POLI 110 weeks 4-6 notes

Belmont Report

Foundational ethical guideline in research involving human subjects; includes principles of respect for persons, beneficence, and justice.

Respect for Persons

Individuals should be treated as autonomous agents and those with diminished autonomy are entitled to protection.

Beneficence

Researchers must maximize benefits and minimize harm to participants.

Justice

Fair distribution of the benefits and burdens of research across groups.

Institutional Review Board (IRB)

Committee that reviews research proposals to ensure ethical standards are met.

Informed Consent

Process by which participants learn about the research and voluntarily agree to participate.

Deception in Research

Withholding or misrepresenting information from participants; must be justified and debriefed.

Treatment

In experimental design, the intervention or condition being tested.

Control Group

Group not receiving the treatment, used for comparison in experiments.

Random Assignment

Method of assigning participants to groups to ensure each has an equal chance of receiving the treatment.

Confounding Variable

A variable that influences both the independent and dependent variables, potentially biasing results.

Internal Validity

The extent to which a study accurately establishes a causal relationship.

External Validity

The extent to which results can be generalized beyond the study.

Placebo Effect

A change in outcome due to participants’ belief in the treatment rather than the treatment itself.

Causal Inference

Determining whether a change in one variable causes a change in another.

Counterfactual

In causal inference, what would have happened to the same unit in the absence of the treatment.

Fundamental Problem of Causal Inference

We can never observe both the treatment and control outcomes for the same unit at the same time.

Selection Bias

Systematic differences between groups being compared, threatening causal inference.

Operationalization

Process of turning abstract concepts into measurable variables.

Concept

A general idea or category being studied (e.g., democracy, inequality).

Variable

A measurable representation of a concept (e.g., income, age, vote choice).

Measurement

The process of assigning numbers or categories to units of analysis.

Reliability

The consistency of a measurement instrument over time.

Validity

The degree to which a measurement accurately reflects the concept.

Hypothesis

A testable statement about the relationship between variables.

Null Hypothesis

The default assumption that there is no relationship between variables.

Type I Error

False positive; rejecting a true null hypothesis.

Type II Error

False negative; failing to reject a false null hypothesis.

Statistical Significance

The likelihood that a result is not due to random chance.

Observational Study

A study where the researcher does not control treatment assignment.

Natural Experiment

An observational study where external circumstances assign treatment as-if randomly.

Cross-sectional Design

A study that examines data at one point in time.

Longitudinal Design

A study that follows the same units over time to assess change.

Independent Variable (IV)

The variable believed to cause a change in another variable.

Dependent Variable (DV)

The variable affected by the independent variable.

Control Variable

A variable that is held constant or accounted for to isolate the relationship between IV and DV.

Interaction Effect

When the effect of one IV on the DV depends on the level of another variable.

Construct Validity

How well a test or measure reflects the concept it’s intended to capture.

Measurement Validity

The extent to which an indicator accurately measures the variable.

Case Selection

Choosing which units (cases) to include in a study.

Most Similar Systems Design (MSSD)

Compares similar cases with different outcomes to find causes of those differences.

Most Different Systems Design (MDSD)

Compares very different cases with the same outcome to identify common causes.

Selection Bias in Case Selection

Occurs when chosen cases systematically differ from the population, leading to skewed results.

Goals of Research Design

To isolate causal effects, control for confounding factors, and ensure valid and reliable inference.