Research Methods Exam 2

Factorial Design

Study with 2 or more independent variables (IVs), including all combinations of their levels

Independent Variable (IV)

Variable manipulated by the researcher to examine its effect on a dependent variable

Dependent Variable (DV)

Outcome measured in a study

Levels of an IV

The different conditions or values of an independent variable

Factorial Notation (e.g., 2×3)

Number of levels for each IV in a factorial design

Main Effect

The overall effect of one independent variable on the dependent variable, ignoring other IVs

Interaction Effect

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

Interaction (example)

Caffeine improves memory only when sleep is low, but not when sleep is high

Three-Way Interaction

When a two-variable interaction changes depending on a third variable

Independent-Groups Factorial Design

All IVs are between-subjects (different participants in each condition)

Within-Groups Factorial Design

All IVs are within-subjects (same participants experience all conditions)

Mixed Factorial Design

Includes at least one between-subjects IV and one within-subjects IV

Participant Variable

A pre-existing characteristic (e.g., age, gender) that is not manipulated

Quasi-Independent Variable

A participant variable used as an IV but not manipulated

Advantage of Factorial Designs

Can test interactions and are more efficient than separate experiments

Disadvantage of Factorial Designs

Can be complex and difficult to interpret

Null Hypothesis (Factorial)

No main effects and no interaction between IVs

Interpreting Main Effect

Compare averages across levels of one IV

Interpreting Interaction

Look at how one IV changes across levels of another IV

Key Sign of Interaction

Lines are NOT parallel on a graph

Range Effect

Distortion in results when DV hits the upper or lower limit

Ceiling Effect

Scores cluster at the top of the scale

Floor Effect

Scores cluster at the bottom of the scale

Why Range Effects Matter

They hide true differences between groups

Respect for Persons

Participants have autonomy and can choose to participate

Informed Consent

Participants are informed about the study before agreeing

Coercion

Pressuring someone to participate (unethical)

Beneficence

Researchers must minimize harm and maximize benefits

Justice

Fair distribution of research benefits and burdens

IRB (Institutional Review Board)

Committee that reviews research for ethical standards

Deception

Withholding or misleading participants about the study

Debriefing

Explaining the true purpose of the study after participation

Research Misconduct

Fabrication, falsification, or plagiarism

Data Fabrication

Making up data

Data Falsification

Altering data

Plagiarism

Using someone else's work as your own

3 Rs

Replacement, Reduction, Refinement

Replacement

Use alternatives instead of animals when possible

Reduction

Use the smallest number of animals necessary

Refinement

Minimize harm and improve animal welfare

Why Use Animals

Greater control and fewer ethical constraints than human research

Non-Experimental Research

No manipulation of variables; used for description or prediction

Why Use Non-Experimental Research

When manipulation is unethical, impractical, or to study real-world settings

External Validity

How well results generalize to the real world

Observational Research

Studying behavior by watching and recording

Naturalistic Observation

Observing behavior in a natural setting

Participant Observation

Researcher becomes part of the group being studied

Reactive Behavior

Participants change behavior because they are being observed

Observer Bias

Researcher's expectations influence observations

Interobserver Reliability

Agreement between multiple observers

Behavioral Categories

Predefined behaviors used for observation

Sampling (Observational)

Strategy for selecting what/when to observe

Time Sampling

Observe at specific time intervals

Event Sampling

Record every occurrence of a behavior

Archival Research

Using existing records/data to study behavior

Content Analysis

Systematic analysis of recorded information

Nonreactive Measure

Data collected without participants knowing

Advantage of Archival Research

High external validity, real-world data

Disadvantage of Archival Research

Data may be incomplete or biased

Case Study

In-depth study of one individual or small group

Why Use Case Studies

Study rare phenomena or generate new hypotheses

Limitation of Case Studies

Cannot generalize findings

Survey

Method of collecting self-reported data from participants

Correlational Analysis

Measures relationship between two variables

Predictor Variable

Variable used to predict another

Criterion Variable

Outcome variable being predicted

Open-Ended Question

Participant responds freely

Closed-Ended Question

Participant selects from given options

Likert Scale

Rating scale (e.g., strongly agree to strongly disagree)

Semantic Differential Scale

Rating between two opposite adjectives

Double-Barreled Question

Asks two things at once (bad design)

Leading Question

Suggests a desired answer (biased)

Good Survey Design

Clear, simple, unbiased, one idea per question

Quasi-Experiment

Study with an IV but no random assignment

Pretest-Posttest Design

Measure participants before and after a treatment

Interrupted Time-Series Design

Multiple measurements before and after an event

Advantage of Quasi-Experiments

Useful when random assignment is not possible

Disadvantage of Quasi-Experiments

Cannot make strong causal claims

What is a quasi-experiment?

A study where the researcher does not randomly assign participants to groups but still examines cause-and-effect relationships.

What is a one-group pretest-posttest design?

A design where one group is measured before (pretest) and after (posttest) a treatment.

Advantages of a one-group pretest-posttest design

Can study real-world changes; high external validity; more ethical when manipulation isn’t possible.

Disadvantages of a one-group pretest-posttest design

No control group; confounding variables; outside factors (history, maturation); natural variability.

What is an interrupted time-series design?

A design with multiple observations before and after a treatment to track changes over time.

Why is an interrupted time-series design stronger than a simple pretest-posttest?

Shows trends over time, making it easier to see if changes are due to the treatment.

What is a multiple time-series design?

An interrupted time-series design that includes a control group.

What is a non-equivalent control group interrupted time-series design?

A design with a treatment group and a similar (not randomly assigned) control group measured over time.

Why is a multiple time-series design useful?

Helps rule out confounding variables by comparing trends between groups.

What is a population?

The entire group of people or items a researcher wants to study.

What is a sample?

A smaller group taken from the population to represent it.

What is external validity?

The extent to which results can be generalized to the larger population.

What is a representative sample?

A sample that accurately reflects the population and gives all members an equal chance of being included.

What is a biased sample?

A sample where some members of the population are more or less likely to be included.

What is probability sampling?

Sampling where every member of the population has a known chance of being selected.

Why is probability sampling important?

It helps create representative samples and improves generalizability.

What is simple random sampling?

Every member of the population has an equal chance of being selected.

What is systematic sampling?

Selecting participants using a system (e.g., every 5th person).

What is stratified random sampling?

Dividing the population into groups and randomly sampling from each group.

What is proportionate sampling?

Sampling so each group is represented in the same proportion as in the population.

What is nonprobability sampling?

Sampling where the chance of selection is unknown.

What is convenience sampling?

Using participants who are easiest to access.