Exam 4 Study Guide: Complex Designs and Research Methods

Flashcards covering complex research designs, small sample research, quasi-experimental designs, and threats to internal validity based on the provided exam study guide.

Complex Design

A study with more than one independent variable (IV).

ex: ● IV1 = caffeine vs no caffeine

● IV2 = sleep vs no sleep

This creates a 2 × 2 design

Conditions

The total number of groups in a complex design, calculated by multiplying the levels of each IV (e.g., a $2 imes 2$ design equals $4$ conditions).

Main Effect

The effect of one IV by itself, ignoring the other IV.

ex: If caffeine improves scores overall regardless of sleep condition → main effect of caffeine.

Interaction

Occurs when the effect of one IV depends on the level of another IV.

Example: Caffeine only helps when participants are sleep deprived.

Ceiling Effect

A phenomenon where scores are clustered near the top of the scale, making differences hard to detect.

Floor Effect

A phenomenon where scores are clustered near the bottom of the scale.

Number of IVs in a $2 imes 3$ design

count the numbers in the design.$2$ independent variables.

How many conditions?

multiply the levels

ex:

Possible Main Effects in a $2 imes 3$ design

$2$ possible main effects, one for each independent variable.

Possible Interactions with $3$ IVs

There are $4$ possible interactions total: $A imes B$, $A imes C$, $B imes C$, and $A imes B imes C$.

Case Study

An in-depth examination of one individual or small group.

Phineas Gage

A famous brain injury case showing the role of the frontal lobe in personality.

Nomothetic

An approach that seeks general laws that apply to many people.

Ideographic

An approach that focuses on understanding one individual deeply.

Testimonial

A personal report or anecdote which constitutes weak scientific evidence.

Single Subject Experiment

Research focused on one participant measured repeatedly over time.

Baseline

The normal behavior of a participant measured before treatment.

ABAB Design

A research pattern consisting of baseline ($A$), treatment ($B$), removal of treatment ($A$), and reintroduction of treatment ($B$).

Reversal

The act of removing a treatment to see whether behavior returns to baseline levels.

Multiple Baseline Design

A design where treatment is introduced at different times across individuals, behaviors, or situations.

Quasi

Defined as "almost" experimental, but lacking full random assignment.

Wait List Control

A control group that receives treatment at a later time.

External Validity

How well results generalize to other people or settings.

Internal Validity Threats

Confounds that make it difficult to determine whether the IV caused the observed results.

History

A threat to internal validity where outside events affect the study participants.

Maturation

A threat to internal validity involving natural changes over time, such as children aging.

Testing

A threat to internal validity where the act of taking a test changes later performance.

Instrumentation

A threat to internal validity where measurement tools or observers change over time.

Regression to the Mean

A threat to internal validity where extreme scores naturally move toward the average upon retesting.

Attrition

A threat to internal validity where participants drop out of a study.

Selection

A threat to internal validity where groups differ before the study begins.

Contamination

A confound that occurs when groups in a study influence each other.

Experimenter Expectancy

When researchers unintentionally influence participants based on their own expectations.

Novelty Effects

Participants improve simply because a treatment or situation is new or interesting.

One-Group Pretest–Posttest Design

A design that measures participants before and after treatment without a control group.

Nonequivalent Control Group Design

A design that uses a comparison group without using random assignment.

Interrupted Time Series

A design involving repeated measurements before and after a treatment or intervention.

Time Series with Nonequivalent Control Groups

A design that adds a comparison group to an interrupted time series design.