What is a research design?
- A logical structure of inquiry that specifies the sort of test or evidence that will convincingly confirm or disconfirm a hypothesis in a coherent and logical way.
- Identifies the type of data necessary, how it will be obtained, what methods of analysis will be applied
What are the various types of research designs used in political research?
- Experimental: closest to true scientific method, most effective to find out if two factors are causally related.
- Cross- sectional: Sample of pop at a single point in time.
- Longitudal: Explores changes or trends over time.
- Comparative: Cross - sectional, case study.
- Historical: Influence of certain events on others through contextualisation, timing, sequence of social actions
Why should researchers be concerned with a research design?
Basic principles:
- Ensuring findings are reliable.
- Decrease of threats to internal and external validity.
- Making explicit the logic, structure, evidence, test.
RD’s allow for in depth and clear use of data in order to investigate a hypothesis without falling into the realm of pure opinion.
Basic principles of a research design.
- Specifies type of research and techniques of data collection.
- Makes explicit the logic which enables you to draw inferences - logical conclusions based on collected info.
- Identifies type of evidence that both confirms and tests hypothesis.
- Decreases threats to internal and external validity.
- Ensures findings are reliable.
What are the potential stakeholder groups in a research project? Ethical concerns?
- Research participants: informed consent, confidentiality, risk/harm cannot be more than daily life.
- Researcher : Bias, incorrect reporting, inappropriate use of information.
3.Sponsoring org : Restrictions imposed/accepted in the way of obtainig/disseminating accurate info, misuse of info.
What is internal validity?
Extent to which we can be confident the independent (causal) variable produced the observed effect.
Is structure of study eliminates alternative interpretations we can draw unambiguous conclusions = has internal validity
What is external validity?
The extent to which results can be generalised beyond the particular study.
What is reliability?
The extent to which a measurement produces the same results on repeated trials.
Different types of questions and how that affects the research design.
- Explanatory: Determine on or more variables causes or affects one or more outcome variables
- Predictive: Forecast future developments based on analysis of current events and relevant theories
Event A causes Event B predicting if A increases we will ahve greater amount of B in future. - Prescriptive: Inventory avalible options, weigh pros and cons they present for acheiving desired outcome.
- Normative: Which view has the most convincing argument?
Test:
1) Logical consistency of ideas
2) Consideration of arguments from other theorists
What are the boundaries for a study, decided in advance of data collection?
- Temporal domain.
- Spatial domain.
- Actors or units that are relevant for focus.
- Variables or factors that are important for arriving at answer.
What are the ethical obligations of researchers?
- Data access: References of the data used and access to data collected.
- Production transparency: Full account of procedures used to generate data.
- Analytic transparency: Full account of how they draw their analytic conclusions from the data.
