before largely institutionalism (study of formal structures of gov't)

> momentum created by sampling + mass surveys

> finding that voters seem less informed than assumend

>>> discovery of uniformity in political behaviour by systematically collecting and precisley recording data that allowed replications

1. there are discoverable regularities in politics, which can lead to theories with a predictive value
2. Such theories must be testable
3. Political Science should be concerned with observable behaviour that can be recorded
4. Findings should be based on quantifiable data
5. Research should be systematic by being theory-oriented
6. Political Science should become self conscious and cirtical about methodology
7. Political Science should aim at applied research that could provide solutions for actual problems
8. Political Science should become interdisciplinary

5,6,8 still considered valid

1. failed to fulfill its promises, even voter behaviour is post-explanation, not prediction
2. undue emphasis on processes, demand for precision narrowed use to issues of no relevance

Regularities and generalization not only proper subjects of science

> chaos theory showed that unlikely events happen relatively often

> before behaviouralism constitution was considered precise depiction of political reality

choices people make are significantly shaped by institution within they operate > institutions limit freedom of choice

broad defintion of institution: set of formal or informal rules, generally followed, whether for normative, cognitive or material reasons

1. Broad understanding of relationship between institution and individual behaviour
2. emphasis on asymmetry of power that arise from work of institution
3. institutional development emphasis path dependendece (current decision influenced by previous choices, not all theoretical options to be taken) and unintended consequences
4. integration of institutional analysis with that of ideas

> structure as shaping and restrainin, but not necessarily dominant, how do institutions affect their members?

Decision makers pursue self interest, subject to information and opportunity costs

> predominant view in USA, focuses on economic influence

> individuals as key factors, collective behaviour less important

> contrary to institutionalism focus on individual decision making rather than collective decision making

capture and understand the meaning of action for the agent performing it > people considered as subjective

contrary to the assumptions of Neorealism and Neoliberalism, international relations are socially constructed, that is, they are given their form by ongoing processes of social practice and interaction. Determined not primarily by material forces.

methodologies should be relevant and appropriate

> theoretical and methodological leverage

> relevant to the problem

1. hypothesis to be tested
2. research questions > important to indicate appropriate research questions
3. evidence needed to provide convincing test (quantitative, qualitative)
4. date to answer research questions
1. reliable (consistent results, which are replicable)
2. validity (measures what you think you measure)

small number of cases (eg. political leaders) in depth, usually at expense of generalization

> unrepresentative

> surveys preferable but often costly and limited in scope

1. ecological fallacy: draw conclusion from group behaviour to an individual
2. individualistic fallacy: draw conclusion from individual to group behaviour

> shows importance to collect several data

1. develop an operational plan
2. assure that procedures provide valid, objective and accurate solutions

> answer questions as unambiguously as possible

1. Theory specification and development of hypothesis
2. Data specification (what data? also pilto samples and restructuring of data)
3. Design of data collection instrument (questionnaire design, survey, random sampling procedures)
4. Pilot study (often omitted, testing of data collection instruments, rehearse actual survey)
5. Data collection
6. Coding (preparing/arrangig data for analyzation)
7. Data analysis (easier with quantitative data)
8. Publication

best way to determine cause and effect in scientific research, possibility to control all variables > manipulation of independet variables to observe effect on dependent variable

> in social sciences difficult a) factual b) ethical

Problems

1. how much can such set-ups explain real-life situations?
2. are people subconciously aware of experiment situation?
3. is such an experiment ethical?

popular in social sciences because experimental design is difficult

1. collection of information on a large number of data at a single point in time to accumulate quantitative or qualitative data
2. checked for patterns of association between variables
3. variables not controlled, causality difficult to determine
4. especially used for voting behaviour
5. can be quickly outdated because collected at one single time (snapshot)

designed to overcome shortcoming of cross-sectional design. Samples are taken at subsequent intervals, giving researcher the ability to evaluate long-term results

1. panel study (sample drawn from population and interviewed on topic)
2. cohort (people with similar characteristics are selected)

Problems:

1. people might be difficult to re-interview
2. people become more knowledgeable on topic, distorts results

focus on individual group or community, in depth study over long period of time

• yield qualitative data (relating to a specific case)
• unique study (difficult to draw generalizations)
• usable for hypothesis which need to be tested
• often strong theoretical dimension to improve generalization
• multiple cases give more robust theories

explanation and generalization usually limited to specific case