Confounding and Internal Validity

Repeated Measures Design

Participants experience all experimental conditions in a study

Biased Language

Use of sensitive terms to refer to groups and individuals

Multiple Baseline Designs

Manipulations introduced at different times to show causation

Program Evaluation

Assessment of program effectiveness through various stages

Interrupted Time Series Design

Examines changes in a dependent variable over time

Regression Equations

Used to predict a variable based on another known variable

Confounding Variable

Variable occurring with the independent variable, affecting results

Posttest-Only Design

Design introducing and measuring effects after independent variable

Pretest-Posttest Design

Design assessing changes from pretest to posttest after manipulation

Mortality

Dropout factor in experiments

Demand Characteristics

Cues altering participant behavior

External Validity

Extent to which results can be generalized to other situations

Independent Groups Design

Participants in only one experimental group

Counterbalancing

Technique to eliminate order effects in repeated measures design

Active Voice

Writing directly to engage readers, avoiding passive constructions

Indent

Space before the start of a paragraph

While vs. Since

Using 'while' for simultaneous events and 'since' for subsequent events

Factorial Designs

Designs with multiple independent variables

Interactions

Relationships between independent variables in a factorial design

Mixed Factorial Design

Combines between-subjects and within-subjects designs

Variability

Spread of scores in a distribution

Effect Size

Magnitude of a relationship between variables

Internal Validity

Results attributed solely to the independent variable

Curvilinear Relationship

Relationship where one variable increases while the other decreases

Inverted-U

Curvilinear relationship where a variable initially increases, then decreases

2 x 2 Factorial Design

Design with two independent variables, each with two levels

Moderator Variables

Variables that influence the strength of a relationship

Single Case Experimental Designs

Formerly single-subject designs, measuring change over time

ABA Design

Baseline → Treatment → Baseline design

Quasi-Experimental Designs

Used when full experimental control is not possible

Cross-Sectional Method

Measuring individuals of different ages at one point in time

Longitudinal Method

Observing the same group at different times

Sequential Method

Combines cross-sectional and longitudinal methods

Descriptive Statistics

Statistics that summarize data's central tendency and variability

Central Tendency

Mean, median, and mode indicating data's center

Correlation Coefficients

Measure strength and direction of relationships between variables

Statistical Significance

Determining if results are likely due to chance

Null Hypothesis

Assumes no significant difference between groups

Research Hypothesis

Predicts a significant difference between groups

Sampling Distributions

Distributions of sample statistics based on repeated sampling

Variance

Square of the standard deviation