Psychology: Heuristics and Research Methods

• Heuristics are mental shortcuts.
• Availability heuristic judges the likelihood of events based on the ease of recalling similar instances.
• Representativeness heuristic assesses likelihood by resemblance to typical examples

Steps of the Scientific Method

• Research question
• Develop a hypothesis
• Test the hypothesis
• Analyze the data
• Report the findings

Conceptual vs. Operational Definitions

• Conceptual definitions explain the abstract meaning of a concept.
• Operational definitions specify how a concept will be measured or observed.

Experimental vs. Non-Experimental Designs

• Experimental designs involve the experimenter controlling and manipulating the independent variable.
• Non-experimental designs lack control/manipulation of the independent variable.
• Independent variables influence the dependent variable and are controlled/manipulated by the researcher.
• Dependent variables are the measured outcomes associated with changes in the independent variable.
• Explanatory/predictor variable is the independent variable in non-experimental designs.
• Criterion/response variable is the dependent variable in non-experimental designs.

Between-Subjects vs. Within-Subjects Designs

• Between-subjects designs assess each participant on the dependent variable, comparing between subjects.
• Within-subjects designs assess each participant on the dependent variable, comparing within subjects.

Principles of the Belmont Report

• Beneficence requires researchers to maximize benefits and minimize harms to participants.
• Respect emphasizes treating individuals as autonomous agents and protecting those with diminished autonomy.
• Justice focuses on fair distribution of research burdens and benefits.

Research Roles and Processes

• A confederate is an actor pretending to be a participant.
• Debriefing provides participants with complete information after their involvement in the study.

IRB Review Levels

• Exempt: no more than minimal risk to participants
• Expedited: minimal risk to participants
• Full review: poses greater than minimal risk to participants

File Drawer Problem

• The tendency to publish positive or statistically significant results more readily.
• This can lead to a skewed representation of research findings.

True vs. Quasi-Experiments

• True experiments use random assignment and can fully manipulate the independent variable.
• Quasi-experiments do not use random assignments and cannot fully manipulate the independent variable.

Measuring Variables

• Self-report involves participants providing information about their thoughts, feelings, etc.
  • Pros: accessibility and subjectivity
  • Cons: biases and interpretation
• Behavioral measures involve observing and recording actions and reactions.
  • Pros: objective data, reduced subject bias, easy administration
  • Cons: observer bias, demand characteristics, labor intensive, and ethical considerations

Scores and Errors

• Raw score: true score + error
• True score: score if the test was a perfect measure.
• Error: extraneous influences that deviate the raw score from the true score
• Random error: variation due to unsystematic or chance factors
• Bias: error that consistently pushes scores in a direction

Keys to Effective Measurement and Considerations

• Standardization: ensures a consistent experiential situation for everyone.
• Sensitivity: the range of data a researcher can gather.
• Avoid ceiling effect by ensuring the upper boundary isn't too low.
• Avoid floor effect by ensuring the lower boundary isn't too high.
• Avoid observer bias, which involves misinterpreting answers.

Reliability and Validity

• Reliability: the consistency of a measure.
• Validity: the accuracy of a measure.

Population vs. Sample

• Population: the entire group about which conclusions are drawn.
• Sample: the specific group from which data is collected.
• A representative sample accurately reflects the characteristics of the larger population.

Types of Probability Sampling

• Probability sampling involves equal chances of selection for all in the population.
• Simple random sampling randomly selects individuals from population members.
• Stratified random sampling includes dividing the population into groups/clusters then using simple random sampling to select clusters.

Types of Nonprobability Sampling

• Nonprobability sampling does not give everyone an equal chance of being sampled.
• Convenience sampling uses readily available participants.
• Quota sampling freely chooses participants as long as they fit the quota.
• Purposive sampling chooses participants based on appropriateness for the study.
• Snowball sampling recruits future participants from existing participants.

Control vs. Experimental Groups

• Control group: does not receive the variable being tested.
• Experimental group: receives the variable being tested.

Establishing Causality

• Covariation: changes in one variable are associated with changes in another.
• Internal validity: rules out possible causal explanations.
• Temporal precedence: changes in the suspected cause occur before changes in the effect.
• Experimental control keeps everything the same except for the test variable.
• Experimental hypothesis: specific prediction of how the IV influences the DV.
• Experimental realism: participant becomes engrossed in the manipulation.
• Extraneous variable: can change the outcome variable.
• Independence: each participant represents a unique data point.
• Manipulation check: measures whether the manipulation changed the IV.
• Mundane realism: parallels everyday situations in the real world.
• Null hypothesis: no difference hypothesis; what the researcher tries to reject.
• Random assignment places participants in groups unsystematically.
• Two-group design: compares two groups or conditions.
• Confound: unintentionally varies along with the manipulation.
• Empty control group: receives no treatment and completes the dependent variable.
• Exploratory analysis: tests for potential differences not anticipated.
• Methodological pluralism: use multiple methods.
• Multigroup design: experimental design with three or more groups.
• Placebo group: believe they are getting the treatment, but aren't.
• Post-hoc test: examines combinations of conditions, statistically accounting for unexpected predictions.