Exercise Psychology Research Design & Interpreting Meta-analyses

Flashcards covering basic concepts of research design, distinctions between observational and experimental designs, mediators, moderators, confounders, and the process and interpretation of meta-analyses based on lecture notes.

Cross-sectional Design

An observational design that captures a 'snap-shot' in time, examining associations between variables at a single point.

Predictive Design

An observational design where measurement of variables is separated by time to observe temporal sequencing.

Longitudinal Design

An observational design that examines change over time by measuring variables repeatedly.

Case Control Study

An observational design that compares 'cases' to 'non-cases' by looking retrospectively to see how they differed on potentially important factors.

Experimental Designs

Research designs that aim to establish cause-and-effect relationships.

True Experiment

An experimental design characterized by experimental manipulation, some degree of randomization, and careful control for extraneous variables.

Quasi-Experiment

An experimental design characterized by experimental manipulation where groups are already formed ('in-tact') and there is less control for extraneous variables.

Randomized Controlled Trial (RCT)

A specific type of true experiment involving experimental manipulation and random assignment, considered the 'gold standard' for inferring causation.

Cross-over Design (Within-subjects or Repeated Measures Experiment)

A special type of true experiment in which each participant completes all experimental conditions, acting as their own controls.

Non-equivalent Groups Design

A quasi-experimental design that compares groups with pre-existing differences, often using a 'pre-post' design.

Static Group Comparison Design

A quasi-experimental design similar to non-equivalent groups, but distinguished by the absence of a pre-intervention measure of the outcome (a 'post-only' design).

Time Series Design

A quasi-experimental design where outcomes are measured repeatedly before, during, and after an intervention, compared to similar measurements in a control group.

Variable

A characteristic of a person, place, or object that can assume more than one value.

Constant

A characteristic that can assume only one value.

Independent Variable (IV)

The variable that is manipulated or changed in an experiment, expected to have an effect on the dependent variable.

Dependent Variable (DV)

The outcome of interest that is measured to see the effect of the independent variable.

Mediator

A variable in the causal pathway between the IV and DV that transfers some or all of the effect of the IV on the DV, representing the 'why?' or 'how?'.

Mechanism

A mediator that can be measured directly.

Moderator

A variable outside of the causal pathway between the IV and DV that modifies how the IV affects the DV; also known as an 'effect modifier,' it represents the 'but' of an effect.

Confounder

An extraneous variable that is related to either the IV, DV, or both, which obscures the true effects observed between the IV and DV.

Meta-analysis

A research method that combines results from multiple independent studies on the same topic to identify patterns, trends, or overall effects, and summarize cumulative scientific evidence.

Effect Size

A quantitative measure produced by meta-analysis that indicates the magnitude and direction of a relationship or effect between variables.

Correlation (Effect Size Type)

An effect size that indicates the strength and direction of a relationship between two variables.

Cohen’s d (Effect Size Type)

An effect size that represents the standardized mean difference between groups.

Odds Ratio (Effect Size Type)

An effect size that indicates the likelihood of an outcome in one group versus another.

Risk Ratio (Effect Size Type)

An effect size that indicates the relative probability of an event in one group versus another.