) leads
to a change in the dependent variable (see slide)
Causality and the logic of experiments
→ causality is inferred by comparing two scenarios in which only the
suspected/hypothesized cause changes while everything else remains the same
→ since we have no access to parallel realities, how can we get to scenarios in which (only)
the hypothesized cause changes?
→ in other words, how to have access to factual and counterfactual versions of the world?
What is a treatment/manipulation?
→ manipulations are deliberate interventions developed to create variation in a putative
causal variable
→ in experiments we manipulate the independent variable
How are experiments different from other research designs?
→ experiments work because random assignment to treatment and control conditions tends
to produce groups that are comparable in all respects but one - the variable manipulated by
the researcher
The logic of experiments
- importance of theory and deriving of hypotheses
- addressing the issue of causality: does change in X cause a change in Y
- standardization: the same intervention (X) applied to everybody in a treatment group
- randomization: random assignment to treatment and control condition
(between-subjects or within-subject design)
- replicability
Survey experiments (Sniderman 2011)
- basic idea: the method combines experiments and survey
- involves experimental intervention in the course of a survey
- this intervention is a measure of the influence of an independent variable of interest:
manipulated experimentally
- participants are assigned randomly to treatment condition(s) and control condition
→ typically involve the creation of experimental scenarios by verbal and/or graphic
narrations in which experimental stimuli are embedded and systematically manipulated
