EBP#2 Chpt 10: Rigor and Validity in Quantitative Research Flashcards

This flashcard set covers the key concepts of rigor, validity, and methods for controlling confounding variables in quantitative research as presented in Chapter 10.

Validity

The approximate truth of inference.

Threats to validity

Reasons that an inference could be wrong.

Statistical conclusion validity

A type of validity concerned with the ability to detect true relationships statistically, affected by factors like low statistical power and restriction of range.

Internal validity

The extent to which it can be inferred that the independent variable is truly responsible for the outcome, rather than confounding factors.

Construct validity

Concerns inferences from the particular exemplars of a study to the higher order constructs that they are intended to represent.

External validity

Concerns the extent to which it can be inferred that relationships observed in a study hold true over variations in people, conditions, and settings.

Randomization

A control method that controls all preintervention confounding variables without requiring advance knowledge of which variables to control, though it faces ethical and practical constraints.

Crossover

A control method that is the strongest possible approach if done with randomization, provided there are no carryover effects from one condition to another.

Homogeneity

A control method that is easy to achieve and enhances interpretability by restricting the sample to certain characteristics, but limits generalizability.

Stratification/blocking

A control method that enhances the ability to detect relationships and offers the opportunity to examine the blocking variable as an independent variable.

Matching

Also known as pair matching; a control method that enhances interpretability but is usually restricted to a few variables and requires a large pool of potential controls.

Statistical control

A relatively economical means of controlling several confounding variables, such as through Analysis of covariance (ANCOVA), requiring statistical sophistication.

Intervention fidelity

Also referred to as treatment fidelity; the reliable implementation of a treatment as planned.

Temporal ambiguity

A threat to internal validity where there is uncertainty about whether the cause preceded the effect.

Selection

A threat to internal validity involving biases resulting from preexisting differences between groups.

History

A threat to internal validity where external factors occurring at the same time as the study change over time and affect the results.

Maturation

A threat to internal validity referring to processes occurring within participants as a result of the passage of time.

Mortality/attrition

A threat to internal validity involving the loss of participants over the course of a study.

Per-protocol analysis

An internal validity data analysis approach that includes only those participants who completed the treatment as specified.

Intention-to-treat analysis

An internal validity data analysis approach that analyzes participants according to the group to which they were originally assigned.

Reactivity to the study situations

A threat to construct validity where participants behave differently because they are aware they are in a study.

Novelty effects

A threat to construct validity where the results are influenced by the newness of the intervention rather than the intervention itself.

Treatment diffusion/contamination

A threat to construct validity occurring when members of the control group receive the intervention intended for the experimental group.

Efficacy studies

Studies designed to evaluate if an intervention works under ideal, highly controlled conditions.

Effectiveness studies

Studies designed to evaluate if an intervention works in real-world clinical settings.